MASARYK UNIVERSITY Faculty of Science

Regional Centre for Applied Molecular Oncology (RECAMO)

INVESTIGATION OF AGR3 PROTEIN FUNCTION AND MECHANISMS TRIGGERING ITS EXPRESSION IN CANCER CELL

Dissertation

Joanna Obacz

Supervisor: Roman Hrstka, Ph.D. Brno 2015 Bibliography

Author’s name and surname: Joanna Obacz, M.Sc.

Dissertation title (Czech): Studium funkce proteinu AGR3 a mechanism ů zodpov ědných za regulaci jeho exprese v nádorové bu ňce.

Dissertation title: Investigation of AGR3 protein function and mechanisms triggering its expression in cancer cell.

Study programme: Biochemistry

Field of study: Genomics and Proteomics

Supervisor: Roman Hrstka, Ph.D.

Year of defence: 2016

Key words (Czech): AGR3, AGR2, biomarker, karcinom mlé čné žlázy, extracelulární protein

Key words: AGR3, AGR2, biomarker, breast cancer, extracellular protein

Acknowledgments

I would like to thank my supervisor Roman Hrstka, Ph.D., for the guidance through the years, and the vital critique of this work; as well as Borivoj Vojtesek, DrSc., for giving me the opportunity to carry out this work in his lab and for many valuable advice. I express my gratitude to prof. Silvia Pastorekova for co-supervision and broad expertise in the field, creating a family atmosphere during my stays in Bratislava and constant supply of encouragement. I would like to express special thanks to Veronika Brychtova, Ph.D., Martina Takacova, Ph.D. and Filippo Iuliano, Ph.D., for friendly and fruitful collaboration in the lab and publications writing. Moreover, I would like to acknowledge all the colleagues from RECAMO and collaborating lab in Bratislava for various ways of helping during my Ph.D. studies, especially Paulina Orzol, Lucia Sommerova, Martin Benej, Ivana Vidlickova and Stela Lausova. Lastly, I would like to thank my family and friends from all over the world for constant support.

This work was supported by the project MEYS – NPS I – LO1413 and European Regional Development Fund and the state budget of the Czech Republic for Regional Centre for Applied Molecular Oncology – RECAMO (CZ.1.05/2.1.00/03.0101).

I hereby declare that this dissertation is my own independent work. I have only used the given sources and materials and I have cited others’ work appropriately.

…………………………......

Abstrakt

Úvod: Anterior gradient protein (AGR) 3 je blízce p říbuzným homologem proonkogenního proteinu AGR2. AGR3 je taxonomicky řazen do rodiny protein disulfid izomeráz. Přestože byla p řítomnost AGR3 detekována u řady malignit, v četn ě nádor ů mlé čné žlázy, vaje čník ů, prostaty a jater, zůstává funkce tohoto proteinu v tumorigenezi stále ne zcela prozkoumána. Cílem p ředložené práce bylo popsat úlohu AGR3 v nádorové buňce, p ředevším pak mechanismy regulující hladinu AGR3, analyzovat vztah mezi proteiny AGR3 a AGR2 a stanovit prognostický význam exprese AGR3 u karcinom ů mlé čné žlázy. Metody: Funkce proteinu AGR3 byla studována na bun ěč ných liniích MCF-7 a T-47D odvozených od karcinomu mlé čné žlázy a H1299 odvozených od karcinomu plic, pomocí jak 2D tak i 3D modelových bun ěč ných systém ů. Vliv extracelulárního proteinu AGR3 na chování nádorové bu ňky byl studován pomocí monitorování bun ěč né proliferace v reálném čase, testu adhezivity, metody zacelování rýhy a β-galaktozidázovým testem. Interakce mezi AGR3 a AGR2 byla analyzována pomocí tzv. „proximity ligation assay“ a imunoprecipitace. Imunohistochemie a kvantitativní PCR byly využity ke stanovení AGR3 exprese u kohorty 129 primárních karcinom ů mlé čné žlázy. Ke statistickým analýzám v četn ě stanovení potenciálního prognostického významu proteinu AGR3 byly použity Fisher ův test, Pearson ův chí-kvadrát test a Breslow ův test. Výsledky: AGR3 přispívá k přežívání nádorových bun ěk b ěhem r ůzných stresových podmínek, v četn ě poškození DNA, a to jak závisle, tak i nezávisle na proteinu p53. V případ ě sekrece tohoto proteinu do extracelulárního prostoru, AGR3 přispívá k bun ěč né adhezi a migraci pravd ěpodobn ě díky modifikaci signálních drah, které se významn ě uplat ňují v nádorových bu ňkách. In silico analýzy odhalily, že AGR3 regulované i regulující signální dráhy se mohou podílet na regulaci bun ěč né proliferace, diferenciace, metastazování a celkového p řežití. Dále bylo zjišt ěno, že protein AGR3 nejen že může p římo interagovat s proteinem AGR2, ale u řady nádorových ale i normálních tkání je jeho expresní profil velmi podobný AGR2. Immunohistochemická analýza u karcinom ů mlé čné žlázy potvrdila korelaci mezi AGR3 expresí a estrogenovými receptory, progesteronovými receptory a nízkým grade. Ačkoli zvýšená exprese AGR3 predikovala lepší bezp říznakové p řežití, vícerozm ěrná statistická analýza nepotvrdila AGR3 jako nezávislý prognostický faktor. Záv ěr: AGR3 představuje potenciální tká ňový a sérový biomarker, který se m ůže podílet respektive p římo ovliv ňovat fenotyp nádorových bun ěk a s tím spojené projevy nádorového onemocn ění.

Abstract

Background: Anterior gradient protein (AGR) 3 is a highly related homologue of pro- oncogenic AGR2 and belongs to the family of protein disulphide isomerases. Although AGR3 was found in breast, ovary, prostate, and liver cancer, it remains of poorly defined function in tumourigenesis. The aim of this study was to elucidate AGR3 role in cancer and mechanisms responsible for its induction in tumour cell, to study putative AGR3- AGR2 cross-talk, and to determine the prognostic significance of AGR3 expression in breast carcinomas. Methods: AGR3 function was studied on breast carcinoma cell lines MCF-7 and T-47D and lung cancer cell line H1299 using both 2D and 3D models. The effect of extracellular AGR3 on cancer cell behaviour was assessed using Real-Time cell proliferation monitoring, detachment, wound-healing and β-galactosidase assays. The interaction between AGR3 and AGR2 was determined using proximity ligation assay and co-immunoprecipitation. Immunohistochemistry and qPCR were used to study AGR3 expression in a cohort of 129 primary breast carcinomas. Statistical analyses used for the determination of AGR3 prognostic significance included the Fisher’s exact, Pearson`s chi-square and Breslow tests. Results: AGR3 promotes survival of tumour cells upon DNA damage and microenvironmantal stresses both in p53-dependent and p53-independent manner. When secreted into extracellular space, it promotes cell adhesion and migration, possibly through the modification of tumour-associated signalling pathways. In silico analyses revealed that AGR3 upstream and downstream signalling may coordinate among others cell proliferation, differentiation, metastasis and survival. Furthermore, AGR3 directly binds to AGR2 and when compared to the latter is expressed in a similar manner in diverse carcinomas and normal tissues. Immunohistochemical analysis of breast cancer specimens showed that AGR3 expression correlated with oestrogen receptor and progesterone receptor positivity, as well as low tumour grade. Although in the whole cohort AGR3 positivity was associated with better progression free survival, the multivariate survival analysis did not found AGR3 as an independent prognostic factor for breast cancer patients outcome. Conclusions: AGR3 is a potential tissue and serum biomarker which modifies tumour- associated phenotype in a context-dependent manner. Table of contents

1 Introduction ...... 10 1.1 AGR protein family ...... 10 1.1.1 AGR proteins in lower vertebrates ...... 10 1.1.2 Human AGR proteins ...... 11 1.2 AGR proteins as PDI family members ...... 12 1.2.1 General features of the PDI family ...... 12 1.2.2 Evidence for the belonging of AGR proteins to PDI family ...... 13 1.3 AGR proteins expression in human tissues ...... 15 1.3.1 AGRs expression in healthy tissues ...... 15 1.3.2 AGRs expression in pathologic conditions ...... 16 1.4 Role of AGR proteins in tumour biology ...... 16 1.5 AGR proteins as components of tumour-associated pathways ...... 18 1.6 Prognostic significance of AGR2 and AGR3 expression in carcinomas ...... 20 1.6.1 Significance of AGRs tissue expression ...... 20 1.6.2 AGRs as secreted biomarkers ...... 22 2 Aims ...... 24 3 Materials and methods ...... 25 3.1 Cell cultures and drug treatments ...... 25 3.1.1 2D models ...... 25 3.1.2 3D models ...... 25 3.2 Generation of stable cell line ...... 25 3.3 Purification of recombinant proteins ...... 26 3.4 SDS-polyacrylamide gel electrophoresis (SDS-PAGE) ...... 26 3.5 Immunoblotting and collection of conditioned media ...... 27 3.6 Co-immunoprecipitation (co-IP) ...... 28 3.7 Cell detachment assay ...... 28 3.8 Wound-healing assay ...... 28 3.9 Proliferation assay ...... 29 3.10 Quantitative polymerase chain reaction (qPCR) ...... 29 3.11 β-galactosidase assay ...... 30 3.12 Flow cytometry ...... 30 3.13 In situ PLA assay ...... 31 3.14 Immunohistochemistry ...... 31 3.15 Patients and tissue specimens ...... 32 3.16 Statistical analysis ...... 32 4 Results ...... 33 4.1 Preliminary characterisation of AGR3 based on data mining ...... 33 4.1.1 AGR3 distribution in human healthy tissues and carcinomas ...... 33 4.1.2 AGR3 co-expressed partners ...... 34 4.1.3 AGR3 in silico promoter analysis ...... 35 4.2 Immunohistochemical analysis of AGR3 expression in primary breast carcinomas .... 36 4.2.1 Association of AGR3 expression with other tumour variables ...... 37 4.2.2 Association of AGR3 with patient survival ...... 39 4.2.3 Association of AGR3 and other tumour variables with patient survival ...... 40 4.3 The extracellular role of AGR3 in breast cancer ...... 42 4.3.1 AGR3 expression in breast cancer cell lines ...... 42 4.3.2 The effect of extracellular AGR3 on tumour cell behaviour ...... 43 4.3.3 Identification of signalling pathways mediating breast cancer cell response to extracellular AGR3 ...... 47 4.3.4 Study of AGR3-AGR2 cross-talk ...... 50 4.4 Role of AGR3 in tumour cell survival ...... 52 4.4.1 Generation of AGR3 stable cell line...... 52 4.4.2 The effect of AGR3 on tumour cell response to DNA damage ...... 53 4.4.3 The role of AGR3 in the adaptation of cancer cell to microenvironmental stresses 54 5 Discussion ...... 57 5.1 Prognostic relevance of AGR3 expression in breast cancer ...... 57 5.2 The intra- and extracellular role of AGR3 in cancer ...... 59 5.3 Mechanisms triggering AGR3 expression in cancer cell ...... 61 5.4 The functional overlap between AGR3 and AGR2 ...... 63 6 Conclusions ...... 65 7 References ...... 66 8 List of figures ...... 77 9 List of tables ...... 78 10 List of abbreviations ...... 79 11 Attachments ...... 81 11.1 Attachment 1 ...... 81 11.2 Attachment 2 ...... 91 11.3 Attachment 3 ...... 102 12 Appendices ...... 113 12.1 Publication activity ...... 113 12.1.1 Articles ...... 113 12.1.2 Conferences & courses attended: ...... 113 12.2 Internships ...... 114 1 INTRODUCTION

1.1 AGR protein family

Anterior gradient (AGR) proteins form an evolutionary conserved family whose members are expressed from lower animals to human and exert prominent, however, ambiguous physiological function within these species. The family “founder”, XAG-1 was firstly described in Xenopus laevis (X. laevis ) and was named according to its specific expression and role during amphibian embryogenesis (Sive et al., 1989). There are three subfamilies of AGRs: AG1, AGR2 and AGR3. Members of AG1 subfamily are restricted to lower vertebrates, characterised by high regenerative capacity (such as fish and amphibians). On the other hand, during the amniotes evolution expression of AG1 genes was lost resulting in an exclusive presence of AGR2 and AGR3 homologues and subsequently decreased ability to regenerate body appendages in higher vertebrates (Ivanova et al., 2013, Ivanova et al., 2015).

1.1.1 AGR proteins in lower vertebrates The most-studied X. laevis AGR homologue, namely XAG-2, is an adhesive molecule expressed in cement gland (Aberger et al., 1998). The latter is a mucus-secreting organ localized at the boundary between dorsal and ventral region of frog embryo. It serves as a solid support for immature tadpole, yet unable to swim and feed as well as sensory device preventing the attached embryo from further movements (Aberger et al., 1998, Sive and Bradley, 1996). Cement gland formation is induced by signalling molecules such as noggin, chordin, follistatin, Xnr3, and members of the Hedgehog family, whereas it is suppressed by secreted factors Xwnt8, bone morphogenetic protein 4 (BMP-4) as well as retinoic acid and embryonic fibroblast growth factor (eFGF) (Sive and Bradley, 1996).

Other AGR genes identified so far in X. laevis include XAG-1, XAgr2 and XAgr3 (Sive et al., 1989, Novoselov et al., 2003, Ivanova et al., 2013). XAgr2 gene, in addition to cement gland, is also present in the otic vesicles within the anterior non-neural ectoderm of Xenopus early neurula embryos (Novoselov et al., 2003). XAgr3 transcript has been described in cement gland and the otic vesicles at the late tailbud stage, though its expression level was much lower than that of XAG-2 and Xagr2 (Ivanova et al., 2013). XAG-1 and XAG-2 share 88% protein sequence identity, however, they do not display overlapping function (Aberger et al., 1998). In contrast to XAG-1, XAG-2 is a secreted protein implicated in the formation of anterior-posterior axis in Xenopus gastrula ectoderm. Overexpression of XAG-2, but not XAG-1, promotes cement gland differentiation and expression of neural markers genes. Given, that XAG-1 starkly differs within the first 40 N-terminal amino acids when compared to XAG-2, it is plausible that XAG-1 underwent mutations within this region which hampered its properties and led to the loss of functionality (Aberger et al., 1998). Furthermore, recent work has unveiled a novel feedback loop mechanism in the development of telencephalon in X. leavis embryos in which both FGF8 and AGR signalling is involved. It was demonstrated

10 that FGF8 produced by anterior neural plate border (ANB) stimulates expression of AGR proteins in the juxtaposed anterior non-neural ectoderm. In turn, AGRs secreted from the latter non-neural ectoderm induce FGF expression in ANB and FGF signalling per se (Tereshina et al., 2014).

Remarkably, in addition to the control of developmental processes AGRs play a significant role in the regeneration of lower vertebrates. For instance, expression of AGR genes is activated to promote regeneration of hindlimb buds and tails of X. laevis tadpoles (Ivanova et al., 2013) as well as fins regeneration of zebrafish Danio rerio (Ivanova et al., 2015). Likewise, AGR2 in newt (nAG) and spotted salamander Ambystoma maculatum is sharply induced after limbs amputation, where it acts as a mitogen promoting blastemal cell proliferation (Kumar et al., 2007). The extracellular effect of secreted nAG can be mediated via binding to the membrane-anchored Prod1 receptor, which leads to the activation of ERK1/2 mitogen-activated protein kinase (MAPK) signalling cascade (Blassberg et al., 2011). Interestingly, nAG is upregulated by nerves in Schwann cells surrounding the regenerating axon (Kumar et al., 2007), whereas it is inhibited in the epidermal cells during development (Kumar et al., 2011). AGR homologues have been also identified in the following species: fish ( Danio rerio , Carassius auratus , Salmo salar , Tetraodon nigroviridis ), amphibians ( Xenopus tropicalis , Anolis carolinensis ), reptiles ( Pelodiscus sinensis ), birds ( Gallus gallus ) and mammals ( Mus musculus ) (Shih et al., 2007, Ivanova et al., 2013).

1.1.2 Human AGR proteins Human members of anterior gradient protein family include two molecules, AGR2 and AGR3. AGR2 sequence shows 54% identity and 71% similarity with Xenopus laevis XAG-2 protein and 91% identity and 96% similarity with mouse homologue mAG-2 (Myung et al., 2008). Alignment of human AGR2 and AGR3 reveals 71% sequence coverage, with major differences occurring within N-terminal cleavable signal sequence (Figure 1) (Fletcher et al., 2003). Both genes lie in the close proximity at chromosomal position 7p21 and compose of seven coding exons with similar exon/intron arrangement, which suggests that they arose from a gene duplication event (Petek et al., 2000, Fletcher et al., 2003).

Figure 1. Sequence alignment of AGR2 and AGR3 proteins Identical amino acids within AGR2 and AGR3 sequences are delineated in yellow, cleavable N- terminal signal peptides are marked in red, active sites are highlighted and C-terminal ER-retention sequences are shown in blue (according to (Fletcher et al., 2003).

There are six protein coding splice variants of AGR2, the most abundant AGR2 001, AGR2 005, AGR2 006, AGR2 007, AGR2 201 and AGR2 202 (Brychtova et al., 2011). AGR2 001 consists of 175 amino acids with a predicted molecular weight of 19979.2 Da and theoretical pI value of 9.03 (Myung et al., 2008) AGR3, on the contrary, has three protein coding splice variants (according to ENSEMBL genomic database, http://www.ensembl.org ). The full length AGR3 variant, AGR3 001, comprises 166 amino acids with a predicted molecular weight of 19171.3 Da and theoretical pI value of 7.76 (calculated in http://expasy.org/tools/pi_tool.html).

1.2 AGR proteins as PDI family members

Protein disulphide isomerase family (PDI) is a part of the thioredoxin (TRX) superfamily, which also includes the glutaredoxins, thioredoxins, ferroredoxins and peroxiredoxins (Jacquot et al., 2002). PDIs mediate the rearrangements of disulphide bonds between cysteine residues of nascent protein, formation of which is a critical step in protein maturation (Appenzeller-Herzog and Ellgaard, 2008). They also function as molecular chaperons involved in the endoplasmic reticulum (ER)-associated degradation (ERAD) mechanisms that lead to protein removal and maintenance of cellular homeostasis (Ni and Lee, 2007). The PDI family members are implicated in a variety of disorders, including neurodegenerative syndromes such as Parkinson’s, Alzheimer’s, Huntington’s diseases, familial amyotrophic lateral sclerosis (ALS) as well as infertility and a diverse range of malignancies (as reviewed by (Benham, 2012).

1.2.1 General features of the PDI family To date, 21 members varying in size, structure, tissue distribution and enzymatic activity were identified (for more detailed overview of PDI family members see Table 1) The common thread of all PDIs is the presence of at least one domain with structural similarity to TRX, which can be either catalytically active (harbouring CXXC motif or its derivatives) or catalytically inactive (Kozlov et al., 2010). Another unifying facet of PDI proteins is a C- terminal ER retention sequence (most typical KDEL sequence), which determines their predominant expression in the lumen of the ER. However, PDIs can be also found in the other subcellular compartments, where they are shown to regulate among others cell adhesion, platelets activation, viral infections and protein–DNA interactions (Turano et al., 2002). Moreover, each protein of the family is characterised by the presence of a short N-terminal signal peptide (15-30 amino acids in length) which is cleaved upon entry into the ER (Rapoport, 1991).

Table 1. Human PDI family Catalytically active domains are denoted as a˚, a or a'; b and b' corresponds to inactive domains; x represents linker region, which regulates ligand binding and homodimerization (according to (Galligan and Petersen, 2012, Benham, 2012). Name(s) Length (in Domain Active site sequence ER- amino acids) arrangement retention sequence PDI (PDIA1, P4HB) 508 abb'xa' CGHC, CGHC KDEL PDIp (PDIA2) 525 abb'xa' CGHC, CTHC KEEL ERp57 (PDIA3) 50 abb'xa' CGHC, CGHC QEDL ERp72 (PDIA4) 5645 a˚ abb'xa' CGHC, CGHC, CGHC KEEL PDIR (PDIA5) 519 ba˚aa' CSMC, CGHC, CPHC KEEL P5 (PDIA6) 440 aa'b CGHC, CGHC KDEL PDILT 584 abb'xa' SKQS, SKKC KEEL ERdJ5 (DNAJC10) 793 abbaaa CSHC, CPPC, CHPC, CGPC KDEL ERp44 406 abb' CRFS RDEL ERp46 (TXNDC5) 432 a˚aa' CGHC, CGHC, CGHC KDEL ERp18 (TXNDC12) 172 a CGAC KTEL ERp27 276 bb' - KTEL ERp29 261 b'D - KEEL TMX1 280 a CPAC - TMX2 296 a SNDC KKEI TMX3 454 abb' CGHC KKKD TMX4 349 a CPSC - AGR2 175 a CPHS KTEL AGR3 166 a CQYS QSEL CASQ1 396 bbb' - - CASQ2 399 bbb' - -

1.2.2 Evidence for the belonging of AGR proteins to PDI family AGR2 and AGR3’s affiliation to the PDI family was based on the phylogenetic analysis performed by Persson et al. (Persson et al., 2005), where it was demonstrated that both proteins have high homology to ERp18/19 protein, also denoted as AGR1. AGR2 and AGR3 show structural organization characteristic for PDI members: N-terminal signal peptide, TRX domain with redox-active centre (CPHS and CPQS, respectively) and ER-retention sequence (KTEL and QSEL, respectively) (Figure 1 and Figure 2). Crystal structure of AGR2 protein has recently been solved depicting that AGR2 forms a dimer through specific intermolecular salt bridges and that interaction between amino acids E60 and K64 is required for proper dimer topography (Patel et al., 2013). On the other hand, based on the PDB-sum database, AGR3 dimer is deemed to occur through specific hydrogen bonds between amino acids Q32 and Q46 (Obacz et al., 2015b). Relying on available dimeric structures of AGR proteins in Protein Data Bank [PDB accession numbers- AGR2: 2LNS, AGR3: 3PH9], it can be concluded that AGR2 secondary structure consists of α-β-α-α-β-α structural motifs, whereas AGR3 consists of α-β-α-α-β-α-β-α-α-α motifs. Moreover, both proteins share high structure homology with the main difference found in their dimerization regions (Figure 2).

Figure 2. Structure of AGR2 and AGR3 proteins AGR2 secondary structure with alpha helices (red), beta strands (yellow), CPHS active site (grey) and dimerization domain (blue) shown (a); Ribbon representation of AGR2 41-175 homodimer with secondary structures, dimerization domain and active site marked correspondingly to 1a (b); AGR2 dimeric interface, interactions between E60 and K64, responsible for dimer stability are delineated in blue (c); AGR3 secondary structure with alpha helices (red), beta strands (yellow), CQYS active site

(grey) and dimerization domain (blue) shown (d); Ribbon representation of AGR3 31-165 homodimer with secondary structures, dimerization domain and active site marked correspondingly to 1d (e); AGR3 dimeric interface, interactions between Q32 and Q46, with putative role in dimer stabilization are delineated in blue (f) (according to (Obacz et al., 2015b).

Analogously to other PDI members, AGR2 and AGR3 are ubiquitous residents of ER lumen (Park et al., 2009, Higa et al., 2011, Myung et al., 2008, Bonser et al., 2015). AGR2, but not AGR3, is induced upon ER stress, driven by an elevated accumulation of misfolded proteins and its basal expression is controlled by the IRE1 α- and ATF6 α-triggered arms of the unfolded protein response (UPR) (Higa et al., 2011, Bonser et al., 2015). Further, it was shown that AGR2 associates with BiP/GRP78, a well-established chaperone involved in the cellular response to many stresses (Ryu et al., 2013). AGR2 client proteins identified so far include intestinal mucin (MUC2) (Park et al., 2009) the airway epithelial MUC5AC and MUC5B (Schroeder et al., 2012) as well as pancreatic MUC1 (Norris et al., 2013). Additionally, by forming a substrate loop between amino acids 104 and 111, AGR2 interacts with ATP-binding protein Reptin and consequently regulates many of its functions such as ATPase activity, ATP binding, helicase functions, telomerase/Pontin binding and others (Maslon et al., 2010). Interestingly, it has been lately demonstrated that AGR2 redox activity is indispensable for the post-translational modification of epidermal growth factor receptor (EGFR) enabling its delivery to the plasma membrane and thus proper signalling (Dong et al., 2015). AGR2 also binds to hypoxia-inducible factor 1 (HIF-1), a master regulator of tumour

14 cell response to diminished oxygen level (hypoxia), and subsequently precludes its proteasomal degradation (Li et al., 2015b). Although AGR3 function as PDI member has been poorly investigated, recent work indicates that AGR3 is required for the regulation of ciliary beat frequency and mucociliary clearance in the airway (Bonser et al., 2015). Further, data derived from Genevestigator platform (Hruz et al., 2008) revealed that AGRs genes may co- express with genes coding for calpain 8 and calpain 9 (Obacz et al., 2015b), crucial components of a protective barrier in the gastric mucosa (Hata et al., 2010), which suggests that AGRs could be responsible for maintenance of calpains physiological properties, similarly to that of MUCs.

1.3 AGR proteins expression in human tissues

Both AGR2 and AGR3 were originally found in breast cancers specimens. AGR2 gene was firstly described in the oestrogen receptor (EsR) positive MCF-7 cell line (Thompson and Weigel, 1998) and AGR3 protein was identified in the membrane of breast cancer cell lines using proteomic screening for putative oncogenic signalling molecules (Adam et al., 2003). However, there is a growing body of evidence that AGR proteins play emerging role both in the physiology of healthy tissues and development of various pathological entities.

1.3.1 AGRs expression in healthy tissues In early stages of low vertebrates development AGR proteins are mainly expressed in ectoderm-derived organs, while in adults can be also found in endoderm-derived organs such as intestine (Chen et al., 2012, Xia et al., 2009). Similarly human AGRs are predominantly distributed in endoderm-derived organs (Xia et al., 2009). As members of PDI family, AGRs participate in the regulation of total protein load in the cell and thus cellular homeostasis. For instance, in normal mammary gland development AGR2 contributes to the regulation of cell proliferation and differentiation, and its maximum expression appeared during late pregnancy and lactation, where there is the highest requirement for milk protein production (Verma et al., 2012). To further evaluate the physiological role of AGR2 protein, a panel of AGR2- deficient mouse models has been generated. As a result, it has been demonstrated that AGR2 is required for mucin and mucus production in the intestine, pancreatic MUC1 expression, proper functioning of different intestinal cell lineages (including goblet and Paneth cells) as well as response to ER stress (Park et al., 2009, Zhao et al., 2010, Norris et al., 2013, Bergstrom et al., 2014). AGR2 is also present in the mucus of all parts of the digestive tract with the highest concentration observed in the proximal colon (Bergstrom et al., 2014); however, its extracellular role has not been validated so far. In addition, AGR2 was found in stomach, trachea, prostate, lungs as well as in adult and foetal liver (Thompson and Weigel, 1998, Schroeder et al., 2012, Lepreux et al., 2011). To date only one report pertaining to AGR3 in healthy tissues was released showing that AGR3 expression is restricted to ciliated cells in airway epithelium where it is required for calcium-mediated regulation of ciliary function (Bonser et al., 2015).

1.3.2 AGRs expression in pathologic conditions The vast majority of available literature describes significance of AGR proteins in the pathology of human carcinomas. They are expressed in both hormone-dependent tumours such as breast, ovarian and prostate cancer (Fletcher et al., 2003, Fritzsche et al., 2006, Gray et al., 2012, Park et al., 2011, Bu et al., 2011, Vaarala et al., 2012) as well as non-hormone one including that of gastro-intestinal tract, oral cavity, lungs and oesophagus (Lepreux et al., 2011, Vivekanandan et al., 2009, Pizzi et al., 2012a, Fritzsche et al., 2007, Pizzi et al., 2012b). AGR2 and AGR3 genes harbour EsR- and androgen receptor (AR)-binding sites within their promoters and are upregulated by both androgen and oestrogen in breast and prostate cancer (Wilson et al., 2006, Welboren et al., 2009, Bu et al., 2013). Moreover, strong positive correlation of both proteins with EsR has been reported (Fletcher et al., 2003), further highlighting AGRs crucial role in hormone-responsive tumours. Expression of AGR2 and AGR3 can be either coincident as is the case of breast and mucinous ovarian carcinomas or relatively uncoupled, as shown in prostate and non-mucinous types of ovarian tumours (Fletcher et al., 2003, Gray et al., 2012), suggesting that mechanism triggering their expression in vivo is a context-dependent event.

In addition to tumourigenesis, AGRs contribute to the development of other pathological entities. Relying on in vivo models, it was demonstrated that AGR2 null mice had an increased incidence of rectal prolapse and were more susceptible to colitis and ileitis (Zhao et al., 2010, Park et al., 2009). This is in line with other study showing that AGR2 genetic variants which dampened its mRNA expression are associated with higher risk of Crohn’s disease and ulcerative colitis, probably due to the disruption of epithelial integrity (Zheng et al., 2006). Moreover, it was found that AGR2 mRNA level is induced in the individuals with asthma and that this increase correlates with MUC5A expression and leads to elevated mucus production (which is a characteristic feature of asthma) (Schroeder et al., 2012).

1.4 Role of AGR proteins in tumour biology

In many cancer studies AGR2 protein was demonstrated to promote aggressive tumour cell phenotype. Liu et al. showed that transfection of AGR2 coding sequence into benign rat mammary cell line, Rama 37, resulted in lung metastasis formation in vivo in the mammary fad pads of syngeneic rats (Liu et al., 2005). The pro-metastatic properties of both AGR2 and AGR3 were also evidenced by their interaction with metastasis-associated genes GPI- anchored C4.4a and extracellular alpha-dystroglycan (DAG-1), the transmembrane proteins engaged in cell-cell and cell-matrix interactions between cancer and non-cancer cells (Fletcher et al., 2003). Similarly, immunohistochemical analysis of vast cohort of primary breast tumours revealed that AGR2 expression significantly correlated with metastasis- inducing proteins such as osteopontin, S100P, and S100A4 (Barraclough et al., 2009). Furthermore, in head and neck squamous cell carcinoma (HNSCC) AGR2 knockdown led to the decreased regional and distant metastasis (Sweeny et al., 2012).

A recent data suggest that AGR2 may influence cancer metastasis by regulating cellular adhesion. It was found that AGR2-deficinient bone metastatic prostate cancer cell line exert reduced attachment ability to extracellular matrix (ECM) components, including fibronectin, collagen I, collagen IV, laminin I and fibrinogen compared to its parental AGR2-expressing counterpart. Moreover, loss of cellular adhesion was associated with sharp decrease in the expression of α4, α5, αV, β3 and β4 integrins (Chanda et al., 2014). Authors further hypothesise that at the initial stage of cancer progression, downregulation of AGR2 expression leads to integrins loss and subsequently breaking of cell-cell contacts and cell detachment from the basement membrane. However, once free tumour cells regain the expression of AGR2 and integrins, which results in the colonization of distant metastatic sites. Similar observation was obtained in breast cancer model, in which AGR2 overexpression altered adhesive properties of tumour cells and increased the number of metastases (Liu et al., 2005).

In addition, numerous gain and loss-of-function approaches revealed that AGR2 augments tumour-associated phenotypes both in vitro and in vivo by increasing proliferation, migration and invasion capacities of various cancers, including that of ovary, oral cavity, pancreas, breast, brain, biliary tract and thyroid (Ramachandran et al., 2008, Vanderlaag et al., 2010, Park et al., 2011, Sweeny et al., 2012, Hong et al., 2013, Kim et al., 2014, Di Maro et al., 2014). Importantly, as demonstrated in different cancer models AGR2 prevents tumour cells from drug- and ER stress-induced apoptosis through the induction of survivin, cyclin D1, Bcl2, Bcl2l1 and c-Myc (Vanderlaag et al., 2010, Di Maro et al., 2014, Kim et al., 2014) as well as inhibition of tumour suppressor p53 (Pohler et al., 2004). Protective properties of AGR2 were also evidenced by its induction upon physiological stresses associated with tumour microenvironment, such as oxygen and nutrients depletion (Zweitzig et al., 2007). Moreover, it was shown that AGR2 homo-dimerization is important for the association with BiP/GRP78, a central regulator for ER stress response (Ryu et al., 2013). Collectively, these data suggest that AGR2 is a specialized member of PDI family that helps excessively growing tumour cells cope with increased protein production and secretion as well as overcome ER stress .

One of the hallmark of solid tumours is resistance to a programmed cell death following detachment from the ECM (known as anoikis) (Paoli et al., 2013). Chanda et al. have recently demonstrated that anoikis resistance in malignant cells can be associated with loss of AGR2 expression and that this effect is mediated by suppression of caspase-3, suggesting that caspase-3 is a direct target of AGR2 (Chanda et al., 2014). Additionally, in prostate cancer knockdown of AGR2 triggered cell cycle arrest and cellular senescence in ERK-p21 CIP1 , Akt- p21 CIP1 or PTEN-Akt-p27 KIP1 -dependent manner (Hu et al., 2012). Moreover, recent report links AGR2 with the regulation of cancer cell self-renewal and epithelial-to-mesenchymal transition (EMT), a fundamental cellular programme controlling embryonic development and homeostasis, as well as tumour progression and metastasis (Thiery et al., 2009). Ma et al. found that in human HNSCC tissue AGR2 protein expression correlated with EMT marker

Slug and that in vitro knockdown of AGR2 significantly decreased the levels of Slug, Snail, Nanog and Oct4 (Ma et al., 2015).

AGR3, on the other hand, is less studied homologue and therefore its role in cancer biology has not been elucidated so far. To date, it was shown that in addition to its putative pro- metastatic properties (Fletcher et al., 2003), AGR3 mediates cisplatin resistance in mouse xenograft models (Gray et al., 2012).

1.5 AGR proteins as components of tumour-associated pathways

There is a growing body of evidence that AGR proteins are components of various signalling pathways governing tumour development and/or progression (Figure 4). Among them, the EsR-dependent regulation of both AGR2 and AGR3 is the best studied pathway so far. It was demonstrated that AGR2 gene is overexpressed in EsR- positive breast cancer cell line (Thompson and Weigel, 1998) as well as in clinical specimens (Fletcher et al., 2003). Concordantly, strong AGR3 co-expression with EsR positivity was observed in breast tumours (Fletcher et al., 2003, Garczyk et al., 2015). Moreover, several independent chromatin immunoprecipitation experiments confirmed that both AGR2 and AGR3 promoters are directly targeted by EsR (Hrstka et al., 2010, Bu et al., 2013, Welboren et al., 2009). AGR2 was found to be induced by oestradiol under normal and pathological conditions (Verma et al., 2012, Wilson et al., 2006) and by hormone therapy, tamoxifen which caused even more robust increase in AGR2 mRNA and protein level compared to oestradiol (Hrstka et al., 2010). Interestingly, in tamoxifen-treated EsR-positive breast cancers, AGR2 predicts poor prognosis, possibly due to the contribution to the intrinsic and/or acquired drug resistance (Hrstka et al., 2010). Later, PDPK1/Akt pathway was deciphered to be involved in tamoxifen-dependent AGR2 protein turnover (Hrstka et al., 2013b). In tamoxifen-resistant cells AGR2 induction loses its dependence on EsR, but requires interaction with forkhead box transcription factor 1 (FOXA1) (Wright et al., 2014). FOXA2, on the other hand, is responsible for AGR2 induction in esophageal squamous cell lines, which occurs downstream of Hedgehog signalling (Wang et al., 2014). Further, it was shown that ErbB3 binding protein 1 (EBP1) represses FOXA1- and FOXA-stimulated AGR2 expression leading to diminished metastatic behaviour of prostate cancer cells (Zhang et al., 2010). In the latter, AGR2 and AGR3 were shown to be stimulated by both oestrogens and androgens and this activation depends on the binding of AR to AGR genes (Bu et al., 2013). Recently, it has been scrutinized that insulin-like growth factor 1 (IGF-1) induces AGR2 expression through ERK1/2, Akt and EsR pathways. Moreover, it was shown that interaction between IGF-1 and AGR2 promoter requires oestrogen responsive elements and a leucin zipper-biding site (Li et al., 2015a).

Figure 3. Schematic overview of AGR2 signalling Pathway intermediates regulating AGR2 expression are shown in light gray ovals, while AGR2 downstream effectors are indicated in dark gray ovals (adapted from (Chevet et al., 2013 )) Abbreviations: Akt, protein kinase B; AR, androgen receptor; AREG, amphiregulin; EGFR, epidermal growth factor receptor; ErbB3, receptor tyrosine -protein kinase; EsR α, oestrogen receptor α; FOXA1/2, forkhead box protein A1/2; HIF -1, hypoxia-inducible factor 1; IGF -1, insulin-like growth factor 1; PDPK1, 3 -phosphoinositide-dependent protein kinase 1; TGF -β transforming growth factor β; YAP1, Yes -associated protein 1.

EGFR signal ling represents another pathway that is affected by AGR2 expression. It was shown that through activation of Hippo pathway co -activator YAP1, AGR2 regulates amphiregulin (AREG) expression, which in turn triggers EGFR phosphorylation and cell proliferation (Dong et al., 2011 ). Given that AGR2 TRX-like properties are required for EGFR transport to plasma membrane, it also directly influences proper EGFR signalling (Dong et al., 2015). AGR2 is also a component of transforming growth factor -beta (TGF-β) signalling as demonstrated in pancreatic carcinomas, where AGR2 expression was suppressed by SMAD4, a downstream target of TGF -β pathway (Norris et al., 2013 ) as well as in HNSCC, where loss of TGF -β per se led to AGR2 induction (Ma et al., 2015 ).

In addition, there have been a number of observations highlighting prominent role of AGR2 in tumour cell adaptation machinery. For instance, it was demonstrated that AGR2 is a HIF-1- responsive gene, which is strongly induced in hypoxia (a hallmark of solid tumours) and influences endothelial cell migration, enhanced tumour angiogenesis and growth (Hong et al.,

2013). Furthermore, by regulating HIF-1α degradation AGR2 mediates hypoxia-induced doxorubicin resistance in breast cancer cells (Li et al., 2015b). It also modulates cyclin D1, survivin and c-Myc expression and in consequence influences growth and survival of breast cancer cells (Vanderlaag et al., 2010). Of particular note is also report showing that p53 phosphorylation upon exposure to UV was attenuated in AGR2-overexpressing cells, suggesting that AGR2 may interfere with p53 signalling and circumvent its response to DNA- damage (Pohler et al., 2004). Subsequently, one of the mechanisms by which AGR2 exerts its inhibitory effect on p53 has been recently described. It was shown that AGR2 induces DUSP10 expression which in turn inhibits p38 MAPK, well known to be involved in the regulation of p53 via posttranslational modifications enabling stabilization and activation of p53 (Hrstka et al., 2015).

1.6 Prognostic significance of AGR2 and AGR3 expression in carcinomas

1.6.1 Significance of AGRs tissue expression AGR2 is overexpressed in the plethora of human tumours, where it correlates with clinicopathologic variables, including tumour size, grade, stage and nodal status, and predicts patient’s outcome (Table 2). Elevated level of AGR2 significantly correlates with worse prognosis of patients suffering from breast, prostate, pancreas, and high grade serous ovarian cancers. In contrast, in endometrioid and mucinous ovarian carcinomas high AGR2 expression predicts more favourable outcome. A trend toward longer survival has been also observed in patients with AGR2-expressing tumours in biliary tract carcinomas. In addition, loss of AGR2 expression when compared to healthy controls is associated with poor survival of lung and colorectal cancer patients. It is noteworthy that different splice variants of AGR2 are selectively expressed in hepatocellular neoplasm as well as prostate cancer (Vivekanandan et al., 2009, Neeb et al., 2014), highlighting an importance of AGRs splice variants evaluation as potential biomarkers in future clinical studies. On the other hand, AGR3 protein has not been identified along with AGR2 using many OMICS approaches and therefore its expression and prognostic significance in human cancers remains poorly-defined (available work are summarized in Table 2). To date, two contradictory reports have been revealed; one depicting AGR3 as an indicator of good prognosis in ovarian cancer (King et al., 2011), while another showing that its presence predicts worse outcome of breast cancer patients (Garczyk et al., 2015). Interestingly, differential expression of AGR3 has been described in liver cancer, where it was shown that intrahepatic cholangiocarcinomas (ICCs) express AGR3 protein, whereas hepatocellular carcinomas are predominantly AGR3 negative. Moreover, it was suggested that together with acid mucopolysaccharides AGR3 could serve as a diagnostic marker of well-differentiated ICCs (Brychtova et al., 2014b).

Table 2. Significance of AGR2 and AGR3 expression in various carcinomas Unless indicated, correlation with clinicopathologic variables and impact of AGR proteins on patient outcome refers to AGR-expressing tumours Abbreviations: AR, androgen receptor; CTC, circulating tumour cell; EGFR, epidermal growth factor receptor; EsR, oestrogen receptor; GS, Gleason score; OS, overall survival; PFS, progression free survival; PR, progesterone receptor; PSA, prostate specific antigen. Tumour Correlation with Prognostic impact References type clinicopathologic variables (Liu et al., 2005) (Fritzsche et al., 2006) (Fletcher et al., correlation with positive EsR 2003) status, positive PR status, worse OS and relapse-free (Innes et al., positive AR status, negative survival, decreased 2006) breast EGFR status, higher pT efficacy of tamoxifen (Wu et al., 2008) status, positive nodal status, treatment (Barraclough et slower proliferation rate and al., 2009) lower tumour grade (Hrstka et al., 2010) (Lacambra et al., 2015) (Zhang et al., correlation with high GS, poor patient survival, 2007) prostate high PSA levels, CTC longer recurrence-free (Kani et al., 2013) enumeration survival (Ho et al., 2013) worse patient outcome in correlation with lower (Brychtova et al., pancreas poorly differentiated tumour grade 2014a) tumours (Chung et al., AGR2 AGR2 loss of AGR2 predicts 2012) correlation with EGFR poor survival; in patient (Narumi et al., lung mutations under 65 correlation with 2015) worse survival (Alavi et al., 2015) loss of AGR2 correlates with loss of AGR2 predicts (Riener et al., higher tumour grade, high reduced OS, high AGR2 2014) colorectal AGR2 mRNA level in CTC mRNA level in CTC (Valladares- correlates with pT3-pT4 and associates with reduced Ayerbes et al., high grade tumours PFS 2012) negative correlation with trend towards longer biliary tract (Kim et al., 2014) tumour size and stage survival correlation with higher pT status, high tumour grade, (Ma et al., 2015) positive lymph node status, trend towards poor (Chen et al., positive correlation with prognosis in head and oral cavity 2013) tumour metastasis, neck squamous cell (Morbini et al., association with tumour carcinomas 2015) origin in the tongue base and p16 expression

better patient outcome in correlation with lower endometrioid and (Armes et al., tumour stage, negative mucinous carcinomas, 2013) ovarian correlation with p53 and p16 shortened OS and PFS in (Darb-Esfahani et expression high grade serous al., 2012) carcinoma (Fletcher et al., correlation with positive EsR unfavourable outcome in 2003) breast status, negative EGFR low grade tumours (Garczyk et al., status, lower tumour grade 2015) AGR3 AGR3 positive correlation with ovarian better patient outcome (King et al., 2011) tumour differentiation

1.6.2 AGRs as secreted biomarkers The extracellular effect of human AGRs is predicted to be akin to that documented in salamanders, where it was shown that through binding to Prod-1 receptor secreted nAG stimulates blastemal cells growth (Kumar et al., 2007, Blassberg et al., 2011). Although there is no human homologue of Prod-1, the structural studies using recombinant monomeric AGR2 protein revealed that AGR2 has putative cell binding sites (Patel et al., 2013). Indeed, recent work has identified C4.4A (LYPD3) as a functional cell surface receptor for the extracellular AGR2 in pancreatic cancer cells. Moreover, using siRNA approach the authors demonstrated that the AGR2/C4.4A receptor complex includes laminins 1 and 5, and integrin β1 (Arumugam et al., 2015). In vitro studies showed that despite the presence of ER-retention signal KDEL, AGR2 was found secreted from various cancer cell lines (Wang et al., 2008, Wayner et al., 2012, Bu et al., 2011, Ramachandran et al., 2008). The mechanisms underlying this phenomenon have not been fully deciphered so far; however, as postulated by Bergstrom et al. AGR2 secretion partially occurs due to the modification of single cysteine residue within TRX-like active site (Bergstrom et al., 2014). Moreover, AGR2 may be secreted as a soluble protein or released from cells in membrane-coated microvesicles (Tsuji et al., 2015), and its O-glycosylation is required upon secretion from mammary epithelial cells (Clarke et al., 2015). The considerable amounts of AGR2 protein were also found in the fluids of patients suffering from prostate, lung, ovarian and pancreatic cancer (Bu et al., 2011, Kani et al., 2013, Chung et al., 2011, Edgell et al., 2010, Chen et al., 2010). Subsequently, extracellular AGR2 can serve as a suitable diagnostic and prognostic marker in these malignancies. For instance, in lung adenocarcinomas high serum AGR2 level was significantly associated with the incidence of recurrence after surgery and with a poor prognosis (Chung et al., 2011). In prostate cancer, on the other hand, the determination of urine AGR2/PSA transcript ratio can help discriminate between benign and cancer patients (Bu et al., 2011). Regarding AGR3 homologue, its elevated concentration was reported in sera from breast cancer patients (Garczyk et al., 2015), highlighting AGR3 utility as a potential biomarker.

There are several lines of evidence that secreted AGRs are functionally active. It was shown that conditioned media from cells expressing AGR2 enhance oesophageal adenocarcinoma cell migration and pancreatic cancer cell growth compared to AGR2-silenced counterparts (Wang et al., 2008, Ramachandran et al., 2008). In addition, in gastric tumour extracellular AGR2 incorporates within adjacent stromal fibroblasts and increases their invasive properties, thus contributing to cancer progression (Tsuji et al., 2015). AGR2 also mediates autocrine effect on pancreatic tumour cells promoting aggressive phenotype and chemoresistance in vitro as well as elevated tumour growth in vivo (Arumugam et al., 2015).

Collectively, these data provide strong evidence for a crucial role of AGR proteins in tumour biology and their utility as tissue and serum biomarkers, as well as potential anti-cancer targets. Although AGR2 has been intensively investigated for some time now and many questions have been resolved, precise functions of AGR3 homologue and its biological pathways remain unknown.

2 AIMS

1. Review of the available literature describing the role of AGR2 and AGR3 in cancer

2. In silico analyses using Genevestigator database and MatInspector enabling preliminary characterisation of AGR3 role in cancer as well as investigation of upstream regulatory pathways potentially triggering AGR3 expression in tumour tissues

3. Immunohistochemical analysis of AGR3 expression in primary breast carcinomas and evaluation of its prognostic significance

4. Study of AGR3 extracellular role in breast cancer

5. Determination of AGR3 role in tumour cell survival

3 MATERIALS AND METHODS

3.1 Cell cultures and drug treatments

3.1.1 2D models The human breast carcinoma cell lines MCF-7, T-47D, BT-474, SK-BR-3 and BT-549, and the human non-small cell lung carcinoma cell line H1299 were purchased from American Type Culture Collection (ATCC). ARN8 cells were derived from A375 malignant melanoma cells by stable transfection with the p53-dependent reporter construct pRGC ∆fos-LacZ (Blaydes and Hupp, 1998). Cells were maintained in Dulbecco’s modified Eagle’s medium- High-Glucose (HG-DMEM) or in McCoy's 5A (modified) medium (SK-BR-3 cells), supplemented with 10% fetal bovine serum (FBS) and 300 µg/ml L-glutamine and incubated at 37°C in humidified atmosphere with 5% CO2.

3.1.2 3D models For spheroids formation cells were detached using trypsin, counted and seeded at the density of 1×10 4 cells/well in a U bottom 96-well plate (CELLSTAR, Greiner Bio-One) pre-coated with 1% agarose. After 4 days of incubation at 37°C and 5% CO 2 medium was refreshed every 48 hrs for the following 4 days. For mammosphere culture, cells were seeded at 2.5×10 4 cells/well in a U bottom 96-well plate (CELLSTAR, Greiner Bio-One) pre-coated with 1% agarose in DMEM/F12 with 5 µg/ml bovine insulin (Sigma-Aldrich), 20 ng/ml recombinant epidermal growth factor (Sigma-Aldrich), 20 ng/ml basic fibroblast growth factor (Gibco), 1× B27 supplement (Invitrogen), 0.5 µg/ml cortisol (Sigma-Aldrich), and 1% gentamicin (Sandoz). Mammospheres were fed every 3 days by adding additional 30 µl of fresh media to each well (without removing the old media). To check integrity and measure spheroids and mammospheres diameter, pictures were regularly taken at 5 × magnification using an Axiovert 40 CFL microscope with AxioVision Rel. 4.8 software (both Zeiss).

Cells were treated with cisplatin ( cis -diamminedichloroplatinum (II); MW 300.1), doxorubicin (both Ebewe Pharma GmbH), etoposide (Sigma-Aldrich) or camptothecin (Sigma-Aldrich) at the noted concentrations and durations.

3.2 Generation of stable cell line

AGR3 stable expressing cell line was generated by Flp-In TM System (Invitrogen) following the manufacturer’s instructions with some minor modifications. Briefly, a host cell line was prepared by transfecting a pFRT/lacZeo vector with Flp Recombination Target (FRT) site into H1299 cell line. The mammalian expression plasmid for AGR3 was produced using Gateway® recombination technology (Invitrogen) following the manufacturer’s instructions. To generate an entry clone, BP recombination reaction was performed between an attB- flanked AGR3 full-length coding sequence and pDORN221 vector (Invitrogen). Expression

25 clone was generated by performing LR recombination reaction between an entry clone and pEF5/FRT/V5-DEST destination vector (Invitrogen). AGR3-expressing plasmid was then co- transfected with pOG44 plasmid which constitutively expresses the Flp recombinase into the H1299 host cell line. Two days post-transfection, cells were resuspended in fresh media containing 250 µg/ml hygromycin (Thermo Fisher Scientific), divided into two 96-well plates and cultivated for 14 days. Drug resistant colonies were then propagated and screened on AGR3 expression using immunoblotting and quantitative Real-Time PCR (as described below). For DNA transfections, 10 6 cells were suspended in 100 µl Nucleofector® solution kit 2 (Lonza) containing 1 µg highly purified plasmid DNA and electroporated with Amaxa Nucleofector Technology (Lonza) according to the manufacturer’s instructions. Following electroporation, cells were transferred to 5 ml fresh medium and incubated for 24 hrs.

3.3 Purification of recombinant proteins

The sequences coding for mature AGR2 protein (NP_006399, amino acids 21-175) were cloned by Gateway® recombination technology (Invitrogen) into a vector containing N- terminal His 6 tag, while the sequences coding for mature AGR3 (NP_789783, amino acids

22-166) were cloned into a vector containing N-terminal His 6-GST tag. Both tags were cleavable by TEV protease. Recombinant fusion proteins His 6-AGR2 and His 6-GST-AGR3 were produced in BL21-CodonPlus (DE3)-RIPL cells (Agilent) according to a protocol described previously (Trcka et al., 2014) with some minor modifications. Briefly, cells were lysed in buffer containing 20 mM Tris-HCl, pH 8.0, 150 mM NaCl, 10% glycerol, 20 mM imidazol, 2 mM MgCl 2, 1 mM PMSF, 1 mg/ml lysozyme (for AGR2) or 50 mM HEPES pH

7.4, 150 mM NaCl, 1 mM PMSF, 1 mg/ml lysozyme (for AGR3). His 6-AGR2 fusion protein was captured on a HisTrap column (GE Healthcare), eluted with 300 mM imidazole and cleaved with tobacco etch virus (TEV) protease. To remove His 6 tag with His 6-TEV protease, protein solution was exchanged into 20 mM Tris-HCl, pH 8.0, 150 mM NaCl, 10% glycerol,

2 mM MgCl 2 and then subjected to second IMAC.

His 6-GST-AGR3 fusion protein was captured on a GSTrap glutathione-agarose column (GE Healthcare), eluted with 20 mM glutathione and subjected to TEV protease cleavage. To remove His 6-GST and His 6-TEV, proteins were applied to HisTrap column (GE Healthcare). The purity and appropriate size of each protein was analysed by Coomassie blue staining of 10% SDS-PAGE gels. In order to exclude the potential impact of bacterial endotoxins on cellular signalling (Raetz et al., 1991), purified glutathione S-transferase (His 6-GST) protein served as a control in all experiments. Tobacco etch virus protease His 6-TEV(S219V)-Arg5 was prepared in-house following the modified method of Tropea et al. (Tropea et al., 2009).

3.4 SDS-polyacrylamide gel electrophoresis (SDS-PAGE)

For cellular proteins detection, cells were detached with cell scraper, washed two times with ice-cold PBS. Afterwards, cells were resuspended in lysis buffer (50 mM TrisHCl, pH 7.4,

150 mM NaCl, 5 mM EDTA, 50 mM NaF, 1 mM Na 3CO 4, 1% Nonidet P40) containing

26 protease inhibitor cocktail and phosphatase inhibitor cocktail 2 (both Sigma-Aldrich) followed by centrifugation at 14 000 rpm for 30 min at 4°C. Protein concentration was measured using Bradford protein assay (Bio-Rad) and proteins were dissolved in lithium dodecyl sulfate (LDS) sample buffer (Pierce, Thermo Scientific) in the presence of reducing agent 100 mM dithiothreitol (DTT). 20 µg of total proteins were separated by SDS-PAGE on

10% gels cast using a mixture of 1.3 ml of dH 2O, 1.5 ml of 30% w/v acrylamide mix, 1.7 ml of 1 M Tris-HCl, pH 8.8, 12 µl of 10% w/v ammonium persulfate (APS), and 3 µl of N,N,N’,N’-tetramethylethylenediamine (TEMED). Electrophoresis was carried out with an electrophoretic buffer pH 7.0 (1 M MOPS; 1 M Tris Base; 70 mM SDS; 20 mM EDTA) and constant 22.5 mA / 1 mm thick gel over 1.5 hrs, or until the dye front eluted off the bottom of the vertically set gel.

3.5 Immunoblotting and collection of conditioned media

SDS-PAGE resolved proteins were transferred onto nitrocellulose membranes (Trans-Blot; 0,22 µm; BioRad) at 100 V over 75 min using cold transfer buffer pH 8.3 (25 mM Tris-Base; 192 mM glycin; 15% methanol). Membranes were blocked in 5% milk (w/v) and 0.1% Tween 20 (v/v) in PBS and probed overnight with specific primary antibodies (Table 3). Peroxidase conjugated rabbit anti-mouse immunoglobulin or porcine anti-rabbit immunoglobulin antisera (DAKO Glostrup) were used as the secondary antibodies and visualized with ECL reagents (Amersham-Pharmacia). The Human Phospho-Kinase Array (ARY003B) assay from R&D Systems was carried out according to manufactures protocol.

Table 3. Primary antibodies used for immunoblotting Antibody Type Supplier Dilution β-actin mouse monoclonal Sigma-Aldrich 1: AGR2 K31 rabbit polyclonal sera Moravian Biotechnology 1:1000 AGR3 clone 1.3 mouse monoclonal Moravian Biotechnology 1:1 (Gray et al., 2012) p44/42 MAPK rabbit monoclonal Cell Signaling 1:1000 Phospho-p44/42 MAPK rabbit monoclonal Cell Signaling 1:1000 (Thr202/Tyr204) FAK rabbit polyclonal Cell Signaling 1:500 phospho-FAK (Tyr397) rabbit polyclonal Merck Millipore 1:250 PARP Rabbit polyclonal Calbiochem 1:250 LC3-B rabbit monoclonal Cell Signaling 1:1000 p27 mouse monoclonal Santa Cruz 1:200

For detection of secreted AGR proteins, cells were grown in serum-deprived DMEM for 48 hrs. Conditioned media were collected, centrifuged at 13 000 rpm for 10 min in order to remove cell debris and included proteins were subjected to overnight precipitation with ice- cold pure acetone. Afterwards, precipitates were resuspended in lysis buffer (50 mM TrisHCl, pH 7.4, 150 mM NaCl, 5 mM EDTA, 50 mM NaF, 1 mM Na 3CO 4, 1% Nonidet P40) containing protease inhibitor cocktail (Sigma-Aldrich), and 20 µl of total volume was subjected to SDS-PAGE followed by immunoblotting as described above.

3.6 Co-immunoprecipitation (co-IP)

T-47D cells were washed twice with cold PBS and cross-linked with 2 mM dithiobis(succinimidyl propionate) (DSP) or disuccinimidyl suberate (DSS) at room temperature for 30 min, or were directly subjected to co-IP. Fresh DSP and DSS stocks solution were prepared in dimethyl sulfoxide (DMSO). The reactions were terminated by the addition of 1 M Tris-HCl, pH 7.5 (final concentration of 20 mM) and incubation at room temperature for 15 min. The cells were washed twice with cold PBS and then lysed with lysis buffer as described in sub-chapter 3.4. For co-IP analysis, total cell lysates was precleared with protein G Sepharose (GE Healthcare) before incubation with specific antibodies, followed by the addition of protein G Sepharose. Total cell lysates were incubated with 2 µg anti-AGR3.1 antibody (in-house) at 4°C for 16 hrs. The precipitated proteins were dissolved in LDS sample buffer (Pierce, Thermo Scientific) in the presence of reducing agent 100 mM DTT and analysed by Western blot with rabbit sera against AGR2 (in-house (Obacz et al., 2015a)).

3.7 Cell detachment assay

Cells were seeded in quadruplet on 24-well plates at the density of 5000 cells/well and incubated with either 5 ng/ml or 50 ng/ml recombinant AGR2 or AGR3 protein for 24 hrs. Corresponding concentrations of GST protein were used as controls. Subsequently, cells were washed with 0.5% EGTA and trypsin was used to detach cells from plate surface. Trypsin concentration and time of incubation were empirically determined for each cell line and were as follows: MCF-7, 0.0025% trypsin for 3 min; T-47D, 0.125% trypsin for 3 min. The remaining cells were washed with PBS, fixed with 4% paraformaldehyde in PBS at 37°C for 20 min, and stained with 5mg/ml crystal violet for 10 min. The trapped crystal violet was subsequently extracted from attached cells with DSMO and quantified by measuring absorbance on a microplate reader at 595 nm (Tecan Group Ltd.).

3.8 Wound-healing assay

Cells grown to 100% confluence were scraped using a sterile micropipette tip to create an in vitro wound. Subsequently, the cells were washed with 0.5% EGTA to remove the detached cells, and then incubated in serum-free DMEM with or without AGR proteins. Time-lapse acquisition of the wound closure was analysed with Nikon eclipse Ti-E system (Bioscience) at 10× magnification. The pictures were captured in three randomly chosen fields within the wounded region every 20 min for 24 hrs. Plates were kept at 37°C and 5% CO 2 throughout the duration of the experiment. The migration rate was assessed using TScratch software (Geback et al., 2009) (CSE Lab, ETH) by quantification of the cell-free area at the different time points.

3.9 Proliferation assay

E-Plate 16 was used for non-invasive real-time measurements with an xCELLigence RTCA SP system and RTCA software version 2.0 (both Roche Applied Science). Prior to analysis, the background was measured using 100 µl of complete DMEM medium. Cells were seeded at the density 5000 cells/well in additional 50 µl cultivation media and recombinant AGR proteins were added to a final concentration of 50 ng/ml or 5 ng/ml after 4 hrs. The impedance was measured continually for a period of 48 hrs. Data are presented as normalized cell index.

3.10 Quantitative polymerase chain reaction (qPCR)

Total cellular RNA was extracted using the RNeasy Mini kit (Qiagen) according to the manufactures protocol. cDNA synthesis was carried out using the M-MuLV reverse transcriptase (RevertAid H Minus; Thermo Fisher Scientific). The triplicate samples were subjected to quantitative polymerase chain reaction (PCR) analysis using Lumino Ct SYBR Green qPCR Ready Mix (Sigma-Aldrich) with ROX (Sigma-Aldrich) as passive reference stain. PCR reaction was performed using default conditions: initial denaturation 95°C, then 40 cycles 95°C 15 seconds and 60°C 1 min, running in 7900 HT Fast Real-Time PCR system (Applied Biosystems). The relative quantification of gene expression was determined based on 2 -∆∆ CT algorithm using β-actin as an endogenous control. To obtain absolute quantification, dilution series of plasmids pDEST12.2 with cloned respective sequences were used in range from 20 to 2 million of copies to generate standard curves. For data normalization 18S rRNA levels were determined using TaqMan assay for 18S rRNA (Applied Biosystems). The sequences of primers used for qPCR are provided in Table 4.

Table 4. List of primers used for qPCR Gene/Protein name Primer Sequence ACTB-F 5’-GGAACGGTGAAGGTGACAGC-3’ actin, beta ACTB-R 5’-ACCTCCCCTGTGTGGACTTG-3’ AGR2-F 5’-GAAGCTCTATATAAATCCAAGACAAGCA-3’ anterior gradient 2 AGR2-R 5’-GCCAATTTCTGGATTTCTTTATTTTC-3’ AGR3-F 5’-CATCACCTGGAGGATTGTCAATAC-3’ anterior gradient 3 AGR3-R 5´-TGAACTTATTCTGAGCCATTTCTTGT-3' BCL2-associated X protein BAX-F 5´-AGAGGATGATTGCCGCCGT-3' (BAX) BAX-R 5´-CAACCACCCTGGTCTTGGATC-3' BCL2 binding component 3 PUMA-F 5´-CCAAACGTGACCACTAGCCT-3' (PUMA) PUMA-R 5´-ACAGGATTCACAGTCTGGGC-3' phorbol-12-myristate-13- NOXA-F 5´-CTGTCCGAGGTGCTCCAGTT-3' acetate-induced protein 1 NOXA-R 5´-TCCTGAGTTGAGTAGCACAC-3' (NOXA) CLDN1-F 5´-AATTCTATGACCCTATGACCC-3' claudin 1 CLDN1-R 5´-GACAGGAACAGCAAAGTAGG-3'

29 carcinoembryonic antigen- CEACAM6-F 5´-GAAATACAGAACCCAGCGAGTGC-3' related cell adhesion CEACAM6-R 5´-CAGTGATGTTGGGGATAAAGAGC-3' molecule 6 (CEACAM6) PAK6-F 5´-TGAGGAGCAGATTGCCACTGTG-3' p21-activated kinase 6 PAK6-R 5´-CTGAGCACAGAATCCGAAGTCC-3' PAK1-F 5´-GCTGTTCTGGATGTGTTGGA-3' p21-activated kinase 1 PAK1-R 5´-TTCTGAAACTGGTGGCACTG-3' TNS3-F 5´-GTTGAAAGGGTGCTCGAATGA-3' tensin 3 TNS3-R 5´-GAACTTTCTGCTATTTCCTCCAATG-3' TNS4-F 5´-CCCACCATGAAGTTCGTGATG-3' tensin 4 TNS4-R 5´-CGGTATGAAGAGCTGTCCCTTATG-3' SVIL-F 5´-TTCACTAAGAGCGGCAGAG-3' supervillin SVIL-R 5´-TTCTCCTGTGGCTGTTCC-3' PXN-F 5´-AACAAGCAGAAGTCAGCAGAGCC-3' paxillin PXN-R 5´-CTAGCTTGTTCAGGTCGGAC-3' LPXN-F 5´-ACGCTCCACCCTTCAGGACA-´3 leupaxin LPXN-R 5´-GACATTGAGCTCCTGGATATTGG-´3 PALLD-F 5' -TGGCCCAGGAGTACAAAGTC -3' palladin PALLD-R 5' -ACATCTGGATCCTGCACCTC -3' RAB7B-F 5´-GGCCAGCATCCTCTCCAAGATTATC-3' Ras-related protein Rab7b RAB7B-R 5´-GATGCAGCCATCGGAGCCCTTGT-3' fascin actin-bundling protein FSCN1-F 5´-GGAGACCGACCAGGAGAC-´3 1 FSCN1-R 5´-CATTGGACGCCCTCAGTG-´3 ITGB4-F 5´-GCGACTACACTATTGGATTTGGC-´3 integrin, beta 4 ITGB4-R 5´-TGTCAGGCTGATGACGTTCTTG-´3 KLK5-F 5′-GCAGGTAGAGACTCCTGCCA-´3 kallikrein-related peptidase 5 KLK5-R 5′-CACAAGGGTAATCTCCCCAG´3

3.11 β-galactosidase assay

ARN8 cells were harvested post-treatment and lysed by repeated cycles of freeze-thawing in 0.25 M Tris-HCl, pH 7.5. To measure β-galactosidase activity, 10 µl of cell lysate (10 µg of total protein) was mixed with 1 µl of buffer 1 (0.1 M MgCl 2, 4.5 M β–mercaptoethanol), 22 µl of ONPG buffer (4 mg/ml o-nitrophenyl-β-D-galactopyranoside in 0.1 M sodium phosphate buffer, pH 7.5) and 67 µl of 0.1 M sodium phosphate buffer, pH 7.5. The reaction was incubated for 30 min at 37°C, stopped with 167 µl of 1M Na 2CO 3 and the absorbance was determined at 420 nm in a microplate reader. Experiment was performed in technical triplicates.

3.12 Flow cytometry

To check cell viability, both adherent cells and spheroids were rinsed in PBS, disrupted by trypsin and stained with 5 µg/ml propidium iodide for 5 min. After intensive washing in PBS,

30 samples were analysed on Guava EasyCyte Plus flow cytometer with Guava Express Pro 2.2.3 software (Millipore). Only single stained cells were analysed.

3.13 In situ PLA assay

Cells were grown in drops on adhesion microscope slides Superfrost Plus (Thermo Scientific) at 37°C for 24 hrs. Thereafter, cells were fixed with 3.7% paraformaldehyde in PBS at 37°C for 20 min and immediately subjected to in situ PLA staining. In situ PLA detection was in brief carried out as follows. Cells were permeabilized with 0.5% Triton-X100 (Sigma-Aldrich) in PBS and subjected to blocking using the DUOLINK (Olink Bioscience) blocking solution at 37°C in a wet chamber for 30 min. The cells were incubated overnight at 4°C with primary antibodies as follows: anti-AGR2 rabbit antibody HPA007912 (Sigma-Aldrich), in-house anti-AGR3.1 mouse antibody (Gray et al., 2012). The incubation was followed by washing in a Tris-buffered saline with Tween 20 (0.05M Tris-base, 0.15M NaCl, pH 7.6, with 0.05% Tween 20) for 5 min and 1 hour of incubation with anti-mouse and anti-rabbit PLA probes (Olink Bioscience). Next, cells were incubated for 30 min at 37°C with ligation mix containing: • connector oligonucleotides: GTTCTGTCATATTTAAGCGTCTTAA; CTATTAGCGTCCAGTGAATGCGAGTCCGTCTAAGAGAGTAGTACAGCAGCCGTCA AGAGTGTCTA (both TriLink BioTechnologies) • 0.05 U/µl T4 DNA ligase (New England Biolabs) • T4 DNA ligase buffer with 1 mM ATP (New England Biolabs) After two washes in a Tris-buffered saline with Tween 20, the ligated circles were amplified with 0.3 U/µl phi29 polymerase (Fermentas) in phi29 polymerase buffer (Fermentas), 0.25 mg/ml BSA, 0.25 mM dNTPs and the single-stranded rolling circle amplification (RCA) products were detected by hybridization with 0.025 µM Texas red labelled detection probe (CAGTGAATGCGAGTCCGTCT; Generi Biotech). Nuclei were counterstained with Hoechst 33342 (Sigma Aldrich). All steps were carried in one reaction for 90 min at 37°C. Slides were analysed using fluorescence microscope (Olympus BX41, Olympus) supplied with 100 W mercury burner and a computer-controlled filter wheel with excitation and emission filters. For observation 40× objective (UPLFLN 40×) and 100× immersion objective (UPLFLN 100×) were used.

3.14 Immunohistochemistry

Tumour samples were fixed in 10% neutral buffered formalin for 24 hrs and then embedded in paraffin wax. Immunohistochemical analysis was performed on 4 µm thick sections cut from formalin-fixed, paraffin-embedded archival tissue blocks, mounted on slides, deparaffinized in xylene and rehydrated into PBS through a graded ethanol series. Endogenous peroxidase activity was quenched in 3% hydrogen peroxide in PBS for 15 min.

Antigen retrieval was performed in citrate buffer pH 6 at 94°C for 20 min. For AGR3 immunodetection, the sections were incubated overnight at 4°C with mouse monoclonal antibody to AGR3 (clone 1.2, in house (Gray et al., 2012)). A streptavidin-biotin peroxidase detection system was used according to the manufacturers’ protocol (Vectastain Ellite ABC Kit, Vector Laboratories). Signal was visualized by 3, 3 ́-diaminobenzidine (Liquid DAB+ Substrate Chromogen System, Dako). Nuclear counterstaining was performed with Gill’s haematoxylin. For immunohistochemical evaluation, 3 conventional categories according to the number of positive cells were assessed: 1 – negative/border (0-5 % of positive cells); 2- weakly/moderately positive (5-50 % of positive cells); 3 – strongly positive (more than 50 % of positive cells) (Taiseer et al., 2014).

3.15 Patients and tissue specimens

The study group consisted of 129 patients undergoing surgical procedure for primary breast cancer at the Masaryk Memorial Cancer Institute (MMCI) between 2003 and 2006. Patient age at the time of diagnosis ranged from 29 - 84 years (median 57 years). Histological typing of tumours was carried out according to the criteria of WHO (Tavassoli FA. Devilee P, 2003). Tumour stage was determined according to the guidelines of the UICC (Sobin and Fleming, 1997). Tumour grade was established according to Bloom and Richardson in the modification of Elston and Ellis (Elston and Ellis, 1991). Oestrogen receptor, progesterone receptor (PR), human epidermal growth factor receptor 2 (Her2/neu) and Ki-67 expression was extracted from pathological records obtained from the MMCI database. For the evaluation of AGR3 prognostic relevance without regard to ER status, additional ER negative group of 90 breast cancer patients treated at MMCI between 1995 and 2006 were included for survival analysis. Informed consent has been obtained from all patients involved in this study. The data used were anonymised and they were handled according to Czech Republic existing legislation.

3.16 Statistical analysis

All statistical analyses were performed using STATISTICA Version 12 (StatSoft, Inc., Tulsa, OK, USA) and IBM SPSS Statistics 20.0. Fisher’s exact test and Pearson’s chi square test were applied to assess the associations of immunochistochemical staining for AGR3 with clinicopathologic variables. Progression free survival (PFS) was defined as the time from the date of surgery to the date of death or relapse of disease. Overall survival (OS) was defined as the time from surgery to death or last record. Patients who had not died or who were lost to follow-up were censored when they were last known to be alive. Differences between survival curves were assessed with the Breslow test. Unadjusted hazard ratios (HRs) ±95% confidence intervals (CIs) were obtained using Cox’s multivariate analysis with backward selection. Differences at P ≤0.05 were considered to be statistically significant.

4 RESULTS

4.1 Preliminary characterisation of AGR3 based on data mining

The precise role of AGR3 in physiology and pathology has not been validated so far. In order to elucidate AGR3 functions, in silico analyses with Genevestigator (Hruz et al., 2008) and MatInspector (Quandt et al., 1995) have been performed to compare AGR3 with its better- described AGR2 homologue.

4.1.1 AGR3 distribution in human healthy tissues and carcinomas To date, AGR3 was found in tumours affecting breast (Fletcher et al., 2003, Adam et al., 2003, Garczyk et al., 2015), liver (Brychtova et al., 2014b), prostate (Bu et al., 2013, Vaarala et al., 2012) and ovary (King et al., 2011, Gray et al., 2012). In order to expand current knowledge regarding AGR3 expression in human healthy tissues and carcinomas, a condition search tool from Genevesigator was used. As shown in Figure 4, both AGR2 and AGR3 are highly expressed in normal stomach, colon, pancreas, female reproductive system and respiratory system tissues as well as the corresponding carcinomas.

Figure 4. Overview of AGR proteins distribution in normal and cancer tissue Four-step grey scale indicates expression levels of AGR2 and AGR3 in normal and corresponding tumour tissue according to data from Genevestigator (lighter shades, lower AGR expression; darker shades, higher AGR expression). Liver tumour is characterized by heterogeneous expression of AGR proteins (shown by light/dark boxes).

In many cases, AGR3 mirrors AGR2 expression; however, they can also be expressed in a tissue-specific manner, such as normal skin or urinary bladder, where AGR3 is not present or expressed to a lesser extent, respectively, when compared to AGR2 that is expressed in both tissues. Interestingly, AGR3 expression is lost in kidney and urinary bladder tumours when compared to healthy tissues.

4.1.2 AGR3 co-expressed partners Further, the co-expression tool from Genevestigator was used to search for AGR3’s putative interacting molecules, downstream partners and/or cross-talking pathways. The results from this analysis are presented in Table 5.

Table 5. AGR3 co-expressed genes Co-expression tool from Genevestigator was used to find genes having an expression profile most similar to AGR3 gene. Each gene description was derived from http://www.genecard.com . Gene symbol Gene name Function AGR2 Anterior gradient homolog 2 proto-oncogene, role in cell migration, differentiation and growth, role in the regulation of mucins production ATP2C2 ATPase, Ca 2+ transporting type 2C, role in the hydrolysis of ATP and calcium member 2 transport C9orf152 Chromosome 9 open reading frame uncharacterized protein 152 CAPN8 Calpain 8 stomach-specific cysteine protease; plays a protective role in the gastric mucosa CAPN9 Calpain 9 stomach-specific cysteine protease; plays a protective role in the gastric mucosa CCL15 Chemokine (C-C motif) ligand 15 chemotactic factor attracting T-cells and monocytes CFTR Cystic fibrosis transmembrane member of the ATP-binding cassette (ABC) conductance regulator transporter superfamily; functions as a chloride channel and controls the regulation of other transport pathways CLDN3 Claudin 3 tight junction protein; role in cell-to-cell adhesion, regulation of tumour growth and metastases hepaCAM2 HEPACAM family member 2 immunoglobulin-like hepatocyte cell adhesion molecule; role in cell cycle/cell division, cell-matrix interactions, adhesion and motility LRRC31 Leucine rich repeat containing 31 uncharacterized protein PIGR Polymeric immunoglobulin member of the immunoglobulin superfamily; receptor mediating the secretion of polymeric immunoglobulins SLC44A4 Solute carrier family 44, member 4 probably a transmembrane transport protein involved in the uptake of choline

ST6GAL Beta-galactoside alpha-2,6- membrane protein, member of sialyltransferase 1 glycosyltransferase family 29, transfers sialic acid from CMP-sialic acid to galactose- containing acceptor substrates TFF1 Trefoil factor 1 membrane protein, member of glycosyltransferase family 29, transfers sialic acid from CMP-sialic acid to galactose- containing acceptor substrates TFF3 Trefoil factor 3 secreted protein; involved in the maintenance and repair of the intestinal mucosa; promotes

the mobility of epithelial cells in healing processes TMC5 Transmembrane channel-like 5 putative role in ion transport TMPRSS2 Transmembrane protease, serine 2 member of serine protease family; contains a type II transmembrane domain, receptor class A domain, a scavenger receptor cysteine-rich domain and a cleavable protease domain

4.1.3 AGR3 in silico promoter analysis The uncoupled expression of AGR2 and AGR3 in different carcinomas has been reported (Gray et al., 2012, Bu et al., 2013), suggesting that both proteins are required in divergent situations and may be components of various signalling pathways. Moreover, except from EsR and AR signalling (Bu et al., 2013, Welboren et al., 2009), little is known about upstream regulators of AGR3 expression. Therefore, in order to analyse the promoter region of both AGR2 and AGR3, MatInspector was used in search for transcription factor (TF) binding sites. TFs were selected according to the matrix similarity and only TFs with score higher than 0.95 were taken into account.

Many of the TFs found in the in silico analysis play significant roles in the developmental processes and differentiation of tissues and/or organs of diverse origins, including among others members of SOX family, PAX4, GATA1, GATA4, NFAT or members of HOX family (Attachment 1, Table S1). Interestingly, most of the TFs potentially binding to AGR2 or AGR3 promoters are exclusive for each protein, with the exception of common FOXP1, ZEB1 (AREB6), SMARCA3, GATA4, NFAT, CEBPB and NKX2.5 (Figure 5), suggesting that their expression may be driven by separate signalling pathways.

Figure 5. Nucleotide sequence of AGR2 and AGR3 promoter with representative potential binding sites for transcription factors Arrows indicate position of the transcription start site determined according to MatInspector. Underlined molecules represent transcription factors which bind to the (+) DNA strand, whereas molecules without highlights correspond to the transcription factors which bind to the (−) DNA strand. Transcription factors common for AGR2 and AGR3 are shown in black boxes.

4.2 Immunohistochemical analysis of AGR3 expression in primary breast carcinomas

Although AGR3 expression in breast carcinoma has been demonstrated in a few independent reports (Adam et al., 2003, Fletcher et al., 2003, Garczyk et al., 2015), there is still a limiting amount of data depicting its prognostic significance in these malignancy. The analysed cohort composed of 95 (73.6%) tumours classified as ductal breast carcinomas, 18 (14%) as lobular type and remaining 16 (12.4%) specimens were either of different or unknown origin. The remaining clinicopathological characteristics of the study group and their distributions are summarized in Table 6. Table 6. Clinicopathological characteristics of primary breast carcinomas aDefined in ‘Materials and Methods’; EsR= oestrogen receptor; PR= progesterone receptor; AGR3 expression: 1 – negative/border, 2- weakly/moderately positive, 3 – strongly positive. bNumber of patients. cPercentage of total patients, out of a total of 129. NA= not available. dCut-off for Ki-67 was used according to St Gallen Consensus in 2009. Variable a Group No b %c Histology Ductal 95 73.6 Lobular 18 14 Other 9 7 NA 7 5.4 Histological grade G1 27 20.9

G2 43 33.4 G3 56 43.4 NA 3 2.3

Tumour size pT 1 44 34.1

pT 2 65 50.3

pT 3 6 4.7

pT 4 9 7 NA 5 3.9 Nodal status Negative 45 34.9 Positive 73 56.6 NA 11 8.5 EsR status Negative 29 22.5 Positive 100 77.5 NA 0 0 PR status Negative 34 26.4 Positive 94 72.8 NA 1 0.8 Her2/neu status Negative 92 71.3 Positive 36 27.9 NA 1 0.8 Ki-67 d <15% 55 42.6 ≥15% 61 47.3 NA 13 10.1 AGR3 expression 1 25 19.4 2 25 19.4 3 79 61.2

4.2.1 Association of AGR3 expression with other tumour variables AGR3 expression was determined using immunohistochemical staining and reverse transcription followed by a quantitative PCR. Staining of primary breast carcinomas for AGR3 varied from tumour to tumour and was mainly cytoplasmic. Overall, of the 129 cases, 25 (19.4%) were classified as negative or borderline stained for AGR3 (< 5% of positive cells) and the remaining 104 (80.6%) showed AGR3 positivity to different degrees (from weak to strong) (Figure 6).

Figure 6. Immunohistochemical staining for AGR3 The level of AGR3 expression in primary breast carcinomas was determined by immunostaining in three-point scale: (A) negative or border; (B) weak to moderate; (C) strong. Scale bars represent a length of 100 µm.

Immunochistochemical staining for AGR3 was then cross-tabulated with selected tumour features including histological type, tumour size, nodal status, histological grade, EsR, PR, Her-2/neu status and Ki-67 expression level. AGR3 positivity was significantly correlated with ductal type and slowly proliferating tumours as measured by expression level of Ki-67 marker ( P <0.0001) as well as lower tumour grade ( P <0.0001). Moreover, the degree of staining for AGR3 was significantly associated with that for the EsR ( P <0.0001) and PR ( P <0.0001). There was no significant correlation between AGR3 positivity and tumour size, nodal status or Her-2/neu status (Table 7). Similar trends for AGR3 in relation to the clinicopathological parameters were found on both protein and mRNA levels.

Table 7. Association of immunohistochemical staining for AGR3 with other tumour variables EsR= oestrogen receptor; PR= progesterone receptor. aNumber (percentage) of patients with tumours characterized by negative/border, weak/moderate or strong expression of AGR3. Probability, P, from bFisher's exact test with the Freeman-Halton extension or cPearson’s chi-square test. Variable No. patients (%) a Statistical Patients AGR3 AGR3 AGR3 significance negative/border weak/moderate strong Histological grade G1 27 3 (11.1) 4 (14.8) 20 (74.1) < 0.0001 b G2 43 3 (7) 7 (16.3) 33 (76.7) G3 56 18 (32.1) 14 (25) 24 (42.9) Tumour size

b pT 1 44 8 (18.2) 9 (20.4) 27 (61.4) 0.664 pT 2 65 11 (16.9) 11 (16.9) 43 (66.2) pT 3 + pT 4 6 5 (33.3) 2 (13.3) 8 (53.3) Nodal status Negative 45 10 (22.2) 10 (22.2) 25 (55.6) 0.332 c Positive 73 13 (17.8) 10 (13.7) 50 (68.5) EsR status Negative 29 20 (69) 8 (27.6) 1 (3.4) < 0.0001 b Positive 100 5 (5) 17 (17) 78 (78) PR status Negative 34 19 (55.9) 9 (26.5) 6 (17.6) < 0.0001 c Positive 94 6 (6.4) 16 (17) 72 (76.6) Her2/neu status Negative 92 18 (19.6) 15 (16.3) 59 (64.1) 0.318 c Positive 36 7 (19.4) 10 (27.8) 19 (52.8) Ki-67 <15% 55 6 (10.9) 9 (16.4) 40 (72.7) < 0.0001 c ≥15% 61 14 (23.0) 14 (23.0) 33 (54.0)

In addition, AGR2 mRNA levels were examined under the same parameters and similar association between AGR2 gene expression and clinicopathological variables were found as seen for AGR3 (see Attachment 2, Table S1). In line with these observations, a strong correlation between AGR2 and AGR3 mRNA levels (p<0.0001, Rs = 0.6327) was confirmed using Spearmann Rank Order correlation.

4.2.2 Association of AGR3 with patient survival For the survival analysis, follow-up was determined for 10 years since surgical removal. Median progression-free survival (PFS) was 92 months (range 1-120) and median overall survival (OS) was 103 months (range 1-120). As there was almost no difference in survival curves between negative/border and weak/moderate subgroups (data not shown), for further statistical analyses the above subgroups were combined (further denoted as AGR3 ”low”) and were compared with patients whose tumours showed strong AGR3 positivity (more than 50% of stained cells, denoted as AGR3 “high”). While OS was not significantly affected by AGR3 expression, despite the fact that Kaplan-Meier curves indicated some trend in favour of increased AGR3 expression, ( P= 0.111), these patients had significantly longer PFS ( P= 0.037) (Figure 7).

Figure 7. Association of immunohistochemical staining for AGR3 with patient survival (A) Determination of progression free survival by Kaplan-Meier analysis in patients with “high” AGR3 expression (more than 50% of positive cells) and patients with “low” AGR3 expression (less than 50% of positive cells) using Breslow test, (P= 0.037 ). (B) Determination of overall survival by Kaplan-Meier analysis in patients with “high” AGR3 expression (and patients with “low” AGR3 expression using Breslow test, (P= 0.111).

4.2.3 Association of AGR3 and other tumour variables with patient survival Patients with larger tumour size, higher histological grade, positive nodal status, and positive Her2/neu status had significantly poorer prognosis at 10 years of follow-up (Breslow test, all P˂0.05). For multivariate survival analysis, the following clinicopathological parameters were included in Cox’s model with backward selection: histological type, tumour grade, tumour size, nodal status, and EsR, PR, Her2/neu, and AGR3 status. As a result, tumour size and Her2/neu status were found to be independent prognostic factors for PFS, whereas tumour size and grade reached statistical significance for OS time in the studied cohort (Table 9). The remaining clinical and histological characteristics, including AGR3, failed statistical significance and were removed from the analysis during the selection process.

Table 8. Independent prognostic factors for the analysed set of tumours according to Cox’s multivariate survival analyses aHR= unadjusted hazard ratio. b 95% CI= ±95% confidence intervals. Variable HR a 95% CI b Statistical significance Progression free survival pT 1 1.00 0.003 pT 2 1.99 0.83-4.73 0.121 pT 3 8.25 2.68-25.44 <0.0001 Her2/neu status 3.60 1.64-7.88 0.001 Overall survival pT 1 1.00 0.006 pT 2 1.38 0.51-3.76 0.531 pT 3 13.56 3.06-60.04 0.001 G1 1.0 0.015 G2 1.87 0.20-17.34 0.582 G3 6.38 0.82-49.42 0.076

When further pair-wised with other variables, AGR3 positivity was associated with better outcome in the subgroup of patients with tumours defined by smaller histological grade (G ≤2; OS: P=0.005; PFS: P=0.024) but not by higher histological grade (G>2; OS: P=0.583; PFS: P=0.945). In Her2/neu negative set of tumours AGR3 expression significantly correlated with longer PFS ( P=0.019) as well as OS ( P=0.009). On the other hand, when ER-positive cases were considered separately, AGR3 expression did not reached statistical significance for improved survival (for PFS: P= 0.228; for OS: P= 0.234) (Table 8).

Table 9. Survival analysis for patients with AGR3-expressing tumours aProbability, P, from Breslow test; EsR= oestrogen receptor; PFS= progression free survival; PR= progesterone receptor; OS= overall survival. Subgroup Statistical significance a PFS OS Histological grade G≤2 0.024 0.005 G>2 0.945 0.583 Her2/neu status Negative 0.019 0.009 Positive 0.781 0.278 PR status Negative 0.669 0.911 Positive 0.448 0.224 EsR status Negative 0.431 0.507 Positive 0.228 0.234

5 DISCUSSION

AGR2 and AGR3 are conserved human homologues of Xenopus laevis XAG-2 protein implicated in developmental processes and regeneration of body appendages (Sive et al., 1989). AGR2 is a considerably scrutinized oncogene and emerging drug target as it promotes growth, survival and dissemination of the metastatic cells as well as resistance to treatment by modulating tumour-associated signalling pathways and overcoming ER stress (Brychtova et al., 2011, Chevet et al., 2013, Salmans et al., 2013). On the other hand, AGR3 function in health and disease remains ambiguous since data published so far are relatively contradictory. Nevertheless, there are reports linking AGR3 with tumour cell metastasis, survival and prediction of patient’s outcome (Fletcher et al., 2003, Gray et al., 2012, King et al., 2011, Garczyk et al., 2015), which provides a clear evidence for its crucial role in cancer biology.

5.1 Prognostic relevance of AGR3 expression in breast cancer

AGR3 expression was demonstrated in various cancers, including that of breast (Adam et al., 2003, Fletcher et al., 2003, Garczyk et al., 2015), prostate (Vaarala et al., 2012, Bu et al., 2013), ovary (Gray et al., 2012, King et al., 2011) and liver (Brychtova et al., 2014b); however, there is a limiting amount of data depicting AGR3 prognostic relevance in these malignancies. In the present work, a cohort of 129 primary breast carcinomas was analysed in order to assess clinical and prognostic relevance of AGR3 expression. AGR3 was detected in 104 (80 %) out of 129 specimens, hence confirming previously reported predominant expression of AGR3 protein in breast tumours (Fletcher et al., 2003, Adam et al., 2003). In the analysed group, AGR3 was significantly associated with EsR and PR positivity and tumour grades G ≤2 but not with tumour size and nodal status, which is consistent with other studies, mostly pertaining to AGR2 homologue (Fletcher et al., 2003, Innes et al., 2006, Fritzsche et al., 2006). Moreover, an increase in AGR3 positivity negatively correlated with the proliferation rate defined by the level of Ki-67 expression. Notably, similar trends in relation to other clinicopathological parameters were also found for AGR3 mRNA level. Correlation with EsR and PR positivity, slowly proliferating and well-differentiated tumours suggests that AGR3 expression is associated with less aggressive tumours that are more prone to effective treatment and therefore favourable outcome. Indeed, in the current work it was demonstrated that the presence of immunohistochemical staining for AGR3 was associated with improved patient progression free survival. Although, in the whole cohort AGR3 expression did not predict longer overall survival, patients whose tumours were characterised by strong AGR3 positivity showed better response to therapy. Moreover, AGR3 predicted better outcome in the subgroup of patients with well-differentiated tumours, which is consistent with previously demonstrated significance of AGR3 expression in ovarian cancers (King et al., 2011). Quite the contrary, AGR2 is often described as an indicator of poor prognosis (Innes et al., 2006, Barraclough et al., 2009) metastasis (Liu et al., 2005, Smirnov et al., 2005), and resistance to commonly used treatments (Hrstka et al., 2010, Hrstka et al., 2013a), indicating divergent and/or context-dependent roles of AGR proteins in breast cancer .

It is of note that similar antagonistic impact of AGR proteins on patient outcome is also observed in ovarian cancers where AGR3 promotes better outcome (King et al., 2011), whereas AGR2 predicts shortened overall survival (Darb-Esfahani et al., 2012), possibly due to the stimulation of cell growth and migration (Park et al., 2011). However, given that AGR3 was also shown to mediate cisplatin resistance and correlate with unfavourable outcome in low and intermediate grade breast tumours (Garczyk et al., 2015), an explicit conclusion of AGR3 protective, anti-tumour role cannot be conclusively drawn. Moreover, comparing AGR2 and AGR3 distribution both in human healthy tissues and carcinomas using Genevestigator, in the present work it was found that AGR3 mirrors AGR2 expression in many cases, such as stomach, colon, pancreas, breast, female reproductive system or respiratory system. In accordance, immunohistochemical analysis demonstrated a strong correlation between AGR3 and AGR2 mRNA levels in breast carcinomas as well as similar associations of both genes with clinicopathological variables, which collectively suggests that AGR proteins may also play convergent role in physiology and pathology.

6 CONCLUSIONS

AGR2 and AGR3 proteins appear to play a significant role in oncology due to their high expression in various tumours. Although AGR2 is linked with elevated proliferation, survival, invasion, metastasis and poor patient’s outcome, information regarding AGR3 function in cancer is limited. The present work demonstrates that AGR3 shares many similarities with AGR2 homologue in terms of expression pattern in normal and tumour tissues, potential interacting molecules as well as cross-talking pathways or exerted functions. It also provides a clear evidence for AGR proteins cross-talk since it was found that AGR3 and AGR2 directly interact and through autocrine fashion regulate the endogenous expression of the other homologue. These findings are of particular importance for clinical oncology applications, as they indicate the necessity of cognate evaluation of AGR2 and AGR3 diagnostic and prognostic significance in tumours.

As shown here, AGR3 acts both intra- and extracellularly; within cancer cell it promotes survival upon tumour-associated stresses including DNA damage and hypoxia, while secreted into the extracellular space, AGR3 regulates cell adhesion and motility and thus may contribute to metastasis development. Secreted AGR3 appears to be more active when compared to AGR2, since it shows more prominent effect on breast cancer cell behaviour and affects wider range of cell signalling pathways. This finding supports the role of AGR3 as a potential serum biomarker and rationalise a need for the development of appropriate assays to detect extracellular AGR3. Further, immunohistochemical analysis of breast cancer specimens revealed that AGR3 expression correlates with differentiation and better response to therapy; suggesting that dependably on cellular context AGR3 may also play a tumour-suppressive role. Mechanisms triggering AGR3 expression in cancer are poorly understood possibly due to the lack of appropriate cellular models, slowing its thorough characterisation. However, the current work suggests that in addition to EsR, AGR3 expression may be controlled by Notch, JAK/STAT, TGF-β, PR and MAPK signalling, as well as by an autoregulatory loop.

Taken together, the findings of this study establish the importance of AGR3 in cancer biology and shed new light on AGR proteins function and their utility as potential biomarkers.

7 REFERENCES

ABERGER, F., WEIDINGER, G., GRUNZ, H. & RICHTER, K. 1998. Anterior specification of embryonic ectoderm: the role of the Xenopus cement gland-specific gene XAG-2. Mech Dev, 72 , 115-30. ADAM, P. J., BOYD, R., TYSON, K. L., FLETCHER, G. C., STAMPS, A., HUDSON, L., POYSER, H. R., REDPATH, N., GRIFFITHS, M., STEERS, G., HARRIS, A. L., PATEL, S., BERRY, J., LOADER, J. A., TOWNSEND, R. R., DAVIET, L., LEGRAIN, P., PAREKH, R. & TERRETT, J. A. 2003. Comprehensive proteomic analysis of breast cancer cell membranes reveals unique proteins with potential roles in clinical cancer. J Biol Chem, 278 , 6482-9. ALAVI, M., MAH, V., MARESH, E. L., BAGRYANOVA, L., HORVATH, S., CHIA, D., GOODGLICK, L. & LIU, A. Y. 2015. High expression of AGR2 in lung cancer is predictive of poor survival. BMC Cancer, 15 , 655. APPENZELLER-HERZOG, C. & ELLGAARD, L. 2008. The human PDI family: versatility packed into a single fold. Biochim Biophys Acta, 1783 , 535-48. ARMES, J. E., DAVIES, C. M., WALLACE, S., TAHERI, T., PERRIN, L. C. & AUTELITANO, D. J. 2013. AGR2 expression in ovarian tumours: a potential biomarker for endometrioid and mucinous differentiation. Pathology, 45 , 49-54. ARUMUGAM, T., DENG, D., BOVER, L., WANG, H., LOGSDON, C. D. & RAMACHANDRAN, V. 2015. New Blocking Antibodies against Novel AGR2- C4.4A Pathway Reduce Growth and Metastasis of Pancreatic Tumors and Increase Survival in Mice. Mol Cancer Ther, 14 , 941-51. BARRACLOUGH, D. L., PLATT-HIGGINS, A., DE SILVA RUDLAND, S., BARRACLOUGH, R., WINSTANLEY, J., WEST, C. R. & RUDLAND, P. S. 2009. The metastasis-associated anterior gradient 2 protein is correlated with poor survival of breast cancer patients. Am J Pathol, 175 , 1848-57. BENHAM, A. M. 2012. The protein disulfide isomerase family: key players in health and disease. Antioxid Redox Signal, 16 , 781-9. BERGSTROM, J. H., BERG, K. A., RODRIGUEZ-PINEIRO, A. M., STECHER, B., JOHANSSON, M. E. & HANSSON, G. C. 2014. AGR2, an endoplasmic reticulum protein, is secreted into the gastrointestinal mucus. PLoS One, 9, e104186. BI, S., HONG, P. W., LEE, B. & BAUM, L. G. 2011. Galectin-9 binding to cell surface protein disulfide isomerase regulates the redox environment to enhance T-cell migration and HIV entry. Proc Natl Acad Sci U S A, 108 , 10650-5. BLASSBERG, R. A., GARZA-GARCIA, A., JANMOHAMED, A., GATES, P. B. & BROCKES, J. P. 2011. Functional convergence of signalling by GPI-anchored and anchorless forms of a salamander protein implicated in limb regeneration. J Cell Sci, 124 , 47-56. BLAYDES, J. P. & HUPP, T. R. 1998. DNA damage triggers DRB-resistant phosphorylation of human p53 at the CK2 site. Oncogene, 17 , 1045-52. BONSER, L. R., SCHROEDER, B. W., OSTRIN, L. A., BAUMLIN, N., OLSON, J. L., SALATHE, M. & ERLE, D. J. 2015. The ER resident protein AGR3 is required for regulation of ciliary beat frequency in the airway. Am J Respir Cell Mol Biol . BROWN, J. M. & GIACCIA, A. J. 1998. The unique physiology of solid tumors: opportunities (and problems) for cancer therapy. Cancer Res, 58 , 1408-16. BRYCHTOVA, V., HERMANOVA, M., KARASEK, P., LENZ, J., SELINGEROVA, I., VOJTESEK, B., KALA, Z. & HRSTKA, R. 2014a. Anterior gradient 2 and mucin 4 expression mirrors tumor cell differentiation in pancreatic adenocarcinomas, but

aberrant anterior gradient 2 expression predicts worse patient outcome in poorly differentiated tumors. Pancreas, 43 , 75-81. BRYCHTOVA, V., VOJTESEK, B. & HRSTKA, R. 2011. Anterior gradient 2: a novel player in tumor cell biology. Cancer Lett, 304 , 1-7. BRYCHTOVA, V., ZAMPACHOVA, V., HRSTKA, R., FABIAN, P., NOVAK, J., HERMANOVA, M. & VOJTESEK, B. 2014b. Differential expression of anterior gradient protein 3 in intrahepatic cholangiocarcinoma and hepatocellular carcinoma. Exp Mol Pathol, 96 , 375-81. BU, H., BORMANN, S., SCHAFER, G., HORNINGER, W., MASSONER, P., NEEB, A., LAKSHMANAN, V. K., MADDALO, D., NESTL, A., SULTMANN, H., CATO, A. C. & KLOCKER, H. 2011. The anterior gradient 2 (AGR2) gene is overexpressed in prostate cancer and may be useful as a urine sediment marker for prostate cancer detection. Prostate, 71 , 575-87. BU, H., SCHWEIGER, M. R., MANKE, T., WUNDERLICH, A., TIMMERMANN, B., KERICK, M., PASQUALINI, L., SHEHU, E., FUCHSBERGER, C., CATO, A. C. & KLOCKER, H. 2013. Anterior gradient 2 and 3--two prototype androgen-responsive genes transcriptionally upregulated by androgens and by oestrogens in prostate cancer cells. FEBS J, 280 , 1249-66. CLARKE, C., RUDLAND, P. & BARRACLOUGH, R. 2015. The metastasis-inducing protein AGR2 is O-glycosylated upon secretion from mammary epithelial cells. Mol Cell Biochem . DARB-ESFAHANI, S., FRITZSCHE, F., KRISTIANSEN, G., WEICHERT, W., SEHOULI, J., BRAICU, I., DIETEL, M. & DENKERT, C. 2012. Anterior gradient protein 2 (AGR2) is an independent prognostic factor in ovarian high-grade serous carcinoma. Virchows Arch, 461 , 109-16. DE SOUZA, W. F., FORTUNATO-MIRANDA, N., ROBBS, B. K., DE ARAUJO, W. M., DE-FREITAS-JUNIOR, J. C., BASTOS, L. G., VIOLA, J. P. & MORGADO-DIAZ, J. A. 2013. Claudin-3 overexpression increases the malignant potential of colorectal cancer cells: roles of ERK1/2 and PI3K-Akt as modulators of EGFR signaling. PLoS One, 8, e74994. DI MARO, G., SALERNO, P., UNGER, K., ORLANDELLA, F. M., MONACO, M., CHIAPPETTA, G., THOMAS, G., OCZKO-WOJCIECHOWSKA, M., MASULLO, M., JARZAB, B., SANTORO, M. & SALVATORE, G. 2014. Anterior gradient protein 2 promotes survival, migration and invasion of papillary thyroid carcinoma cells. Mol Cancer, 13 , 160. DONG, A., GUPTA, A., PAI, R. K., TUN, M. & LOWE, A. W. 2011. The human adenocarcinoma-associated gene, AGR2, induces expression of amphiregulin through Hippo pathway co-activator YAP1 activation. J Biol Chem, 286 , 18301-10. DONG, A., WODZIAK, D. & LOWE, A. W. 2015. Epidermal growth factor receptor (EGFR) signaling requires a specific endoplasmic reticulum thioredoxin for the posttranslational control of receptor presentation to the cell surface. J Biol Chem, 290 , 8016-27. DUMARTIN, L., WHITEMAN, H. J., WEEKS, M. E., HARIHARAN, D., DMITROVIC, B., IACOBUZIO-DONAHUE, C. A., BRENTNALL, T. A., BRONNER, M. P., FEAKINS, R. M., TIMMS, J. F., BRENNAN, C., LEMOINE, N. R. & CRNOGORAC-JURCEVIC, T. 2011. AGR2 is a novel surface antigen that promotes the dissemination of pancreatic cancer cells through regulation of cathepsins B and D. Cancer Res, 71 , 7091-102. EDGELL, T. A., BARRACLOUGH, D. L., RAJIC, A., DHULIA, J., LEWIS, K. J., ARMES, J. E., BARRACLOUGH, R., RUDLAND, P. S., RICE, G. E. & AUTELITANO, D. J.

2010. Increased plasma concentrations of anterior gradient 2 protein are positively associated with ovarian cancer. Clin Sci (Lond), 118 , 717-25. ELSTON, C. W. & ELLIS, I. O. 1991. Pathological prognostic factors in breast cancer. I. The value of histological grade in breast cancer: experience from a large study with long- term follow-up. Histopathology, 19 , 403-10. FENG, L., HOLLSTEIN, M. & XU, Y. 2006. Ser46 phosphorylation regulates p53-dependent apoptosis and replicative senescence. Cell Cycle, 5, 2812-9. FLETCHER, G. C., PATEL, S., TYSON, K., ADAM, P. J., SCHENKER, M., LOADER, J. A., DAVIET, L., LEGRAIN, P., PAREKH, R., HARRIS, A. L. & TERRETT, J. A. 2003. hAG-2 and hAG-3, human homologues of genes involved in differentiation, are associated with oestrogen receptor-positive breast tumours and interact with metastasis gene C4.4a and dystroglycan. Br J Cancer, 88 , 579-85. FRAM, S., KING, H., SACKS, D. B. & WELLS, C. M. 2014. A PAK6-IQGAP1 complex promotes disassembly of cell-cell adhesions. Cell Mol Life Sci, 71 , 2759-73. FRITZSCHE, F. R., DAHL, E., DANKOF, A., BURKHARDT, M., PAHL, S., PETERSEN, I., DIETEL, M. & KRISTIANSEN, G. 2007. Expression of AGR2 in non small cell lung cancer. Histol Histopathol, 22 , 703-8. FRITZSCHE, F. R., DAHL, E., PAHL, S., BURKHARDT, M., LUO, J., MAYORDOMO, E., GANSUKH, T., DANKOF, A., KNUECHEL, R., DENKERT, C., WINZER, K. J., DIETEL, M. & KRISTIANSEN, G. 2006. Prognostic relevance of AGR2 expression in breast cancer. Clin Cancer Res, 12 , 1728-34. GALLIGAN, J. J. & PETERSEN, D. R. 2012. The human protein disulfide isomerase gene family. Hum Genomics, 6, 6. GARCZYK, S., VON STILLFRIED, S., ANTONOPOULOS, W., HARTMANN, A., SCHRAUDER, M. G., FASCHING, P. A., ANZENEDER, T., TANNAPFEL, A., ERGONENC, Y., KNUCHEL, R., ROSE, M. & DAHL, E. 2015. AGR3 in Breast Cancer: Prognostic Impact and Suitable Serum-Based Biomarker for Early Cancer Detection. PLoS One, 10 , e0122106. GEBACK, T., SCHULZ, M. M., KOUMOUTSAKOS, P. & DETMAR, M. 2009. TScratch: a novel and simple software tool for automated analysis of monolayer wound healing assays. Biotechniques, 46 , 265-74. GRAY, T. A., MACLAINE, N. J., MICHIE, C. O., BOUCHALOVA, P., MURRAY, E., HOWIE, J., HRSTKA, R., MASLON, M. M., NENUTIL, R., VOJTESEK, B., LANGDON, S., HAYWARD, L., GOURLEY, C. & HUPP, T. R. 2012. Anterior Gradient-3: a novel biomarker for ovarian cancer that mediates cisplatin resistance in xenograft models. J Immunol Methods, 378 , 20-32. HATA, S., ABE, M., SUZUKI, H., KITAMURA, F., TOYAMA-SORIMACHI, N., ABE, K., SAKIMURA, K. & SORIMACHI, H. 2010. Calpain 8/nCL-2 and calpain 9/nCL-4 constitute an active protease complex, G-calpain, involved in gastric mucosal defense. PLoS Genet, 6, e1001040. HIGA, A., MULOT, A., DELOM, F., BOUCHECAREILH, M., NGUYEN, D. T., BOISMENU, D., WISE, M. J. & CHEVET, E. 2011. Role of pro-oncogenic protein disulfide isomerase (PDI) family member anterior gradient 2 (AGR2) in the control of endoplasmic reticulum homeostasis. J Biol Chem, 286 , 44855-68. HO, M. E., QUEK, S. I., TRUE, L. D., MORRISSEY, C., COREY, E., VESSELLA, R. L., DUMPIT, R., NELSON, P. S., MARESH, E. L., MAH, V., ALAVI, M., KIM, S. R., BAGRYANOVA, L., HORVATH, S., CHIA, D., GOODGLICK, L. & LIU, A. Y. 2013. Prostate cancer cell phenotypes based on AGR2 and CD10 expression. Mod Pathol, 26 , 849-59.

HONG, X. Y., WANG, J. & LI, Z. 2013. AGR2 expression is regulated by HIF-1 and contributes to growth and angiogenesis of glioblastoma. Cell Biochem Biophys, 67 , 1487-95. HRSTKA, R., BOUCHALOVA, P., MICHALOVA, E., MATOULKOVA, E., MULLER, P., COATES, P. J. & VOJTESEK, B. 2015. AGR2 oncoprotein inhibits p38MAPK and p53 activation through a DUSP10-mediated regulatory pathway. Molecular Oncology (accepted). HRSTKA, R., BRYCHTOVA, V., FABIAN, P., VOJTESEK, B. & SVOBODA, M. 2013a. AGR2 predicts tamoxifen resistance in postmenopausal breast cancer patients. Dis Markers, 35 , 207-12. HRSTKA, R., MURRAY, E., BRYCHTOVA, V., FABIAN, P., HUPP, T. R. & VOJTESEK, B. 2013b. Identification of an AKT-dependent signalling pathway that mediates tamoxifen-dependent induction of the pro-metastatic protein anterior gradient-2. Cancer Lett, 333 , 187-93. HRSTKA, R., NENUTIL, R., FOURTOUNA, A., MASLON, M. M., NAUGHTON, C., LANGDON, S., MURRAY, E., LARIONOV, A., PETRAKOVA, K., MULLER, P., DIXON, M. J., HUPP, T. R. & VOJTESEK, B. 2010. The pro-metastatic protein anterior gradient-2 predicts poor prognosis in tamoxifen-treated breast cancers. Oncogene, 29 , 4838-47. HRUZ, T., LAULE, O., SZABO, G., WESSENDORP, F., BLEULER, S., OERTLE, L., WIDMAYER, P., GRUISSEM, W. & ZIMMERMANN, P. 2008. Genevestigator v3: a reference expression database for the meta-analysis of transcriptomes. Adv Bioinformatics, 2008 , 420747. HU, Z., GU, Y., HAN, B., ZHANG, J., LI, Z., TIAN, K., YOUNG, C. Y. & YUAN, H. 2012. Knockdown of AGR2 induces cellular senescence in prostate cancer cells. Carcinogenesis, 33 , 1178-86. CHANDA, D., LEE, J. H., SAWANT, A., HENSEL, J. A., ISAYEVA, T., REILLY, S. D., SIEGAL, G. P., SMITH, C., GRIZZLE, W., SINGH, R. & PONNAZHAGAN, S. 2014. Anterior gradient protein-2 is a regulator of cellular adhesion in prostate cancer. PLoS One, 9, e89940. CHEN, R., PAN, S., DUAN, X., NELSON, B. H., SAHOTA, R. A., DE RHAM, S., KOZAREK, R. A., MCINTOSH, M. & BRENTNALL, T. A. 2010. Elevated level of anterior gradient-2 in pancreatic juice from patients with pre-malignant pancreatic neoplasia. Mol Cancer, 9, 149. CHEN, Y. C., LU, Y. F., LI, I. C. & HWANG, S. P. 2012. Zebrafish Agr2 is required for terminal differentiation of intestinal goblet cells. PLoS One, 7, e34408. CHEN, Y. T., HO, C. L., CHEN, P. K., CHEN, Y. L. & CHANG, C. F. 2013. Anterior gradient 2: a novel sensitive tumor marker for metastatic oral cancer. Cancer Lett, 339 , 270-8. CHEVET, E., FESSART, D., DELOM, F., MULOT, A., VOJTESEK, B., HRSTKA, R., MURRAY, E., GRAY, T. & HUPP, T. 2013. Emerging roles for the pro-oncogenic anterior gradient-2 in cancer development. Oncogene, 32 , 2499-509. CHO, J., KENNEDY, D. R., LIN, L., HUANG, M., MERRILL-SKOLOFF, G., FURIE, B. C. & FURIE, B. 2012. Protein disulfide isomerase capture during thrombus formation in vivo depends on the presence of beta3 integrins. Blood, 120 , 647-55. CHUNG, K., NISHIYAMA, N., WANIBUCHI, H., YAMANO, S., HANADA, S., WEI, M., SUEHIRO, S. & KAKEHASHI, A. 2012. AGR2 as a potential biomarker of human lung adenocarcinoma. Osaka City Med J, 58 , 13-24. CHUNG, K., NISHIYAMA, N., YAMANO, S., KOMATSU, H., HANADA, S., WEI, M., WANIBUCHI, H., SUEHIRO, S. & KAKEHASHI, A. 2011. Serum AGR2 as an early

diagnostic and postoperative prognostic biomarker of human lung adenocarcinoma. Cancer Biomark, 10 , 101-7. INNES, H. E., LIU, D., BARRACLOUGH, R., DAVIES, M. P., O'NEILL, P. A., PLATT- HIGGINS, A., DE SILVA RUDLAND, S., SIBSON, D. R. & RUDLAND, P. S. 2006. Significance of the metastasis-inducing protein AGR2 for outcome in hormonally treated breast cancer patients. Br J Cancer, 94 , 1057-65. IVANOVA, A. S., SHANDARIN, I. N., ERMAKOVA, G. V., MININ, A. A., TERESHINA, M. B. & ZARAISKY, A. G. 2015. The secreted factor Ag1 missing in higher vertebrates regulates fins regeneration in Danio rerio. Sci Rep, 5, 8123. IVANOVA, A. S., TERESHINA, M. B., ERMAKOVA, G. V., BELOUSOV, V. V. & ZARAISKY, A. G. 2013. Agr genes, missing in amniotes, are involved in the body appendages regeneration in frog tadpoles. Sci Rep, 3, 1279. JACQUOT, J. P., GELHAYE, E., ROUHIER, N., CORBIER, C., DIDIERJEAN, C. & AUBRY, A. 2002. Thioredoxins and related proteins in photosynthetic organisms: molecular basis for thiol dependent regulation. Biochem Pharmacol, 64 , 1065-9. JANKNECHT, R., ERNST, W. H., PINGOUD, V. & NORDHEIM, A. 1993. Activation of ternary complex factor Elk-1 by MAP kinases. EMBO J, 12 , 5097-104. JARRIAULT, S., BROU, C., LOGEAT, F., SCHROETER, E. H., KOPAN, R. & ISRAEL, A. 1995. Signalling downstream of activated mammalian Notch. Nature, 377 , 355-8. KANI, K., MALIHI, P. D., JIANG, Y., WANG, H., WANG, Y., RUDERMAN, D. L., AGUS, D. B., MALLICK, P. & GROSS, M. E. 2013. Anterior gradient 2 (AGR2): blood-based biomarker elevated in metastatic prostate cancer associated with the neuroendocrine phenotype. Prostate, 73 , 306-15. KAUFMANN, S. H., DESNOYERS, S., OTTAVIANO, Y., DAVIDSON, N. E. & POIRIER, G. G. 1993. Specific proteolytic cleavage of poly(ADP-ribose) polymerase: an early marker of chemotherapy-induced apoptosis. Cancer Res, 53 , 3976-85. KHAITAN, D. & DWARAKANATH, B. S. 2009. Endogenous and induced oxidative stress in multi-cellular tumour spheroids: implications for improving tumour therapy. Indian J Biochem Biophys, 46 , 16-24. KIM, S. J., KIM, D. H., KANG, D. & KIM, J. H. 2014. Expression of anterior gradient 2 is decreased with the progression of human biliary tract cancer. Tohoku J Exp Med, 234 , 83-8. KING, E. R., TUNG, C. S., TSANG, Y. T., ZU, Z., LOK, G. T., DEAVERS, M. T., MALPICA, A., WOLF, J. K., LU, K. H., BIRRER, M. J., MOK, S. C., GERSHENSON, D. M. & WONG, K. K. 2011. The anterior gradient homolog 3 (AGR3) gene is associated with differentiation and survival in ovarian cancer. Am J Surg Pathol, 35 , 904-12. KITISIN, K., SAHA, T., BLAKE, T., GOLESTANEH, N., DENG, M., KIM, C., TANG, Y., SHETTY, K., MISHRA, B. & MISHRA, L. 2007. Tgf-Beta signaling in development. Sci STKE, 2007 , cm1. KLIJN, C., DURINCK, S., STAWISKI, E. W., HAVERTY, P. M., JIANG, Z., LIU, H., DEGENHARDT, J., MAYBA, O., GNAD, F., LIU, J., PAU, G., REEDER, J., CAO, Y., MUKHYALA, K., SELVARAJ, S. K., YU, M., ZYNDA, G. J., BRAUER, M. J., WU, T. D., GENTLEMAN, R. C., MANNING, G., YAUCH, R. L., BOURGON, R., STOKOE, D., MODRUSAN, Z., NEVE, R. M., DE SAUVAGE, F. J., SETTLEMAN, J., SESHAGIRI, S. & ZHANG, Z. 2015. A comprehensive transcriptional portrait of human cancer cell lines. Nat Biotechnol, 33 , 306-12. KOUL, H. K., PAL, M. & KOUL, S. 2013. Role of p38 MAP Kinase Signal Transduction in Solid Tumors. Genes Cancer, 4, 342-59.

KOZLOV, G., MAATTANEN, P., THOMAS, D. Y. & GEHRING, K. 2010. A structural overview of the PDI family of proteins. FEBS J, 277 , 3924-36. KUMAR, A., DELGADO, J. P., GATES, P. B., NEVILLE, G., FORGE, A. & BROCKES, J. P. 2011. The aneurogenic limb identifies developmental cell interactions underlying vertebrate limb regeneration. Proc Natl Acad Sci U S A, 108 , 13588-93. KUMAR, A., GODWIN, J. W., GATES, P. B., GARZA-GARCIA, A. A. & BROCKES, J. P. 2007. Molecular basis for the nerve dependence of limb regeneration in an adult vertebrate. Science, 318 , 772-7. LACAMBRA, M. D., TSANG, J. Y., NI, Y. B., CHAN, S. K., TAN, P. H. & TSE, G. M. 2015. Anterior Gradient 2 is a Poor Outcome Indicator in Luminal Breast Cancer. Ann Surg Oncol . LAI, E. C. 2004. Notch signaling: control of cell communication and cell fate. Development, 131 , 965-73. LEPREUX, S., BIOULAC-SAGE, P. & CHEVET, E. 2011. Differential expression of the anterior gradient protein-2 is a conserved feature during morphogenesis and carcinogenesis of the biliary tree. Liver Int, 31 , 322-8. LEWIS-WAMBI, J. S., CUNLIFFE, H. E., KIM, H. R., WILLIS, A. L. & JORDAN, V. C. 2008. Overexpression of CEACAM6 promotes migration and invasion of oestrogen- deprived breast cancer cells. Eur J Cancer, 44 , 1770-9. LI, Z., WU, Z., CHEN, H., ZHU, Q., GAO, G., HU, L., NEGI, H., KAMLE, S. & LI, D. 2015a. Induction of anterior gradient 2 (AGR2) plays a key role in insulin-like growth factor-1 (IGF-1)-induced breast cancer cell proliferation and migration. Med Oncol, 32 , 577. LI, Z., ZHU, Q., HU, L., CHEN, H., WU, Z. & LI, D. 2015b. Anterior gradient 2 is a binding stabilizer of HIF-1alpha that enhances CoCl -induced doxorubicin resistance in breast cancer cells. Cancer Sci . LIU, D., RUDLAND, P. S., SIBSON, D. R., PLATT-HIGGINS, A. & BARRACLOUGH, R. 2005. Human homologue of cement gland protein, a novel metastasis inducer associated with breast carcinomas. Cancer Res, 65 , 3796-805. MA, S. R., WANG, W. M., HUANG, C. F., ZHANG, W. F. & SUN, Z. J. 2015. Anterior gradient protein 2 expression in high grade head and neck squamous cell carcinoma correlated with cancer stem cell and epithelial mesenchymal transition. Oncotarget, 6, 8807-21. MASLON, M. M., HRSTKA, R., VOJTESEK, B. & HUPP, T. R. 2010. A divergent substrate-binding loop within the pro-oncogenic protein anterior gradient-2 forms a docking site for Reptin. J Mol Biol, 404 , 418-38. MI, H., MURUGANUJAN, A. & THOMAS, P. D. 2013. PANTHER in 2013: modeling the evolution of gene function, and other gene attributes, in the context of phylogenetic trees. Nucleic Acids Res, 41 , D377-86. MORBINI, P., CAPELLO, G. L., ALBERIZZI, P., BENAZZO, M., PAGLINO, C., COMOLI, P. & PEDRAZZOLI, P. 2015. Markers of squamocolumnar junction cells in normal tonsils and oropharyngeal cancer with and without HPV infection. Histol Histopathol, 30 , 833-9. MYUNG, J. K., FRISCHER, T., AFJEHI-SADAT, L., POLLAK, A. & LUBEC, G. 2008. Mass spectrometrical analysis of the processed metastasis-inducing anterior gradient protein 2 homolog reveals 100% sequence coverage. Amino Acids, 35 , 485-94. NARUMI, S., MIKI, Y., HATA, S., EBINA, M., SAITO, M., MORI, K., KOBAYASHI, M., SUZUKI, T., IWABUCHI, E., SATO, I., MAEMONDO, M., ENDO, C., INOUE, A., KONDO, T., YAMADA-OKABE, H., ICHINOSE, M. & SASANO, H. 2015.

Anterior gradient 2 is correlated with EGFR mutation in lung adenocarcinoma tissues. Int J Biol Markers, 30 , e234-42. NEEB, A., HEFELE, S., BORMANN, S., PARSON, W., ADAMS, F., WOLF, P., MIERNIK, A., SCHOENTHALER, M., KROENIG, M., WILHELM, K., SCHULTZE- SEEMANN, W., NESTEL, S., SCHAEFER, G., BU, H., KLOCKER, H., NAZARENKO, I. & CATO, A. C. 2014. Splice variant transcripts of the anterior gradient 2 gene as a marker of prostate cancer. Oncotarget, 5, 8681-9. NI, M. & LEE, A. S. 2007. ER chaperones in mammalian development and human diseases. FEBS Lett, 581 , 3641-51. NORRIS, A. M., GORE, A., BALBONI, A., YOUNG, A., LONGNECKER, D. S. & KORC, M. 2013. AGR2 is a SMAD4-suppressible gene that modulates MUC1 levels and promotes the initiation and progression of pancreatic intraepithelial neoplasia. Oncogene, 32 , 3867-76. NOVOSELOV, V. V., ALEXANDROVA, E. M., ERMAKOVA, G. V. & ZARAISKY, A. G. 2003. Expression zones of three novel genes abut the developing anterior neural plate of Xenopus embryo. Gene Expr Patterns, 3, 225-30. OBACZ, J., BRYCHTOVA, V., PODHOREC, J., FABIAN, P., DOBES, P., VOJTESEK, B. & HRSTKA, R. 2015a. Anterior gradient protein 3 is associated with less aggressive tumors and better outcome of breast cancer patients. Onco Targets Ther, 8, 1523-32. OBACZ, J., PASTOREKOVA, S., VOJTESEK, B. & HRSTKA, R. 2013. Cross-talk between HIF and p53 as mediators of molecular responses to physiological and genotoxic stresses. Mol Cancer, 12 , 93. OBACZ, J., TAKACOVA, M., BRYCHTOVA, V., DOBES, P., PASTOREKOVA, S., VOJTESEK, B. & HRSTKA, R. 2015b. The role of AGR2 and AGR3 in cancer: similar but not identical. Eur J Cell Biol, 94 , 139-47. PAOLI, P., GIANNONI, E. & CHIARUGI, P. 2013. Anoikis molecular pathways and its role in cancer progression. Biochim Biophys Acta, 1833 , 3481-98. PARK, K., CHUNG, Y. J., SO, H., KIM, K., PARK, J., OH, M., JO, M., CHOI, K., LEE, E. J., CHOI, Y. L., SONG, S. Y., BAE, D. S., KIM, B. G. & LEE, J. H. 2011. AGR2, a mucinous ovarian cancer marker, promotes cell proliferation and migration. Exp Mol Med, 43 , 91-100. PARK, S. W., ZHEN, G., VERHAEGHE, C., NAKAGAMI, Y., NGUYENVU, L. T., BARCZAK, A. J., KILLEEN, N. & ERLE, D. J. 2009. The protein disulfide isomerase AGR2 is essential for production of intestinal mucus. Proc Natl Acad Sci U S A, 106 , 6950-5. PATEL, P., CLARKE, C., BARRACLOUGH, D. L., JOWITT, T. A., RUDLAND, P. S., BARRACLOUGH, R. & LIAN, L. Y. 2013. Metastasis-promoting anterior gradient 2 protein has a dimeric thioredoxin fold structure and a role in cell adhesion. J Mol Biol, 425 , 929-43. PERSSON, S., ROSENQUIST, M., KNOBLACH, B., KHOSRAVI-FAR, R., SOMMARIN, M. & MICHALAK, M. 2005. Diversity of the protein disulfide isomerase family: identification of breast tumor induced Hag2 and Hag3 as novel members of the protein family. Mol Phylogenet Evol, 36 , 734-40. PETEK, E., WINDPASSINGER, C., EGGER, H., KROISEL, P. M. & WAGNER, K. 2000. Localization of the human anterior gradient-2 gene (AGR2) to chromosome band 7p21.3 by radiation hybrid mapping and fluorescencein situ hybridisation. Cytogenet Cell Genet, 89 , 141-2. PIZZI, M., FASSAN, M., BALISTRERI, M., GALLIGIONI, A., REA, F. & RUGGE, M. 2012a. Anterior gradient 2 overexpression in lung adenocarcinoma. Appl Immunohistochem Mol Morphol, 20 , 31-6.

PIZZI, M., FASSAN, M., REALDON, S., BALISTRERI, M., BATTAGLIA, G., GIACOMETTI, C., ZANINOTTO, G., ZAGONEL, V., DE BONI, M. & RUGGE, M. 2012b. Anterior gradient 2 profiling in Barrett columnar epithelia and adenocarcinoma. Hum Pathol, 43 , 1839-44. POHLER, E., CRAIG, A. L., COTTON, J., LAWRIE, L., DILLON, J. F., ROSS, P., KERNOHAN, N. & HUPP, T. R. 2004. The Barrett's antigen anterior gradient-2 silences the p53 transcriptional response to DNA damage. Mol Cell Proteomics, 3, 534-47. QUANDT, K., FRECH, K., KARAS, H., WINGENDER, E. & WERNER, T. 1995. MatInd and MatInspector: new fast and versatile tools for detection of consensus matches in nucleotide sequence data. Nucleic Acids Res, 23 , 4878-84. RAETZ, C. R., ULEVITCH, R. J., WRIGHT, S. D., SIBLEY, C. H., DING, A. & NATHAN, C. F. 1991. Gram-negative endotoxin: an extraordinary lipid with profound effects on eukaryotic signal transduction. FASEB J, 5, 2652-60. RAMACHANDRAN, V., ARUMUGAM, T., WANG, H. & LOGSDON, C. D. 2008. Anterior gradient 2 is expressed and secreted during the development of pancreatic cancer and promotes cancer cell survival. Cancer Res, 68 , 7811-8. RAPOPORT, T. A. 1991. Protein transport across the endoplasmic reticulum membrane: facts, models, mysteries. FASEB J, 5, 2792-8. RAWLINGS, J. S., ROSLER, K. M. & HARRISON, D. A. 2004. The JAK/STAT signaling pathway. J Cell Sci, 117 , 1281-3. REINHARDT, H. C. & SCHUMACHER, B. 2012. The p53 network: cellular and systemic DNA damage responses in aging and cancer. Trends Genet, 28 , 128-36. RIENER, M. O., THIESLER, T., HELLERBRAND, C., AMANN, T., CATHOMAS, G., FRITZSCHE, F. R., DAHL, E., BAHRA, M., WEICHERT, W., TERRACCIANO, L. & KRISTIANSEN, G. 2014. Loss of anterior gradient-2 expression is an independent prognostic factor in colorectal carcinomas. Eur J Cancer, 50 , 1722-30. RYU, J., PARK, S. G., LEE, P. Y., CHO, S., LEE DO, H., KIM, G. H., KIM, J. H. & PARK, B. C. 2013. Dimerization of pro-oncogenic protein Anterior Gradient 2 is required for the interaction with BiP/GRP78. Biochem Biophys Res Commun, 430 , 610-5. SALMANS, M. L., ZHAO, F. & ANDERSEN, B. 2013. The estrogen-regulated anterior gradient 2 (AGR2) protein in breast cancer: a potential drug target and biomarker. Breast Cancer Res, 15 , 204. SHAO, H., GU, Y., DING, J., LU, P., RUAN, T. & LU, W. 2015. HEPACAM inhibited the growth and migration of cancer cells in the progression of non-small cell lung cancer. Tumour Biol . SHI, T., GAO, Y., QUEK, S. I., FILLMORE, T. L., NICORA, C. D., SU, D., ZHAO, R., KAGAN, J., SRIVASTAVA, S., RODLAND, K. D., LIU, T., SMITH, R. D., CHAN, D. W., CAMP, D. G., 2ND, LIU, A. Y. & QIAN, W. J. 2014. A highly sensitive targeted mass spectrometric assay for quantification of AGR2 protein in human urine and serum. J Proteome Res, 13 , 875-82. SHIH, L. J., LU, Y. F., CHEN, Y. H., LIN, C. C., CHEN, J. A. & HWANG, S. P. 2007. Characterization of the agr2 gene, a homologue of X. laevis anterior gradient 2, from the zebrafish, Danio rerio. Gene Expr Patterns, 7, 452-60. SCHIER, A. F. & SHEN, M. M. 2000. Nodal signalling in vertebrate development. Nature, 403 , 385-9. SCHROEDER, B. W., VERHAEGHE, C., PARK, S. W., NGUYENVU, L. T., HUANG, X., ZHEN, G. & ERLE, D. J. 2012. AGR2 is induced in asthma and promotes allergen- induced mucin overproduction. Am J Respir Cell Mol Biol, 47 , 178-85.

SIDIROPOULOS, K. G., WHITE, N. M., BUI, A., DING, Q., BOULOS, P., PAMPALAKIS, G., KHELLA, H., SAMUEL, J. N., SOTIROPOULOU, G. & YOUSEF, G. M. 2014. Kallikrein-related peptidase 5 induces miRNA-mediated anti-oncogenic pathways in breast cancer. Oncoscience, 1, 709-24. SIVE, H. & BRADLEY, L. 1996. A sticky problem: the Xenopus cement gland as a paradigm for anteroposterior patterning. Dev Dyn, 205 , 265-80. SIVE, H. L., HATTORI, K. & WEINTRAUB, H. 1989. Progressive determination during formation of the anteroposterior axis in Xenopus laevis. Cell, 58 , 171-80. SMIRNOV, D. A., ZWEITZIG, D. R., FOULK, B. W., MILLER, M. C., DOYLE, G. V., PIENTA, K. J., MEROPOL, N. J., WEINER, L. M., COHEN, S. J., MORENO, J. G., CONNELLY, M. C., TERSTAPPEN, L. W. & O'HARA, S. M. 2005. Global gene expression profiling of circulating tumor cells. Cancer Res, 65 , 4993-7. SNEL, B., LEHMANN, G., BORK, P. & HUYNEN, M. A. 2000. STRING: a web-server to retrieve and display the repeatedly occurring neighbourhood of a gene. Nucleic Acids Res, 28 , 3442-4. SOBIN, L. H. & FLEMING, I. D. 1997. TNM Classification of Malignant Tumors, fifth edition (1997). Union Internationale Contre le Cancer and the American Joint Committee on Cancer. Cancer, 80 , 1803-4. SUI, X., KONG, N., YE, L., HAN, W., ZHOU, J., ZHANG, Q., HE, C. & PAN, H. 2014. p38 and JNK MAPK pathways control the balance of apoptosis and autophagy in response to chemotherapeutic agents. Cancer Lett, 344 , 174-9. SWEENY, L., LIU, Z., BUSH, B. D., HARTMAN, Y., ZHOU, T. & ROSENTHAL, E. L. 2012. CD147 and AGR2 expression promote cellular proliferation and metastasis of head and neck squamous cell carcinoma. Exp Cell Res, 318 , 1788-98. SWISSHELM, K., MACEK, R. & KUBBIES, M. 2005. Role of claudins in tumorigenesis. Adv Drug Deliv Rev, 57 , 919-28. TAISEER, I. R., SAMAR, A. R. M. & ABDELMONEM, H. A. 2014. Immunohistochemical Expression of Aldehyde Dehydrogenase-1 and Hypoxia- Inducible Factor-1α in Breast Cancer. IJAR, 2, 822-830. TAVASSOLI FA. DEVILEE P, E. 2003. WHO classification of tumours. Pathology and genetics of tumours of the breast and female genital organs. Lyon (France): IARC Press. TERESHINA, M. B., ERMAKOVA, G. V., IVANOVA, A. S. & ZARAISKY, A. G. 2014. Ras-dva1 small GTPase regulates telencephalon development in Xenopus laevis embryos by controlling Fgf8 and Agr signaling at the anterior border of the neural plate. Biol Open, 3, 192-203. THIERY, J. P., ACLOQUE, H., HUANG, R. Y. & NIETO, M. A. 2009. Epithelial- mesenchymal transitions in development and disease. Cell, 139 , 871-90. THOMPSON, D. A. & WEIGEL, R. J. 1998. hAG-2, the human homologue of the Xenopus laevis cement gland gene XAG-2, is coexpressed with estrogen receptor in breast cancer cell lines. Biochem Biophys Res Commun, 251 , 111-6. TRCKA, F., DURECH, M., MAN, P., HERNYCHOVA, L., MULLER, P. & VOJTESEK, B. 2014. The assembly and intermolecular properties of the Hsp70-Tomm34-Hsp90 molecular chaperone complex. J Biol Chem, 289 , 9887-901. TROPEA, J. E., CHERRY, S. & WAUGH, D. S. 2009. Expression and purification of soluble His(6)-tagged TEV protease. Methods Mol Biol, 498 , 297-307. TSUJI, T., SATOYOSHI, R., AIBA, N., KUBO, T., YANAGIHARA, K., MAEDA, D., GOTO, A., ISHIKAWA, K., YASHIRO, M. & TANAKA, M. 2015. Agr2 mediates paracrine effects on stromal fibroblasts that promote invasion by gastric signet-ring carcinoma cells. Cancer Res, 75 , 356-66.

TURANO, C., COPPARI, S., ALTIERI, F. & FERRARO, A. 2002. Proteins of the PDI family: unpredicted non-ER locations and functions. J Cell Physiol, 193 , 154-63. VAARALA, M. H., HIRVIKOSKI, P., KAUPPILA, S. & PAAVONEN, T. K. 2012. Identification of androgen-regulated genes in human prostate. Mol Med Rep, 6, 466- 72. VALLADARES-AYERBES, M., BLANCO-CALVO, M., REBOREDO, M., LORENZO- PATINO, M. J., IGLESIAS-DIAZ, P., HAZ, M., DIAZ-PRADO, S., MEDINA, V., SANTAMARINA, I., PERTEGA, S., FIGUEROA, A. & ANTON-APARICIO, L. M. 2012. Evaluation of the adenocarcinoma-associated gene AGR2 and the intestinal stem cell marker LGR5 as biomarkers in colorectal cancer. Int J Mol Sci, 13 , 4367-87. VANDERLAAG, K. E., HUDAK, S., BALD, L., FAYADAT-DILMAN, L., SATHE, M., GREIN, J. & JANATPOUR, M. J. 2010. Anterior gradient-2 plays a critical role in breast cancer cell growth and survival by modulating cyclin D1, estrogen receptor- alpha and survivin. Breast Cancer Res, 12 , R32. VERMA, S., SALMANS, M. L., GEYFMAN, M., WANG, H., YU, Z., LU, Z., ZHAO, F., LIPKIN, S. M. & ANDERSEN, B. 2012. The estrogen-responsive Agr2 gene regulates mammary epithelial proliferation and facilitates lobuloalveolar development. Dev Biol, 369 , 249-60. VIVEKANANDAN, P., MICCHELLI, S. T. & TORBENSON, M. 2009. Anterior gradient-2 is overexpressed by fibrolamellar carcinomas. Hum Pathol, 40 , 293-9. WANG, D. H., TIWARI, A., KIM, M. E., CLEMONS, N. J., REGMI, N. L., HODGES, W. A., BERMAN, D. M., MONTGOMERY, E. A., WATKINS, D. N., ZHANG, X., ZHANG, Q., JIE, C., SPECHLER, S. J. & SOUZA, R. F. 2014. Hedgehog signaling regulates FOXA2 in esophageal embryogenesis and Barrett's metaplasia. J Clin Invest, 124 , 3767-80. WANG, Z., HAO, Y. & LOWE, A. W. 2008. The adenocarcinoma-associated antigen, AGR2, promotes tumor growth, cell migration, and cellular transformation. Cancer Res, 68 , 492-7. WAYNER, E. A., QUEK, S. I., AHMAD, R., HO, M. E., LOPRIENO, M. A., ZHOU, Y., ELLIS, W. J., TRUE, L. D. & LIU, A. Y. 2012. Development of an ELISA to detect the secreted prostate cancer biomarker AGR2 in voided urine. Prostate, 72 , 1023-34. WELBOREN, W. J., VAN DRIEL, M. A., JANSSEN-MEGENS, E. M., VAN HEERINGEN, S. J., SWEEP, F. C., SPAN, P. N. & STUNNENBERG, H. G. 2009. ChIP-Seq of ERalpha and RNA polymerase II defines genes differentially responding to ligands. EMBO J, 28 , 1418-28. WILSON, C. L., SIMS, A. H., HOWELL, A., MILLER, C. J. & CLARKE, R. B. 2006. Effects of oestrogen on gene expression in epithelium and stroma of normal human breast tissue. Endocr Relat Cancer, 13 , 617-28. WRIGHT, T. M., WARDELL, S. E., JASPER, J. S., STICE, J. P., SAFI, R., NELSON, E. R. & MCDONNELL, D. P. 2014. Delineation of a FOXA1/ERalpha/AGR2 regulatory loop that is dysregulated in endocrine therapy-resistant breast cancer. Mol Cancer Res, 12 , 1829-39. WU, Z. S., WU, Q., DING, X. D., WANG, H. Q., SHEN, Y. X. & FANG, S. Y. 2008. [Expression of a novel metastasis-inducing protein human anterior gradient-2 (AGR2) in breast cancer and its clinical and prognostic significance]. Zhonghua Bing Li Xue Za Zhi, 37 , 109-13. XIA, J. H., JIANG, J., SHI, Y. H. & GUI, J. F. 2009. Predominant expression and cellular distribution of fish Agr2 in renal collecting system. Comp Biochem Physiol B Biochem Mol Biol, 152 , 397-404.

YU, H., ZHAO, J., LIN, L., ZHANG, Y., ZHONG, F., LIU, Y., YU, Y., SHEN, H., HAN, M., HE, F. & YANG, P. 2012. Proteomic study explores AGR2 as pro-metastatic protein in HCC. Mol Biosyst, 8, 2710-8. ZHANG, Q. L., LUO, C. L., WU, X. H., WANG, C. Y., XU, X., ZHANG, Y. Y., LIU, Q. & SHEN, S. L. 2011. HepaCAM induces G1 phase arrest and promotes c-Myc degradation in human renal cell carcinoma. J Cell Biochem, 112 , 2910-9. ZHANG, Y., ALI, T. Z., ZHOU, H., D'SOUZA, D. R., LU, Y., JAFFE, J., LIU, Z., PASSANITI, A. & HAMBURGER, A. W. 2010. ErbB3 binding protein 1 represses metastasis-promoting gene anterior gradient protein 2 in prostate cancer. Cancer Res, 70 , 240-8. ZHANG, Y., FOROOTAN, S. S., LIU, D., BARRACLOUGH, R., FOSTER, C. S., RUDLAND, P. S. & KE, Y. 2007. Increased expression of anterior gradient-2 is significantly associated with poor survival of prostate cancer patients. Prostate Cancer Prostatic Dis, 10 , 293-300. ZHAO, F., EDWARDS, R., DIZON, D., AFRASIABI, K., MASTROIANNI, J. R., GEYFMAN, M., OUELLETTE, A. J., ANDERSEN, B. & LIPKIN, S. M. 2010. Disruption of Paneth and goblet cell homeostasis and increased endoplasmic reticulum stress in Agr2-/- mice. Dev Biol, 338 , 270-9. ZHENG, W., ROSENSTIEL, P., HUSE, K., SINA, C., VALENTONYTE, R., MAH, N., ZEITLMANN, L., GROSSE, J., RUF, N., NURNBERG, P., COSTELLO, C. M., ONNIE, C., MATHEW, C., PLATZER, M., SCHREIBER, S. & HAMPE, J. 2006. Evaluation of AGR2 and AGR3 as candidate genes for inflammatory bowel disease. Genes Immun, 7, 11-8. ZWEITZIG, D. R., SMIRNOV, D. A., CONNELLY, M. C., TERSTAPPEN, L. W., O'HARA, S. M. & MORAN, E. 2007. Physiological stress induces the metastasis marker AGR2 in breast cancer cells. Mol Cell Biochem, 306 , 255-60.

8 LIST OF FIGURES

Figure 1. Sequence alignment of AGR2 and AGR3 proteins...... 12 Figure 2. Structure of AGR2 and AGR3 proteins ...... 14 Figure 3. Schematic overview of AGR2 signalling ...... 19 Figure 4. Overview of AGR proteins distribution in normal and cancer tissue ...... 33 Figure 5. Nucleotide sequence of AGR2 and AGR3 promoter with representative potential binding sites for transcription factors ...... 36 Figure 6. Immunohistochemical staining for AGR3 ...... 38 Figure 7. Association of immunohistochemical staining for AGR3 with patient survival ...... 40 Figure 8. AGR3 and AGR2 expression in a panel of breast cancer cell lines ...... 42 Figure 9. AGR3 and AGR2 regulate breast cancer cell adhesion ...... 44 Figure 10. Extracellular AGR3 promotes migration of breast cancer cells...... 45 Figure 11. Extracellular AGR proteins have no effect on breast cancer cell proliferation ...... 46 Figure 12. AGR3 regulate different signalling pathways ...... 48 Figure 13. STRING analysis of AGR3-deregulated kinases ...... 49 Figure 14. AGR3-AGR2 cross-talk on the level of transcriptional regulation ...... 50 Figure 15. AGR proteins form heterodimers in breast cancer cells ...... 51 Figure 16. Stable AGR3 expression in H1299 cell line ...... 51 Figure 17. Extracellular AGR3 downregulates p53 activation in DDR ...... 52 Figure 18. AGR3 significantly attenuates tumour cell response to genotoxic stress ...... 53 Figure 19. The effect of AGR3 on cancer cell growth and survival in spheroid model...... 54 Figure 20. Identification of the cell death signalling affected by AGR3 expression ...... 55 Figure 21. AGR3 diminishes p53 activation both in normoxic and hypoxic conditions ...... 55 Figure 22. Overview of AGR3 functions in cancer ...... 59 Figure 23 Mutual relations between AGR3 and AGR2 on the level of their regulation ...... 62

9 LIST OF TABLES

Table 1. Human PDI family ...... 13 Table 2. Significance of AGR2 and AGR3 expression in various carcinomas ...... 21 Table 3. Primary antibodies used for immunoblotting ...... 27 Table 4. List of primers used for qPCR ...... 29 Table 5. AGR3 co-expressed genes ...... 34 Table 6. Clinicopathological characteristics of primary breast carcinomas ...... 36 Table 7. Association of immunohistochemical staining for AGR3 with other tumour variables ...... 38 Table 8. Independent prognostic factors for the analysed set of tumours according to Cox’s multivariate survival analyses ...... 40 Table 9. Survival analysis for patients with AGR3-expressing tumours ...... 41 Table 10. qPCR analysis of AGR-dependent changes in the expression of CAMs ...... 44 Table 11. AGR3-dependent changes in key cellular phosphoproteins analysed by human phospho-kinase array...... 46

10 LIST OF ABBREVIATIONS

AGR anterior gradient protein family AGR2 human anterior gradient protein 2 AGR2 gene coding for human anterior gradient protein 2 AGR3 human anterior gradient protein 3 AGR3 gene coding for human anterior gradient protein 3 AGRs anterior gradient proteins AMPK AMP-activated protein kinase ANB anterior neural plate border AR androgen receptor CAMs cell adhesion molecules CEACAM6 carcinoembryonic antigen-related cell adhesion molecule 6 CLDN claudin co-IP co-immunoprecipitation DMSO dimethyl sulfoxide DSP dithiobis(succinimidylpropionate) DSS disuccinimidyl suberate DTT dithiothreitol ECM extracellular matrix EGFR epidermal growth factor receptor EMT epithelial-to-mesenchymal transition ER endoplasmic reticulum EsR oestrogen receptor FGF fibroblast growth factor FOXA forkhead box transcription factor HEPACAM hepatocyte cell adhesion molecule HIF-1 hypoxia-inducible factor 1 HNSCC head and neck squamous cell carcinoma hrs hours ICC intrahepatic cholangiocarcinoma IGF-1 insulin-like growth factor 1 IPA Ingenuity Pathway Analysis JAK/STAT Janus kinase-signal transducer and activator of transcription KLK5 kallikrein-related peptidase 5 LDS dodecyl sulphate MAPK mitogen-activated protein kinase min minutes MUC mucin nAG newt anterior gradient protein OS overall survival

PAK6 p21-activated kinase 6 PARP poly(ADP-ribose) polymerase PDI protein disulphide isomerise family PFS progression free survival PR progesterone receptor PFS progression free survival PSA prostate specific antigen qPCR quantitative polymerase chain reaction rAGR recombinant anterior gradient protein STAT signal transducer and activator of transcription TF transcription factor TGF-β transforming growth factor β wt wild-type X. laexis Xenopus laexis

11 ATTACHMENTS

11.1 Attachment 1

OBACZ, J., TAKACOVA, M., BRYCHTOVA, V., DOBES, P., PASTOREKOVA, S., VOJTESEK, B. & HRSTKA, R. 2015b. The role of AGR2 and AGR3 in cancer: similar but not identical. Eur J Cell Biol, 94 , 139-47.

G Model

EJCB-50808; No. of Pages 9 ARTICLE IN PRESS

European Journal of Cell Biology xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

European Journal of Cell Biology

jo urnal homepage: www.elsevier.com/locate/ejcb

Mini Review

The role of AGR2 and AGR3 in cancer: Similar but not identical

a,b a,b a a

Joanna Obacz , Martina Takacova , Veronika Brychtova , Petr Dobes ,

a,b a a,∗

Silvia Pastorekova , Borivoj Vojtesek , Roman Hrstka

Masaryk Memorial Cancer Institute, RECAMO, Zluty kopec 7, 65653 Brno, Czech Republic

Department of Molecular Medicine, Institute of Virology, Slovak Academy of Sciences, Dubravska cesta 9, 84505 Bratislava, Slovak Republic

a r t i c l e i n f o a b s t r a c t

Article history: In the past decades, highly related members of the protein disulphide isomerase family, anterior gradient

Received 12 December 2014

protein AGR2 and AGR3, attracted researchers’ attention due to their putative involvement in develop-

Received in revised form 8 January 2015

mental processes and carcinogenesis. While AGR2 has been widely demonstrated as a metastasis-related

Accepted 9 January 2015

protein whose elevated expression predicts worse patient outcome, little is known about AGR3’s role

in tumour biology. Thus, we aim to confront the issue of AGR3 function in physiology and pathology

Keywords:

in the following review by comparing this protein with the better-described homologue AGR2. Rely-

AGR2

ing on available data and in silico analyses, we show that AGR proteins are co-expressed or uncoupled in

AGR3

context-dependent manners in diverse carcinomas and healthy tissues. Further, we discuss plausible roles

PDI family

Cancer of both proteins in tumour-associated processes such as differentiation, proliferation, migration, inva-

sion and metastasis. This work brings new hints and stimulates further thoughts on hitherto unresolved

conundrum of anterior gradient protein function.

Introduction functions in vertebrates. The ﬁrst identiﬁed was a secreted protein

XAG-2, originally discovered in Xenopus laevis and implicated in the

Anterior gradient (AGR) proteins form an evolutionarily broad formation of the anteroposterior axis during embryogenesis (Sive

family with prominent, however, poorly understood physiological et al., 1989). AGR2 extracellular role was also documented dur-

ing limb regeneration in salamanders by stimulation of blastemal

growth (Kumar et al., 2007). In addition, it was recently demon-

strated that AGR proteins in X. laevis promote regeneration of

Abbreviations: AA, amino acids; ALS, familial amyotrophic lateral sclerosis;

AGR, anterior gradient; AhR, aryl hydrocarbon receptor; AR, androgen receptor; hindlimb buds and tails of tadpoles (Ivanova et al., 2013). Dur-

CAPN8, calpain 8; CAPN9, calpain 9; CDDP, cisplatin; CFTR, cystic ﬁbrosis trans- ing the last decades, intense research has commenced in order to

membrane conductance regulator; CLDN3, claudin 3; DAG-1, alpha-dystroglycan;

elucidate biological function of human homologues, namely AGR2

EGFR, epidermal growth factor receptor; ELK1, ETS-domain containing protein; EMT,

and AGR3 both in health and disease. Strikingly, although both

epithelial-to-mesenchymal transition; ER, endoplasmic reticulum; ERAD, endoplas-

molecules share 71% sequence identity and lie adjacent to one

mic reticulum-associated degradation; EsR, oestrogen receptor; FOX, forkhead box;

GC-SBE, glycine-rich SMAD binding elements; HCC, hepatocellular carcinomas; another at chromosomal position 7p21 (Fletcher et al., 2003; Petek

HGSC, high-grade serous ovarian carcinomas; ICC, intrahepatic cholangiocarcino-

et al., 2000), AGR2, but not AGR3, is a dominant factor identiﬁed in

mas; JAK-STAT, Janus kinase-signal transducer and activator of transcription; LEF-1,

many OMICS screens, and therefore, many more reports have been

lymphoid enhancer factor-1; LGSC, low-grade serous ovarian carcinomas; MAPK,

published in relation to its characterization. Thus, relying also on

mitogen-activated protein kinases; MICA, MHC I-related chain A; MUCs, mucins;

NK, natural killer; PDB, Protein Data Bank; PDIs, protein disulphide isomerases; our recent observation of uncoupled AGR2 and AGR3 expression in

␤

SLC44A4, protein member 4 of solute carrier family 44; TGF- , transforming growth various tumour tissues (unpublished data), we sought to compare

factor-beta; TGIF1, TG-interacting factor 1; TM4SF, transmembrane 4 superfamily;

both proteins by analysing their expression pattern and regulatory

TMC4, transmembrane channel-like protein 4; TMC5, transmembrane channel-like

mechanisms in this review. For this reason, we have reviewed all

protein 5; TMPRSS2, transmembrane protease serine 2; TRX, thioredoxin; TSPAN1,

the published data and used the Genevestigator tool (Hruz et al.,

tetraspanin 1; TSS, transcription start site; UPR, unfolded protein response; ZEB1,

zinc-ﬁnger enhancer binding-1. 2008), which enables not only prediction of AGRs tissue distribu-

∗

Corresponding author. Tel.: +420 543 133 306. tions but also their co-expressed partners. Moreover, we performed

E-mail addresses: [email protected] (J. Obacz), [email protected]

in silico promoter analysis to ﬁnd upstream regulatory pathways

(M. Takacova), [email protected] (V. Brychtova), [email protected] (P. Dobes),

potentially triggering AGR2 and AGR3 expression. All the analyses

[email protected] (S. Pastorekova), [email protected] (B. Vojtesek), [email protected]

were conducted on the data available online in May 2014. (R. Hrstka).

http://dx.doi.org/10.1016/j.ejcb.2015.01.002

Please cite this article in press as: Obacz, J., et al., The role of AGR2 and AGR3 in cancer: Similar but not identical. Eur. J. Cell Biol. (2015), http://dx.doi.org/10.1016/j.ejcb.2015.01.002

G Model

EJCB-50808; No. of Pages 9 ARTICLE IN PRESS

2 J. Obacz et al. / European Journal of Cell Biology xxx (2015) xxx–xxx

AGR2 and AGR3 as PDI family members Role of PDI family in cell homeostasis

Structural features of the PDI family The PDI family members are implicated in a variety of disor-

ders, including neurodegenerative syndromes such as Parkinson’s,

The protein disulphide isomerase (PDI) family is part of the Alzheimer’s, Huntington’s diseases, familial amyotrophic lateral

thioredoxin (TRX) superfamily, which also includes the glutaredox- sclerosis (ALS) as well as infertility and a diverse range of malig-

ins, thioredoxins, ferroredoxins and peroxiredoxins (Jacquot et al., nancies (as reviewed by Benham, 2012). Of note is also a report

2002). Recent work has revealed that there are three subfamilies of showing decrease of AGR2 expression in ulcerative colitis, where it

AGRs: AG1, AGR2 and AGR3, all showing the highest homology to was postulated to play a crucial role in the maintenance of epithelial

non-secreted PDI of the TLP19 subfamily. Remarkably, members of integrity (Zheng et al., 2006).

AGR2 and AGR3 subfamilies are present in amniotes, while mem- The main function of the PDI family members is to form/disrupt,

bers of AG1 subfamily (to which family founder, XAG-2 belongs) are oxidize/reduce and isomerize the disulphide bonds between the

restricted to lower vertebrates (Ivanova et al., 2013). Human AGR2 cysteine residues of nascent proteins in the lumen of the endo-

and AGR3’s afﬁliation to the PDI family was based on the phylo- plasmic reticulum (ER) in order to provide their proper folding and

genetic analysis performed by Persson et al. (2005), where it was maturation prior to the release for cellular transport (Hatahet and

shown that both proteins have high homology to ERp18/19 pro- Ruddock, 2009). Thus, they play a pivotal role in the maintenance of

tein, also denoted as AGR1. To date, 21 members varying in size, cellular homeostasis. They can also serve as molecular chaperones

structure, tissue distribution and enzymatic activity were identi- involved in the ER-associated degradation (ERAD) mechanisms that

ﬁed (Galligan and Petersen, 2012). The common feature of all the lead to protein removal (Ni and Lee, 2007). Apart from their ubiq-

PDI members is the presence of at least one domain with struc- uitous expression in ER lumen, PDIs can be found in other cellular

tural similarity to TRX, which can be either active or enzymatically compartments, where they are shown to regulate among others

inactive (Kozlov et al., 2010). Protein activity depends on the pres- cell adhesion, platelets activation, viral infections and protein–DNA

ence of CXXC motif, which determines the reaction with thiols of interactions (Turano et al., 2002).

newly synthesized proteins. However, some of the proteins vary Although oxidative properties of AGR2 protein have not yet been

in their active site composition, including AGR2 and AGR3 pro- validated in vitro or in vivo, some AGR2 client proteins have been

teins with CXXS motif (Galligan and Petersen, 2012); therefore, identiﬁed, supporting its role in the protein folding machinery. For

it is possible that in order to act as PDIs AGR proteins cooper- instance, AGR2 was demonstrated to regulate production of mucins

ate with other redox-active molecules. Within the family, ERp18, (MUCs), including intestinal MUC2 (Park et al., 2009), the airway

TMX, TMX2, TMX4, TMX5, AGR2 and AGR3 proteins possess a epithelial MUC5AC and MUC5B (Schroeder et al., 2012) as well as

single active domain, while ERp27, ERp29, CASQ1 and CASQ2 pro- pancreatic MUC1 (Norris et al., 2013). Additionally, by forming a

teins contain only inactive domains (Benham, 2012; Galligan and substrate loop between amino acids 104 and 111, it interacts with

Petersen, 2012). Moreover, each protein of the family is character- ATP-binding protein Reptin and consequently regulates many of its

ized by the presence of a short NH2-terminal signal peptide and the functions such as ATPase activity, ATP binding, helicase functions,

COOH-terminal endoplasmic reticulum (ER) retention sequence telomerase/Pontin binding and others (Maslon et al., 2010). Fur-

(with the exceptions of TMX1, TMX4, CASQ1 and CASQ2 that ther, AGR2-interacting proteins were identiﬁed in yeast two-hybrid

lack the ER retention sequence) (Appenzeller-Herzog and Ellgaard, screen, including neurexin 3, cytoskeleton-associated protein 2 or

2008; Galligan and Petersen, 2012). An ample insight into the Ly6/PLAUR domain-containing protein 3, linking AGR2 with cell

sequence characterization as well as structural overview of the PDI adhesion, division and migration (see review by Chevet et al., 2013).

family can be found in the following works (Appenzeller-Herzog Perturbation of ER homeostasis leads to the accumulation of

and Ellgaard, 2008; Galligan and Petersen, 2012; Kozlov et al., unfolded or mis-folded proteins, ER stress and consequently activa-

2010). tion of the unfolded protein response (UPR). UPR signalling results

The structure of the mature AGR2 protein, characterized by the either in the degradation of mis-folded proteins by upregulating

lack of the ﬁrst 20-amino acid signal peptide, was recently char- PDIs and molecular chaperones or in the attenuation of protein

acterized. It was shown that unfolded N-terminal 21–40 amino synthesis (Ron and Walter, 2007). However, if ER homeostasis can-

acid region determines adhesion properties of AGR2, whereas the not be restored, apoptotic pathways are induced (Tabas and Ron,

folded domain forms a dimer through speciﬁc intermolecular salt 2011). Using both proteomic and biochemical approaches, Higa

bridges. Moreover, in that work, the authors demonstrated that et al. identiﬁed AGR2 as one of the ER proteins that associates with

the proper topography of the dimeric structure relies on interac- membrane-bound ribosomes through nascent protein chains. They

tion between amino acids E60 and K64 (Patel et al., 2013). Relying found that AGR2 is induced upon ER stress and that its basal expres-

on available dimeric structures of AGR proteins in Protein Data sion is controlled by the IRE1␣- and ATF6␣-triggered arms of the

Bank [PDB accession numbers—AGR2: 2LNS, AGR3: 3PH9], it can be UPR. They also showed that AGR2 silencing altered the expression

concluded that AGR2 secondary structure consists of ␣-␤-␣-␣-␤-␣ of ERAD components, resulting in cell survival under stress condi-

structural motifs, whereas AGR3 consists of ␣-␤-␣-␣-␤-␣-␤-␣-␣- tion (Higa et al., 2011). Additionally, an independent study showed

␣

motifs. AGR2 and AGR3 thioredoxin domain with active site CPHS that AGR2 homo-dimerization is important for the association with

and CQYS, respectively, is situated on ␣2 helix. The main difference BiP/GRP78, a well-established chaperone involved in the cellular

between AGR2 and AGR3 structure lies in the dimerization region response to many stresses (Ryu et al., 2013).

(Fig. 1). The dimerization of AGR2 arises from interaction between

random coils (residues 45–54) and ␣1 helices (between residues

60–67), whereas AGR3 is predicted to form dimer through ran- AGR2 and AGR3 expression in tumour cells and tissues

dom coils corresponding to 32–36, 42–46 and 100–105 residues as

well as part of ␣1 helix (residues 47–53). Moreover, based on PDB- Uncoupled expression of AGR proteins in carcinomas

sum database, we suggest that AGR3 dimerization occurs through

speciﬁc hydrogen bonds between amino acids Q32 and Q46. How- Both AGR2 and AGR3 were originally found in breast cancer

ever, to verify whether AGR homologues are able to dimerize with specimens. AGR2 gene was ﬁrst described in the oestrogen recep-

each other through random coils and/or ␣1 helices warrants further tor (EsR)-positive MCF-7 cell line (Thompson and Weigel, 1998),

investigation. and AGR3 protein was identiﬁed in the membrane of breast cancer

Please cite this article in press as: Obacz, J., et al., The role of AGR2 and AGR3 in cancer: Similar but not identical. Eur. J. Cell Biol. (2015), http://dx.doi.org/10.1016/j.ejcb.2015.01.002

G Model

EJCB-50808; No. of Pages 9 ARTICLE IN PRESS

J. Obacz et al. / European Journal of Cell Biology xxx (2015) xxx–xxx 3

Fig. 1. Structure of AGR2 and AGR3 proteins (derived from Protein Data Bank, http://www.rcsb.org, accession numbers 2LNS and 3PH9, respectively). (a) AGR2 secondary

structure with alpha helices (red), beta strands (yellow), CPHS active site (grey) and dimerization domain (blue) shown; (b) ribbon representation of AGR241–175 homodimer

with secondary structures, dimerization domain and active site marked correspondingly to 1a; (c) AGR2 dimeric interface, interactions between E60 and K64, responsible for

dimer stability are delineated in blue; (d) AGR3 secondary structure with alpha helices (red), beta strands (yellow), CQYS active site (grey) and dimerization domain (blue)

shown; (e) ribbon representation of AGR331–165 homodimer with secondary structures, dimerization domain and active site marked correspondingly to 1d and (f) AGR3

dimeric interface, interactions between Q32 and Q46, with putative role in dimer stabilization are delineated in blue.

cell lines using proteomic screening (Adam et al., 2003). The latter associated with longer patient survival. On the other hand, high-

study showed that AGR3 was present in the majority of analysed grade serous ovarian carcinomas (HGSCs) showed AGR3 positivity

breast tumours, while in normal tissue, it was found only in the to a much lesser extent (King et al., 2011). Conversely, strong AGR2

epithelial lining of the colonic mucosa (Adam et al., 2003). The same expression was reported in HGSCs and was statistically signiﬁcantly

group reported concordant expression of both AGR2 and AGR3 in linked to shorter overall survival and progression-free survival

breast tumours, which positively correlated with EsR status and (Darb-Esfahani et al., 2012).

negatively with epidermal growth factor receptor (EGFR) expres- Liver cancer represents another example of malignancy in which

sion (Fletcher et al., 2003). Regulation of AGR2 protein expression AGR proteins exhibit a cognate expression pattern. Recently, our

by oestrogen was later conﬁrmed by both in vivo (Wilson et al., group has shown that intrahepatic cholangiocarcinomas (ICCs)

2006) and in vitro experiments (Hrstka et al., 2010). Furthermore, express AGR3 protein, whereas hepatocellular carcinomas (HCCs)

based on ChIP-Seq experiments, it was validated that both AGR2 and are predominantly AGR3-negative. Due to the statistically signiﬁ-

AGR3 possess a single EsR-binding site in their regulatory regions cant association between AGR3 expression and the presence of acid

(Welboren et al., 2009). mucopolysaccharide, it was postulated that together they could

AGR proteins are also involved in response to other sex serve as diagnostic markers of well-differentiated ICCs (Brychtova

hormones. Using a suppression subtractive hybridization-based et al., 2014). In agreement with this report is work demonstrat-

technique, Zhang et al. (2005) demonstrated that AGR2 is an ing AGR2 overexpression in ﬁbrolamellar carcinomas compared

androgen-inducible gene. On the other hand, comparing benign, to other HCCs (Vivekanandan et al., 2009). AGR2 expression also

malignant prostate tissue and samples obtained from prostate can- mirrors the level of differentiation of biliary tract cholangiocytes,

cer patients after surgical castration, Vaarala et al. (2012) found similar to AGR3 (Lepreux et al., 2011). AGR2 expression in other

AGR3 among other androgen-regulated genes, expression of which carcinomas is out of scope of this review and can be found else-

was highly elevated in human prostate cancer. This report is in where (Brychtova et al., 2011; Chevet et al., 2013; Salmans et al.,

agreement with recent work by Bu and co-workers in which they 2013).

depicted androgen receptor (AR)-dependent up-regulation of both The above data indicate that the mechanism of either coin-

AGR2 and AGR3 genes by androgen and oestrogen (speciﬁcally cident or separate in vivo expression of AGR proteins is a

17␤-oestrogen) in prostate cancer cell lines. Moreover, they charac- context-dependent event. Further investigation is required to ver-

terized four AR-binding sites in the distal promoter region of AGR2, ify whether similar phenomena can be observed in different types

located 5.5, 8 and 11 and 17 kb upstream of the transcription start of tumours. Indeed, recently, our group observed different expres-

site (TSS), and one strong AR-binding site in AGR3, localized to the sion patterns of both proteins in various tumours (complete data

ﬁrst intron approximately 1 kb downstream of TSS (Bu et al., 2013). not shown) including breast carcinomas, in some of which AGR

Furthermore, uncoupled expression of both AGR proteins was proteins correlated while in others were uncoupled (Fig. 2).

described in ovarian cancers. Gray et al. found that AGR3 is over-

expressed in four divergent types of ovarian cancers, including AGRs expression based on Genevestigator data

serous papillary, endometrioid, clear cell (non-mucinous types)

and mucinous carcinomas. In non-mucinous types, AGR3 expres- In order to expand current knowledge regarding expression of

sion was found to be heterogeneous, EsR-independent and not AGRs in different types of cancer, we aimed to compare their distri-

related to AGR2 expression. However, mucinous ovarian can- bution both in human healthy tissues and carcinomas (Fig. 3) using

cers showed corresponding positive staining of both AGR2 and Genevestigator platform, a tool that integrates high-quality public

AGR3 (Gray et al., 2012). In another study, AGR3 staining corre- microarray data (Hruz et al., 2008).

lated with the level of differentiation in serous borderline ovarian It is noteworthy that AGR3 mirrors AGR2 expression in many

tumours and low-grade serous ovarian carcinomas (LGSCs) and was cases, suggesting their cognate physiological function and role in

Please cite this article in press as: Obacz, J., et al., The role of AGR2 and AGR3 in cancer: Similar but not identical. Eur. J. Cell Biol. (2015), http://dx.doi.org/10.1016/j.ejcb.2015.01.002

G Model

EJCB-50808; No. of Pages 9 ARTICLE IN PRESS

4 J. Obacz et al. / European Journal of Cell Biology xxx (2015) xxx–xxx

Fig. 2. Expression patterns of AGR proteins in breast cancer. Immunohistochemical detection of AGR2 and AGR3 proteins showed discordant expression pattern in triple-

negative breast cancer. In some tumour samples, AGR2 and AGR3 proteins showed similar cytoplasmic positivity (1a, 1b) in identical tumour tissue, while in other samples,

AGR2 and AGR3 expression was uncoupled in the corresponding tissue sections (2a, 2b; 3a, 3b) (AGR2 protein was detected using HPA007912 antibody, Sigma-Aldrich, St

Louis, MO, USA; AGR3 protein was detected using in-house anti-AGR3.1 antibody (Gray et al., 2012) magniﬁcation 400×).

pathological conditions. However, they can also be expressed in adenoma (Lee et al., 2006). Moreover, according to our analyses,

a tissue-speciﬁc manner, such as normal skin or urinary bladder, expression of both AGR2 and AGR3 genes was elevated in pancreatic

where AGR3 is not present or expressed to a lesser extent, respec- and breast carcinomas in comparison to healthy tissues, which is

tively, when compared to AGR2 that is expressed in both tissues. in line with published reports (Adam et al., 2003; Brychtova et al.,

In early stages of low vertebrates development, AGR proteins are 2014).

mainly expressed in ectoderm-derived organs, while in adults they Strikingly, unlike the increased levels of AGR2 in urinary blad-

can also be found in endoderm-derived organs such as intestine der carcinoma compared to normal tissues, we found AGR3 to be

(Chen et al., 2012; Xia et al., 2009). Similarly, human AGRs are downregulated in this type of tumour. To date, only one report

predominantly distributed in endoderm-derived organs (Xia et al., describing AGR2 expression in bladder urothelial carcinoma was

2009). Hence, as expected, we found that both AGR2 and AGR3 are released, showing downregulation of this gene in tumours when

highly expressed in normal stomach, colon, female reproductive compared to healthy controls (Izquierdo et al., 2010), which is in

system and respiratory system tissues as well as the corresponding disagreement with our ﬁndings based on Genevestigator. However,

carcinomas. further investigation is required to elucidate AGRs function in blad-

In most of the cases, our ﬁndings are in concordance with the der tumours. According to Genevestigator data, AGRs expression in

literature regarding AGRs expression in ovarian, gastric and lung liver carcinomas is either elevated or downregulated, which could

carcinomas (Armes et al., 2013; Bai et al., 2011; Darb-Esfahani et al., be explained by the fact that AGRs are differentially expressed in

2012; Gray et al., 2012; King et al., 2011; Park et al., 2011; Pizzi different histological types of these tumours (Brychtova et al., 2014;

et al., 2012), with the exception of AGR2 in colon cancers, where Vivekanandan et al., 2009). Although AGR3 levels are increased

it was found among down-regulated genes when compared with in various carcinomas, as shown above, there are still a limited

number of reports regarding its role in tumours, as well as its clin-

icopathological signiﬁcance.

In normal tissues, AGR proteins could participate in the regu-

lation of the total protein load in the cell. There are some lines of

evidence to support this hypothesis. For instance, AGR2 is required

for the production of airway epithelial MUC5AC and MUC5B in

respiratory system tissues (Schroeder et al., 2012), whereas AGR2

was found to regulate cell proliferation and differentiation during

normal mammary gland development, and its maximum expres-

sion appeared during late pregnancy and lactation, where there is

the highest requirement for milk protein production (Verma et al.,

2012). Further work is warranted to identify AGRs role in the physi-

ology of other tissues, including those found in our in silico analysis,

such as prostate, stomach, colon or urinary tract.

AGR2 and AGR3 in silico promoter analysis

Fig. 3. Overview of AGR proteins distribution in normal and cancer tissues. Four-

Uncoupled expression of AGR2 and AGR3 in different carci-

step grey scale indicates expression levels of AGR2 and AGR3 in normal and

nomas has already been reported (Bu et al., 2013; Gray et al.,

corresponding tumour tissue according to data from Genevestigator (lighter shades,

2012), which indicates that their functions are required in diver-

lower AGR expression; darker shades, higher AGR expression). Liver tumour is char-

acterized by heterogeneous expression of AGR proteins (shown by light/dark boxes). gent situations. Therefore, in order to elucidate the mechanisms

Please cite this article in press as: Obacz, J., et al., The role of AGR2 and AGR3 in cancer: Similar but not identical. Eur. J. Cell Biol. (2015), http://dx.doi.org/10.1016/j.ejcb.2015.01.002

G Model

EJCB-50808; No. of Pages 9 ARTICLE IN PRESS

J. Obacz et al. / European Journal of Cell Biology xxx (2015) xxx–xxx 5

Fig. 4. Nucleotide sequence of AGR2 and AGR3 promoter with representative potential binding sites for transcription factors. Arrows indicate position of the transcription

start site determined according to MatInspector. Underlined molecules represent transcription factors which bind to the (+) DNA strand, whereas molecules without highlights

correspond to the transcription factors which bind to the (−) DNA strand. Transcription factors common for AGR2 and AGR3 are shown in black boxes.

triggering differential expression of these proteins, we aimed to is a hallmark of EMT) and inﬂuences other key cellular processes

analyse the promoter region of both AGR2 and AGR3 using MatIn- such as cell cycle control, senescence and apoptosis (Brabletz and

spector (Cartharius et al., 2005; Quandt et al., 1995) in search for Brabletz, 2010; Browne et al., 2010). AGR proteins could act as

transcription factor (TF) binding sites. TFs were selected accord- downstream effectors of ZEB1-induced metastasis and promote

ing to the matrix similarity, and only TFs with score higher than tumour cell dissemination through modulation of their adhesion

0.95 were taken into account. Many of the TFs found in the in silico properties and interaction with the extracellular matrix (Dumartin

analysis play signiﬁcant roles in developmental processes and dif- et al., 2011; Fletcher et al., 2003).

ferentiation of tissues and/or organs of diverse origins (Fig. 4 and Several pieces of evidence suggest that AGR2 is a mediator of

Table S1). This is consistent with the report demonstrating that various cancer signalling pathways including Hippo, EGFR, EsR,

anterior gradient proteins in lower vertebrates (particularly XAG- cyclin D1, Src, c-Myc, survivin, aryl hydrocarbon receptor (AhR) and

2) play a putative role in ectodermal patterning during amphibian transforming growth factor-beta (TGF-␤) (see review by Salmans

embryogenesis, triggering cement gland differentiation as well as et al., 2013). Further evidence for AGR2 involvement in TGF-␤

the expression of anterior neuronal marker genes (Aberger et al., signalling is provided by the presence of a binding site for TG-

1998). Moreover, AGRs correlation with differentiation lineages has interacting factor 1 (TGIF1) within the AGR2 promoter. TGIF1 is

been shown in liver carcinomas (Brychtova et al., 2014; Lepreux a well-documented co-repressor of this pathway (Wotton et al.,

et al., 2011). 1999) and therefore can be expected to inhibit AGR2. Moreover, in

Interestingly, most of the TFs potentially binding to AGR2 or pancreatic carcinomas, AGR2 expression is regulated by SMAD4, a

AGR3 promoters are exclusive for each protein, with the excep- downstream target of TGF-␤ signalling (Norris et al., 2013), which

tion of common FOXP1, ZEB1 (AREB6), SMARCA3, GATA4, NFAT, is in agreement with our ﬁnding of glycine-rich SMAD binding

CEBPB and NKX2.5. Forkhead box (FOX) family members, namely elements (GC-SBE) in the AGR2 promoter. Additionally, we iden-

FOXA1 and FOXA2, have been implicated in the regulation of tiﬁed a lymphoid enhancer factor-1 (LEF-1)-binding site in the

AGR2 expression both in prostate cancer and in human embryonic AGR2 promoter region, indicating a possible functional involve-

kidney-derived HEK293 cell lines (Zhang et al., 2010; Zheng et al., ment of AGR2 in Wnt/␤-catenin/LEF-1 signalling. This pathway is

2006). Recently, it has also been shown that FOXA1 drives expres- activated in many types of cancers (Reya and Clevers, 2005) and

sion of AGR2 in both tamoxifen-sensitive and -resistant breast regulates among others expression of c-Myc and cyclin D1 (He

cancer cells (Wright et al., 2014). On the other hand, FOXP1 is con- et al., 1998; Shtutman et al., 1999). Interestingly, AGR2 has been

sidered to act as a tumour suppressor due to the fact that it is also demonstrated to modulate cyclin D and c-Myc expression and

often downregulated in cancers and its loss correlates with worse in consequence inﬂuence growth and survival of breast cancer cells

patient outcome (Koon et al., 2007). Furthermore, elevated nuclear (Vanderlaag et al., 2010).

FOXP1 activity is associated with more favourable prognosis of Although AGR3’s role as a component of tumour signalling is

breast cancer patients treated with tamoxifen (Shigekawa et al., poorly understood, in silico promoter analysis revealed a binding

2011). Following this, it is plausible that FOXP1 inhibits AGR2 acti- site for the ETS-domain containing protein ELK1, which is a direct

vation, which in consequence leads to prolonged patient survival target for the mitogen-activated protein kinase (MAPK) signalling

by reduction of tumour aggressiveness. It is therefore necessary pathway (Janknecht et al., 1993), implicated in many aspects of

to verify whether FOXP1 is indeed able to bind the AGR2 promoter tumour biology including proliferation, apoptosis, EMT, invasion

and control its activity. Whether FOXP1 has similar effects on AGR3, and migration (Koul et al., 2013; Sui et al., 2014). Particularly,

which is also expressed in EsR-positive breast cancers, remains to activation of ELK1 leads to the enhanced cell proliferation and sur-

be established as well (Fletcher et al., 2003). vival (Huynh, 2002; Koul et al., 2013). Moreover, relying on our

Another noteworthy transcription factor that potentially trigg- in silico analysis, we hypothesize that AGR3 may work as a down-

ers expression of AGRs is zinc-ﬁnger enhancer binding-1 (ZEB1). stream effector of other signal transduction pathways such as Notch

ZEB1 is a master regulator of epithelial-to-mesenchymal transition (due to the presence of RBPj␬ binding site (Jarriault et al., 1995)),

(EMT), a fundamental cellular programme controlling embryonic TGF-␤/activin/Nodal (presence of FOXH1 binding site (Schier and

development and homeostasis, as well as tumour progression and Shen, 2000)) and Janus kinase-signal transducer and activator

metastasis (Thiery et al., 2009). ZEB1 suppresses the epithelial of transcription (JAK-STAT) signalling (presence of STAT binding

cell–cell adhesion molecule E-cadherin (downregulation of which site). These pathways coordinate signalling cascades controlling

Please cite this article in press as: Obacz, J., et al., The role of AGR2 and AGR3 in cancer: Similar but not identical. Eur. J. Cell Biol. (2015), http://dx.doi.org/10.1016/j.ejcb.2015.01.002

G Model

EJCB-50808; No. of Pages 9 ARTICLE IN PRESS

6 J. Obacz et al. / European Journal of Cell Biology xxx (2015) xxx–xxx

Some of the transporter molecules identiﬁed in gene expression

database (Table S2) could participate in AGRs export from the cell,

including transmembrane channel-like protein 4 (TMC4), trans-

membrane channel-like protein 5 (TMC5), protein member 4 of

solute carrier family 44 (SLC44A4) or cystic ﬁbrosis transmembrane

conductance regulator (CFTR).

Presence of AGR proteins outside the cells suggests their possi-

ble role in an autocrine and paracrine signalling. We also found

tetraspanin 1 (TSPAN1) as a potential AGR2 co-expressed part-

ner. Tetraspanins, also known as transmembrane 4 superfamily

(TM4SF) members, mediate outside-in signal transduction events

and consequently regulate cell differentiation, migration and pro-

liferation (Richardson et al., 2011).

Yeast two-hybrid screen identiﬁed metastasis-associated GPI-

anchored C4.4a protein and extracellular alpha-dystroglycan

(DAG-1) as binding proteins for both AGR2 and AGR3. It was fur-

Fig. 5. AGR2 and AGR3 co-expressed genes. Co-expression tool from Genevestigator

ther discussed that through these interactions, AGR proteins can

was used to ﬁnd genes having an expression proﬁle most similar to AGR genes.

be involved in many pivotal processes promoting tumour progres-

sion including metastasis, migration and invasion (Fletcher et al.,

2003). AGR2 role as a metastasis-related protein is intensively stud-

cell proliferation, differentiation, migration, apoptosis, cell-to-cell

ied in many cancer types including breast carcinomas (Liu et al.,

communication as well as developmental processes (Kitisin et al.,

2005), prostate adenocarcinomas (Zhang et al., 2010), head and

2007; Lai, 2004; Rawlings et al., 2004).

neck squamous cell carcinomas (Sweeny et al., 2012) or gastroin-

In addition, given that during brain development in X. lae-

testinal tumours (Valladares-Ayerbes et al., 2008).

vis amphibian’s AGRs are induced in the FGF-dependent manner

AGR3 function in cancer development remains ambiguous as

(Tereshina et al., 2014), it is plausible that expression of human

yet; nevertheless, it was shown that AGR3 mediates resistance to

AGRs is also driven through this signalling pathway.

cisplatin (CDDP) in a mouse xenograft model (Gray et al., 2012) and

Interestingly, although regulation of AGR2 and AGR3 expression

can bind to metastasis-related proteins C4.4a and DAG-1 (Fletcher

by EsR has been demonstrated several times (Hrstka et al., 2010;

et al., 2003). Co-expression analysis using Genevestigator depicted

Welboren et al., 2009; Wilson et al., 2006), we have not identi-

other possible AGR3 partners linking AGR3 with cell adhesion,

ﬁed EsR-binding sites in the analysed promoter region of either of

motility, metastasis and regulation of cell cycle. We found that

these two genes. It is thus possible that (i) EsR-binding sites are

AGR3 is co-expressed with a gene coding for tight junction protein,

located in the distant promoter regions or in introns, (ii) EsR acts

claudin 3 (CLDN3), responsible for control of tumour growth and

through interaction with other transcription factors, (iii) EsR is an

metastasis (Shang et al., 2012) as well as suppression of epithelial-

enhancer or (iv) EsR signalling pathways activate AGR transcription

to-mesenchymal transition (Lin et al., 2013). On the other hand,

or remove inhibition of their induction.

overexpression of claudin 3 promotes the malignant potential of

Although the number of in silico identiﬁed binding sites for spe-

colorectal cancer cells, which is regulated by ERK1/2 and PI3K-Akt

ciﬁc transcription factors is in substance consistent with previously

pathways (de Souza et al., 2013). Interestingly, it was also demon-

published data, it is clear that real understanding of involvement

strated that CLDN3 affects sensitivity of ovarian cancer cells to

of these identiﬁed putative transcription factors and hierarchy of

CDDP through regulation of copper inﬂux transporter CTR1 func-

their interplay in regulating transcription of AGR genes requires

tion (Shang et al., 2013). Therefore, it is tempting to speculate that

further experimental evidence.

insensitivity of AGR3-positive tumours to the growth-inhibiting

effect of CDDP (Gray et al., 2012) would be due to cooperation with

AGR2 and AGR3 co-expressed partners claudin 3.

Another noteworthy AGR3 co-expression partner is the gene

In order to provide deeper insight into potential AGR func- coding for immunoglobulin-like hepatocyte cell adhesion molecule

tions, with a focus on the poorly characterized AGR3 protein, (hepaCAM), which is frequently downregulated in diverse human

we performed co-expression analysis using Genevestigator (Hruz cancers (Chung Moh et al., 2005; Moh et al., 2005). There is mount-

et al., 2008). We found several shared and unique genes, which ing evidence for hepaCAM role as a tumour suppressor as it inhibits

could exemplify AGRs interacting molecules, downstream partners proliferation of human bladder cancer cells (Wang et al., 2013),

and/or cross-talking pathways (Fig. 5 and Table S2). causes G1 phase arrest and promotes c-Myc degradation in human

There are several lines of evidence that AGRs are secreted renal cell carcinomas (Zhang et al., 2011), whereas in MCF7 breast

proteins. Although both AGR2 and AGR3 harbour ER-retention cancer cells, it induces cellular senescence (Moh et al., 2008). It

sequences (KTEL and QSEL. respectively) in their C-terminal parts, would be interesting to evaluate whether AGR3 may suppress

they can escape ER retrieval machinery and can be found in the hepaCAM activity and, in consequence, promote tumour aggres-

cytoplasm as well as in the extracellular environment (Fletcher siveness.

et al., 2003). In both T47-D and MDA-MB-468 cells, AGR3 is local- In our analysis, both AGR2 and AGR3 were found to co-express

ized in secretory or endosome-like vesicles (Adam et al., 2003), with genes coding two gastrointestinal (GI) track-speciﬁc mem-

while AGR2 was found to be secreted during the development of bers of the calpain family, calpain 8 (CAPN8) and calpain 9 (CAPN9)

pancreatic cancer (Ramachandran et al., 2008) and was detected in (Lee et al., 1998; Sorimachi et al., 1993). Most calpains are ubiqui-

the urine of prostate cancer patients (Bu et al., 2011) as well as in tous cytosolic proteases involved in a plethora of cellular processes

the blood of ovarian cancer patients (Edgell et al., 2010). Remark- such as signal transduction, cell cycle progression, apoptosis or

ably, Bergstrom et al. (2014) have recently demonstrated that in cytoskeletal remodelling (Storr et al., 2011). CAPN8 and CAPN9

addition to the KTEL ER-retention signal, the presence of a sin- were demonstrated to play a protective role in the gastric mucosa

gle cysteine within the AGR2 TRX-like domain is required for the by forming an active protease complex (Hata et al., 2010). In addi-

control of protein secretion. tion, downregulation of CAPN9 has been reported in a subset of

Please cite this article in press as: Obacz, J., et al., The role of AGR2 and AGR3 in cancer: Similar but not identical. Eur. J. Cell Biol. (2015), http://dx.doi.org/10.1016/j.ejcb.2015.01.002

G Model

EJCB-50808; No. of Pages 9 ARTICLE IN PRESS

J. Obacz et al. / European Journal of Cell Biology xxx (2015) xxx–xxx 7

gastric cancer patients and gastric cancer cell lines (Yoshikawa Bai, Z., Ye, Y., Liang, B., Xu, F., Zhang, H., Zhang, Y., et al., 2011. Proteomics-based

identiﬁcation of a group of apoptosis-related proteins and biomarkers in gastric

et al., 2000). Therefore, it is possible that AGRs are essential for the

cancer. Int. J. Oncol. 38, 375–383.

maintenance of calpains physiological properties, similar to that of

Benham, A.M., 2012. The protein disulﬁde isomerase family: key players in health

intestinal MUC2 (Park et al., 2009). and disease. Antioxid. Redox Signal. 16, 781–789.

Bergstrom, J.H., Berg, K.A., Rodriguez-Pineiro, A.M., Stecher, B., Johansson, M.E.,

Hansson, G.C., 2014. AGR2, an endoplasmic reticulum protein, is secreted into

Conclusions and perspectives

the gastrointestinal mucus. PLoS ONE 9, e104186.

Brabletz, S., Brabletz, T., 2010. The ZEB/miR-200 feedback loop—a motor of cellular

plasticity in development and cancer? EMBO Rep. 11, 670–677.

In conclusion, AGR2 and AGR3 proteins due to their high

Browne, G., Sayan, A.E., Tulchinsky, E., 2010. ZEB proteins link cell motility with cell

sequence homology are similar but not identical and therefore can

cycle control and cell survival in cancer. Cell Cycle 9, 886–891.

be expressed in cognate or uncoupled manner in divergent carci- Brychtova, V., Vojtesek, B., Hrstka, R., 2011. Anterior gradient 2: a novel player in

tumor cell biology. Cancer Lett. 304, 1–7.

nomas and normal tissues.

Brychtova, V., Zampachova, V., Hrstka, R., Fabian, P., Novak, J., Hermanova, M.,

Mechanisms triggering context-dependent expression of AGR

et al., 2014. Differential expression of anterior gradient protein 3 in intrahep-

proteins remain poorly deﬁned and thus need deeper investiga- atic cholangiocarcinoma and hepatocellular carcinoma. Exp. Mol. Pathol. 96,

375–381.

tion. As a part of the PDI family, AGRs are predicted to control

Bu, H., Bormann, S., Schafer, G., Horninger, W., Massoner, P., Neeb, A., et al., 2011. The

proper protein load in the endoplasmic reticulum, and indeed, there

anterior gradient 2 (AGR2) gene is overexpressed in prostate cancer and may be

are reports showing AGR2’s essential role, for instance, in mucin useful as a urine sediment marker for prostate cancer detection. Prostate 71,

575–587.

production. The information depicted in this review indicates that

Bu, H., Schweiger, M.R., Manke, T., Wunderlich, A., Timmermann, B., Kerick, M., et al.,

AGR2 and AGR3 plausibly control similar aspects of tumour biol-

2013. Anterior gradient 2 and 3—two prototype androgen-responsive genes

ogy including cell cycle control, differentiation, migration, invasion transcriptionally upregulated by androgens and by oestrogens in prostate cancer

and metastasis. Moreover, relying on available data and on our in cells. FEBS J. 280, 1249–1266.

Cartharius, K., Frech, K., Grote, K., Klocke, B., Haltmeier, M., Klingenhoff, A., et al.,

silico analysis, it can be speculated that, in addition to oestrogen

2005. MatInspector and beyond: promoter analysis based on transcription factor

receptor signalling pathway, they are components of additional

binding sites. Bioinformatics 21, 2933–2942.

signal transduction cascades. Following this, it would be interest- Chen, Y.C., Lu, Y.F., Li, I.C., Hwang, S.P., 2012. Zebraﬁsh Agr2 is required for terminal

differentiation of intestinal goblet cells. PLoS ONE 7, e34408.

ing to verify whether a functional cross-talk between AGR proteins

Chevet, E., Fessart, D., Delom, F., Mulot, A., Vojtesek, B., Hrstka, R., et al., 2013. Emerg-

exists in order to regulate cell fate processes in health and disease.

ing roles for the pro-oncogenic anterior gradient-2 in cancer development.

Taking into account their high expression and emerging roles, it Oncogene 32, 2499–2509.

Chung Moh, M., Hoon Lee, L., Shen, S., 2005. Cloning and characterization of

is highly possible that they can act as either tumour suppressors

hepaCAM, a novel Ig-like cell adhesion molecule suppressed in human hepa-

or oncogenes, similar to TGF-␤ (Imamura et al., 2012). However,

tocellular carcinoma. J. Hepatol. 42, 833–841.

mechanisms that enable switch between these two contradictory Darb-Esfahani, S., Fritzsche, F., Kristiansen, G., Weichert, W., Sehouli, J., Braicu, I.,

et al., 2012. Anterior gradient protein 2 (AGR2) is an independent prognostic

actions require further investigation. In the light of recent develop-

factor in ovarian high-grade serous carcinoma. Virch. Arch. 461, 109–116.

ments in cancer research, proper understanding of AGR functions

de Souza, W.F., Fortunato-Miranda, N., Robbs, B.K., de Araujo, W.M., de-Freitas-

would be inevitable for new drugs development aiming both at Junior, J.C., Bastos, L.G., et al., 2013. Claudin-3 overexpression increases the

malignant potential of colorectal cancer cells: roles of ERK1/2 and PI3K-Akt as

these proteins and pathways driving their expression. Moreover, as

modulators of EGFR signaling. PLoS ONE 8, e74994.

secreted proteins they could serve as markers for cancer detection

Dumartin, L., Whiteman, H.J., Weeks, M.E., Hariharan, D., Dmitrovic, B., Iacobuzio-

and/or prediction of patient outcome. Lastly, as suggested in this Donahue, C.A., et al., 2011. AGR2 is a novel surface antigen that promotes the

work, AGRs could cooperate with other druggable targets, which dissemination of pancreatic cancer cells through regulation of cathepsins B and

D. Cancer Res. 71, 7091–7102.

makes them even more attractive for thorough characterization.

Edgell, T.A., Barraclough, D.L., Rajic, A., Dhulia, J., Lewis, K.J., Armes, J.E., et al., 2010.

Increased plasma concentrations of anterior gradient 2 protein are positively

Acknowledgements associated with ovarian cancer. Clin. Sci. (Lond.) 118, 717–725.

Fletcher, G.C., Patel, S., Tyson, K., Adam, P.J., Schenker, M., Loader, J.A., et al., 2003.

hAG-2 and hAG-3, human homologues of genes involved in differentiation, are

We thank Dr P.J. Coates for critical reading of the manuscript. associated with oestrogen receptor-positive breast tumours and interact with

metastasis gene C4.4a and dystroglycan. Br. J. Cancer 88, 579–585.

The work was supported by European Regional Development Fund

Galligan, J.J., Petersen, D.R., 2012. The human protein disulﬁde isomerase gene fam-

and the state budget of the Czech Republic for Regional Centre for

ily. Hum. Genom. 6, 6.

Applied Molecular Oncology—RECAMO (CZ.1.05/2.1.00/03.0101), Gray, T.A., MacLaine, N.J., Michie, C.O., Bouchalova, P., Murray, E., Howie, J., et al.,

GACR P206/12/G151, GACR 13-00956S and by the 7th Framework 2012. Anterior gradient-3: a novel biomarker for ovarian cancer that mediates

cisplatin resistance in xenograft models. J. Immunol. Methods 378, 20–32.

Programme (a part of the EU Marie Curie Initial Training Networks

Hata, S., Abe, M., Suzuki, H., Kitamura, F., Toyama-Sorimachi, N., Abe, K., et al., 2010.

(ITN) Biomedical engineering for cancer and brain disease diagnosis

Calpain 8/nCL-2 and calpain 9/nCL-4 constitute an active protease complex, G-

and therapy development: EngCaBra. Project no. PITN-GA-2010- calpain, involved in gastric mucosal defense. PLoS Genet. 6, e1001040.

Hatahet, F., Ruddock, L.W., 2009. Protein disulﬁde isomerase: a critical evalua-

264417). The authors declare that they have no conﬂict of interests.

tion of its function in disulﬁde bond formation. Antioxid. Redox Signal. 11,

2807–2850.

He, T.C., Sparks, A.B., Rago, C., Hermeking, H., Zawel, L., da Costa, L.T., et al., 1998.

Appendix A. Supplementary data

Identiﬁcation of c-MYC as a target of the APC pathway. Science 281, 1509–1512.

Higa, A., Mulot, A., Delom, F., Bouchecareilh, M., Nguyen, D.T., Boismenu, D., et al.,

Supplementary data associated with this article can be found, in 2011. Role of pro-oncogenic protein disulﬁde isomerase (PDI) family member

anterior gradient 2 (AGR2) in the control of endoplasmic reticulum homeostasis.

the online version, at http://dx.doi.org/10.1016/j.ejcb.2015.01.002.

J. Biol. Chem. 286, 44855–44868.

Hrstka, R., Nenutil, R., Fourtouna, A., Maslon, M.M., Naughton, C., Langdon, S., et al.,

References 2010. The pro-metastatic protein anterior gradient-2 predicts poor prognosis in

tamoxifen-treated breast cancers. Oncogene 29, 4838–4847.

Hruz, T., Laule, O., Szabo, G., Wessendorp, F., Bleuler, S., Oertle, L., et al., 2008.

Aberger, F., Weidinger, G., Grunz, H., Richter, K., 1998. Anterior speciﬁcation of

Genevestigator v3: a reference expression database for the meta-analysis of

embryonic ectoderm: the role of the Xenopus cement gland-speciﬁc gene XAG-2.

transcriptomes. Adv. Bioinformatics 2008, 420747.

Mech. Dev. 72, 115–130.

Huynh, H., 2002. Induction of apoptosis in rat ventral prostate by ﬁnasteride is

Adam, P.J., Boyd, R., Tyson, K.L., Fletcher, G.C., Stamps, A., Hudson, L., et al., 2003. Com-

associated with alteration in MAP kinase pathways and Bcl-2 related family

prehensive proteomic analysis of breast cancer cell membranes reveals unique

of proteins. Int. J. Oncol. 20, 1297–1303.

proteins with potential roles in clinical cancer. J. Biol. Chem. 278, 6482–6489.

Imamura, T., Hikita, A., Inoue, Y., 2012. The roles of TGF-beta signaling in carcino-

Appenzeller-Herzog, C., Ellgaard, L., 2008. The human PDI family: versatility packed

genesis and breast cancer metastasis. Breast Cancer 19, 118–124.

into a single fold. Biochim. Biophys. Acta 1783, 535–548.

Ivanova, A.S., Tereshina, M.B., Ermakova, G.V., Belousov, V.V., Zaraisky, A.G., 2013. Agr

Armes, J.E., Davies, C.M., Wallace, S., Taheri, T., Perrin, L.C., Autelitano, D.J., 2013.

genes, missing in amniotes, are involved in the body appendages regeneration

AGR2 expression in ovarian tumours: a potential biomarker for endometrioid

in frog tadpoles. Sci. Rep. 3, 1279.

and mucinous differentiation. Pathology 45, 49–54.

Please cite this article in press as: Obacz, J., et al., The role of AGR2 and AGR3 in cancer: Similar but not identical. Eur. J. Cell Biol. (2015), http://dx.doi.org/10.1016/j.ejcb.2015.01.002

G Model

EJCB-50808; No. of Pages 9 ARTICLE IN PRESS

8 J. Obacz et al. / European Journal of Cell Biology xxx (2015) xxx–xxx

Izquierdo, L., Mengual, L., Gazquez, C., Ingelmo-Torres, M., Alcaraz, A., 2010. Molec- Richardson, M.M., Jennings, L.K., Zhang, X.A., 2011. Tetraspanins and tumor progres-

ular characterization of upper urinary tract tumours. BJU Int. 106, 868–872. sion. Clin. Exp. Metastasis 28, 261–270.

Jacquot, J.P., Gelhaye, E., Rouhier, N., Corbier, C., Didierjean, C., Aubry, A., 2002. Ron, D., Walter, P., 2007. Signal integration in the endoplasmic reticulum unfolded

Thioredoxins and related proteins in photosynthetic organisms: molecular basis protein response. Nat. Rev. Mol. Cell Biol. 8, 519–529.

for thiol dependent regulation. Biochem. Pharmacol. 64, 1065–1069. Ryu, J., Park, S.G., Lee, P.Y., Cho, S., Lee do, H., Kim, G.H., et al., 2013. Dimerization of

Janknecht, R., Ernst, W.H., Pingoud, V., Nordheim, A., 1993. Activation of ternary pro-oncogenic protein anterior gradient 2 is required for the interaction with

complex factor Elk-1 by MAP kinases. EMBO J. 12, 5097–5104. BiP/GRP78. Biochem. Biophys. Res. Commun. 430, 610–615.

Jarriault, S., Brou, C., Logeat, F., Schroeter, E.H., Kopan, R., Israel, A., 1995. Signalling Salmans, M.L., Zhao, F., Andersen, B., 2013. The estrogen-regulated anterior gradient

downstream of activated mammalian Notch. Nature 377, 355–358. 2 (AGR2) protein in breast cancer: a potential drug target and biomarker. Breast

King, E.R., Tung, C.S., Tsang, Y.T., Zu, Z., Lok, G.T., Deavers, M.T., et al., 2011. The Cancer Res. 15, 204.

anterior gradient homolog 3 (AGR3) gene is associated with differentiation and Schier, A.F., Shen, M.M., 2000. Nodal signalling in vertebrate development. Nature

survival in ovarian cancer. Am. J. Surg. Pathol. 35, 904–912. 403, 385–389.

Kitisin, K., Saha, T., Blake, T., Golestaneh, N., Deng, M., Kim, C., et al., 2007. Tgf-Beta Schroeder, B.W., Verhaeghe, C., Park, S.W., Nguyenvu, L.T., Huang, X., Zhen, G., et al.,

signaling in development. Sci. STKE 2007, cm1. 2012. AGR2 is induced in asthma and promotes allergen-induced mucin over-

Koon, H.B., Ippolito, G.C., Banham, A.H., Tucker, P.W., 2007. FOXP1: a potential ther- production. Am. J. Respir. Cell Mol. Biol. 47, 178–185.

apeutic target in cancer. Expert Opin. Ther. Targets 11, 955–965. Shang, X., Lin, X., Alvarez, E., Manorek, G., Howell, S.B., 2012. Tight junction proteins

Koul, H.K., Pal, M., Koul, S., 2013. Role of p38 MAP kinase signal transduction in solid claudin-3 and claudin-4 control tumor growth and metastases. Neoplasia 14,

tumors. Genes Cancer 4, 342–359. 974–985.

Kozlov, G., Maattanen, P., Thomas, D.Y., Gehring, K., 2010. A structural overview of Shang, X., Lin, X., Manorek, G., Howell, S.B., 2013. Claudin-3 and claudin-4 regu-

the PDI family of proteins. FEBS J. 277, 3924–3936. late sensitivity to cisplatin by controlling expression of the copper and cisplatin

Kumar, A., Godwin, J.W., Gates, P.B., Garza-Garcia, A.A., Brockes, J.P., 2007. Molecu- inﬂux transporter CTR1. Mol. Pharmacol. 83, 85–94.

lar basis for the nerve dependence of limb regeneration in an adult vertebrate. Shigekawa, T., Ijichi, N., Ikeda, K., Horie-Inoue, K., Shimizu, C., Saji, S., et al., 2011.

Science 318, 772–777. FOXP1, an estrogen-inducible transcription factor, modulates cell prolifera-

Lai, E.C., 2004. Notch signaling: control of cell communication and cell fate. Devel- tion in breast cancer cells and 5-year recurrence-free survival of patients with

opment 131, 965–973. tamoxifen-treated breast cancer. Horm. Cancer 2, 286–297.

Lee, H.J., Sorimachi, H., Jeong, S.Y., Ishiura, S., Suzuki, K., 1998. Molecular cloning and Shtutman, M., Zhurinsky, J., Simcha, I., Albanese, C., D’Amico, M., Pestell, R., et al.,

characterization of a novel tissue-speciﬁc calpain predominantly expressed in 1999. The cyclin D1 gene is a target of the beta-catenin/LEF-1 pathway. Proc.

the digestive tract. Biol. Chem. 379, 175–183. Natl Acad. Sci. U.S.A. 96, 5522–5527.

Lee, S., Bang, S., Song, K., Lee, I., 2006. Differential expression in normal-adenoma- Sive, H.L., Hattori, K., Weintraub, H., 1989. Progressive determination during forma-

carcinoma sequence suggests complex molecular carcinogenesis in colon. Oncol. tion of the anteroposterior axis in Xenopus laevis. Cell 58, 171–180.

Rep. 16, 747–754. Sorimachi, H., Ishiura, S., Suzuki, K., 1993. A novel tissue-speciﬁc calpain species

Lepreux, S., Bioulac-Sage, P., Chevet, E., 2011. Differential expression of the anterior expressed predominantly in the stomach comprises two alternative splic-

gradient protein-2 is a conserved feature during morphogenesis and carcino- ing products with and without Ca(2+)-binding domain. J. Biol. Chem. 268,

genesis of the biliary tree. Liver Int. 31, 322–328. 19476–19482.

Lin, X., Shang, X., Manorek, G., Howell, S.B., 2013. Regulation of the epithelial- Storr, S.J., Carragher, N.O., Frame, M.C., Parr, T., Martin, S.G., 2011. The calpain system

mesenchymal transition by claudin-3 and claudin-4. PLoS ONE 8, e67496. and cancer. Nat. Rev. Cancer 11, 364–374.

Liu, D., Rudland, P.S., Sibson, D.R., Platt-Higgins, A., Barraclough, R., 2005. Human Sui, X., Kong, N., Ye, L., Han, W., Zhou, J., Zhang, Q., et al., 2014. p38 and JNK

homologue of cement gland protein, a novel metastasis inducer associated with MAPK pathways control the balance of apoptosis and autophagy in response

breast carcinomas. Cancer Res. 65, 3796–3805. to chemotherapeutic agents. Cancer Lett. 344, 174–179.

Maslon, M.M., Hrstka, R., Vojtesek, B., Hupp, T.R., 2010. A divergent substrate-binding Sweeny, L., Liu, Z., Bush, B.D., Hartman, Y., Zhou, T., Rosenthal, E.L., 2012. CD147 and

loop within the pro-oncogenic protein anterior gradient-2 forms a docking site AGR2 expression promote cellular proliferation and metastasis of head and neck

for Reptin. J. Mol. Biol. 404, 418–438. squamous cell carcinoma. Exp. Cell Res. 318, 1788–1798.

Moh, M.C., Zhang, C., Luo, C., Lee, L.H., Shen, S., 2005. Structural and functional anal- Tabas, I., Ron, D., 2011. Integrating the mechanisms of apoptosis induced by endo-

yses of a novel Ig-like cell adhesion molecule, hepaCAM, in the human breast plasmic reticulum stress. Nat. Cell Biol. 13, 184–190.

carcinoma MCF7 cells. J. Biol. Chem. 280, 27366–27374. Tereshina, M.B., Ermakova, G.V., Ivanova, A.S., Zaraisky, A.G., 2014. Ras-dva1 small

Moh, M.C., Zhang, T., Lee, L.H., Shen, S., 2008. Expression of hepaCAM is GTPase regulates telencephalon development in Xenopus laevis embryos by con-

downregulated in cancers and induces senescence-like growth arrest via a trolling Fgf8 and Agr signaling at the anterior border of the neural plate. Biol.

p53/p21-dependent pathway in human breast cancer cells. Carcinogenesis 29, Open 3, 192–203.

2298–2305. Thiery, J.P., Acloque, H., Huang, R.Y., Nieto, M.A., 2009. Epithelial–mesenchymal

Ni, M., Lee, A.S., 2007. ER chaperones in mammalian development and human dis- transitions in development and disease. Cell 139, 871–890.

eases. FEBS Lett. 581, 3641–3651. Thompson, D.A., Weigel, R.J., 1998. hAG-2, the human homologue of the Xenopus

Norris, A.M., Gore, A., Balboni, A., Young, A., Longnecker, D.S., Korc, M., 2013. AGR2 laevis cement gland gene XAG-2, is coexpressed with estrogen receptor in breast

is a SMAD4-suppressible gene that modulates MUC1 levels and promotes the cancer cell lines. Biochem. Biophys. Res. Commun. 251, 111–116.

initiation and progression of pancreatic intraepithelial neoplasia. Oncogene 32, Turano, C., Coppari, S., Altieri, F., Ferraro, A., 2002. Proteins of the PDI family: unpre-

3867–3876. dicted non-ER locations and functions. J. Cell. Physiol. 193, 154–163.

Park, K., Chung, Y.J., So, H., Kim, K., Park, J., Oh, M., et al., 2011. AGR2, a mucinous Vaarala, M.H., Hirvikoski, P., Kauppila, S., Paavonen, T.K., 2012. Identiﬁca-

ovarian cancer marker, promotes cell proliferation and migration. Exp. Mol. Med. tion of androgen-regulated genes in human prostate. Mol. Med. Rep. 6,

43, 91–100. 466–472.

Park, S.W., Zhen, G., Verhaeghe, C., Nakagami, Y., Nguyenvu, L.T., Barczak, A.J., et al., Valladares-Ayerbes, M., Diaz-Prado, S., Reboredo, M., Medina, V., Iglesias-Diaz, P.,

2009. The protein disulﬁde isomerase AGR2 is essential for production of intesti- Lorenzo-Patino, M.J., et al., 2008. Bioinformatics approach to mRNA markers dis-

nal mucus. Proc. Natl Acad. Sci. U.S.A. 106, 6950–6955. covery for detection of circulating tumor cells in patients with gastrointestinal

Patel, P., Clarke, C., Barraclough, D.L., Jowitt, T.A., Rudland, P.S., Barraclough, R., cancer. Cancer Detect. Prev. 32, 236–250.

et al., 2013. Metastasis-promoting anterior gradient 2 protein has a dimeric Vanderlaag, K.E., Hudak, S., Bald, L., Fayadat-Dilman, L., Sathe, M., Grein, J., et al.,

thioredoxin fold structure and a role in cell adhesion. J. Mol. Biol. 425, 2010. Anterior gradient-2 plays a critical role in breast cancer cell growth and

929–943. survival by modulating cyclin D1, estrogen receptor-alpha and survivin. Breast

Persson, S., Rosenquist, M., Knoblach, B., Khosravi-Far, R., Sommarin, M., Michalak, Cancer Res. 12, R32.

M., 2005. Diversity of the protein disulﬁde isomerase family: identiﬁcation of Verma, S., Salmans, M.L., Geyfman, M., Wang, H., Yu, Z., Lu, Z., et al., 2012. The

breast tumor induced Hag2 and Hag3 as novel members of the protein family. estrogen-responsive Agr2 gene regulates mammary epithelial proliferation and

Mol. Phylogenet. Evol. 36, 734–740. facilitates lobuloalveolar development. Dev. Biol. 369, 249–260.

Petek, E., Windpassinger, C., Egger, H., Kroisel, P.M., Wagner, K., 2000. Localization Vivekanandan, P., Micchelli, S.T., Torbenson, M., 2009. Anterior gradient-2 is over-

of the human anterior gradient-2 gene (AGR2) to chromosome band 7p21.3 by expressed by ﬁbrolamellar carcinomas. Hum. Pathol. 40, 293–299.

radiation hybrid mapping and ﬂuorescence in situ hybridisation. Cytogenet. Cell Wang, Q., Luo, C., Wu, X., Du, H., Song, X., Fan, Y., 2013. hepaCAM and p-mTOR

Genet. 89, 141–142. closely correlate in bladder transitional cell carcinoma and hepaCAM expression

Pizzi, M., Fassan, M., Balistreri, M., Galligioni, A., Rea, F., Rugge, M., 2012. Anterior inhibits proliferation via an AMPK/mTOR dependent pathway in human bladder

gradient 2 overexpression in lung adenocarcinoma. Appl. Immunohistochem. cancer cells. J. Urol. 190, 1912–1918.

Mol. Morphol. 20, 31–36. Welboren, W.J., van Driel, M.A., Janssen-Megens, E.M., van Heeringen, S.J., Sweep,

Quandt, K., Frech, K., Karas, H., Wingender, E., Werner, T., 1995. MatInd and MatIn- F.C., Span, P.N., et al., 2009. ChIP-Seq of ERalpha and RNA polymerase II deﬁnes

spector: new fast and versatile tools for detection of consensus matches in genes differentially responding to ligands. EMBO J. 28, 1418–1428.

nucleotide sequence data. Nucleic Acids Res. 23, 4878–4884. Wilson, C.L., Sims, A.H., Howell, A., Miller, C.J., Clarke, R.B., 2006. Effects of oestrogen

Ramachandran, V., Arumugam, T., Wang, H., Logsdon, C.D., 2008. Anterior gradient on gene expression in epithelium and stroma of normal human breast tissue.

2 is expressed and secreted during the development of pancreatic cancer and Endocr. Relat. Cancer 13, 617–628.

promotes cancer cell survival. Cancer Res. 68, 7811–7818. Wotton, D., Lo, R.S., Lee, S., Massague, J., 1999. A Smad transcriptional corepressor.

Rawlings, J.S., Rosler, K.M., Harrison, D.A., 2004. The JAK/STAT signaling pathway. J. Cell 97, 29–39.

Cell Sci. 117, 1281–1283. Wright, T.M., Wardell, S.E., Jasper, J.S., Stice, J.P., Saﬁ, R., Nelson, E.R., et al., 2014.

Reya, T., Clevers, H., 2005. Wnt signalling in stem cells and cancer. Nature 434, Delineation of a FOXA1/ERalpha/AGR2 regulatory loop that is dysregulated in

843–850. endocrine therapy-resistant breast cancer. Mol. Cancer Res.

Please cite this article in press as: Obacz, J., et al., The role of AGR2 and AGR3 in cancer: Similar but not identical. Eur. J. Cell Biol. (2015), http://dx.doi.org/10.1016/j.ejcb.2015.01.002

G Model

EJCB-50808; No. of Pages 9 ARTICLE IN PRESS

J. Obacz et al. / European Journal of Cell Biology xxx (2015) xxx–xxx 9

Xia, J.H., Jiang, J., Shi, Y.H., Gui, J.F., 2009. Predominant expression and cellular dis- Zhang, Q.L., Luo, C.L., Wu, X.H., Wang, C.Y., Xu, X., Zhang, Y.Y., et al., 2011. HepaCAM

tribution of ﬁsh Agr2 in renal collecting system. Comp. Biochem. Physiol. B: induces G1 phase arrest and promotes c-Myc degradation in human renal cell

Biochem. Mol. Biol. 152, 397–404. carcinoma. J. Cell. Biochem. 112, 2910–2919.

Yoshikawa, Y., Mukai, H., Hino, F., Asada, K., Kato, I., 2000. Isolation of two Zhang, Y., Ali, T.Z., Zhou, H., D’Souza, D.R., Lu, Y., Jaffe, J., et al., 2010. ErbB3 binding

novel genes, down-regulated in gastric cancer. Jpn. J. Cancer Res. 91, protein 1 represses metastasis-promoting gene anterior gradient protein 2 in

459–463. prostate cancer. Cancer Res. 70, 240–248.

Zhang, J.S., Gong, A., Cheville, J.C., Smith, D.I., Young, C.Y., 2005. AGR2, an Zheng, W., Rosenstiel, P., Huse, K., Sina, C., Valentonyte, R., Mah, N., et al., 2006. Eval-

androgen-inducible secretory protein overexpressed in prostate cancer. Genes uation of AGR2 and AGR3 as candidate genes for inﬂammatory bowel disease.

Chromosomes Cancer 43, 249–259. Genes Immun. 7, 11–18.

Please cite this article in press as: Obacz, J., et al., The role of AGR2 and AGR3 in cancer: Similar but not identical. Eur. J. Cell Biol. (2015), http://dx.doi.org/10.1016/j.ejcb.2015.01.002 11.2 Attachment 2

OBACZ, J., BRYCHTOVA, V., PODHOREC, J., FABIAN, P., DOBES, P., VOJTESEK, B. & HRSTKA, R. 2015a. Anterior gradient protein 3 is associated with less aggressive tumors and better outcome of breast cancer patients. Onco Targets Ther, 8, 1523-32.

Journal name: OncoTargets and Therapy Article Designation: Original Research Year: 2015 Volume: 8 OncoTargets and Therapy Dovepress Running head verso: Obacz et al Running head recto: Significance of AGR3 in breast cancer open access to scientific and medical research DOI: http://dx.doi.org/10.2147/OTT.S82235

Open Access Full Text Article Original Research Anterior gradient protein 3 is associated with less aggressive tumors and better outcome of breast cancer patients

Joanna Obacz1 Abstract: Anterior gradient protein (AGR) 3 is a highly related homologue of pro-oncogenic Veronika Brychtova1 AGR2 and belongs to the family of protein disulfide isomerases. Although AGR3 was found Jan Podhorec1 in breast, ovary, prostate, and liver cancer, it remains of yet poorly defined function in tumo- Pavel Fabian2 rigenesis. This study aimed to determine AGR3 expression in a cohort of 129 primary breast Petr Dobes1 carcinomas and evaluate the clinical and prognostic significance of AGR3 in these tumors. Borivoj Vojtesek1 The immunohistochemical analysis revealed the presence of AGR3 staining to varying degrees in 80% of analyzed specimens. The percentage of AGR3-positive cells significantly Roman Hrstka1 correlated with estrogen receptor, progesterone receptor (both P,0.0001) as well as low 1 Regional Centre for Applied histological grade (P=0.003), and inversely correlated with the level of Ki-67 expression Molecular Oncology (RECAMO), 2Department of Pathology, Masaryk (P,0.0001). In the whole cohort, AGR3 expression was associated with longer progression- Memorial Cancer Institute, Brno, free survival (PFS), whereas AGR3-positive subgroup of low-histological grade tumors showed Czech Republic both significantly longer PFS and overall survival. In conclusion, AGR3 is associated with the level of differentiation, slowly proliferating tumors, and more favorable prognosis of breast cancer patients. Keywords: AGR3, patient survival, protein disulfide isomerase, ER-positive breast cancer, immunohistochemistry

Introduction Breast cancer is the most common female malignancy and a leading cause of deaths among women worldwide. Only in 2012, in Europe, roughly 464,000 new incidences were registered, and 131,000 women died from breast cancer.1 Despite intensive research on various diagnostic and/or prognostic markers, thorough understanding of factors affecting breast cancer patients’ outcome remains of great importance. In recent years, an increasing number of reports have linked anterior gradient protein (AGR) 2 with many aspects of breast tumor biology. AGR2 is a human homologue of Xenopus laevis-secreted protein XAG-2 and belongs to an evolutionary broad family with prominent role in developmental processes and regeneration of body appendages.2,3 There are three subfamilies of AGRs: AGR1, AGR2, and AGR3, all showing the highest homology to non-secreted protein disulfide isomerase (PDI) of the TLP19 subfamily.3 PDIs are involved in proper folding and maturation of newly synthesized proteins and the regulation of endoplasmic reticulum homeostasis.4 Following the first characterization of AGR2 in the estrogen receptor (ER)- positive breast cancer cell line MCF-7,5 AGR2 has been frequently shown as an Correspondence: Roman Hrstka RECAMO, Masaryk Memorial Cancer estrogen-responsive gene/protein. It was demonstrated that AGR2 is upregulated Institute, Zluty kopec 7, 65653 Brno, in response to estradiol treatment both in vitro5 and in vivo,6 and its high expres- Czech Republic Email [email protected] sion correlates with ER status7 and predicts poor prognosis in ER-positive breast submit your manuscript | www.dovepress.com OncoTargets and Therapy 2015:8 1523–1532 1523 Dovepress © 2015 Obacz et al. This work is published by Dove Medical Press Limited, and licensed under Creative Commons Attribution – Non Commercial (unported, v3.0) License. The full terms of the License are available at http://creativecommons.org/licenses/by-nc/3.0/. Non-commercial uses of the work are permitted without any further http://dx.doi.org/10.2147/OTT.S82235 permission from Dove Medical Press Limited, provided the work is properly attributed. Permissions beyond the scope of the License are administered by Dove Medical Press Limited. Information on how to request permission may be found at: http://www.dovepress.com/permissions.php Obacz et al Dovepress cancers8,9 as well as resistance to tamoxifen.10 Moreover, Memorial Cancer Institute (MMCI) between 2003 and 2006. chromatin immunoprecipitation (ChIP) confirmed direct Patient age at the time of diagnosis ranged from 29 years AGR2 regulation by ER.10–12 In normal mammary gland, to 84 years (median 57 years). The clinical, histological, AGR2 induces cell proliferation and differentiation as and molecular characteristics of the analyzed set of tumors shown in the mouse models,13 whereas in breast tumors, are summarized in Table 1. Histological typing of tumors it promotes cell progression and survival through, among was carried out according to the criteria of World Health others, ER, cyclin D1, c-Myc, and survivin signaling Organization.23 Tumor stage was determined according to pathways.14 Furthermore, when introduced into benign rat the guidelines of the Union for International Cancer Control mammary epithelial cell line, AGR2 was found to contribute (UICC).24 Tumor grade was established according to Bloom to metastasis development.15 and Richardson in the modification of Elston and Ellis.25 ER, Closely related AGR2 homologue, AGR3,7 has also been progesterone receptor (PR), human epidermal growth factor identified in breast cancer cell lines using proteomics screen receptor 2 (Her2/neu), and Ki-67 statuses were extracted from as one of the membrane-associated proteins.16 Although both pathological records obtained from the MMCI database. For molecules share 71% sequence identity and lie adjacent to the evaluation of AGR3 prognostic relevance without regard one another at chromosomal position 7p21,7,17 AGR2, but to ER status, additional ER-negative group of 90 breast not AGR3, is a dominant factor identified in many OMICS screens. Thus, to date, only few reports describing AGR3 Clinicopathological characteristics of primary breast car expression in various tumors were published, and there are Table 1 cinomas limiting amount of data depicting AGR3 prognostic rele Variablea Group Nb %c vance in these malignancies. It has been shown that AGR3 Histology Ductal 95 73.6 is strongly expressed in breast carcinomas when compared Lobular 18 14 16 to healthy tissues and that its expression correlates with ER Other 9 7 status in breast tumors.7 In another study, single ER-binding NA 7 5.4 site on AGR3 promoter has been found using ChIP-Seq Histological grade G1 27 20.9 G2 43 33.4 12 approach. Our group has recently demonstrated that intra- G3 56 43.4 hepatic cholangiocarcinomas (ICCs) express AGR3 protein, NA 3 2.3 while hepatocellular carcinomas are predominantly AGR3 Tumor size pT1 44 34.1 pT 65 50.3 negative. Furthermore, we postulated that together with acid 2 pT 6 4.7 mucopolysaccharides, AGR3 could serve as a diagnostic 3 pT4 9 7 marker of well-differentiated ICCs.18 It has also been shown NA 5 3.9 that AGR3 is overexpressed in different histological types Nodal status Negative 45 34.9 Positive 73 56.6 of ovarian cancers. In non-mucinous types (including serous NA 11 8.5 papillary, endometrioid, and clear cell), AGR3 expression ER status Negative 29 22.5 was found to be ER independent and uncoupled with AGR2 Positive 100 77.5 expression, whereas in mucinous ovarian cancers, both NA 0 0 PR status Negative 34 26.4 19 AGR2 and AGR3 showed cognate expression patterns. Positive 94 72.8 In serous type, AGR3 staining correlated with the level of dif- NA 1 0.8 ferentiation and was associated with longer patient survival.20 Her2/neu status Negative 92 71.3 Additionally, AGR3 was found to be androgen-regulated Positive 36 27.9 NA 1 0.8 21,22 gene, expression of which was highly elevated in human Ki-67d ,15% 55 42.6 21 prostate cancer. The aim of this study is to examine the $15% 61 47.3 significance between AGR3 expression, clinicopathologi- NA 13 10.1 cal characteristics, and patient outcome in primary breast AGR3 expression 1 25 19.4 2 25 19.4 carcinomas. 3 79 61.2 Notes: aDefined in the “Materials and methods” section. bNumber of patients. Materials and methods cPercentage of total patients, out of a total of 129. dCut-off for Ki-67 was used according to St Gallen Consensus in 2009. AGR3 expression: 1 – negative/border, Study group and tissue specimens 2 – weakly/moderately positive, and 3 – strongly positive. The study group consisted of 129 patients undergoing sur- Abbreviations: NA, not available; ER, estrogen receptor; PR, progesterone receptor; Her2/neu, human epidermal growth factor receptor 2; AGR3, anterior gradient gical procedure for primary breast cancer at the Masaryk protein 3.

1524 submit your manuscript | www.dovepress.com OncoTargets and Therapy 2015:8 Dovepress Dovepress Significance of AGR3 in breast cancer cancer patients treated at MMCI between 1995 and 2006 95°C, and then 40 cycles at 95°C for 15 seconds and at 60°C were included for survival analysis. Informed consent has for 1 minute. To obtain absolute quantification, dilution series been obtained from all patients involved in this study. The of plasmids pDEST12.2 with cloned respective sequences study was approved by ethical committee of MMCI, and the were used in range from 20 to 2 millions of copies to generate data used were anonymized and were handled according to standard curves. For data normalization, 18S rRNA levels Czech Republic existing legislation. were determined using TaqMan assay for 18S rRNA (Thermo Fisher Scientific, Waltham, MA, USA). Immunohistochemistry Tumor samples were fixed in 10% neutral buffered formalin Statistical analysis for 24 hours and then embedded in paraffin wax. Immuno- All statistical analyses were performed using STATISTICA histochemical analysis was performed on 4 μm thick sections Version 12 (StatSoft, Inc., Tulsa, OK, USA) and IBM SPSS cut from formalin-fixed, paraffin-embedded archival tissue Statistics 20.0. Fisher’s exact test and Pearson’s chi-squared blocks, mounted on slides, deparaffinized in xylene, and rehy- test were applied to assess the associations of immuno- drated in phosphate-buffered saline through a graded ethanol histochemical staining for AGR3 with clinicopathological series. Endogenous peroxidase activity was quenched in 3% variables. Progression-free survival (PFS) was defined as the hydrogen peroxide in phosphate-buffered saline for 15 min- time from the date of surgery to the date of death or relapse of utes. Antigen retrieval was performed in citrate buffer pH 6 disease. Overall survival (OS) was defined as the time from at 94°C for 20 minutes. For AGR3 immunodetection, the sec- surgery to death or last record. Patients who had not died or tions were incubated overnight at 4°C with mouse monoclonal who were lost to follow-up were censored when they were last antibody to AGR3 (clone 1, in house).19 A streptavidin–biotin known to be alive. Differences between survival curves were peroxidase detection system was used according to the manu- assessed with the Breslow test. Unadjusted hazard ratios (HRs) facturer’s protocol (Vectastain Elite ABC Kit; Vector Labo- ±95% confidence intervals (CIs) were obtained using Cox’s ratories, Burlingame, CA, USA). Signal was visualized by multivariate analysis with backward selection. Differences at 3,3′-diaminobenzidine (Liquid DAB+ Substrate Chromogen P#0.05 were considered to be statistically significant. System; Dako Denmark A/S, Glostrup, Denmark). Nuclear counterstaining was performed with Gill’s hematoxylin. For Results immunohistochemical evaluation, three conventional catego- Association of AGR3 expression ries according to the number of positive cells were assessed: with other tumor variables 1 – negative/border (0%–5% of positive cells); 2 – weakly/ Due to the high homology between AGR2 and AGR3, protein moderately positive (5%–50% of positive cells); 3 – strongly specificity of the anti-AGR3 antibody was tested (Figure S1). positive (more than 50% of positive cells).26 The analyzed cohort composed of 95 (73.6%) tumors classified as ductal breast carcinomas, 18 (14%) as lobular type, Reverse transcription and quantitative and remaining 16 (12.4%) specimens were either of differ- PCR ent or unknown origin. The remaining clinicopathological Under the supervision of a pathologist, correspond- characteristics of the study group and their distributions are ing samples of tumor tissue were collected and used for summarized in Table 1. Staining of primary breast carcino- extraction of total cellular RNA by TRI Reagent (MRC, mas for AGR3 varied from tumor to tumor and was mainly Cincinnati, OH, USA). cDNA synthesis was carried out cytoplasmic. Overall, of the 129 cases, 25 (19.4%) were using the M-MLV reverse transcriptase (Thermo Fisher classified as negative or borderline stained for AGR3 (,5% Scientific, Waltham, MA, USA). Triplicate samples were of positive cells), and the remaining 104 (80.6%) showed subjected to quantitative polymerase chain reaction (PCR) AGR3 positivity to different degrees (from weak to strong) analysis using SYBR Green (Sigma-Aldrich, St Louis, MO, (Figure 1). Immunohistochemical staining for AGR3 was USA) for AGR2 and AGR3. The primer pairs used were then cross-tabulated with selected tumor features including as follows: for AGR2 – forward: 5′-GGAGCTCTATAT histological type, tumor size, nodal status, histological grade, AAATCCAAGACAAGCA-3′ and reverse: 5′-GCCAAT ER, PR, and Her2/neu status, and Ki-67 expression level. TTCTGGATTTCTTTATTTTC-3′; for AGR3 – forward: AGR3 positivity was significantly correlated with ductal type 5′-GCCTAGAATCATGTTTGTAGACC-3′ and reverse: and slowly proliferating tumors as measured by expression 5′-GCTTTCTTCATGTTTTCTATCAAT-3′. PCR was level of Ki-67 marker (P,0.0001) as well as lower tumor performed using default conditions: initial denaturation at grade (P,0.0001). Moreover, the degree of staining for

OncoTargets and Therapy 2015:8 submit your manuscript | www.dovepress.com 1525 Dovepress Obacz et al Dovepress

Figure 1 Immunohistochemical staining for AGR3. Notes: The level of AGR3 expression in primary breast carcinomas was determined by immunostaining in 3-point scale: (A) negative or border; (B) weak to moderate; and (C) strong. Scale bars represent a length of 100 µm. Abbreviation: AGR3, anterior gradient protein 3.

AGR3 was significantly associated with that for the ER AGR3 expression determined by immunohistoche (P,0.0001) and PR (P,0.0001). There was no significant mistry was also compared with AGR3 mRNA levels and correlation between AGR3 positivity and tumor size, nodal evaluated in relation to other clinicopathological variables. status, or Her2/neu status (Table 2). Interestingly, except Ki-67, whose elevated expression was

Table 2 Association of immunohistochemical staining for AGR3 with other tumor variables Variable N (%)a Statistical Patients AGR3 negative/border AGR3 weak/moderate AGR3 strong significance Histological grade G1 27 3 (11.1) 4 (14.8) 20 (74.1) ,0.0001b G2 43 3 (7) 7 (16.3) 33 (76.7) G3 56 18 (32.1) 14 (25) 24 (42.9) Tumor size b pT1 44 8 (18.2) 9 (20.4) 27 (61.4) 0.664

pT2 65 11 (16.9) 11 (16.9) 43 (66.2) 6 5 (33.3) 2 (13.3) 8 (53.3) pT3 + pT4 Nodal status Negative 45 10 (22.2) 10 (22.2) 25 (55.6) 0.332c Positive 73 13 (17.8) 10 (13.7) 50 (68.5) ER status Negative 29 20 (69) 8 (27.6) 1 (3.4) ,0.0001b Positive 100 5 (5) 17 (17) 78 (78) PR status Negative 34 19 (55.9) 9 (26.5) 6 (17.6) ,0.0001c Positive 94 6 (6.4) 16 (17) 72 (76.6) Her2/neu status Negative 92 18 (19.6) 15 (16.3) 59 (64.1) 0.318c Positive 36 7 (19.4) 10 (27.8) 19 (52.8) Ki-67 ,15% 55 6 (10.9) 9 (16.4) 40 (72.7) ,0.0001c $15% 61 14 (23.0) 14 (23.0) 33 (54.0) Notes: aNumber (percentage) of patients with tumors characterized by negative/border, weak/moderate, or strong expression of AGR3. bProbability, P, from Fisher’s exact test with the Freeman–Halton extension. cProbability, P, from Pearson’s chi-squared test. Abbreviations: AGR3, anterior gradient protein 3; ER, estrogen receptor; PR, progesterone receptor; Her2/neu, human epidermal growth factor receptor 2.

1526 submit your manuscript | www.dovepress.com OncoTargets and Therapy 2015:8 Dovepress Dovepress Significance of AGR3 in breast cancer associated predominantly with negative or weak AGR3 (data not shown), for further statistical analyses, the above expression (Table 2), we found similar trends for AGR3 on subgroups were combined (further denoted as AGR3 “low”) both protein and mRNA level in relation to other clinico- and were compared with patients whose tumors showed pathological parameters (Tables 2 and S1). strong AGR3 positivity (more than 50% of stained cells, We also examined AGR2 mRNA levels under the same denoted as AGR3 “high”). While OS was not significantly parameters and found almost similar association between affected by AGR3 expression, despite the fact that Kaplan– AGR2 gene expression and clinicopathological variables as Meier curves indicated some trend in favor of increased seen for AGR3 (Table S1). In line with these observations, AGR3 expression (P=0.111), these patients had significantly we also confirmed a strong correlation between AGR2 and longer PFS (P=0.037) (Figure 2). AGR3 mRNA levels (P,0.0001, R=0.6327) according to Spearman Rank Order correlation. On the other hand, we Association of AGR3 and other tumor also observed several statistically significant differences variables with patient survival in the association between AGR2 expression and clinico- As expected, patients with larger tumor size, higher histologi- pathological variables with respect to AGR3 indicating that cal grade, positive nodal status, and positive Her2/neu status the expression of these genes is similar but not identical. had significantly poorer prognosis at 10 years of follow-up The evaluation of AGR2 and AGR3 mRNA levels revealed (Table S2). For multivariate survival analysis, the follow- only marginal correlation of AGR2 mRNA levels with ER ing clinicopathological parameters were included in Cox’s (P=0.083) in comparison with AGR3 and ER (P,0.001). model with backward selection: histological type, tumor In accordance with immunohistochemical staining (P=0.003), grade, tumor size, nodal status, and ER, PR, Her2/neu, and determination of AGR3 transcription levels showed signifi- AGR3 status. As a result, tumor size and Her2/neu status were cant association (P=0.037) with grade as well. Conversely, found to be independent prognostic factors for PFS, whereas determination of AGR2 mRNA levels did not show this trend tumor size and grade reached statistical significance for OS (P=0.166; Table S1). time in the studied cohort (Table 3). The remaining clinical and histological characteristics, including AGR3, failed Association of AGR3 with patient survival statistical significance and were removed from the analysis For the survival analysis, follow-up was determined for during the selection process. When further pairwised with 10 years since surgical removal. Median PFS was 92 months other variables (Table S3), AGR3 positivity was associated (range 1–120), and median OS was 103 months (range with better outcome in the subgroup of patients with tumors 1–120). As there was almost no difference in survival curves defined by smaller histological grade (G#2; OS: P=0.005; between negative/border and weak/moderate subgroups PFS: P=0.024) but not by higher histological grade (G.2;

$ %

2YHUDOOVXUYLYDO 3URJUHVVLRQIUHHVXUYLYDO

)ROORZXS PRQWKV )ROORZXS PRQWKV

$*5³KLJK´ $*5³ORZ´

Figure 2 Association of immunohistochemical staining for AGR3 with patient survival. Notes: (A) Determination of progression-free survival by Kaplan–Meier analysis in patients with “high” AGR3 expression (more than 50% of positive cells) and patients with “low” AGR3 expression (less than 50% of positive cells) using Breslow test (P=0.037). (B) Determination of overall survival by Kaplan–Meier analysis in patients with “high” AGR3 expression and patients with “low” AGR3 expression using Breslow test (P=0.111). Abbreviation: AGR3, anterior gradient protein 3.

OncoTargets and Therapy 2015:8 submit your manuscript | www.dovepress.com 1527 Dovepress Obacz et al Dovepress

Table 3 Independent prognostic factors for the analyzed set of binds to metastasis-associated GPI-anchored C4.4a protein tumors according to Cox’s multivariate survival analysis and extracellular alpha-dystroglycan (DAG-1)7 and mediates Variable HR 95% CI Statistical resistance to cisplatin in mouse xenograft model,19 providing significance clear evidence for its important involvement in tumor bio Progression-free survival logy. In our descriptive study, we analyzed a cohort of 129 pT 1.00 0.003 1 primary breast carcinomas in order to assess clinical and pT2 1.99 0.83–4.73 0.121 prognostic relevance of AGR3 expression. We have detected pT3 8.25 2.68–25.44 ,0.0001 Her2/neu status 3.60 1.64–7.88 0.001 AGR3 in 104 (80%) out of 129 specimens, hence confirming Overall survival previously reported predominant expression of AGR3 pro- pT 1.00 0.006 7,16 1 tein in breast tumors. In the analyzed group, AGR3 was pT 1.38 0.51–3.76 0.531 2 significantly associated with ER and PR positivity and tumor pT 13.56 3.06–60.04 0.001 3 grades G#2 but not with tumor size and nodal status, which G1 1.0 0.015 7,8,30 G2 1.87 0.20–17.34 0.582 is consistent with other studies. Moreover, we observed G3 6.38 0.82–49.42 0.076 that increase in AGR3 positivity negatively correlated with Abbreviations: HR, hazard ratio; CI, confidence interval; Her2/neu, human the proliferation rate defined by the level of Ki-67 expression. epidermal growth factor receptor 2. Notably, similar trends in relation to other clinicopathological parameters were also found for AGR3 mRNA level. Correla- OS: P=0.583; PFS: P=0.945). In Her2/neu-negative set of tion with ER and PR positivity and slowly proliferating and tumors, AGR3 expression significantly correlated with lon- well-differentiated tumors suggests that AGR3 expression is ger PFS (P=0.019) as well as OS (P=0.009). On the other associated with less aggressive tumors that are more prone hand, when ER-positive cases were considered separately, to effective treatment and therefore favorable outcome. AGR3 expression did not reach statistical significance for Indeed, in our work, we demonstrated for the first time that improved survival (for PFS: P=0.228; for OS: P=0.234). the presence of immunohistochemical staining for AGR3 Therefore, the subgroup of ER- and PR-negative patients is associated with improved patient PFS. Although, in the was extended to determine the impact of AGR3 on patients’ whole cohort, AGR3 expression did not predict longer OS, outcome. However, within the additional ER-negative group patients whose tumors were characterized by strong AGR3 of 90 patients, no significant association between AGR3 positivity showed better response to therapy. Moreover, expression and patient outcome was observed as well with AGR3 predicted better outcome in the subgroup of patients regard to both PFS and OS (P=0.282 and P=0.867, respec- with well-differentiated tumors, which is consistent with tively; Figure S2). Statistical analysis of AGR3 IHC staining previously demonstrated significance of AGR3 expression patterns with other clinicopathological parameters in cohort in ovarian cancers.20 Quite the contrary, AGR2 is often of ER- and PR-negative breast tumors revealed significant described as an indicator of poor prognosis,8,9 metastasis,15,31 association between AGR3 expression and presence and and resistance to commonly used treatments,10,32 indicating Her2/neu status only (Table S4). divergent and/or context-dependent roles of AGR proteins in breast cancer. It is of note that similar antagonistic impact Discussion of AGR proteins on patient outcome is also observed in AGR2 and AGR3 are conserved human homologues of ovarian cancers where AGR3 promotes better outcome,20 X. laevis XAG-2 protein implicated in development and whereas AGR2 predicts shortened OS,33 possibly due to the regeneration.2 AGR2 and AGR3 share high-sequence homo stimulation of cell growth and migration.34 However, given logy, localize to the same chromosomal position 7p21,7 and that AGR3 was also shown to mediate cisplatin resistance, both respond to estrogen12 and androgen stimulation,21,22 an explicit conclusion of AGR3-protective, antitumor role which suggests their possible functional overlap. AGR2 is cannot be conclusively drawn. Moreover, in our recent work, a well-studied pro-oncogene, promoting aggressive tumor we have compared AGRs distribution both in human healthy phenotype and less favorable patient outcome in various tissues and carcinomas using Genevestigator platform,35 and malignancies.27–29 On the other hand, AGR3 function in we found that AGR3 mirrors AGR2 expression in many cases, health and disease remains ambiguous, since data published such as stomach, colon, pancreas, breast, female reproduc- so far are relatively contradictory. AGR3 expression was tive system, or respiratory system.36 In accordance, here, we demonstrated in various cancers, including breast,7 prostate,21 have demonstrated strong correlation between AGR2 and ovary,19,20 and liver.18 Moreover, it was shown that AGR3 AGR3 mRNA levels in breast carcinomas as well as similar

1528 submit your manuscript | www.dovepress.com OncoTargets and Therapy 2015:8 Dovepress Dovepress Significance of AGR3 in breast cancer associations of both genes with clinicopathological variables, Acknowledgments which suggests their cognate physiological function and role We thank Pavlina Zatloukalova, PhD, for testing the antibody in pathological conditions. cross-reactivity. The work was supported by MH CZ – DRO In the present work, we observed that better outcome in (MMCI, 00209805), European Regional Development Fund AGR3-positive group was independent of ER status (consi and the State Budget of the Czech Republic for Regional dered separately, neither ER-positive nor negative-subgroups Centre for Applied Molecular Oncology – RECAMO had significantly longer survival time when pairwised with (CZ.1.05/2.1.00/03.0101), the projects MEYS-NPS I-LO1413 AGR3). These findings suggest more complex control of GACR 13-00956S, and IGA NT/13794-4/2012. AGR3 expression in breast carcinomas, not solely dependent on ER, similarly to that of AGR2.8 Thus, some clues Disclosure regarding AGR3 regulation could be derived from the studies The authors declare that they have no competing interest. focusing on AGR2 homologue. For instance, in addition to ER, AGR2 was reported to be a component of, among others, References EGFR, cyclin D1, survivin, AKT, and transforming growth 1. Ferlay J, Steliarova-Foucher E, Lortet-Tieulent J, et al. Cancer incidence factor-beta signaling pathway.14,29,37,38 However, mechanisms and mortality patterns in Europe: estimates for 40 countries in 2012. Eur J Cancer. 2013;49(6):1374–1403. triggering expression of AGR2 and AGR3 could be relatively 2. Sive HL, Hattori K, Weintraub H. Progressive determination during unrelated as manifested by the uncoupled expression of formation of the anteroposterior axis in Xenopus laevis. Cell. 1989; 7,19 58(1):171–180. both proteins in prostate and ovarian cancers, and thus, 3. Ivanova AS, Tereshina MB, Ermakova GV, Belousov VV, Zaraisky AG. further in vitro and in vivo studies are warranted to under- Agr genes, missing in amniotes, are involved in the body appendages stand AGR3’s function(s) in tumor biology. Relying on our regeneration in frog tadpoles. Sci Rep. 2013;3:1279. 4. Hatahet F, Ruddock LW. Protein disulfide isomerase: a critical evalua- in silico analyses, we have recently shown that AGR2 and tion of its function in disulfide bond formation.Antioxid Redox Signal. AGR3 plausibly control similar aspects of tumor biology 2009;11(11):2807–2850. 5. Thompson DA, Weigel RJ. hAG-2, the human homologue of the Xeno- including cell cycle control, differentiation, migration, inva- pus laevis cement gland gene XAG-2, is coexpressed with estrogen sion, and metastasis.36 Additionally, we performed promoter receptor in breast cancer cell lines. Biochem Biophys Res Commun. analysis and demonstrated that most of the transcription 1998;251(1):111–116. 6. Wilson CL, Sims AH, Howell A, Miller CJ, Clarke RB. Effects of factors potentially binding to AGR2 or AGR3 promoters are oestrogen on gene expression in epithelium and stroma of normal human exclusive for each protein,36 which could partially elucidate breast tissue. Endocr Relat Cancer. 2006;13(2):617–628. 7. Fletcher GC, Patel S, Tyson K, et al. hAG-2 and hAG-3, human their uncoupled expression. One possible explanation of homologues of genes involved in differentiation, are associated with observed AGR3 ambiguity is that dependent on the cellular oestrogen receptor-positive breast tumours and interact with metastasis context, it could support different phenotypes leading either gene C4.4a and dystroglycan. Br J Cancer. 2003;88(4):579–585. 8. Innes HE, Liu D, Barraclough R, et al. Significance of the metastasis- to tumor progression or to regression. inducing protein AGR2 for outcome in hormonally treated breast cancer In the light of what has been reported to date, it would patients. Br J Cancer. 2006;94(7):1057–1065. 9. Barraclough DL, Platt-Higgins A, de Silva Rudland S, et al. The be necessary not only to verify whether AGR3 plays tumor- metastasis-associated anterior gradient 2 protein is correlated with poor suppressive or tumor-promoting role but also to evaluate survival of breast cancer patients. Am J Pathol. 2009;175(5):1848–1857. the plausible relevance of AGR3 presence in patient’s 10. Hrstka R, Nenutil R, Fourtouna A, et al. The pro-metastatic protein anterior gradient-2 predicts poor prognosis in tamoxifen-treated breast fluids. AGR3 was firstly characterized in breast cancer cancers. Oncogene. 2010;29(34):4838–4847. cell membranes and was found to localize in secretory or 11. Fullwood MJ, Liu MH, Pan YF, et al. An oestrogen-receptor-alpha-bound human chromatin interactome. Nature. 2009;462(7269):58–64. endosome-like vesicles in both T47-D and MDA-MB-468 12. Welboren WJ, van Driel MA, Janssen-Megens EM, et al. ChIP-Seq of cells,16 suggesting more prominent role of secreted form of ERalpha and RNA polymerase II defines genes differentially responding AGR3. Indeed, recent works have depicted emerging role of to ligands. EMBO J. 2009;28(10):1418–1428. 13. Verma S, Salmans ML, Geyfman M, et al. The estrogen-responsive extracellular AGR2 in the control of tumor aggressiveness Agr2 gene regulates mammary epithelial proliferation and facilitates through both autocrine and paracrine effects,39,40 indicating lobuloalveolar development. Dev Biol. 2012;369(2):249–260. 14. Vanderlaag KE, Hudak S, Bald L, et al. Anterior gradient-2 plays a critical that similar mechanism can also be valid for AGR3. Lastly, role in breast cancer cell growth and survival by modulating cyclin D1, taking into account cognate expression pattern of AGR estrogen receptor-alpha and survivin. Breast Cancer Res. 2010; proteins in different carcinomas,7,19 it can be speculated 12(3):R32. 15. Liu D, Rudland PS, Sibson DR, Platt-Higgins A, Barraclough R. Human that there is a functional cross talk between these proteins. homologue of cement gland protein, a novel metastasis inducer associ- However, whether they compete with each other, compen- ated with breast carcinomas. Cancer Res. 2005;65(9):3796–3805. 16. Adam PJ, Boyd R, Tyson KL, et al. Comprehensive proteomic analysis sate for one’s lost, or support one another requires further of breast cancer cell membranes reveals unique proteins with potential investigation. roles in clinical cancer. J Biol Chem. 2003;278(8):6482–6489.

OncoTargets and Therapy 2015:8 submit your manuscript | www.dovepress.com 1529 Dovepress Obacz et al Dovepress

17. Petek E, Windpassinger C, Egger H, Kroisel PM, Wagner K. Localiza- 29. Salmans ML, Zhao F, Andersen B. The estrogen-regulated anterior tion of the human anterior gradient-2 gene (AGR2) to chromosome band gradient 2 (AGR2) protein in breast cancer: a potential drug target and 7p21.3 by radiation hybrid mapping and fluorescencein situ hybridisa- biomarker. Breast Cancer Res. 2013;15(2):204. tion. Cytogenet Cell Genet. 2000;89(3–4):141–142. 30. Fritzsche FR, Dahl E, Pahl S, et al. Prognostic relevance of AGR2 expres- 18. Brychtova V, Zampachova V, Hrstka R, et al. Differential expression sion in breast cancer. Clin Cancer Res. 2006;12(6):1728–1734. of anterior gradient protein 3 in intrahepatic cholangiocarcinoma and 31. Smirnov DA, Zweitzig DR, Foulk BW, et al. Global gene expres- hepatocellular carcinoma. Exp Mol Pathol. 2014;96(3):375–381. sion profiling of circulating tumor cells. Cancer Res. 2005;65(12): 19. Gray TA, MacLaine NJ, Michie CO, et al. Anterior gradient-3: 4993–4997. a novel biomarker for ovarian cancer that mediates cisplatin resistance 32. Hrstka R, Brychtova V, Fabian P, Vojtesek B, Svoboda M. AGR2 in xenograft models. J Immunol Methods. 2012;378(1–2):20–32. predicts tamoxifen resistance in postmenopausal breast cancer patients. 20. King ER, Tung CS, Tsang YT, et al. The anterior gradient homolog 3 Dis Markers. 2013;35(4):207–212. (AGR3) gene is associated with differentiation and survival in ovarian 33. Darb-Esfahani S, Fritzsche F, Kristiansen G, et al. Anterior gradient cancer. Am J Surg Pathol. 2011;35(6):904–912. protein 2 (AGR2) is an independent prognostic factor in ovarian high- 21. Vaarala MH, Hirvikoski P, Kauppila S, Paavonen TK. Identification grade serous carcinoma. Virchows Arch. 2012;461(2):109–116. of androgen-regulated genes in human prostate. Mol Med Rep. 2012; 34. Park K, Chung YJ, So H, et al. AGR2, a mucinous ovarian cancer marker, 6(3):466–472. promotes cell proliferation and migration. Exp Mol Med. 2011;43(2): 22. Bu H, Schweiger MR, Manke T, et al. Anterior gradient 2 and 3 – two 91–100. prototype androgen-responsive genes transcriptionally upregulated by 35. Hruz T, Laule O, Szabo G, et al. Genevestigator v3: a reference expres- androgens and by oestrogens in prostate cancer cells. FEBS J. 2013; sion database for the meta-analysis of transcriptomes. Adv Bioinformat- 280(5):1249–1266. ics. 2008;2008:420747. 23. Tavassoli FA, Devilee P. WHO Classification of Tumours. Pathology 36. Obacz J, Takacova M, Brychtova V, et al. The role of AGR2 and AGR3 and Genetics of Tumours of the Breast and Female Genital Organs. in cancer: similar but not identical. Eur J Cell Biol. 2015;94(3–4): editors ed. Lyon, France: IARC Press; 2003. 139–147. 24. Sobin LH, Fleming ID. TNM classification of malignant Tumors, fifth 37. Norris AM, Gore A, Balboni A, Young A, Longnecker DS, Korc M. edition (1997). Union Internationale Contre le Cancer and the American AGR2 is a SMAD4-suppressible gene that modulates MUC1 levels Joint Committee on Cancer. Cancer. 1997;80(9):1803–1804. and promotes the initiation and progression of pancreatic intraepithelial 25. Elston CW, Ellis IO. Pathological prognostic factors in breast cancer I. neoplasia. Oncogene. 2013;32(33):3867–3876. The value of histological grade in breast cancer: experience from a 38. Hrstka R, Murray E, Brychtova V, Fabian P, Hupp TR, Vojtesek B. large study with long-term follow-up. Histopathology. 1991;19(5): Identification of an AKT-dependent signalling pathway that mediates 403–410. tamoxifen-dependent induction of the pro-metastatic protein anterior 26. Taiseer IR, Samar ARM, Abdelmonem HA. Immunohistochemi- gradient-2. Cancer Lett. 2013;333(2):187–193. cal expression of aldehyde dehydrogenase-1 and hypoxia-inducible 39. Tsuji T, Satoyoshi R, Aiba N, et al. Agr2 mediates paracrine effects factor-1α in breast cancer. IJAR. 2014;2(7):822–830. on stromal fibroblasts that promote invasion by gastric signet-ring 27. Brychtova V, Vojtesek B, Hrstka R. Anterior gradient 2: a novel player carcinoma cells. Cancer Res. 2015;75(2):356–366. in tumor cell biology. Cancer Lett. 2011;304(1):1–7. 40. Arumugam T, Deng D, Bover L, Wang H, Logsdon CD, Ramachandran V. 28. Chevet E, Fessart D, Delom F, et al. Emerging roles for the pro- New blocking antibodies against novel AGR2-C4.4A pathway reduce oncogenic anterior gradient-2 in cancer development. Oncogene. 2013; growth and metastasis of pancreatic tumors and increase survival in 32(20):2499–2509. mice. Mol Cancer Ther. 2015;14(4):941–951.

1530 submit your manuscript | www.dovepress.com OncoTargets and Therapy 2015:8 Dovepress Dovepress Significance of AGR3 in breast cancer

Supplementary materials Table S2 Univariate survival analyses for the analyzed set of tumors Variablea Statistical significanceb PFS OS AGR3 expression 0.037 0.111 Histology + $ 7' $*5SURWHLQ$*5SURWHLQ DU vs LO 0.665 0.393 $*5 DU vs OTH 0.537 0.980 LO vs OTH 0.851 0.675 $*5 Grade G1 vs G2 0.159 0.137 Figure S1 Determination of anti-AGR3 antibody cross-reactivity. G1 vs G3 0.002 0.009 Notes: The specificity of our in-house anti-AGR3 antibody (AGR3.1) was confirmed G2 vs G3 0.020 0.055 by Western blot (upper panel). For comparison, we added testing of rabbit polyclonal sera raised against AGR2 protein, which recognizes AGR3 as well (bottom panel). Tumor size Abbreviation: AGR, anterior gradient protein. pT1 vs pT2 0.249 0.333 pT1 vs pT3 0.000 0.000 pT2 vs pT3 0.000 0.000 Nodal status 0.043 0.415 ER status 0.110 0.344 PR status 0.023 0.145 Her2/neu status 0.000 0.000 Ki-67 expression 0.004 0.018 Table S1 Association of AGR2 and AGR3 mRNA levels with Notes: aAGR3 expression, AGR3 “low” (less than 50% of stained cells) vs AGR3 other tumor variables “high” (more than 50% of stained cells); histology, ductal vs lobular vs others; nodal status, negative vs positive; estrogen receptor status, negative vs positive; Patients (n) AGR2 mRNA AGR3 mRNA progesterone receptor status, negative vs positive; Her2/neu status, negative vs Histological grade positive; Ki-67 expression, ,15% vs $15%. bProbability, P, from Breslow test. G1 26 0.166a 0.037a Abbreviations: PFS, progression-free survival; OS, overall survival; AGR3, anterior gradient protein 3; DU, ductal; LO, lobular; OTH, others; ER, estrogen receptor; G2 33 PR, progesterone receptor; Her2/neu, human epidermal growth factor receptor 2. G3 34 Tumor size a a pT1 35 0.774 0.990 pT 50 2 Table S3 Survival analysis of patients with AGR3-expressing pT + pT 8 3 4 tumors Nodal status a Negative 37 0.822b 0.541b Subgroup Statistical significance Positive 52 PFS OS ER status Histological grade b b Negative 17 0.081 ,0.001 G#2 0.024 0.005 Positive 76 G.2 0.945 0.583 PR status Her2/neu status b b Negative 19 0.124 ,0.001 Negative 0.019 0.009 Positive 73 Positive 0.781 0.278 Her2/neu status PR status b b Negative 57 0.364 0.603 Negative 0.669 0.911 Positive 36 Positive 0.448 0.224 Ki-67 ER status b b ,15% 44 0.169 0.494 Negative 0.431 0.507 $15% 48 Positive 0.228 0.234 Notes: aDetermination of P-level using Kruskal–Wallis analysis of variance. bDetermination Note: aProbability, P, from Breslow test. of P-level using Mann–Whitney U-test. Abbreviations: AGR3, anterior gradient protein 3; PFS, progression-free survival; Abbreviations: AGR, anterior gradient protein; ER, estrogen receptor; PR, proges- OS, overall survival; Her2/neu, human epidermal growth factor receptor 2; ER, estrogen terone receptor; Her2/neu, human epidermal growth factor receptor 2. receptor; PR, progesterone receptor.

OncoTargets and Therapy 2015:8 submit your manuscript | www.dovepress.com 1531 Dovepress Obacz et al Dovepress

$ %

2YHUDOOVXUYLYDO

3URJUHVVLRQIUHHVXUYLYDO

)ROORZXS PRQWKV )ROORZXS PRQWKV

$*5³KLJK´ $*5³ORZ´

Figure S2 Survival analysis of cohort of ER- and PR-negative breast cancer patients. Notes: Kaplan–Meier analysis of (A) progression-free survival in relation to AGR3 expression (P=0.282, Breslow test) and (B) overall survival in relation to AGR3 expression (P=0.867, Breslow test). Abbreviations: ER, estrogen receptor; PR, progesterone receptor; AGR3, anterior gradient protein 3.

Table S4 Association of immunohistochemical staining for AGR3 with other tumor variables in a cohort of ER-negative breast cancer patients Variable N (%)a Statistical b Patients AGR3 negative/border AGR3 weak/moderate AGR3 strong significance Histological grade G1 1 1 (100.0) 0 (0.0) 0 (0.0) 0.917 G2 10 6 (60.0) 3 (30.0) 1 (10.0) G3 71 47 (66.2) 15 (21.1) 9 (12.7) Tumor size

pT1 32 24 (75.0) 5 (15.6) 3 (9.4) 0.729

pT2 44 26 (59.1) 11 (25.0) 7 (15.9) 11 8 (72.8) 2 (18.2) 1 (9.0) pT3 + pT4 Nodal status Negative 34 25 (73.5) 7 (20.6) 2 (5.9) 0.282 Positive 53 31 (58.5) 13 (24.5) 9 (17.0) Her2/neu status Negative 40 34 (85.0) 5 (12.5) 1 (2.5) 0.001 Positive 49 24 (49.0) 15 (30.6) 10 (20.4) Ki-67 ,15% 6 4 (66.7) 2 (33.3) 0 (0.0) 0.626 $15% 42 30 (71.4) 7 (16.7) 5 (11.9) Notes: aNumber (percentage) of patients with tumors characterized by negative/border, weak/moderate, or strong expression of AGR3. Probability, bP, was calculated using Fisher’s exact test with the Freeman–Halton extension. Abbreviations: AGR3, anterior gradient protein 3; ER, estrogen receptor; Her2/neu, human epidermal growth factor receptor 2.

OncoTargets and Therapy Dovepress Publish your work in this journal OncoTargets and Therapy is an international, peer-reviewed, open patient perspectives such as quality of life, adherence and satisfaction. access journal focusing on the pathological basis of all cancers, potential The manuscript management system is completely online and includes targets for therapy and treatment protocols employed to improve the a very quick and fair peer-review system, which is all easy to use. Visit management of cancer patients. The journal also focuses on the impact http://www.dovepress.com/testimonials.php to read real quotes from of management programs and new therapeutic agents and protocols on published authors.

Submit your manuscript here: http://www.dovepress.com/oncotargets-and-therapy-journal

1532 submit your manuscript | www.dovepress.com OncoTargets and Therapy 2015:8 Dovepress 11.3 Attachment 3

OBACZ, J., PASTOREKOVA, S., VOJTESEK, B. & HRSTKA, R. 2013. Cross-talk between HIF and p53 as mediators of molecular responses to physiological and genotoxic stresses. Mol Cancer, 12 , 93.

100

Obacz et al. Molecular Cancer 2013, 12:93 http://www.molecular-cancer.com/content/12/1/93

REVIEW Open Access Cross-talk between HIF and p53 as mediators of molecular responses to physiological and genotoxic stresses Joanna Obacz1,2, Silvia Pastorekova2, Borek Vojtesek1 and Roman Hrstka1*

Abstract Abnormal rates of growth together with metastatic potential and lack of susceptibility to cellular signals leading to apoptosis are widely investigated characteristics of tumors that develop via genetic or epigenetic mechanisms. Moreover, in the growing tumor, cells are exposed to insufficient nutrient supply, low oxygen availability (hypoxia) and/or reactive oxygen species. These physiological stresses force them to switch into more adaptable and aggressive phenotypes. This paper summarizes the role of two key mediators of cellular stress responses, namely p53 and HIF, which significantly affect cancer progression and compromise treatment outcomes. Furthermore, it describes cross-talk between these factors. Keywords: p53, HIF-1, Hypoxia, DNA damage, Cancer

HIF-mediated responses to hypoxia and HIF-2α. These subunits contain similar oxygen- Important consequences of rapid tumor growth include dependent degradation domains, but play different roles in poor vascularization and insufficient oxygen delivery that hypoxic tumor growth and progression (for extended re- together lead to formation of hypoxic (poorly oxygenated) view see Keith et al.) [4]. Whereas HIF-1 mediates acute areas [1]. Adaptation to hypoxia is facilitated by the activa- responses to hypoxia, HIF-2 is more involved in adaptation tion of transcriptional machinery, in which hypoxia indu- to chronic hypoxia and is functionally implicated in tumor cible factor (HIF) plays a pivotal role. HIF is a progression [5]. heterodimeric transcription factor composed of an oxygen- In situations of insufficient oxygen levels, PHDs and dependent α-subunit and constitutively expressed β- FIH remain inactive, while HIF-1α is no longer hydrox- subunit. Regulation of the α-subunit is driven by enzymes ylated and escapes recognition by pVHL. This results in of the prolyl hydroxylase family (PHDs) and by the factor its stabilization, accumulation and translocation to the inhibiting HIF (FIH) [2,3]. Under normoxia, PHDs hydrox- nucleus, where it interacts with a β-subunit leading to ylate prolines at positions 564 and 402 (in HIF-1α isoform) creation of an active heterodimeric form of the tran- and FIH hydroxylates asparagine at position 803 [3]. Hy- scription factor. This heterodimer binds to specific cis- droxylation of prolines is required for recognition of HIF- acting hypoxia responsive elements (HREs) in the pro- 1α by the ubiquitin ligase complex via von Hippel-Lindau moters of target genes [6]. (pVHL) tumor suppressor protein, which in consequence Several recent reports point out novel molecular leads to HIF-1α ubiquitination followed by its proteasomal mechanisms that affect HIF-1α levels in normoxia. An degradation [2]. Simultaneously, FIH prevents interaction inhibitor of Janus Activated Kinase (JAK2), AG490, pre- between HIF-1α and the transcriptional co-activator, p300. vents HIF-1α hydroxylation and thus interferes with Although there are three isoforms of the α-subunit: HIF- VHL-mediated degradation resulting in increased HIF- 1α,HIF-2α and HIF-3α, most attention is drawn to HIF-1α 1α protein half-life [7]. Another mechanism by which HIF-1α can be rescued from degradation is via inter- * Correspondence: [email protected] action with ubiquitin-specific protease 19 (USP19) [8]. 1Masaryk Memorial Cancer Institute, Regional Centre for Applied Molecular Epigenetic mechanisms such as histone methylation can Oncology, Zluty kopec 7, 65653 Brno, Czech Republic also be involved in HIF-1α regulation, which was studied Full list of author information is available at the end of the article

© 2013 Obacz et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Obacz et al. Molecular Cancer 2013, 12:93 Page 2 of 10 http://www.molecular-cancer.com/content/12/1/93

in clear cell renal cell carcinoma (ccRCC) [9]. Moreover, senescence [27-29]. Comparably to HIF-1α, the basal HIF-1 activity is phosphorylation-dependent and thus level of wild-type p53 is kept low due to murine double requires engagement of signaling such as mitogen- minute 2 (MDM2)-dependent ubiquitination [30]. In re- activated protein kinase (MAPK), PI3K/Akt and mam- sponse to DNA damage p53 is stabilized and phosphory- malian target of rapamycin (mTOR), amongst others lated by ataxia telangiectasia mutated (ATM) protein, (see review by Dimova et al.) [10]. which leads to its activation and binding to the regula- HIF-induced cascades of events allow cells to survive tory region of target genes [31,32]. Moreover, p53 can be and overcome unfavorable conditions during hypoxia by regulated through methylation caused by MDM2- transcriptional reprogramming that leads to modulated dependent recruitment of methyltransferases [32]. In proliferation, angiogenesis, cell metabolism and many contrast, MDM2 can also act as a p53 inducer. This is other features of tumor phenotype. One of the promin- mediated through the interaction of p53 mRNA region ent HIF-1 downstream genes involved in this process is containing the MDM2-binding site with the RING do- the gene coding for carbonic anhydrase IX (CA IX). CA main of MDM2, which impairs the E3 ligase activity of IX is a member of the family of zinc metalloenzymes in- MDM2 and promotes p53 mRNA translation [33]. This volved in regulation of cellular pH by reversible conver- interaction depends on ATM-mediated phosphorylation sion of CO2 to bicarbonate and proton [11,12]. Its of MDM2 at Ser395 [34]. Finally, activated p53 can then activity is regulated by hypoxia through protein kinase A start the machinery leading either to cell cycle arrest and and leads to acidosis of the tumor milieu, which is DNA repair or to apoptosis. For example, p53- known to be one of the hallmarks of solid tumors dependent upregulation of genes involved in inhibition [13,14]. CA IX also promotes tumor cell growth and sur- of IGF-1/AKT and mTOR pathways prevents cell growth vival and helps to eliminate the surplus of intracellular and division [29,35,36]. On the other hand, inhibition of acids generated through oncogenic metabolism [15,16]. DNA damage-activated kinases leads to switch of the Moreover, it facilitates migration and invasiveness of p53-dependent growth arrest to apoptosis [37]. tumor cells and thereby supports tumor progression ATF3 gene, a downstream target of p53, encodes a [17]. transcription factor involved in adaptation to hypoxia, To satisfy the need for nutrients, tumor cells are ER stress, oxidative stress and genotoxic stress [38]. forced to create an extensive net of new vessels through ATF3 acts both as an effector of p53-mediated cell death increased expression of pro-angiogenic molecules, in- and a regulator of p53 signaling. A recent report indi- cluding vascular endothelial growth factor (VEGF), cates that ATF3 has opposing effects on apoptotic tran- which is also a well-known HIF target gene [18,19]. scriptome in stress response and in cancer, where it was Additionally, VEGF can promote both angiogenesis and found to be over-expressed [39]. Zhang and colleagues metastasis via up-regulation of matrix metalloproteinase [40] developed a four-module model to investigate p53 28 and matrix metalloproteinase 14 [20]. dynamics and the DNA damage response. They found Due to lack of oxygen, a key factor for respiration, that primary modifications such as phosphorylation at hypoxia is also known to induce a shift to glycolytic me- Ser-15 and Ser-20 cause cell cycle arrest, whereas further tabolism [21]. HIF-1 plays a growth factor-dependent modifications such as phosphorylation at Ser-46 fully ac- role in the regulation of glycolysis in hematopoietic cells tivate p53 which can then induce apoptosis. This report even in the absence of hypoxia [22] and reduces mito- more clearly elucidates how p53 converts between the chondrial respiration in RCC lacking VHL [23]. HIF was cell cycle arrester and the killer, which was previously also shown to be responsible for expression of specific shown to be controlled by Wip1 (wild-type p53-induced isoforms of glycolytic enzymes and transporters via alter- phosphatase 1) [41]. native splicing [24]. p53 does not only act as a transcriptional factor in the There are many other molecular targets of HIF that nucleus, but also can move to the mitochondria where it execute multiple adaptive responses to hypoxia depend- induces permeabilization of the mitochondrial outer ing on the cell type and physiological context as de- membrane consequently releasing pro-apoptotic factors scribed elsewhere [25,26]. [28]. Suppression of autophagy via inhibition of AMP- dependent kinase and/or activation of mTOR is another p53-mediated responses to genotoxic stress cytoplasmic p53 function [42]. For the extensive insight Tumor suppressor p53, which shows many similarities into the cytoplasmic functions of p53, see the review by to HIF-1 in terms of protein control by degradation, is Green and Kroemer [28]. predominantly involved in adaptation of cells to p53 as a tumor suppressor plays an important role in genotoxic stresses. p53 is a well-characterized transcrip- maintaining of genome stability thus it is not surprising tion factor that plays a crucial role in responses to DNA that is mutated in more than 50% of cancers in which its damage, aberrant cell cycle control, apoptosis, and loss facilitates malignant transformation [43]. The Obacz et al. Molecular Cancer 2013, 12:93 Page 3 of 10 http://www.molecular-cancer.com/content/12/1/93

majority of p53 mutations represent missense mutations between p53 and RPA70, dissociation of RPA70 and acti- located in the DNA-binding core domain of p53, produ- vation of RPA70-mediated nucleotide excision repair cing a full-length protein that is incapable of binding and non-homologous end-joining repair, which cause re- DNA and is therefore nonfunctional as a transcriptional sistance to apoptosis in hypoxic cancer cells [61]. That activator/repressor. Compared to wild-type p53, mis- report poses a new insight into impairment of the p53- sense mutant proteins show increased stability, which is mediated apoptosis and consequent insensitivity of can- partly caused by their inability to induce MDM2 but also cer cells to treatment. However, it is still hard to eluci- by the formation of complexes with HSP90 and HSP70 date what starts the p53 and/or HIF-1 machinery for the [44]. adaptation of cells to unfavorable conditions. Thomas et al. [62] focused on tumor response to nitric Cross-talk between HIF-1 and p53 oxide (NO) exposure and proposed that both p53 and In addition, p53 participates in responses to hypoxia by HIF-1 are stabilized by NO in a dose- and time- regulating expression of genes involved in cell cycle con- dependent manner, with a higher NO concentration re- trol. This happens via a pathway that is different than quired for p53 stabilization. They suggested that cells lo- that involved in the DNA damage response [45]. There calized closer to the source of NO production can are many contradictory reports on mutual influence of undergo p53-dependent cell arrest and death, while p53 and hypoxic signaling. Some of them claim that more distant cells respond with increased HIF-1 levels. hypoxia causes accumulation and increase in p53 protein Additionally, their results indicated that HIF-1 level [46,47], whereas others postulate degradation- stabilization by NO was independent of p53 status. mediated decrease in p53 level [48,49] or no effect at all Altered metabolism is one of the prominent features [50]. These intricate relations have been extensively that promote tumor survival. The first who discovered reviewed by Sermeus and Michiels [51]. One explanation that tumors rely on anaerobic glycolysis even in the of these contradictory statements can be found in the presence of sufficient oxygen and produce large amount phosphorylation status of HIF-1. It was shown that of lactate was Otto Warburg [63]. Later this dephosphorylated HIF-1 is a major form binding to p53, phenomenon was named after him. The consequences precluding downregulation of p53 by MDM-2, and thus of this effect have been previously reviewed [64]. An- enabling it to conduct apoptosis [52]. As both p53 and other tumor characteristic is increased uptake of nutri- HIF-1 are mediators of cell adaptation to many stresses, ents that as stated by Vander Heiden et al. [65] is due to they are known to be involved in similar processes such oncogenic mutations mainly in Akt, Myc and Ras [66]. as apoptosis, cell cycle control, metabolism etc. A multitude of mutations of genes encoding enzymes (Figure 1). Severe and/or prolonged hypoxia activates participating in glycolysis, tricarboxylic acid cycle, mito- p53-dependent apoptosis, which is initiated by chondrial oxidative phosphorylation and other molecular stabilization of 53 by HIF-1 [53]. In contrast, another re- pathways underlying the advantageous metabolism of port states that hypoxia causes growth arrest by decreas- cancers have been already characterized [67-71]. Com- ing p53 phosphorylation, but has no impact on either prehensive insights into this phenomenon can be found p21WAF1 or HIF-1 protein stabilization [54]. One of the in recent works [72-75]. In this respect HIF-1 and p53 possible explanations is that these convergences can be play crucial, but usually competing, roles. HIF-1 controls due to cancer cell type [55]. Opposite effects can be ob- expression of genes encoding e.g. glucose transporters, served upon genotoxic stress, where wild-type 53 abro- glycolytic enzymes, lactate dehydrogenase etc. [25,76]. gates HIF-1 activity triggering its proteasomal Interestingly, inactivating mutations in fumarate degradation [56]. hydratase and succinate dehydrogenase cause accumula- However, there is a line of evidence that HIF-1 can tion of their substrates, which interfere with HIF-1α deg- also impair p53 activity, through the downregulation of radation leading to its accumulation [77]. On the other the tumor suppressor homeodomain-interacting protein hand, loss of p53 contributes to enhancement of glucose kinase-2 (HIPK2) [57]. HIPK2 phosphorylates p53 at transport and metabolism through NF-κB pathway [78]. serine 46 in response to DNA damage and subsequently Furthermore, it increases lactate production, diminishes activates its apoptotic function [58]. Moreover, HIPK2 oxygen consumption and enhances hypoxia-induced cell inhibition can result from the hypoxia-induced death. Disruption of p53 function reduces the expression upregulation of MDM2 [59]. of cytochrome c oxidase 2 (SCO2), which is necessary p53 can respond to DNA damage in cooperation with for the respiratory chain function [79]. This indicates 70 kDa subunit of the replication protein A (RPA70). that mutations in the TP53 gene contribute to Warburg Under hypoxia, wild-type p53 undergoes a conform- effect. ational change and acquires mutant conformation [60]. In order to eliminate damaged proteins and organelles Furthermore, hypoxia leads to disruption of the complex as well as to fulfill requirements for high ATP level, Obacz et al. Molecular Cancer 2013, 12:93 Page 4 of 10 http://www.molecular-cancer.com/content/12/1/93

Figure 1 HIF-1 and/or p53 regulated genes mediating adaptation to cellular stresses through activation of different pathways. Upon hypoxia, the interaction between HIF-1α and von Hippel Lindau protein (pVHL) is disrupted, leading to HIF-1α translocation into nucleus, dimerization with HIF-1β subunit and formation of HIF-1 active form, which can regulate transcription of target genes . HIF-1 activates lactate dehydrogenase (LDH-A), pyruvate dehydrogenase kinase 1 (PDK1), phosphoglycerate mutase (PGM) and glucose transporter 1 (GLUT-1) to switch into more glycolytic phenotype [25]. To prevent apoptosis, it induces survivin expression [25] and downregulates BAX, BID and caspases activity [26]. HIF-1 can also induce autophagy by upregulation of beclin-1, BNIP3 and NIX [81]. Through modulating vascular endothelial growth factor (VEGF) [18], angiopioetin-2 (Ang-2) [25], carbonic anhydrase IX (CA IX) [12] and p21WAF1 [90] expression, HIF-1 triggers activation of pro-survival pathways. Different molecular stresses (including DNA damage, hypoxia, oxidative stress), cause dissociation of p53 from murine double minute 2 (MDM2) complex, enabling its binding to regulatory elements of target genes [31]. Thereby p53 can repress glycolysis by altering expression of GLUT-1, PGM, TP53-induced glycolysis and apoptosis regulator (TIGAR) and inhibits pentose phosphate pathway by downregulating glucose-6 -phosphate dehydrogenase (G6PDH) [36]. p53 regulates expression of many pro-apoptotic proteins, including PUMA, NOXA, CD95, Apaf1, BAX, BID and caspases [28]. Induction of autophagy by p53 relies on activation of damage-regulated autophagy modulator (DRAM) [83], sestrin 1, sestrin 2 and AMP-dependent kinase (AMPK) [84], but depending on cellular localization it can also inhibit this process [86]. Regulation the expression of transcription factor ATF3 enables adaptation to hypoxia, ER stress, oxidative stress and genotoxic stress [38], whereas during hypoxia induction of p21WAF1 causes cell cycle arrest [102]. p53 suppresses Akt-mTOR axis by transactivation of PTEN, TSC2 and AMPKβ1 [36]. cancer cells utilize the machinery of autophagy, a cata- p53 involvement in autophagy appears to rely on two bolic process in which cytoplasmic cargos are embedded contradictory functions. On one hand, p53 facilitates au- in double-membrane structures called autophagosomes tophagy by inducing expression of a damage-regulated to digest their content [80]. Among proteins involved in autophagy modulator (DRAM) [83], sestrin 1, sestrin 2, triggering autophagy, BCL2/adenovirus E1B 19kDa- AMP-dependent kinase (AMPK) [84] and/or inhibiting interacting protein 3 (BNIP3), BCL2/Adenovirus E1B mTOR pathway [85]. On the other, Tasdemir et al. [86] 19kDa Interacting Protein 3-Like (BNIP3L, NIX), to- postulate that cytoplasmic fraction of p53 can repress gether with Beclin-1 are induced under hypoxia in HIF- autophagy through a transcription-independent effect 1-dependent manner (see review by Mazure and and that p53 inactivation enhances this process. On the Pouyssegur) [81]. Moreover, HIF-1 promotes the pro- contrary, Naves et al. [87] found that neuroblastoma autophagic signaling pathways in adjacent tumor stroma, cells with the mutated p53 undergo autophagy when ex- which not only provides cancer cells with necessary posed to hypoxia mimetic CoCl2, but this pathway is ac- chemical building blocks but also renders them less sus- tivated when p53 localizes to the nucleus. The studies ceptible to apoptosis [82]. quoted above show that the ‘self-digestion’ is another Obacz et al. Molecular Cancer 2013, 12:93 Page 5 of 10 http://www.molecular-cancer.com/content/12/1/93

example of the mutual communication between HIF-1 of caveolin-1 in the cancer-associated fibroblasts causes and p53 in regulation of the tumor cells survival. induction of their senescence and supports tumor Recent developments in the field of senescence, a growth due to HIF-1α stabilization by ROS increase process leading to elimination of damaged cells from the [93]. In addition, VHL loss induces senescence in an growing population and subsequently preventing cancer oxygen-dependent manner by increasing the level of occurrence, reveal a dual role for hypoxia. Leontieva p27, which regulates cell cycle. However, these effects do et al. [88] found that hypoxia inhibits a conversion from not rely on HIF-1α or HIF-2α activity [94]. p53 involve- the reversible cell cycle arrest to senescence (known as ment in senescence has been intensively studied till geroconversion), nutlin-induced senescence and mTOR nowadays and recent achievements in that field have activity. Additionally, in marrow-derived mesenchymal been profoundly reviewed [95-97]. It is noteworthy that stem cells (MSCs) hypoxia promotes proliferation [89] p53 induction together with the prolonged p21WAF1 and causes downregulation of p21WAF1 expression in a overexpression can suppress senescence in favor of qui- HIF-1α-dependent manner [90]. On the other hand, escence [98]. many of HIF-1- regulated genes are associated with the Importantly, the cross-talk between p53 and HIF-1 can senescence induction, including plasminogen activator be observed at the level of their regulation, within a com- inhibitor (PAI1), cell cycle regulators, glycolytic enzymes plex molecular loop which involves both factors (Figure 2). and secreted molecules (see review by Welford et al.) As mentioned before, ATM mediates a DNA double [91]. The classic model of senescence shows that strand break signaling and repair via phosphorylation of hyperoxia can induce senescence through reactive oxy- p53. Ousset et al. [99] used various cellular models where gen species (ROS). In accordance, senescence is ATM was disrupted and demonstrated that the absence inhibited under low oxygen conditions simply due to de- of ATM increases expression of both subunits of HIF-1 as creased production of the mitochondrial ROS [92]. well as protein biosynthesis, through oxidative stress. Interestingly, recent report indicates that overexpression However, ATM is also responsible for the

VHL ATM 2 3

Hypoxia 1 HIF-1

98 ? DNA p300 mTOR pathway Damage p53 HIF Cell cycle control 7 4 Apoptosis 11 Metabolism VHL PCAF 10 p53 HIF ?

ATM p53

6 5 MDM2 Figure 2 Schematic characterization of mutual relations between HIF-1 and p53 pathways under different stress conditions. 1. Activation of hypoxia-inducible factor (HIF-1) during hypoxia; 2. Suppression of HIF-1 by von Hippel Lindau protein (VHL) during normoxia; 3. Downregulation of HIF-1 expression and protein biosynthesis by ataxia-telangiectasia mutated protein (ATM); 4. Stabilization and phosphorylation of p53 triggered by ATM in response to DNA damage; 5. Murine double minute 2 (MDM2)-dependent ubiqutination of p53; 6. Activation of p53 dependent on ATM-mediated phosphorylation of MDM2; 7. Positive regulation of p53 during DNA damage by nucleating ATM mediated by VHL; 8. Downregulation of p53 by HIF-1 under mild hypoxia; 9. Activation of p53 by HIF-1 under severe or/and prolonged hypoxia; 10. Suppression of HIF-1 by p53 under anoxia; 11. Competition of p53 and HIF-1 for binding the cofactors p300 and PCAF during hypoxia. Obacz et al. Molecular Cancer 2013, 12:93 Page 6 of 10 http://www.molecular-cancer.com/content/12/1/93

phosphorylation of HIF-1 on Ser-696, which causes a genes in solid tumors [45]. Conformational changes re- downregulation of mTORC1 signaling that regulates lated to missense mutations in the DNA-binding domain a translational efficiency [100]. Not only hypoxia sup- disrupt p53 transcriptional activity resulting in impaired presses the mTOR pathway; p53 in response to stress also ability of p53 to regulate the cellular response to hypoxia negatively regulates mTORC1 by inducing the expression in an effective way [105,106]. It was also established that of a plethora of target genes in the IGF-1/AKT and low oxygen pressure selects cells carrying p53 mutation mTOR pathways. This intrinsic regulation was reviewed and due to that contributes to metastatic potential and previously [29]. diminished apoptosis [46,107]. Interestingly, Gogna et al. Another crosstalk between HIF-1 and p53 is observed [60] using in-vivo electron paramagnetic resonance ox- on the level of trans-activation. During hypoxia, these imetry 3D imaging found that conformationally mutated transcription factors compete for the binding to the p53 appears in tumor hypoxic core and that its conform- CH1 domain of p300 cofactor [101]. Furthermore, it was ation is oxygen-dependent. found that another cofactor, p300/CBP Associated Factor Furthermore, not only p53 mutations act in favor of (PCAF) is involved in this regulatory mechanism. A cancer progression. Also hypoxia correlates with more study carried out by Xenaki et al. [102] focused on the aggressive tumor phenotypes and poor responses to expression of the pro-apoptotic p53 target BID and re- therapy [108]. This mainly involves stabilization of HIF- vealed a molecular mechanism underlying the regulation 1 and overexpression of its target genes [109]. For in- of p53 transcriptional activity in hypoxia. They have stance, expression of a HIF-1 target CA IX has been in- shown that hypoxia not only enables preferential direc- vestigated in various types of cancers, including breast, tion of p53 to the promoter of p21WAF1 cell cycle ar- colorectal, pancreatic etc. [110-112]. In these reports rester via PCAF, but also decreases PCAF-dependent overexpression of this hypoxic marker was associated acetylation of p53, which disrupts binding to its pro- with poorer patient survival, less differentiated tumors apoptotic targets. They found that PCAF is also a HIF-1 of higher grade and worse response to therapy. Similar cofactor involved in HIF-1- mediated apoptosis, whereas effects were described for VEGF in lung and gastric can- PCAF histone acetyltransferase (HAT) activity regulates cers [20,113]. Interestingly, high expression of HIF hy- transcriptional selectivity. droxylases, which negatively regulate HIF-1 and are Additional convergences are visible on the level of themselves regulated by hypoxia were postulated as poor regulation of these two transcription factors by VHL, prognostic factors in non small cell type lung cancers which as mentioned above, is a well-documented [114], whereas their inhibition reduced survival of glio- ubiquitin-dependent executer of HIF-1 degradation blastoma cells [115]. Concurrent overexpression of both [2,103]. However, it was also reported that VHL posi- HIF-1 and p53 was found in many cancers as well [116]. tively regulates p53 activity, preceded by DNA damage, An in vivo study, based on an experimental model of via nucleating ATM and histone acetyltransferase to chick embryo chorioallantoic membrane, revealed that p53. It also influences cell cycle arrest and apoptosis HIF-1α increases invasiveness of human small cell lung triggered by p53 due to upgrading the p53-p300 inter- carcinoma via promoting angiogenesis not only due to action and p53 acetylation [103]. Moreover, ATF3 links overexpression of VEGF but also due to secretion of the molecular pathways of HIF-1 and p53 in response to pro-inflammatory factors [20]. Moreover, Khromova DNA-damage, where both transcription factors are over- et al. [117] found that accelerated growth of cancer cells represented, which can be explained by the suggestion is associated with p53 mutations and caused by ROS- that ATF3 synergizes with these transcription factors to mediated activation of the HIF-1/VEGF-A pathway, modulate their target gene expression [39]. Recently, which links both factors with neovascularization. In a FIH was added to an even more complicated network in large cohort of colorectal cancers, HIF-1α but not HIF- which p53 and HIF-1 are involved: FIH silencing in 2α was shown to have an important negative prognostic colon adenocarcinomas and melanoma cells greatly role in cancer aggressiveness and overall survival of pa- abolishes cell proliferation and, more importantly, in- tients [118]. Contradictory to that, Cleven et al. [110] creases both p53 and p21WAF1 protein levels [104]. suggested that in the stroma of these tumors HIF-2α These results support the role of FIH in the suppression and CA IX serve as poor prognostic factors in tumors of the p53-p21WAF1 axis. expressing wild-type p53 compared with tumors with mutant form. Regarding p53, some studies join its ex- Impact of the p53 and HIF-1 interplay on cancer pression with patient survival [119] another with inva- progression sion depth [120] and poor differentiation [111] or worse Despite the fact that p53 is known to prevent mutations distant survival [121]. Moreover, another report indicates which cause genome instability and can lead to carcino- no significant survival difference between wild-type and genesis, it represents one of the most frequently mutated mutant p53 [110]. This leaves an open question on how Obacz et al. Molecular Cancer 2013, 12:93 Page 7 of 10 http://www.molecular-cancer.com/content/12/1/93

hypoxia selects for mutated p53 and thereby impacts on FIH: The factor inhibiting HIF; HIF: hypoxia inducible factor; patient outcome. HIPK2: Homeodomein-interacting protein kinase-2; HREs: Hypoxia responsive elements; JAK2: Janus Activated Kinase; MAPK: Mitogen-activated protein Hypoxia causes resistance to commonly used anti- kinase; MDM2: Murine double minute 2; MDR: Multidrug resistance; cancer agents either due to downregulation of genes that MSCs: Marrow-derived mesenchymal stem cells; mTOR: mammalian target of are drug targets or because oxygen deprivation abrogates rapamycin; NO: Nitric oxide; PAI1: Plasminogen activator inhibitor; PCAF: p300/CBP Associated Factor; PHDs: Prolyl hydroxylase family; Pgp: P- activity of the drugs. Chemotherapeutics of the first glycoprotein; pVHL: Von Hippel-Lindau tumor suppressor protein; choice (doxorubicin, etoposide, cisplatin) cause DNA ROS: Reactive oxygen species; RPA70: 70 kDa subunit of replication protein damage and therefore activate p53 to conduct apoptosis. A; SCO2: Cytochrome c oxidase 2; USP19: Ubiquitin-specific protease 19; Wip1: Wild-type p53-induced phosphatase 1; VEGF: Vascular endothelial HIF-1 by modulating expression of its target genes, ren- growth factor. der the cells less prone to treatment, although this effect is cell type-dependent [55]. Insensitivity can be HIF-1 in- Competing interests dependent as well, but relies on p53 suppression [122]. The authors declare they have no competing interests. Moreover, hypoxic cells divide less rapidly and are local- Authors’ contributions ized further from functional blood vessels. Due to that, JO reviewed the literature, and wrote and edited the manuscript. SP drugs are unable to reach poorly oxygenated areas and contributed to study conception and critically revised the paper. BV critically revised the paper. RH contributed to study conception, revised and finalized work less efficiently than in highly proliferating cells the manuscript. All authors read and approved the final manuscript. [123]. Last but not least, overexpression of P-glycoprotein Acknowledgements This work was supported by the European Regional Development Fund and (Pgp), a member of ATP-binding cassette (ABC) protein the State Budget of the Czech Republic RECAMO CZ.1.05/2.1.00/03.0101, MH superfamily has been reported to cause multidrug resist- CZ-DRO (MMCI, 00209805), GACR P206/12/G151, GACR P301/13/00956S and ance (MDR) of tumors [124,125]. Other studies eluci- by the 7th Framework Program (ITN project ENGCABRA). dated that increase in Pgp abundance is due to Author details transactivation by HIF-1 recruited to the MDR-1 gene in 1Masaryk Memorial Cancer Institute, Regional Centre for Applied Molecular MCF-7 spheroids and hypoxic cells. Importantly, both Oncology, Zluty kopec 7, 65653 Brno, Czech Republic. 2Department of Molecular Medicine, Institute of Virology, Slovak Academy of Sciences, MCF-7 spheroids and hypoxic cells show lower suscepti- Dubravska cesta 9, 84505 Bratislava, Slovak Republic. bility to doxorubicin treatment and reduced accumulation of drugs [126]. Received: 23 April 2013 Accepted: 10 August 2013 Published: 14 August 2013

Conclusions References It is well known that hypoxia and genome instability are 1. Liao D, Johnson RS: Hypoxia: a key regulator of angiogenesis in cancer. Cancer Metastasis Rev 2007, 26(2):281–290. intrinsic tumor characteristics, which influence cancer 2. Ivan M, Kondo K, Yang H, Kim W, Valiando J, Ohh M, Salic A, Asara JM, Lane progression and hence patient outcome. This report de- WS, Kaelin WG Jr: HIFalpha targeted for VHL-mediated destruction by scribes mutual relations between p53 and HIF-1 as me- proline hydroxylation: implications for O2 sensing. Science 2001, 292(5516):464–468. diators of adaptation to diverse cellular stresses, 3. Lando D, Peet DJ, Gorman JJ, Whelan DA, Whitelaw ML, Bruick RK: FIH-1 is including DNA damage and hypoxia. Although they an asparaginyl hydroxylase enzyme that regulates the transcriptional share many similarities, they can either act in parallel or activity of hypoxia-inducible factor. Genes Dev 2002, 16(12):1466–1471. 4. Keith B, Johnson RS, Simon MC: HIF1α and HIF2α: sibling rivalry in compete with each other in regulation of diverse mo- hypoxic tumour growth and progression. Nat Rev Cancer 2012, 12(1):9–22. lecular pathways. These discrepancies have been exten- 5. Uchida T, Rossignol F, Matthay MA, Mounier R, Couette S, Clottes E, Clerici sively studied, but there are still many gaps in C: Prolonged hypoxia differentially regulates hypoxia-inducible factor (HIF)-1alpha and HIF-2alpha expression in lung epithelial cells: understanding what triggers pro-survival or lethal activ- implication of natural antisense HIF-1alpha. J Biol Chem 2004, ity of these transcription factors. This work highlights 279(15):14871–14878. the importance of further investigation of this loop as 6. Ratcliffe PJ, O'Rourke JF, Maxwell PH, Pugh CW: Oxygen sensing, hypoxia- inducible factor-1 and the regulation of mammalian gene expression. the data mentioned above indicate that it involves both J Exp Biol 1998, 201(Pt 8):1153–1162. positive and negative regulators as well as epigenetic 7. Fernandez-Sanchez R, Berzal S, Sanchez-Nino MD, Neria F, Goncalves S, mechanisms. This knowledge is indispensable not only Calabia O, Tejedor A, Calzada MJ, Caramelo C, Deudero JJ, et al: AG490 Promotes HIF-1alpha accumulation by inhibiting its hydroxylation. for proper patient treatment, which as reported here can Curr Med Chem 2012, 19(23):4014–4023. be influenced by both cancer cell type and tumor envir- 8. Altun M, Zhao B, Velasco K, Liu H, Hassink G, Paschke J, Pereira T, Lindsten K: onment, but also for development of new drugs Ubiquitin-specific protease 19 (USP19) regulates hypoxia-inducible factor 1α (HIF-1α) during hypoxia. J Biol Chem 2012, 287(3):1962–1969. targeting p53 and/or HIF-1 pathways. 9. Xu J, Wang B, Xu Y, Sun L, Tian W, Shukla D, Barod R, Grillari J, Grillari- Voglauer R, Maxwell PH, et al: Epigenetic regulation of HIF-1α in renal Abbreviations cancer cells involves HIF-1α/2α binding to a reverse hypoxia-response ABC: ATP-binding cassette protein superfamily; ATM: Ataxia telangiectasia element. Oncogene 2012, 31(8):1065–1072. mutated protein; BNIP3: BCL2/adenovirus E1B 19kDa-interacting protein 3; 10. Dimova EY, Michiels C, Kietzmann T: Kinases as upstream regulators of the BNIP3L: BCL2/Adenovirus E1B 19kDa Interacting Protein 3-Like; CA HIF system: their emerging potential as anti-cancer drug targets. IX: Carbonic anhydrase IX; DRAM: Damage-regulated autophagy modulator; Curr Pharm Des 2009, 15(33):3867–3877. Obacz et al. Molecular Cancer 2013, 12:93 Page 8 of 10 http://www.molecular-cancer.com/content/12/1/93

11. Pastorek J, Pastorekova S, Callebaut I, Mornon JP, Zelnik V, Opavsky R, methyltransferases regulates p53 transcriptional output. EMBO J 2010, Zat'ovicova M, Liao S, Portetelle D, Stanbridge EJ, et al: Cloning and 29(15):2538–2552. characterization of MN, a human tumor-associated protein with a 33. Candeias MM, Malbert-Colas L, Powell DJ, Daskalogianni C, Maslon MM, domain homologous to carbonic anhydrase and a putative helix-loop Naski N, Bourougaa K, Calvo F, Fahraeus R: P53 MRNA controls p53 activity -helix DNA binding segment. Oncogene 1994, 9(10):2877–2888. by managing Mdm2 functions. Nat Cell Biol 2008, 10(9):1098–1105. 12. Wykoff CC, Beasley NJ, Watson PH, Turner KJ, Pastorek J, Sibtain A, 34. Gajjar M, Candeias MM, Malbert-Colas L, Mazars A, Fujita J, Olivares-Illana V, Wilson GD, Turley H, Talks KL, Maxwell PH, et al: Hypoxia-inducible Fahraeus R: The p53 mRNA-Mdm2 interaction controls Mdm2 nuclear expression of tumor-associated carbonic anhydrases. Cancer Res 2000, trafficking and is required for p53 activation following DNA damage. 60(24):7075–7083. Cancer Cell 2012, 21(1):25–35. 13. Ditte P, Dequiedt F, Svastova E, Hulikova A, Ohradanova-Repic A, Zatovicova 35. Feng Z, Zhang H, Levine AJ, Jin S: The coordinate regulation of the p53 M, Csaderova L, Kopacek J, Supuran CT, Pastorekova S, et al: and mTOR pathways in cells. Proc Natl Acad Sci USA 2005, Phosphorylation of carbonic anhydrase IX controls its ability to mediate 102(23):8204–8209. extracellular acidification in hypoxic tumors. Cancer Res 2011, 36. Sen N, Satija YK, Das S: p53 and metabolism: old player in a new game. 71(24):7558–7567. Transcription 2012, 3(3):119–123. 14. Svastova E, Hulikova A, Rafajova M, Zat'ovicova M, Gibadulinova A, Casini A, 37. Hublarova P, Greplova K, Holcakova J, Vojtesek B, Hrstka R: Switching p53- Cecchi A, Scozzafava A, Supuran CT, Pastorek J, et al: Hypoxia activates the dependent growth arrest to apoptosis via the inhibition of DNA capacity of tumor-associated carbonic anhydrase IX to acidify damage-activated kinases. Cell Mol Biol Lett 2010, 15(3):473–484. extracellular pH. FEBS Lett 2004, 577(3):439–445. 38. Chen BP, Wolfgang CD, Hai T: Analysis of ATF3, a transcription factor 15. Chiche J, Ilc K, Brahimi-Horn MC, Pouyssegur J: Membrane-bound carbonic induced by physiological stresses and modulated by gadd153/Chop10. anhydrases are key pH regulators controlling tumor growth and cell Mol Cell Biol 1996, 16(3):1157–1168. migration. Adv Enzyme Regul 2010, 50(1):20–33. 39. Tanaka Y, Nakamura A, Morioka MS, Inoue S, Tamamori-Adachi M, Yamada 16. Swietach P, Wigfield S, Cobden P, Supuran CT, Harris AL, Vaughan-Jones RD: K, Taketani K, Kawauchi J, Tanaka-Okamoto M, Miyoshi J, et al: Systems Tumor-associated carbonic anhydrase 9 spatially coordinates analysis of ATF3 in stress response and cancer reveals opposing effects intracellular pH in three-dimensional multicellular growths. J Biol Chem on pro-apoptotic genes in p53 pathway. PLoS One 2011, 6(10):e26848. 2008, 283(29):20473–20483. 40. Zhang X-P, Liu F, Wang W: Two-phase dynamics of p53 in the DNA 17. Svastova E, Witarski W, Csaderova L, Kosik I, Skvarkova L, Hulikova A, damage response. Proc Natl Acad Sci USA 2011, 108(22):8990–8995. Zatovicova M, Barathova M, Kopacek J, Pastorek J, et al: Carbonic 41. Takekawa M, Adachi M, Nakahata A, Nakayama I, Itoh F, Tsukuda H, Taya Y, anhydrase IX interacts with bicarbonate transporters in lamellipodia and Imai K: p53-Inducible wip1 phosphatase mediates a negative feedback increases cell migration via its catalytic domain. J Biol Chem 2012, regulation of p38 MAPK-p53 signaling in response to UV radiation. EMBO 287(5):3392–3402. J 2000, 19(23):6517–6526. 18. Buchler P, Reber HA, Buchler M, Shrinkante S, Buchler MW, Friess H, 42. Tasdemir E, Maiuri MC, Galluzzi L, Vitale I, Djavaheri-Mergny M, D'Amelio M, Semenza GL, Hines OJ: Hypoxia-inducible factor 1 regulates vascular Criollo A, Morselli E, Zhu C, Harper F, et al: Regulation of autophagy by endothelial growth factor expression in human pancreatic cancer. cytoplasmic p53. Nat Cell Biol 2008, 10(6):676–687. Pancreas 2003, 26(1):56–64. 43. Prives C, Hall PA: The p53 pathway. J Pathol 1999, 187(1):112–126. 19. Forsythe JA, Jiang BH, Iyer NV, Agani F, Leung SW, Koos RD, Semenza GL: 44. Muller P, Hrstka R, Coomber D, Lane DP, Vojtesek B: Chaperone-dependent Activation of vascular endothelial growth factor gene transcription by stabilization and degradation of p53 mutants. Oncogene 2008, hypoxia-inducible factor 1. Mol Cell Biol 1996, 16(9):4604–4613. 27(24):3371–3383. 20. Wan J, Chai H, Yu Z, Ge W, Kang N, Xia W, Che Y: HIF-1α effects on 45. Royds JA, Dower SK, Qwarnstrom EE, Lewis CE: Response of tumour cells angiogenic potential in human small cell lung carcinoma. J Exp Clin to hypoxia: role of p53 and NFkB. Mol Pathol 1998, 51(2):55–61. Cancer Res 2011, 30(1):77. 46. Graeber TG, Peterson JF, Tsai M, Monica K, Fornace AJ Jr, Giaccia AJ: 21. Brahimi-Horn MC, Bellot G, Pouyssegur J: Hypoxia and energetic tumour Hypoxia induces accumulation of p53 protein, but activation of a G1- metabolism. Curr Opin Genet Dev 2011, 21(1):67–72. phase checkpoint by low-oxygen conditions is independent of p53 22. Lum JJ, Bui T, Gruber M, Gordan JD, DeBerardinis RJ, Covello KL, Simon MC, status. Mol Cell Biol 1994, 14(9):6264–6277. Thompson CB: The transcription factor HIF-1alpha plays a critical role in 47. Hammond EM, Denko NC, Dorie MJ, Abraham RT, Giaccia AJ: Hypoxia links the growth factor-dependent regulation of both aerobic and anaerobic ATR and p53 through replication arrest. Mol Cell Biol 2002, glycolysis. Genes Dev 2007, 21(9):1037–1049. 22(6):1834–1843. 23. Zhang H, Gao P, Fukuda R, Kumar G, Krishnamachary B, Zeller KI, Dang CV, 48. Hubert A, Paris S, Piret JP, Ninane N, Raes M, Michiels C: Casein kinase 2 Semenza GL: HIF-1 inhibits mitochondrial biogenesis and cellular inhibition decreases hypoxia-inducible factor-1 activity under hypoxia respiration in VHL-deficient renal cell carcinoma by repression of C-MYC through elevated p53 protein level. J Cell Sci 2006, 119(Pt 16):3351–3362. activity. Cancer Cell 2007, 11(5):407–420. 49. Zhang L, Hill RP: Hypoxia enhances metastatic efficiency by up-regulating 24. Marin-Hernandez A, Gallardo-Perez JC, Ralph SJ, Rodriguez-Enriquez S, Mdm2 in KHT cells and increasing resistance to apoptosis. Moreno-Sanchez R: HIF-1alpha modulates energy metabolism in Cancer Res 2004, 64(12):4180–4189. cancer cells by inducing over-expression of specific glycolytic 50. Wouters A, Pauwels B, Lambrechts HAJ, Pattyn GGO, Ides J, Baay M, isoforms. Mini Rev Med Chem 2009, 9(9):1084–1101. Meijnders P, Dewilde S, Vermorken JB, Lardon F: Chemoradiation 25. Semenza GL: Defining the role of hypoxia-inducible factor 1 in cancer interactions under reduced oxygen conditions: cellular characteristics of biology and therapeutics. Oncogene 2010, 29(5):625–634. an in vitro model. Cancer Lett 2009, 286(2):180–188. 26. Rohwer N, Cramer T: Hypoxia-mediated drug resistance: novel 51. Sermeus A, Michiels C: Reciprocal influence of the p53 and the hypoxic insights on the functional interaction of HIFs and cell death pathways. Cell Death Dis 2011, 2:e164. pathways. Drug Resist Updat 2011, 14(3):191–201. 52. Suzuki H, Tomida A, Tsuruo T: Dephosphorylated hypoxia-inducible factor 27. Riley T, Sontag E, Chen P, Levine A: Transcriptional control of human 1alpha as a mediator of p53-dependent apoptosis during hypoxia. p53-regulated genes. NatRevMolCellBiol2008, 9(5):402–412. Oncogene 2001, 20(41):5779–5788. 28. Green DR, Kroemer G: Cytoplasmic functions of the tumour suppressor 53. An WG, Kanekal M, Simon MC, Maltepe E, Blagosklonny MV, Neckers LM: p53. Nature 2009, 458(7242):1127–1130. Stabilization of wild-type p53 by hypoxia-inducible factor 1alpha. 29. Feng Z, Levine AJ: The regulation of energy metabolism and the IGF-1 Nature 1998, 392(6674):405–408. /mTOR pathways by the p53 protein. Trends Cell Biol 2010, 20(7):427–434. 54. Achison M, Hupp TR: Hypoxia attenuates the p53 response to cellular 30. Wu X, Bayle JH, Olson D, Levine AJ: The p53-mdm-2 autoregulatory damage. Oncogene 2003, 22(22):3431–3440. feedback loop. Genes Dev 1993, 7(7A):1126–1132. 55. Cosse J-P, Sermeus A, Vannuvel K, Ninane N, Raes M, Michiels C: Differential 31. Prives C: Signaling to p53: breaking the MDM2-p53 circuit. Cell 1998, effects of hypoxia on etoposide-induced apoptosis according to the 95(1):5–8. cancer cell lines. Mol Cancer 2007, 6:61. 32. Chen L, Li Z, Zwolinska AK, Smith MA, Cross B, Koomen J, Yuan Z-M, 56. Kaluzova M, Kaluz S, Lerman MI, Stanbridge EJ: DNA damage is a Jenuwein T, Marine J-C, Wright KL, et al: MDM2 Recruitment of lysine prerequisite for p53-mediated proteasomal degradation of HIF-1alpha in Obacz et al. Molecular Cancer 2013, 12:93 Page 9 of 10 http://www.molecular-cancer.com/content/12/1/93

hypoxic cells and downregulation of the hypoxia marker carbonic 81. Mazure NM, Pouyssegur J: Hypoxia-induced autophagy: cell death or cell anhydrase IX. Mol Cell Biol 2004, 24(13):5757–5766. survival? Curr Opin Cell Biol 2010, 22(2):177–180. 57. Nardinocchi L, Puca R, D'Orazi G: HIF-1 alpha antagonizes p53-mediated 82. Martinez-Outschoorn UE, Trimmer C, Lin Z, Whitaker-Menezes D, Chiavarina apoptosis by triggering HIPK2 degradation. Aging-Us 2011, 3(1):33–43. B, Zhou J, Wang CW, Pavlides S, Martinez-Cantarin MP, Capozza F, et al: 58. D'Orazi G, Cecchinelli B, Bruno T, Manni I, Higashimoto Y, Saito S, Gostissa Autophagy in cancer associated fibroblasts promotes tumor cell survival M, Coen S, Marchetti A, Del Sal G, et al: Homeodomain-interacting protein role of hypoxia, HIF1 induction and NF kappa B activation in the tumor kinase-2 phosphorylates p53 at Ser 46 and mediates apoptosis. Nat Cell stromal microenvironment. Cell Cycle 2010, 9(17):3515–3533. Biol 2002, 4(1):11–19. 83. Crighton D, Wilkinson S, O'Prey J, Syed N, Smith P, Harrison PR, Gasco M, 59. Nardinocchi L, Puca R, Sacchi A, Rechavi G, Givol D, D'Orazi G: Targeting Garrone O, Crook T, Ryan KM: DRAM, a p53-induced modulator of hypoxia in cancer cells by restoring homeodomain interacting protein- autophagy, is critical for apoptosis. Cell 2006, 126(1):121–134. kinase 2 and p53 activity and suppressing HIF-1 alpha. PLoS One 2009, 84. Xu-Monette ZY, Young KH: The TP53 tumor suppressor and autophagy in 4(8):e6819. malignant lymphoma. Autophagy 2012, 8(5):842–845. 60. Gogna R, Madan E, Kuppusamy P, Pati U: Chaperoning of mutant p53 85. Boya P, Gonzalez-Polo RA, Casares N, Perfettini JL, Dessen P, Larochette N, protein by wild-type p53 protein causes hypoxic tumor regression. Metivier D, Meley D, Souquere S, Yoshimori T, et al: Inhibition of J Biol Chem 2012, 287(4):2907–2914. macroautophagy triggers apoptosis. Mol Cell Biol 2005, 25(3):1025–1040. 61. Madan E, Gogna R, Pati U: p53 Ser15 Phosphorylation disrupts the 86. Tasdemir E, Maiuri MC, Galluzzi L, Vitale I, Djavaheri-Mergny M, D'Amelio M, p53-RPA70 complex and induces RPA70-mediated DNA repair in Criollo A, Morselli E, Zhu CL, Harper F, et al: Regulation of autophagy by hypoxia. Biochem J 2012, 443(3):811–820. cytoplasmic p53. Nat Cell Biol 2008, 10(6):676–687. 62. Thomas DD, Espey MG, Ridnour LA, Hofseth LJ, Mancardi D, Harris CC, 87. Naves T, Jawhari S, Jauberteau MO, Ratinaud MH, Verdier M: Autophagy Wink DA: Hypoxic inducible factor 1alpha, extracellular signal- takes place in mutated p53 neuroblastoma cells in response to hypoxia regulated kinase, and p53 are regulated by distinct threshold mimetic CoCl2. Biochem Pharmacol 2013, 85(8):1153–1161. concentrations of nitric oxide. Proc Natl Acad Sci USA 2004, 88. Leontieva OV, Natarajan V, Demidenko ZN, Burdelya LG, Gudkov AV, 101(24):8894–8899. Blagosklonny MV: Hypoxia suppresses conversion from proliferative arrest 63. Warburg O: On the origin of cancer cells. Science 1956, 123(3191):309–314. to cellular senescence. Proc Natl Acad Sci USA 2012, 109(33):13314–13318. 64. Yalcin A, Telang S, Clem B, Chesney J: Regulation of glucose metabolism 89. Dos Santos F, Andrade PZ, Boura JS, Abecasis MM, da Silva CL, Cabral JM: Ex by 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatases in cancer. vivo expansion of human mesenchymal stem cells: a more effective cell Exp Mol Pathol 2009, 86(3):174–179. proliferation kinetics and metabolism under hypoxia. J Cell Physiol 2010, 65. Vander Heiden MG, Cantley LC, Thompson CB: Understanding the 223(1):27–35. Warburg effect: the metabolic requirements of cell proliferation. 90. Tsai CC, Chen YJ, Yew TL, Chen LL, Wang JY, Chiu CH, Hung SC: Hypoxia Science 2009, 324(5930):1029–1033. inhibits senescence and maintains mesenchymal stem cell properties 66. Hsu PP, Sabatini DM: Cancer cell metabolism: Warburg and beyond. through down-regulation of E2A-p21 by HIF-TWIST. Blood 2011, Cell 2008, 134(5):703–707. 117(2):459–469. 67. Ferguson EC, Rathmell JC: New roles for pyruvate kinase M2: working out 91. Welford SM, Giaccia AJ: Hypoxia and senescence: the impact of the Warburg effect. Trends Biochem Sci 2008, 33(8):359–362. oxygenation on tumor suppression. Mol Cancer Res 2011, 9(5):538–544. 68. Hommes FA, Everts RS: Particulate and free hexokinase in fetal rat liver. 92. Chen Q, Fischer A, Reagan JD, Yan LJ, Ames BN: Oxidative DNA damage Biol Neonate 1978, 33(3–4):193–200. and senescence of human diploid fibroblast cells. Proc Natl Acad Sci USA 69. Kuhajda FP: Fatty-acid synthase and human cancer: new perspectives on 1995, 92(10):4337–4341. its role in tumor biology. Nutrition 2000, 16(3):202–208. 93. Capparelli C, Whitaker-Menezes D, Guido C, Balliet R, Pestell TG, Howell A, 70. Migita T, Narita T, Nomura K, Miyagi E, Inazuka F, Matsuura M, Ushijima M, Sneddon S, Pestell RG, Martinez-Outschoorn U, Lisanti MP, et al: CTGF Mashima T, Seimiya H, Satoh Y, et al: ATP citrate lyase: activation and drives autophagy, glycolysis and senescence in cancer-associated therapeutic implications in non-small cell lung cancer. Cancer Res 2008, fibroblasts via HIF1 activation, metabolically promoting tumor growth. 68(20):8547–8554. Cell Cycle 2012, 11(12):2272–2284. 71. Yan H, Parsons DW, Jin G, McLendon R, Rasheed BA, Yuan W, Kos I, Batinic- 94. Welford SM, Dorie MJ, Li X, Haase VH, Giaccia AJ: Renal oxygenation Haberle I, Jones S, Riggins GJ, et al: IDH1 and IDH2 mutations in gliomas. suppresses VHL loss-induced senescence that is caused by increased N Engl J Med 2009, 360(8):765–773. sensitivity to oxidative stress. Mol Cell Biol 2010, 30(19):4595–4603. 72. Young CD, Anderson SM: Sugar and fat - that's where it's at: metabolic 95. Dulic V: Senescence regulation by mTOR. Methods Mol Biol 2013, changes in tumors. Breast Cancer Res 2008, 10(1):202. 965:15–35. 73. Mentis A-FA, Kararizou E: Metabolism and cancer: an up-to-date review of 96. Hasty P, Sharp ZD, Curiel TJ, Campisi J: mTORC1 and p53: clash of the a mutual connection. Asian Pac J Cancer Prev 2010, 11(6):1437–1444. gods? Cell Cycle 2013, 12(1):20–25. 74. Israel M, Schwartz L: The metabolic advantage of tumor cells. Mol Cancer 97. Rufini A, Tucci P, Celardo I, Melino G: Senescence and aging: the critical 2011, 10:70. roles of p53. Oncogene 2013. doi:10.1038/onc.2012.640. Epub ahead of 75. Munoz-Pinedo C, El Mjiyad N, Ricci JE: Cancer metabolism: current print. perspectives and future directions. Cell Death Dis 2012, 3:e248. 98. Blagosklonny MV: Tumor suppression by p53 without apoptosis and 76. Guccini I, Serio D, Condo I, Rufini A, Tomassini B, Mangiola A, Maira G, Anile senescence: conundrum or rapalog-like gerosuppression? Aging 2012, C, Fina D, Pallone F, et al: Frataxin participates to the hypoxia-induced 4(7):450–455. response in tumors. Cell Death Dis 2011, 2:e123. 99. Ousset M, Bouquet F, Fallone F, Biard D, Dray C, Valet P, Salles B, Muller C: 77. Selak MA, Armour SM, MacKenzie ED, Boulahbel H, Watson DG, Loss of ATM positively regulates the expression of hypoxia inducible Mansfield KD, Pan Y, Simon MC, Thompson CB, Gottlieb E: Succinate factor 1 (HIF-1) through oxidative stress: role in the physiopathology of links TCA cycle dysfunction to oncogenesis by inhibiting HIF-alpha the disease. Cell Cycle 2010, 9(14):2814–2822. prolyl hydroxylase. Cancer Cell 2005, 7(1):77–85. 100. Cam H, Easton JB, High A, Houghton PJ: mTORC1 Signaling under hypoxic 78. Kawauchi K, Araki K, Tobiume K, Tanaka N: p53 Regulates glucose conditions is controlled by ATM-dependent phosphorylation of HIF-1α. metabolism through an IKK-NF-kappaB pathway and inhibits cell Mol Cell 2010, 40(4):509–520. transformation. Nat Cell Biol 2008, 10(5):611–618. 101. Freedman SJ, Sun Z-YJ, Poy F, Kung AL, Livingston DM, Wagner G, Eck MJ: 79. Wanka C, Brucker DP, B√§hr O, Ronellenfitsch M, Weller M, Steinbach Structural basis for recruitment of CBP/p300 by hypoxia-inducible factor- JP, Rieger J: Synthesis of cytochrome c oxidase 2: a p53-dependent 1 alpha. Proc Natl Acad Sci U S A 2002, 99(8):5367–5372. metabolic regulator that promotes respiratory function and protects 102. Xenaki G, Ontikatze T, Rajendran R, Stratford IJ, Dive C, Krstic-Demonacos M, glioma and colon cancer cells from hypoxia-induced cell death. Demonacos C: PCAF is an HIF-1alpha cofactor that regulates p53 Oncogene 2012, 31(33):3764–3776. transcriptional activity in hypoxia. Oncogene 2008, 27(44):5785–5796. 80. White E, Karp C, Strohecker AM, Guo YX, Mathew R: Role of autophagy in 103. Roe JS, Kim H, Lee SM, Kim ST, Cho EJ, Youn HD: p53 Stabilization and suppression of inflammation and cancer. Curr Opin Cell Biol 2010, transactivation by a von hippel-lindau protein. Mol Cell 2006, 22(2):212–217. 22(3):395–405. Obacz et al. Molecular Cancer 2013, 12:93 Page 10 of 10 http://www.molecular-cancer.com/content/12/1/93

104. Pelletier J, Dayan F, Durivault J, Ilc K, Pecou E, Pouyssegur J, Mazure NM: 123. Brown JM: Exploiting the hypoxic cancer cell: mechanisms and The asparaginyl hydroxylase factor-inhibiting HIF is essential for tumor therapeutic strategies. Mol Med Today 2000, 6(4):157–162. growth through suppression of the p53-p21 axis. Oncogene 2012, 124. Fardel O, Lecureur V, Guillouzo A: The P-glycoprotein multidrug 31(24):2989–3001. transporter. Gen Pharmacol 1996, 27(8):1283–1291. 105. Ashcroft M, Taya Y, Vousden KH: Stress signals utilize multiple pathways 125. Tiwari AK, Sodani K, Dai C-L, Ashby CR Jr, Chen Z-S: Revisiting the ABCs of to stabilize p53. Mol Cell Biol 2000, 20(9):3224–3233. multidrug resistance in cancer chemotherapy. Curr Pharm Biotechnol 106. Koumenis C, Alarcon R, Hammond E, Sutphin P, Hoffman W, Murphy M, 2011, 12(4):570–594. Derr J, Taya Y, Lowe SW, Kastan M, et al: Regulation of p53 by hypoxia: 126. Doublier S, Belisario DC, Polimeni M, Annaratone L, Riganti C, Allia E, Ghigo dissociation of transcriptional repression and apoptosis from p53- D, Bosia A, Sapino A: HIF-1 activation induces doxorubicin resistance in dependent transactivation. Mol Cell Biol 2001, 21(4):1297–1310. MCF7 3-D spheroids via P-glycoprotein expression: a potential model of 107. Koshikawa N, Maejima C, Miyazaki K, Nakagawara A, Takenaga K: Hypoxia the chemo-resistance of invasive micropapillary carcinoma of the breast. selects for high-metastatic Lewis lung carcinoma cells overexpressing BMC Cancer 2012, 12:4. Mcl-1 and exhibiting reduced apoptotic potential in solid tumors. Oncogene 2006, 25(6):917–928. doi:10.1186/1476-4598-12-93 108. Hockel M, Schlenger K, Aral B, Mitze M, Schaffer U, Vaupel P: Association Cite this article as: Obacz et al.: Cross-talk between HIF and p53 as between tumor hypoxia and malignant progression in advanced cancer mediators of molecular responses to physiological and genotoxic of the uterine cervix. Cancer Res 1996, 56(19):4509–4515. stresses. Molecular Cancer 2013 12:93. 109. Cleven AHG, van Engeland M, Wouters BG, de Bruine AP: Stromal expression of hypoxia regulated proteins is an adverse prognostic factor in colorectal carcinomas. Cell Oncol 2007, 29(3):229–240. 110. Cleven AH, Wouters BG, Schutte B, Spiertz AJ, van Engeland M, de Bruine AP: Poorer outcome in stromal HIF-2 alpha- and CA9-positive colorectal adenocarcinomas is associated with wild-type TP53 but not with BNIP3 promoter hypermethylation or apoptosis. Br J Cancer 2008, 99(5):727–733. 111. O'Toole D, Couvelard A, Rebours V, Zappa M, Hentic O, Hammel P, Levy P, Bedossa P, Raymond E, Ruszniewski P: Molecular markers associated with response to chemotherapy in gastro-entero-pancreatic neuroendocrine tumors. Endocr Relat Cancer 2010, 17(4):847–856. 112. Tan EY, Yan M, Campo L, Han C, Takano E, Turley H, Candiloro I, Pezzella F, Gatter KC, Millar EK, et al: The key hypoxia regulated gene CAIX is upregulated in basal-like breast tumours and is associated with resistance to chemotherapy. Br J Cancer 2009, 100(2):405–411. 113. Maeda K, Chung YS, Ogawa Y, Takatsuka S, Kang SM, Ogawa M, Sawada T, Sowa M: Prognostic value of vascular endothelial growth factor expression in gastric carcinoma. Cancer 1996, 77(5):858–863. 114. Andersen S, Donnem T, Stenvold H, Al-Saad S, Al-Shibli K, Busund L-T, Bremnes RM: Overexpression of the HIF hydroxylases PHD1, PHD2, PHD3 and FIH are individually and collectively unfavorable prognosticators for NSCLC survival. PLoS One 2011, 6(8):e23847. 115. Henze A-T, Riedel J, Diem T, Wenner J, Flamme I, Pouyseggur J, Plate KH, Acker T: Prolyl hydroxylases 2 and 3 act in gliomas as protective negative feedback regulators of hypoxia-inducible factors. Cancer Res 2010, 70(1):357–366. 116. Baas IO, Hruban RH, Offerhaus GJ: Clinical applications of detecting dysfunctional p53 tumor suppressor protein. Histol Histopathol 1999, 14(1):279–284. 117. Khromova NV, Kopnin PB, Stepanova EV, Agapova LS, Kopnin BP: p53 hot- spot mutants increase tumor vascularization via ROS-mediated activation of the HIF1/VEGF-a pathway. Cancer Lett 2009, 276(2):143–151. 118. Baba Y, Nosho K, Shima K, Irahara N, Chan AT, Meyerhardt JA, Chung DC, Giovannucci EL, Fuchs CS, Ogino S: HIF1A Overexpression is associated with poor prognosis in a cohort of 731 colorectal cancers. Am J Pathol 2010, 176(5):2292–2301. 119. Fondevila C, Metges JP, Fuster J, Grau JJ, Palacín A, Castells A, Volant A, Pera M: p53 and VEGF expression are independent predictors of tumour recurrence and survival following curative resection of gastric cancer. Submit your next manuscript to BioMed Central Br J Cancer 2004, 90(1):206–215. 120. Oh SY, Kwon H-C, Kim S-H, Jang JS, Kim MC, Kim KH, Han J-Y, Kim CO, Kim and take full advantage of: S-J, Jeong J-s, et al: Clinicopathologic significance of HIF-1alpha, p53, and VEGF expression and preoperative serum VEGF level in gastric cancer. • Convenient online submission BMC Cancer 2008, 8:123. • Thorough peer review 121. Gryko M, Pryczynicz A, Guzinska-Ustymowicz K, Kamocki Z, Zareba K, Kemona A, Kedra B: Immunohistochemical assessment of apoptosis- • No space constraints or color ﬁgure charges associated proteins: p53, Bcl-xL, Bax and Bak in gastric cancer cells in • Immediate publication on acceptance correlation with clinical and pathomorphological factors. Adv Med Sci • Inclusion in PubMed, CAS, Scopus and Google Scholar 2012, 57(1):77–83. 122. Wang J, Biju MP, Wang M-H, Haase VH, Dong Z: Cytoprotective effects of • Research which is freely available for redistribution hypoxia against cisplatin-induced tubular cell apoptosis: involvement of mitochondrial inhibition and p53 suppression. J Am Soc Nephrol 2006, Submit your manuscript at 17(7):1875–1885. www.biomedcentral.com/submit 12 APPENDICES

12.1 Publication activity

12.1.1 Articles Obacz J, Sommerova L, Pastorek M, Durech M, Iuliano F, Pastorekova S, Vojtesek B, Hrstka R, Brychtova V. 2015. Extracellular AGR3, a closely related homologue of pro-oncogenic AGR2, modifies adhesive and migratory properties of breast cancer cell. (submitted to Mol Cancer )

Obacz J, Brychtova V, Podhorec J, Fabian P, Dobes P, Vojtesek B, Hrstka R. 2015a. Anterior gradient protein 3 is associated with less aggressive tumors and better outcome of breast cancer patients. Onco Targets Ther, 8, 1523-32. doi: 10.2147/OTT.S82235

Obacz J, Takacova M, Brychtova V, Dobes P, Pastorekova S, Vojtesek B, Hrstka R. 2015b. The role of AGR2 and AGR3 in cancer: similar but not identical. Eur J Cell Biol, 94 , 139-147. doi: 10.1016/j.ejcb.2015.01.002

Obacz J, Hrstka R. Vojtesek B. Pastorekova S. 2013. Tumor hypoxia and its clinical significance. Folia Mendeliana, 49 , 69-73

Obacz J, Pastorekova S, Vojtesek B, Hrstka R. 2013. Cross-talk between HIF and p53 as mediators of molecular responses to physiological and genotoxic stresses. Mol Cancer , 12 , 93. doi: 10.1186/1476-4598-12-93

12.1.2 Conferences & courses attended: Obacz J, Brychtova V, Podhorec J, Fabian P, Dobes P, Vojtesek B, Hrstka R, 2015. Anterior gradient protein 3 (AGR3) is associated with less aggressive tumours and better outcome of breast cancer patients. XXXIX. Brno Oncology days and XXIX. Conference for medical professionals, Brno, Czech Republic

Obacz J, Pastorekova S, Takacova M, Brychtova V, Vojtesek B, Hrstka R, 2014. Anterior gradient protein 3 (AGR3) is a new intracellular signaling molecule, which influences tumor cell response to apoptosis-inducing agents. XVI. Meeting of Biochemists and Molecular Biologists, Brno, Czech Republic

Brychtova V, Blahak J, Obacz J, Zelinka J, Fabian P, Hrstka R, Pavlovsky P, Bulik O, Vojtesek B. 2014. Evaluation of AGR2, AGR3 and MUC4 expression in oral cavity carcinomas. XXXVIII. Brno Oncology days and XXVIII. Conference for medical professionals, Brno, Czech Republic

Workshops: Regional Centre for Applied Molecular Oncology, 2011-2015, Brno, Czech Republic

111

Summer school: Applied and clinical oncology research, 2014, Moravec, Czech Republic

Ph.D. summer school: Label free sensing, 2013, Copenhagen, Denmark

Obacz J. 2013. Detecting cancer even earlier to beat the unbeatable. Copenhagen, Denmark (public event)

Zulfiqar A, Pfreundt A, Kwasny D, Obacz J, Svendsen W. 2012. Fabrication of Polysilicon Nanowire Field Effect Transistors for Biosensor Applications. Semiconductor Nanowires Based Sensors, Rennes, France

12.2 Internships

Slovak Academy of Sciences, Institute of Virology, Department of Molecular Medicine, Bratislava, Slovak Republic (couple of months between 2011-2014)

University of Copenhagen, Department of Veterinary Disease Biology, Copenhagen, Denmark, (Aug/Sep 2012)

Technical University of Denmark DTU, Department of Micro- and Nanotechnology , Copenhagen, Denmark, (Aug/Sep 2012)

112