Tumor Suppressor Activity of Klotho in Breast Cancer Is ... · Oncogenes and Tumor Suppressors...

Oncogenes and Tumor Suppressors

Tumor Suppressor Activity of Klotho in BreastCancer Is Revealed by Structure–FunctionAnalysisHagai Ligumsky1,2, Tami Rubinek1, Keren Merenbakh-Lamin1,2, Adva Yeheskel3,Rotem Sertchook4, Shiri Shahmoon1,2, Sarit Aviel-Ronen5, and Ido Wolf1,2

Abstract

Klotho is a transmembrane protein containing two internalrepeats, KL1 and KL2, both displaying significant homology tomembers of the b-glycosidase family. Klotho is expressed in thekidney, brain, and various endocrine tissues, but can also becleaved and act as a circulating hormone. Klotho is an essentialcofactor for binding of fibroblast growth factor 23 (FGF23) tothe FGF receptor and can also inhibit the insulin-like growthfactor-1 (IGF-1) pathway. Data from a wide array of malignan-cies indicate klotho as a tumor suppressor; however, the struc-ture–function relationships governing its tumor suppressoractivities have not been deciphered. Here, the tumor suppressoractivities of the KL1 and KL2 domains were examined. Over-expression of either klotho or KL1, but not of KL2, inhibitedcolony formation by MCF-7 and MDA-MB-231 cells. Moreover,in vivo administration of KL1 was not only well tolerated but

significantly slowed tumor formation in nude mice. Furtherstudies indicated that KL1, but not KL2, interacted with the IGF-1R and inhibited the IGF-1 pathway. Based on computerizedstructural modeling, klotho constructs were generated in whichcritical amino acids have been mutated. Interestingly, themutated proteins retained their tumor suppressor activity butshowed reduced ability to modulate FGF23 signaling. Thesedata indicate differential activity of the klotho domains, KL1and KL2, in breast cancer and reveal that the tumor suppressoractivities of klotho can be dissected from its physiologicactivities.

Implications: These findings pave the way for a rational designof safe klotho-based molecules for the treatment of breast cancer.Mol Cancer Res; 13(10); 1398–407. �2015 AACR.

IntroductionKlotho (a-klotho) is a transmembrane protein expressed in the

distal convoluted tubules of the kidneys and the choroid plexus, aswell as in various endocrine and exocrine tissues (1), but it can alsobe cleaved, shed, and act as a circulating hormone (2, 3). Klotho-deficient mice manifest a syndrome resembling human aging,whereas overexpression of klotho extends lifespan (1, 4). Majorphysiologic activities of klotho include regulation of phosphateand calcium homeostasis as well as of metabolic activity (5–7). Todate, several mechanisms of action of klotho have been described.Klotho is an essential cofactor for binding of FGF23 to the FGFreceptor (FGFR; refs. 5, 8), it modulates bFGF signaling (9, 10),

inhibits the insulin and the insulin-like growth factor (IGF)-1pathways (4, 9), and regulates the activity of the transient receptorpotential vanilloid type 5 (TRPV5) calcium channel (11, 12). Thecommon link between these diverse activities has not been eluci-dated yet.

In the past years klotho has emerged as a potent tumorsuppressor in a wide array of malignancies, including breast,pancreas, colon, gastric, and lung cancers, as well as inmelanoma(9, 10, 13–16). We have previously shown that klotho is abun-dantly expressed in normal breast and pancreatic tissues and thatupon malignant transformation its expression is reduced (9,10, 17). Restoring klotho expression, or treatment with thesoluble protein, inhibited proliferation of breast and pancreaticcancer cells (9, 10), and further studies indicated it as an inhibitorof the IGF-1 pathway (4, 9).

Mouse klotho is a 1,014-amino-acid–long protein (1). Itsintracellular domain is composed of 10 amino acids and has noknown functional role (1). The extracellular domain is composedof a signal sequence (SS), responsible for directing the protein tothe cell surface, and two internal repeats, KL1 andKL2, each about450-amino acids long (1), which exhibit only weak similarityto each other (21% amino-acid identity; ref. 1). The discreteactivities of each domain have not been well-characterized yet.Cleavage of the extracellular domainof klotho can yield either theentire extracellular domain (130 kDa) or the discrete domainsKL1 and KL2 (2, 18, 19). Although only the full-length proteincan serve as a cofactor for the activity of FGF23 (5, 8), either thefull-length protein or KL1 can inhibit proliferation of pancreaticcancer cells (10).

1The Oncology Division, Tel Aviv Sourasky Medical Center, Tel Aviv,Israel. 2Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel.3Bioinformatics Unit, George S.Wise Faculty of Life Sciences,Tel AvivUniversity, Tel Aviv, Israel. 4Bioinformatics and Biological ComputingUnit,Weizmann Institute of Science, Rehovot, Israel. 5Department ofPathology,Talpiot Medical Leadership Program,Chaim ShebaMedicalCenter, Ramat Gan, Israel.

Note: Supplementary data for this article are available at Molecular CancerResearch Online (http://mcr.aacrjournals.org/).

H. Ligumsky performed this work in partial fulfillment of the requirements for aPh.D. degree.

CorrespondingAuthor: IdoWolf, Tel Aviv Medical Center, 6Weizmann Tel Aviv,Israel, 64239. Phone: 972-52-736-0558; Fax: 972-3-697-3030; E-mail:[email protected]

doi: 10.1158/1541-7786.MCR-15-0141

�2015 American Association for Cancer Research.

MolecularCancerResearch

Mol Cancer Res; 13(10) October 20151398

on August 29, 2020. © 2015 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst June 25, 2015; DOI: 10.1158/1541-7786.MCR-15-0141

http://mcr.aacrjournals.org/

Each klotho domain displays significant homology to mem-bers of the glycoside hydrolase family 1 enzymes (1). Twoglutamic acid residues are required for the enzymatic activity offamily 1 b-glycosidases, one acts as a nucleophile and the other asan acid/base catalyst (20). Although these critical amino acids arenot conserved in klotho, klotho was shown to hydrolyze b-gly-coside and b-d-glucuronide (20). In support of these activities,mutations of critical amino acids, considered to be essential for itsenzymatic activity, abolished the ability of klotho to increaseTRPV5 currents (11). Yet, these mutations did not affect its abilityto inhibit Wnt signaling. To our knowledge, the role of theseresidues in mediating klotho tumor suppressor activities has notbeen studied yet.

We aimed to decipher klotho structure–function relationshipsthat are involved in mediating its tumor suppressor activities inbreast cancer. Our findings demonstrate that KL1, but not KL2,can inhibit growth of breast cancer cells in vitro and in vivo.Structural modeling and site-directed mutagenesis indicate thatamino acids, essential formodulation of FGF23 signaling, are notrequired for klotho tumor suppressor activity.

Materials and MethodsChemicals and antibodies

The chemicals usedwere FGF23, IGF-1 and soluble humanKL1(hKL1) (PeproTech Inc.), andG418 (Invitrogen). Antibodies usedwere total-IGF-1bR (Santa Cruz Biotechnology), phospho-AKT1(S473), phospho-IGF-IR (Y1131), total pan-AKT (Cell SignalingTechnology), diphosphorylated-Thr183/Tyr185 and total-ERK1/2 (Sigma), and HA (Covance). Anti-klotho antibodies directedagainst KL1 domain (KM2076) and KL2 domain (KM2119) werea kind gift from Kyowa Hakko Kogyo Co., Ltd.

Cells and transfectionsMCF-7, MDA-MB-231, and HEK-293 cells were obtained

from the ATCC. Cells were received from Dr. Hadassa Degani'slab at Weizmann Institute of Science, which obtained themdirectly from the ATCC and were not passaged more than 6months after resuscitation. All transfections used Lipofecta-mine 2000 (Invitrogen).

Structural model construction of KL1 and KL2 domainsTemplate structures for molecular modeling of human klotho

were selected by BLAST search against the Protein Data Base(PDB). The BLAST search suggested the structures of the humanCytosolic Neutral beta-Glycosylceramidase (klotho-related Pro-tein, KLrP) as best template (42% sequence identity to KL1domain and 33% to KL2 domain). A crystal structure of KLrPwith highest resolution (1.60 Å) and good stereochemistry wasselected (PDB entry 2E9L). Structural model of KL1 was builtusing PDB 2E9L (21) as a template. The template was selectedusing HHPRED (22). The model was built using Modeller (23).The model structure was refined using the Rosetta Relax protocol(v. 3.2) (24). KL1 model was assessed using ModFold (25) andConSurf (26), using a multiple sequence alignment of 150 KL1homologs collected from Uniprot (27) using Blast (28) andaligned with Mafft (29). Structural model of KL2 was based ona sequence alignment between KL2 domain, the template struc-ture, and 14 representative sequences of family 1 Glycosidehydrolase, whichwas carried out using severalmultiple alignmentalgorithms (Expresso, Promals3D, and FFAS03). A set of 20 all-

atoms model of KL2 was built using the restrained-based model-ing approach as implemented in the program MODELLER 9V6based on the different alignments. Themodel was evaluated usingMODELLER objective function, DOPE, ProSA, and Anolea. Themodel showing the best score as judged by these evaluationmethods was saved for further studies and presented here.

Generation of klotho expression constructsMouseHA-tagged full-length klotho (FL-KL) in pcDNA3.1 was a

generous gift from Y.Nabeshima (KyotoUniversity, Japan).MouseKL1 and KL2 constructs were prepared in two steps, in order tocontain the original FL-KL SS. For the first step, the SS was PCRamplified and restriction sites were introduced using specific pri-mers: 50-ATCGAATTCATGCTAGCCCGCGCCCCTCCT-3 (EcoRI)and 50-ATCAAGCTTGGCGCGCAGGGCGAGCAGCAG-3 (Hin-dIII). KL1 and KL2 were PCR amplified and restriction sites wereintroduced using specific primers. For KL1, 5-ATCAAGCTTATG-CGCTGCCTGAGCGC-3 (HindIII) and 5-ATCCTCGAGTCTCTT-CTTGGCTACAACCC-3 (Xho). For KL2, 5-ATCAAGCTTAAACCT-TACTGTGTTGATTTC-3 (HindIII) and 5-ATCCTCGAGTCTCTTC-TTGGCTACAACCC-3 (Xho). Plasmids were digested with the suit-able enzymes, and the SSwas ligated to KL1 and KL2. Subsequently,the constructs were sequenced for verification and subcloned intopcDNA3.1 backbone.

Mouse KL1 with intra-cytoplasmic/transmembrane (IC/TM)was prepared as followed: at first, an IC/TM fragment was PCRamplified and restriction sites were introduced using specificprimers: 50-ATCCTCGAGTCTTTGCTGGTCTTCATC-30 (XhoI)and 50-GATCTCGAGCTTATAACTTCTCTGGCC-30 (XhoI). Thenthe fragment and KL1 plasmid were digested with the suitableenzymes, and the IC/TM was ligated to KL1.

Human FL-KL (hFL-KL) in pEF and human full-length withoutIC/TM (hFL-KL DIC/TM) was a generous gift fromM. Kuro-O (UTSouthwestern, Dallas, TX).

Directed mutagenesis of klotho putative enzymatic active sitePrior to mutagenesis, constructs were transferred to a Topo

plasmid to preserve the HA tag. The selected amino acids forsubstitution were chosen in order to preserve the three-dimen-sional conformation of the molecule. First, a single mutation wasmade at amino acid 416 by converting Glu to Gln. FL-KL and KL1constructs were PCR amplifiedusing specific primers formutationinsertion: 5-GGAATATTTATTGTGCAAAATGGCTGG-3 (muta-tion insertion) and 5-CAGGAAACAGCTATGAC-3.

Afterwards, doublemutation constructs weremade by adding asecondmutation at residue 240 by converting Asp to Asn. For thesecond mutation, site-directed mutagenesis technique wasemployed. Constructs were PCR amplified using specific primersfor mutation insertion: 5-CACCATTAACAACCCCTACGTGGT-GGCCTGGCACG-3 and 5-AGGGGTTGTTAATGGTGATCCAG-TACTTGACCTGACCACCG-3.

Colony assaysTwo days following transfection with the indicated plasmids,

G418 (750 mg/mL for all cell lines) was added to the culturemedia. At day 14, cells were stained using crystal violet (Sigma)and quantitated using ImageJ software.

Western blot analysisCells were harvested, lysed, and total proteinwas extractedwith

RIPAbuffer (50mmol/L Tris–HCl, pH7.4, 150mmol/LNaCl, 1%

Klotho Structure–Function Relationships in Breast Cancer

www.aacrjournals.org Mol Cancer Res; 13(10) October 2015 1399




NP-40, 0.25%Na-deoxycholate, 1mmol/L EDTA, 1mmol/LNaF)together with a protease inhibitor cocktail (Sigma). Lysates wereresolved on 10% SDS-PAGE and immunoblotted with the indi-cated antibodies. Band intensities were quantified using ImageJsoftware.

ImmunoprecipitationMCF-7 cells were harvested as described above, and 400 mg

protein lysates were incubated with protein A/G (Pierce) beadsat 4�C for 1 hour to eliminate nonspecifically bound proteins.These beads were subsequently used as a negative control.The supernatants were then incubated with 2 ml anti total IGF-1R for 4 hours and then protein A/G was added. Following anovernight incubation, the beads were washed six times in RIPAbuffer, and the immunoprecipitated materials were separated bySDS–polyacrylamide gel electrophoresis and detected byWesternblotting.

Mice tumor xenograft studiesMice maintenance and experiments were carried out under

institutional guidelines of the ShebaMedical Center. Experimentswere conducted as described (30). Six-week-old female athymicnude mice were injected with 0.75 � 106 MDA-MB-231 cellssubcutaneously into both flanks. After 5 days, mice were treatedwithdaily i.p. injections of soluble hKL1: 10mg/kg, or saline (n¼6per group). Tumor sizewasmeasuredweeklywith adigital caliper,and the volume was estimated using the equation V¼ (a� b2)�0.5236, where a is the larger dimension and b is the perpendiculardiameter.

Immunohistochemistry analysisAfter mice were euthanized, tumors were removed and fixed in

4% paraformaldehyde. Paraffin-embedded tumors were seriallysectioned and stained with hematoxylin & eosin (H&E), andimmunohistochemically stained for Ki-67 and vimentin. Stainedsections were examined microscopically by an experiencedpathologist. For plasma analyses, immediately after mice wereeuthanized, blood was drawn from the heart and plasma wasanalyzed at the automated blood lab, Sheba Medical Center.

Statistical analysisResults are presented as mean � SD. Continuous variables

were compared using the t test. All significance tests were two-tailed, and a P value of <0.05 was considered as statisticallysignificant.

ResultsStructural modeling of the KL1 and KL2 domains

We generated a computational structuremodel of KL1 and KL2domains based on similarity of klotho to KLrP (Fig. 1A). Crys-tallographic studies indicated KLrP as a classic family 1 glycosidehydrolases that display the Koshland-retaining mechanism. Twoglutamic residues serve as its catalytic residues: Glu-373 is thenucleophile and Glu-165 is the acid/base catalyst. The modelindicates that both KL1 domain (residues 57–506) and KL2domain (residues 515–953) share significant homology to gly-coside hydrolase family 1 and are predicted to have similar TIM-Barrel fold (ref. 31; Fig. 1A). According to the model, Glu-414 inKL1 domain can serve as the nucleophilic residue (Fig. 1B).Although the corresponding residue to the acid/base catalystGlu-165 is Asn-239, which cannot serve as an acid/base catalyst,

Asp-238may function as an alternative acid/base catalyst and it isin the requireddistance from thenucleophile (2Å). Thus, KL1mayserve as an active enzyme. In KL2, on the other hand, the criticalnucleophilic residue (Glu-373) is replaced by Ser-872, whichcannot serve as the nucleophilic residue (Fig. 1C). Thus, themodel suggests that KL2 is not an active glycoside hydrolase.

KL1 inhibits colony formation and suppresses in vivo growth ofbreast cancer cells

In order to evaluate the differential activities of KL1 andKL2,weconstructed expression vectors of KL1 and KL2 domains, bothwith an SS, directing them to the membrane (Fig. 2A and B) andenabling them to be secreted (Supplementary Fig. S1). The effectof overexpression of these domains on growth of breast cancercells was assessed using colony formation assays. MCF-7 andMDA-MB-231 breast cancer cells were transfected with eitherklotho, KL1, or KL2 expression vectors or an empty vector.Transfected cells were cultured in media containing G418 for 2weeks and stained to determine the number of surviving colonies.Klotho and KL1 expression significantly reduced the number andsize of surviving colonies, whereas KL2 showed only a modesteffect (Fig. 2C and D). The tumor suppressor activity of KL1 wasalso verified in vivo. For this study, MDA-MB-231 cells wereinjected to both flanks of athymic mice, and mice were treatedwithdaily i.p. injections of either control vehicle (n¼6)or solublehKL1 (10 mg/kg, n ¼ 6) for 4 weeks. Already after 12 days oftreatment, a significant inhibition of tumor growth was notedamong KL1-treated mice (P < 0.001, for comparison of tumorvolume between the control and hKL1-treated group), lasting tothe endof the experiment (Fig. 3A). Tumorweight at the endof theexperiment was significantly lower (P < 0.01) in hKL1-treatedgroup (Fig. 3B andC). Immunohistochemistry analysis revealed asignificant reduction in the number of cells expressing the pro-liferation marker Ki-67 and the mesenchymal marker vimentin,suggesting effect of KL1 on proliferation, as well as differentiation(Fig. 3D). Importantly, KL1 treatment was safe and did not affectthe weight of mice (Fig. 3A) and was associated with higheralbumin levels and reduced liver function tests and lactate dehy-drogenase (LDH; Table 1).

Differential modulation of IGF-1 signaling by KL1 and KL2We have previously shown that klotho efficiently suppresses

activation of the IGF-1 pathway and interacts with the IGF-1R inbreast cancer cells (9). In order to analyze the differential effects ofKL1 and KL2 on IGF-1 signaling, MCF-7 cells were transfectedwith either klotho, KL1, or KL2 expression vectors or anempty vector, starved for 48 hours, treated with IGF-1 (100 ng/mL, 15minutes) and analyzed usingWestern blotting for the levelof expression and phosphorylation of IGF-1R, AKT, and ERK1/2.KL1 as well as klotho reduced IGF-1–induced phosphorylationof these proteins, whereas KL2 had only minor effect (Fig. 4A).Co-immunoprecipitation (Co-IP) assays using MCF-7 cellsexpressing either klotho, KL1, or KL2 indicated that only klothoand KL1 can interact with the endogenous IGF-1R (Fig. 4B).

Only full-length, membrane-bound klotho can transduceFGF23 signaling

Membrane-bound klotho is an essential cofactor for activationof the FGFRs by FGF23 (8). To our knowledge, the ability of KL1and KL2 to modulate FGF23 signaling has not been determined

Ligumsky et al.

Mol Cancer Res; 13(10) October 2015 Molecular Cancer Research1400




yet. In order to test this, MCF-7 andHEK293 cells were transfectedwith either klotho, KL1, or KL2 expression vectors or an emptyvector, starved for 48 hours, treated with FGF23 (100 ng/mL, 15minutes), and analyzed using Western blotting for the level ofexpression and phosphorylation of ERK1/2. Neither KL1 nor KL2were able tomediate FGF23-induced phosphorylation of ERK1/2(Fig. 4C). Removing the IC/TM domain (FL-KL-DIC/TM), whichanchors klotho to the cell membrane, also impaired its ability totransduce FGF23 signal (Fig. 4D).

In order to test if membrane-bound KL1 can mediate FGF23signaling, we generated a KL1 domain fusedwith the IC/TM regionof klotho (KL1-IC/TM, Fig. 2A). HEK 293 cells were transfectedwith these constructs, serum starved for 48 hours, stimulated witheither a vehicle or 100 ng/mL FGF23 for 15minutes, and analyzedusingWesternblotting for thephosphorylationofERK1/2.KL1-IC/TM failed to transduce the FGF23 signal (Fig. 4D).

Glu-416 and Asp-240 are not required for klotho tumorsuppressor activity

Mutations of klotho putative active sites abolished its effect onTRPV5 and on the renal phosphate transporter NaPi-2a

(11, 32, 33), but did not affect its ability to inhibit Wnt signaling(34). To our knowledge, the role of these residues in mediatingklotho tumor suppressor activities has not been studied yet. Inorder to decipher the role of these amino acids in mediating thetumor suppressor activity of klotho, a series of mouse klotho andKL1 constructs were designed, in which amino acids have beenmutated. Glu-416 (putative nucleophile, corresponds to humanGlu-414)was converted to glutamine andAsp-240 (putative acid/base catalyst, corresponds to human Asp-238) was converted toasparagine (Fig. 1B). Thesemutations are not expected to alter thetertiary structure of the protein. The ability of the variousmutatedproteins to inhibit colony formation ofMCF-7 andMDA-MB-231cells was determined as described above. In both breast cancer celllines, the mutated klotho constructs, FL-KL or KL1, harboringsingle (416) or double (240-416) mutations, retained the anti-proliferative effect exhibited by klotho. Surprisingly, the mutatedklotho and KL1 proteins displayed even a more potent antipro-liferative activity compared with their wild-type counterparts. Forexample, in MCF-7 cells, FL-KL, FL-KL-416, and FL-KL-240-416reduced number of colonies to 61%, 37%, and 45% of control,respectively (Fig. 5A). In MDA-MB 231 cells, FL-KL, FL-KL-416,

Figure 1.Structure–function relationships ofklotho internal repeats. A, predictedstructure of klotho internal repeatswithputative active site residues (in yellow),responsible for the catalyticmechanism of a family 1 glycosidase. B,superimposition of KLrP and KL1 activesites. KLrP active site side chains arepresented in pink and KL1 domain incyan. KLrP was crystallized withpart of its product, glucose. C,superimposition of KLrP and KL2 activesites. KLrP active site side chains arepresented in pink and KL2 domain ingreen. KLrP was crystallized with partof its product, glucose.






and FL-KL-240-416 reduced number of colonies to 32%, 18%,and 16% of control, respectively (Fig. 5B).

Dissecting klotho tumor suppressor activities from its physio-logic activities may enable the development of safe klotho-basedtherapies. We, therefore, aimed to study whether klotho mutant,which potently slows proliferation of cancer cells, can modulateFGF23 signaling. As only FL-KL can function as a cofactor forFGF23 (Fig. 4C and D), we analyzed only FL-KL mutants. To thisaim, HEK293 cells were transfected with either FL-KL, FL-KL-416,and FL-KL-240-416 expression vectors or an empty vector, starvedfor 48 hours, treated with FGF23 (100 ng/mL, 15 minutes), andanalyzed using Western blotting for the level of expression andphosphorylation of ERK1/2. Although the WT-FL and KL-416were able tomediate FGF23-induced phosphorylation of ERK1/2(Fig. 5C), transfection with FL-KL-240-416 only partiallyincreased ERK phosphorylation (Fig. 5C).

DiscussionThis structure–function analysis of klotho in breast cancer

indicates KL1 as sufficient to mediate growth inhibition, whereas

KL2 does not have a significant tumor suppressor activity. Fur-thermore, our data suggest that the activity of KL1, as well as of thefull-length protein, is independent of the amino acids of itsputative catalytic site.

Distinct activities of klotho, KL1, and KL2 have been demon-strated in several systems. The FGF23 coreceptor function ismediated only by full-length, membrane-anchored klotho(refs. 5, 8; Fig. 3C and D). On the other hand, either KL1 or KL2were sufficient to activate TRPV5 (11). In contrast, several klothoactivities were shown to be KL1 dependent. Thus, it was demon-strated that KL1 is essential and sufficient to mediate the anti-inflammatory activity of klotho in senescent cells, whereas KL2did not show any such activity (35). Similarly, KL1 is mandatoryand sufficient to inhibit Wnt signaling by klotho, whereas KL2 isdevoid of this activity (34). Our data suggest that klotho tumorsuppressor activities are KL1 dependent and that KL2 by itselfcannot significantly inhibit proliferation of breast cancer cells.

Administration of KL1 slowed tumor formation by breastcancer cells in nude mice. Tumors from KL1-treated mice weresmaller, showed reduced proliferation, and reduced expression ofthe EMT marker vimentin. Furthermore, their blood albumin

Figure 2.Klotho internal repeats exhibit differential effect on colony formation in breast cancer cells. A, schematic illustration of klotho constructs used in the study.B, MCF-7 and MDA-MB-231 cells were transfected with FL-KL and indicated truncated KL1 and KL2 HA-tagged constructs. Total cell lysates were analyzed byimmunoblotting with anti-HA antibody. C, MCF-7 and MDA-MB-231 cells were transfected with either a full-length klotho expression vector (FL-KL), KL1,KL2, or an empty vector (pcDNA3; all constructs depicted inA). Transfected cellswere cultured inmedia containingG418 for 2weeks. Colonieswere fixed and stainedwith crystal violet and photographed. All experiments were conducted at least 3 times, and representative wells for each condition are shown. D, quantitationof the colonies was performed using ImageJ (� , P < 0.05, FL-KL and KL1 vs. control).

Ligumsky et al.





levels were higher and LDH levels lower compared with control-treated mice. Treatment did not adversely affect either weight orkidney functions. This favorable toxicity profile can be attributed

to the inability of KL1 to serve as a coreceptor for FGF23. Thesedata support our previous observation using pancreatic cancercells (10) and suggest that treatment with KL1 is safe and may

Table 1. Comparison of blood chemistry parameters between the control and mice treated with soluble hKL1 (10 mg/kg)

ParameterControl(N ¼ 6)

Soluble hKL1(N ¼ 6)

P value (con. vs.soluble hKL1)

Urea (mg/dL) 56.6 � 5.03 57.33 � 7.06 0.85Creatinine (mg/dL) 0.24 � 0.02 0.25 � 0.02 0.46Glucose (mg/dL) 205.2 � 26.57 204.5 � 24.92 0.96Phosphorous (mg/dL) 7.8 � 0.39 7.4 � 0.89 0.48Sodium (mg/dL) 142.6 � 5.73 148 � 2 0.057Calcium mg/dL) 9.04 � 0.67 9.42 � 0.58 0.15Chloride (mg/dL) 107.8 � 3.83 110.67 � 2.16 0.34Potassium (mg/dL) 7.14 � 1.07 6.53 � 1.57 0.37Uric acid (mg/dL) 1.92 � 0.28 1.65 � 0.43 0.25SGOT (AST), IU/L 574.80 � 323.23 232.67 � 122.01 0.04a

SGPT (ALT), IU/L 273.75 � 199.85 75.83 � 55.36 0.04a

LDH, IU/L 2534.33 � 1348.54 698.75 � 251.42 0.04a

Alkaline phosphatase (IU/L) 49.8 � 14.34 81.17 � 16.94 0.01a

Protein (g/dL) 5.02 � 0.5 5.53 � 0.37 0.08Albumin (g/dL) 2.66 � 0.21 3.02 � 0.29 0.05a

Globulin (g/dL) 2.36 � 0.32 2.5 � 0.14 0.35

NOTE: Blood chemistry parameters fromvehicle and soluble hKL1-treated nudemice harboringMDA-MB-231 humanbreast cancer cells tumors. Plasmawas analyzedby automated blood lab. SGOT, Serum glutamic oxaloacetic transaminase; SGPT, Serum glutamic pyruvic transaminase.a, P < 0.05, soluble hKL1-treated mice versus control vehicle. Significant values are in bold.

Figure 3.Soluble hKL1 inhibits growth of breast tumors in nude mice. A, MDA-MB-231 cells were injected into both flanks of 6-week-old female athymic nude mice. The mice(N ¼ 6 per group) were treated with daily i.p. injections of soluble hKL1 (10 mg/kg) or vehicle control (saline) for 3 weeks. Tumors volumes were measuredweekly (�� ,P <0.001, for tumor volume between soluble hKL1-treated and control group). B, representative photographs ofmicewith tumors in soluble hKL1-treatedgroup and control group before they were sacrificed. C, tumor weight was measured on day of sacrifice (�� , P < 0.01 for comparison of tumor weightbetween the control group and the mice treated with soluble hKL1). D, H&E, Ki67, vimentin, and cytokeratin AE1/AE3 immunostaining magnification, �400).E, mice weight of soluble hKL1-treated group and control group at the beginning and the end of the i.p. injections.






serve as an effective novel strategy for the treatment of breastcancer.

Although ample data indicate klotho as a potent tumor sup-pressor in a wide array of malignancies (9, 10, 15, 16, 36), itsprecise mechanism of action is still a matter of debate. Mechan-isms implicated in its tumor suppressor activities include inhibi-tion of insulin and IGF-1 signaling, modulation of FGF signaling,and inhibition ofWnt signaling (4, 5, 8, 9, 12, 15, 37).We noted acorrelation between the ability to inhibit the IGF-1 pathway,interactionwith the IGF-1R, and inhibition of proliferation. Thus,klotho and KL1 slowed proliferation, interacted with the IGF-1R,and inhibited the IGF-1 pathway. Conversely, KL2, which did notslow proliferation, did not inhibit the pathway or interacted withthe receptor. These data provide an indirect support to thehypothesis that inhibition of the IGF-1 pathway is a majormediator of klotho activities in breast cancer. Interestingly, thesame klotho region required for tumor suppression is the one

required for the inhibition of the IGF-1 pathway and of Wntsignaling (10, 34). Thus, it is possible that a commonmechanismthat involves inhibition of the Wnt and IGF-1 signaling pathwaysis responsible to klotho tumor suppressor activities.

While klotho enables activation of the FGFR pathway byFGF23, it inhibits activation of the IGF-1 pathway. Thus, treat-ment with klotho can lead to either activation or inhibition of theERK pathway, the downstream effector of both pathways.Although this may seem to be counterintuitive, it is importantto note that the activity of the ERK pathway is tissue and cell typedependent (38). The ERK pathway regulates a wide array ofphysiologic and pathologic processes, including proliferation,development, and metabolism, and while in the kidneys, as adownstream effector of FGF23 signaling, it participates in regu-lating phosphorus homeostasis (39), in cancer cells it enhancesproliferation as a downstream effector of multiple growth factors,including IGF-1 (40). Furthermore, the effect of klotho on various

Figure 4.Only KL1 domain suppresses IGF-1 signaling and interacts with the IGF-1R in breast cancer cells. A, MCF-7 cells were transfected as described above, starved for 48hours, and then stimulatedwith IGF-1 for 15minutes. Cell lysateswere immunoblotted using anti-phosphorylated (p) IGF-1R, p- and total (t)-AKT, andp- and t-ERK1/2antibodies. Phosphorylated-to-total ratio is calculated relative to untreated pcDNA3. Band quantification was performed using ImageJ. B, MCF-7 cellsoverexpressing FL-KL KL1 and KL2 or a control vector were lysed, and 400 mg protein was immunoprecipitated for total IGF-1R with anti–IGF-1R antibody. Afteraddition of protein A/G, immunocomplexes were washed, resolved on SDS–polyacrylamide gel electrophoresis gels, and immunoblotted with anti-HAantibody. Preliminary unprocessed cell lysate (input) is also shown (� , IgG heavy chain). C, MCF-7 and HEK-293 were transfected as described above, starved for 48hours, and then stimulatedwith FGF23 (100 ng/mL). Cell lysateswere immunoblotted using phospho (p) and total (t) ERK1/2. D, HEK-293 cellswere transfectedwitheither FL-KL, FL-KL without IC/TM, pEF (vector control), KL1, KL1 with IC/TM or pcDNA3 (vector control). Following 48-hour serum starvation, cells werestimulated with vehicle or 100 ng/mL FGF23 for 15 minutes and snap-frozen in liquid nitrogen. Cell lysates were prepared for Western blot using antibodies againstphosphorylated ERK1/2 (pERK1/2) and total ERK1/2.

Ligumsky et al.





tyrosine kinase receptors is also complex. Although klothoenhances activation of FGFRs by FGF23 (5, 8), either klotho orKL1 inhibit activation of the FGFRs by bFGF (10). As klothoregulates glycosylation pattern of various membrane proteins(33, 41), it is possible that some of its activities are mediated byregulating glycosylation pattern of the IGF-1R and FGFRs.

MCF-7 cells express somewhat higher klotho levels com-pared with MDA-MB-231 cells (9). Yet, MCF-7 cells are moresensitive to the tumorigenic effects of IGF-1 (42–45). Thesedata suggest a resistance to endogenous klotho in MCF-7 cells.The mechanisms mediating this resistance to endogenousklotho remain to be elucidated. The IGF-1 signaling may playa less prominent role in MDA-MB-231 cells (44), and othersignaling pathways, including Wnt signaling, are importantmediators of tumorigenesis in these cells. As klotho is a potentinhibitor of Wnt signaling in other malignancies, it is possible

that some of its effects in breast cancer are mediating byaffecting Wnt signaling (37, 46).

Several studies demonstrated a critical role for residues withinklotho's putative active site in mediating some of the activities ofklotho. Thus, mutations of these residues abolished its effect onTRPV5 and on the renal phosphate transporter NaPi-2a(11, 32, 33), but did not affect its ability to inhibit Wnt signaling(34). Our studies indicate that these amino acids are not requiredfor klotho tumor suppressor activity. These data strengthen apossible relationship between klotho tumor suppressor activityand Wnt inhibition and suggest that enzymatic activity of klothomay not be required for its activity against cancer cells. On theother hand, the klotho double mutant showed reduced ability tomediate FGF23 signaling. This observation indicates that klotho'sdiverse activities are mediated by different regions and may leadfor the development of klotho-based molecules showing

Figure 5.Mutating klotho putative enzymatic active site does not affect klotho anticancerous activity while compromising FGF23 signaling. A and B, MCF-7 and MDA-MB-231cells were transfected as described above. Transfected cells were cultured in media containing G418 for 2 weeks. Colonies were fixed and stained withcrystal violet andphotographed. All experimentswere conducted at least 3 times, and representativewells for each condition are shown.Quantitation of the colonieswas performed using ImageJ (� ,P<0.05, FL-KL, FL-KL-416, FL-KL-240-416, KL1, KL1-416, andKL1-240-416 vs.matched control; #,P<0.05, FL-KL-416, FL-KL-240-416vs. WT FL-KL and KL1, KL1-416 and KL1-240-416 vs. WT KL1). C, HEK 293 cells were transfected with either FL-KL, FL-KL-416, FL-KL-240-416, KL1, KL1-416,and KL1-240-416 or pcDNA3 (vector control). Following 48-hour serum starvation, cells were stimulated with vehicle or 100 ng/mL FGF23 for 15 minutes and snap-frozen in liquid nitrogen. Cell lysates were prepared for Western blot using antibodies against phosphorylated ERK1/2 (pERK1/2) and total ERK1/2.






differential activities. The design of such molecules may enablethedevelopment of rational novel active and safe cancer therapies.

Taken together, our data indicate differential activity of KL1and KL2 domains of klotho in breast cancer cells and suggestthat klotho tumor suppressor activities are not mediated byenzymatic activity. Importantly, our studies indicate that it ispossible to dissociate klotho tumor suppressor activities fromother activities of it. These observations may pave the way for arational design of klotho-based molecules for the treatment ofbreast cancer.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: H. Ligumsky, T. Rubinek, I. WolfDevelopment of methodology: H. Ligumsky, T. Rubinek, I. WolfAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): H. Ligumsky, K. Merenbakh-Lamin, S. Aviel-Ronen,I. Wolf

Analysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): H. Ligumsky, T. Rubinek, A. Yeheskel, R. Sertchook,S. Aviel-Ronen, I. WolfWriting, review, and/or revision of the manuscript: H. Ligumsky, T. Rubinek,A. Yeheskel, I. WolfAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): H. Ligumsky, S. Shahmoon, I. WolfStudy supervision: T. Rubinek, I. WolfOther (performed structural modeling Of Kl1): A. YeheskelOther (performed histopathologic evaluation): S. Aviel-Ronen

Grant SupportThis study was supported in part by the Israeli Science Foundation (grant

number 1112/09), the Israel Cancer Association Research Grant by Peter &Nancy Brown in memory of Eric & Melvin Brown, and the Sackler Faculty ofMedicine, Tel Aviv University, Tel-Aviv, Israel.

The costs of publication of this articlewere defrayed inpart by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received March 23, 2015; revised May 21, 2015; accepted June 10, 2015;published OnlineFirst June 25, 2015.

References1. Kuro-o M, Matsumura Y, Aizawa H, Kawaguchi H, Suga T, Utsugi T, et al.

Mutation of themouse klotho gene leads to a syndrome resembling ageing.Nature 1997;390:45–51.

2. Matsumura Y, Aizawa H, Shiraki-Iida T, Nagai R, Kuro-o M, Nabeshima Y.Identification of the human klotho gene and its two transcripts encodingmembrane and secreted klotho protein. Biochem Biophys Res Commun1998;242:626–30.

3. Chen CD, Podvin S, Gillespie E, Leeman SE, Abraham CR. Insulin stimu-lates the cleavage and release of the extracellular domain of klotho byADAM10 and ADAM17. Proc Natl Acad Sci U S A 2007;104:19796–801.

4. Kurosu H, Yamamoto M, Clark JD, Pastor JV, Nandi A, Gurnani P, et al.Suppression of aging in mice by the hormone klotho. Science 2005;309:1829–33.

5. Urakawa I, Yamazaki Y, Shimada T, Iijima K, Hasegawa H, Okawa K, et al.Klotho converts canonical FGF receptor into a specific receptor for FGF23.Nature 2006;444:770–4.

6. Imura A, Tsuji Y, Murata M, Maeda R, Kubota K, Iwano A, et al. Alpha-klotho as a regulator of calcium homeostasis. Science 2007;316:1615–8.

7. Razzaque MS. The FGF23-klotho axis: endocrine regulation of phosphatehomeostasis. Nat Rev Endocrinol 2009;5:611–9.

8. Kurosu H, Ogawa Y, Miyoshi M, Yamamoto M, Nandi A, Rosenblatt KP,et al. Regulation of fibroblast growth factor-23 signaling by klotho. J BiolChem 2006;281:6120–3.

9. Wolf I, Levanon-Cohen S, Bose S, Ligumsky H, Sredni B, Kanety H, et al.Klotho: a tumor suppressor and a modulator of the IGF-1 and FGF path-ways in human breast cancer. Oncogene 2008;27:7094–105.

10. Abramovitz L, Rubinek T, LigumskyH, Bose S, Barshack I, Avivi C, et al. KL1internal repeat mediates klotho tumor suppressor activities and inhibitsbFGF and IGF-I signaling in pancreatic cancer. Clin Cancer Res2011;17:4254–66.

11. Cha SK,Ortega B, KurosuH, Rosenblatt KP, KuroOM,Huang CL. Removalof sialic acid involving klotho causes cell-surface retention of TRPV5channel via binding to galectin-1. Proc Natl Acad Sci U S A 2008;105:9805–10.

12. Chang Q, Hoefs S, van der Kemp AW, Topala CN, Bindels RJ, HoenderopJG. The beta-glucuronidase klotho hydrolyzes and activates the TRPV5channel. Science 2005;310:490–3.

13. Li XX, Huang LY, Peng JJ, Liang L, Shi DB, Zheng HT, et al. Klothosuppresses growth and invasion of colon cancer cells through inhibitionof IGF1R-mediated PI3K/AKT pathway. Int J Oncol 2014;45:611–8.

14. Xie B, Zhou J, Shu G, Liu DC, Zhou J, Chen J, et al. Restoration of klothogene expression induces apoptosis and autophagy in gastric cancer cells:tumor suppressive role of klotho in gastric cancer. Cancer Cell Int 2013;13:18.

15. Chen B, Wang X, Zhao W, Wu J. Klotho inhibits growth and promotesapoptosis in human lung cancer cell line A549. J Exp Clin Cancer Res2010;29:99.

16. Camilli TC, Xu M, O'Connell MP, Chien B, Frank BP, Subaran S, et al. Lossof Klotho during melanoma progression leads to increased filamin cleav-age, increased Wnt5A expression, and enhanced melanoma cell motility.Pigment Cell Melanoma Res 2011;24:175–86.

17. Rubinek T, Shulman M, Israeli S, Bose S, Avraham A, Zundelevich A, et al.Epigenetic silencing of the tumor suppressor klotho in human breastcancer. Breast Cancer Res Treat 2012;133:649–57.

18. Imura A, Iwano A, Tohyama O, Tsuji Y, Nozaki K, Hashimoto N, et al.Secreted Klotho protein in sera and CSF: implication for post-translationalcleavage in release of Klotho protein from cell membrane. FEBS Lett2004;565:143–7.

19. Shiraki-Iida T, Aizawa H, Matsumura Y, Sekine S, Iida A, Anazawa H, et al.Structure of the mouse Klotho gene and its two transcripts encodingmembrane and secreted protein. FEBS Lett 1998;424:6–10.

20. Tohyama O, Imura A, Iwano A, Freund JN, Henrissat B, Fujimori T, et al.Klotho is a novel beta-glucuronidase capable of hydrolyzing steroid beta-glucuronides. J Biol Chem 2004;279:9777–84.

21. Hayashi Y, Okino N, Kakuta Y, Shikanai T, Tani M, Narimatsu H, et al.Klotho-related protein is a novel cytosolic neutral beta-glycosylceramidase.J Biol Chem 2007;282:30889–900.

22. Soding J. Protein homology detection by HMM-HMM comparison. Bio-informatics 2005;21:951–60.

23. Sali A, Potterton L, Yuan F, van Vlijmen H, Karplus M. Evaluation ofcomparative protein modeling by MODELLER. Proteins 1995;23:318–26.

24. Chaudhury S, Berrondo M, Weitzner BD, Muthu P, Bergman H, Gray JJ.Benchmarking and analysis of protein docking performance in Rosettav3.2. PLoS One 2011;6:e22477.

25. Roche DB, Buenavista MT, McGuffin LJ. Assessing the quality of modelled3D protein structures using the ModFOLD server. Methods Mol Biol2014;1137:83–103.

26. Ashkenazy H, Erez E, Martz E, Pupko T, Ben-Tal N. ConSurf 2010:calculating evolutionary conservation in sequence and structure of proteinsand nucleic acids. Nucleic Acids Res 2010;38(Web Server issue):W529–33.

27. UniProt C. Activities at the Universal Protein Resource (UniProt). NucleicAcids Res 2014;42:7486.

28. Altschul SF, Wootton JC, Gertz EM, Agarwala R, Morgulis A, Schaffer AA,et al. Protein database searches using compositionally adjusted substitu-tion matrices. FEBS J 2005;272:5101–9.

29. Katoh K, Misawa K, Kuma K, Miyata T. MAFFT: a novel method for rapidmultiple sequence alignment based on fast Fourier transform. NucleicAcids Res 2002;30:3059–66.

Ligumsky et al.





30. Wolf I, O'Kelly J, Rubinek T, Tong M, Nguyen A, Lin BT, et al. 15-hydroxyprostaglandin dehydrogenase is a tumor suppressor of humanbreast cancer. Cancer Res 2006;66:7818–23.

31. Mian IS. Sequence, structural, functional, and phylogenetic analyses ofthree glycosidase families. Blood Cells Mol Dis 1998;24:83–100.

32. HuMC, ShiM, Zhang J, Pastor J, Nakatani T, Lanske B, et al. Klotho: a novelphosphaturic substance acting as an autocrine enzyme in the renal prox-imal tubule. FASEB J 2010;24:3438–50.

33. Wolf MT, An SW, Nie M, Bal MS, Huang CL. Klotho up-regulates renalcalcium channel transient receptor potential vanilloid 5 (TRPV5) by intra-and extracellular N-glycosylation-dependent mechanisms. J Biol Chem2014;289:35849–57.

34. Liu H, Fergusson MM, Castilho RM, Liu J, Cao L, Chen J, et al. AugmentedWnt signaling in a mammalian model of accelerated aging. Science2007;317:803–6.

35. Liu F, Wu S, Ren H, Gu J. Klotho suppresses RIG-I-mediated senescence-associated inflammation. Nat Cell Biol 2011;13:254–62.

36. Doi S, Zou Y, Togao O, Pastor JV, John GB, Wang L, et al. Klotho inhibitstransforming growth factor-beta1 (TGF-beta1) signaling and suppressesrenalfibrosis and cancermetastasis inmice. J Biol Chem2011;286:8655–65.

37. Chen B, Ma X, Liu S, Zhao W, Wu J. Inhibition of lung cancer cells growth,motility and induction of apoptosis by Klotho, a novel secreted Wntantagonist, in a dose-dependent manner. Cancer Biol Ther 2012;13:1221–8.

38. Plotnikov A, Zehorai E, Procaccia S, Seger R. The MAPK cascades: signalingcomponents, nuclear roles and mechanisms of nuclear translocation.Biochim Biophys Acta 2011;1813:1619–33.

39. Andrukhova O, Zeitz U, Goetz R, Mohammadi M, Lanske B, Erben RG.FGF23 acts directly on renal proximal tubules to induce phosphaturiathrough activation of the ERK1/2-SGK1 signaling pathway. Bone 2012;51:621–8.

40. Santarpia L, Lippman SM, El-Naggar AK. Targeting the MAPK-RAS-RAFsignaling pathway in cancer therapy. Expert Opin Ther Targets 2012;16:103–19.

41. Cha SK, Hu MC, Kurosu H, Kuro-o M, Moe O, Huang CL. Regulation ofrenal outer medullary potassium channel and renal K(þ) excretion byKlotho. Mol Pharmacol 2009;76:38–46.

42. Huff KK, Kaufman D, Gabbay KH, Spencer EM, Lippman ME, Dickson RB.Secretionof an insulin-like growth factor-I-relatedprotein byhumanbreastcancer cells. Cancer Res 1986;46:4613–9.

43. de Blaquiere GE, May FE, Westley BR. Increased expression of both insulinreceptor substrates 1 and 2 confers increased sensitivity to IGF-1 stimulatedcell migration. Endocr Relat Cancer 2009;16:635–47.

44. BartucciM,Morelli C,Mauro L, Ando S, Surmacz E.Differential insulin-likegrowth factor I receptor signaling and function in estrogen receptor (ER)-positive MCF-7 and ER-negative MDA-MB-231 breast cancer cells. CancerRes 2001;61:6747–54.

45. Stewart AJ, Johnson MD, May FE, Westley BR. Role of insulin-like growthfactors and the type I insulin-like growth factor receptor in the estrogen-stimulated proliferation of human breast cancer cells. J Biol Chem 1990;265:21172–8.

46. King TD, Suto MJ, Li Y. The Wnt/beta-catenin signaling pathway: apotential therapeutic target in the treatment of triple negative breast cancer.J Cell Biochem 2012;113:13–8.






2015;13:1398-1407. Published OnlineFirst June 25, 2015.Mol Cancer Res Hagai Ligumsky, Tami Rubinek, Keren Merenbakh-Lamin, et al.

Function Analysis−by Structure Tumor Suppressor Activity of Klotho in Breast Cancer Is Revealed

Updated version

10.1158/1541-7786.MCR-15-0141doi:

Access the most recent version of this article at:

Material

Supplementary

http://mcr.aacrjournals.org/content/suppl/2015/06/26/1541-7786.MCR-15-0141.DC1

Access the most recent supplemental material at:

Cited articles

http://mcr.aacrjournals.org/content/13/10/1398.full#ref-list-1

This article cites 46 articles, 18 of which you can access for free at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://mcr.aacrjournals.org/content/13/10/1398To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-15-0141

http://mcr.aacrjournals.org/content/suppl/2015/06/26/1541-7786.MCR-15-0141.DC1

http://mcr.aacrjournals.org/content/13/10/1398.full#ref-list-1

http://mcr.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://mcr.aacrjournals.org/content/13/10/1398


Tumor Suppressor Activity of Klotho in Breast Cancer Is ... · Oncogenes and Tumor Suppressors...

Documents

Transcript of Tumor Suppressor Activity of Klotho in Breast Cancer Is ... · Oncogenes and Tumor Suppressors...