Folliculin-interacting Fnip1 and Fnip2 play critical roles in kidney tumor suppression in cooperation with Flcn

Hisashi Hasumia,b, Masaya Babaa,c, Yukiko Hasumia, Martin Langa, Ying Huanga, HyoungBin F. Oha, Masayuki Matsuod, Maria J. Merinoe, Masahiro Yaob, Yusuke Itob, Mitsuko Furuyaf, Yasuhiro Iribef, Tatsuhiko Kodamag, Eileen Southonh,i, Lino Tessarolloh, Kunio Nagashimaj, Diana C. Hainesk, W. Marston Linehana, and Laura S. Schmidta,l,1

aUrologic Oncology Branch, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892; Departments of bUrology and Molecular Genetics and fMolecular Pathology, Yokohama City University, Yokohama 236-0004, Japan; cInternational Research Center for Medical Sciences, Priority Organization for Innovation and Excellence, Kumamoto University, Kumamoto 860-0811, Japan; dRadiation Biology Branch and eLaboratory of Pathology, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892; gLaboratory for Systems Biology and Medicine, Research Center for Advanced Science and Technology, University of Tokyo, Tokyo 153-8904, Japan; hMouse Cancer Genetics Program, Center for Cancer Research, National Cancer Institute, Frederick, MD 21702; and iLaboratory of Animal Sciences Program, jElectron Microscope Laboratory, Cancer Technology Group, kVeterinary Pathology Section, Pathology/Histotechnology Laboratory, and lBasic Science Program, Leidos Biomedical Research, Inc., Frederick National Laboratory for Cancer Research, Frederick, MD 21702

Edited by Bert Vogelstein, Johns Hopkins University, Baltimore, MD, and approved February 19, 2015 (received for review October 10, 2014) Folliculin (FLCN)-interacting proteins 1 and 2 (FNIP1, FNIP2) are Ppargc1a-driven mitochondrial biogenesis (5–7). Importantly, mice homologous binding partners of FLCN, a tumor suppressor for heterozygous for Flcn inactivation develop renal tumors at 24 mo of kidney cancer. Recent studies have revealed potential functions age with increased mTORC1/2 activity and Ppargc1a expression, for Flcn in kidney; however, kidney-specific functions for Fnip1 and which mimics the human BHD tumor (6, 8–10), sug- Fnip2 are unknown. Here we demonstrate that Fnip1 and Fnip2 gesting potential therapeutic targets in metabolic pathways for play critical roles in kidney tumor suppression in cooperation with BHD-associated kidney cancer. Flcn. We observed no detectable phenotype in Fnip2 knockout The first FLCN interacting FNIP1 was identified mice, whereas Fnip1 deficiency produced similar to through protein–protein interaction studies of the FLCN protein those seen in Flcn-deficient mice in multiple organs, but not in (11). FNIP1 binds to the C terminus of FLCN and to AMP- kidneys. We found that absolute Fnip2 mRNA copy number was activated protein kinase (AMPK) (11), a critical molecule for low relative to Fnip1 in organs that showed phenotypes under energy sensing, further underscoring a central role for the FLCN/ Fnip1 Fnip1 deficiency but was comparable to mRNA copy number FNIP1 pathway in cellular metabolism. A second folliculin- Fnip1 Fnip2 in mouse kidney. Strikingly, kidney-targeted / double interacting protein FNIP2 was discovered through bioinformatics inactivation produced enlarged polycystic kidneys, as was previ- searches for sequences similar to FNIP1 (12, 13). Similar to ously reported in Flcn-deficient kidneys. Kidney-specific Flcn inac- FNIP1, FNIP2 was found to bind to the C terminus of FLCN and tivation did not further augment kidney size or cystic histology of Fnip1/Fnip2 double-deficient kidneys, suggesting pathways dys- to AMPK (12), suggesting a potential functional redundancy regulated in Flcn-deficient kidneys and Fnip1/Fnip2 double-deficient with FNIP1. Recent studies with Fnip1 knockout mouse models kidneys are convergent. Heterozygous Fnip1/homozygous Fnip2 have demonstrated that Fnip1 is required for B-cell development double-knockout mice developed kidney cancer at 24 mo of age, analogous to the heterozygous Flcn knockout mouse model, fur- Significance ther supporting the concept that Fnip1 and Fnip2 are essential for the tumor-suppressive function of Flcn and that kidney tumorigen- The role of FLCN as a tumor suppressor in kidney cancer has esis in human Birt–Hogg–Dubé syndrome may be triggered by loss been well documented, whereas the functional roles of follic- of interactions among Flcn, Fnip1, and Fnip2. Our findings uncover ulin (FLCN)-interacting proteins 1 and 2 (FNIP1 and FNIP2) in important roles for Fnip1 and Fnip2 in kidney tumor suppression kidney are unknown. In this study, we demonstrate that and may provide molecular targets for the development of novel double inactivation of Fnip1 and Fnip2 leads to enlarged therapeutics for kidney cancer. polycystic kidneys or kidney cancer, which mimics the pheno- types seen in Flcn-deficient kidneys and underscores the sig- folliculin | FNIP1 | FNIP2 | Birt–Hogg–Dubé syndrome | kidney tumor nificance of Fnip1 and Fnip2 in kidney tumor suppression. Moreover, we found that Fnip1/Fnip2 mRNA ratios differ ermline alteration of the folliculin (FLCN) , a novel among organs, which may reflect tissue-specific roles for each Gtumor suppressor, is responsible for Birt–Hogg–Dubé (BHD) Fnip. Our findings define Fnip1 and Fnip2 as critical compo- syndrome, an inherited kidney cancer syndrome characterized by nents of the Flcn complex that are essential for its tumor cutaneous fibrofolliculomas, pulmonary cysts, and an increased suppressive function and will aid in the development of novel risk for the development of kidney cancer (1–4). Genetic studies therapeutics for kidney cancer. using Flcn knockout mice have defined important roles for Flcn in metabolism. Kidney-targeted Flcn knockout mice developed Author contributions: H.H., M.B., T.K., W.M.L., and L.S.S. designed research; H.H., M.B., Y. Hasumi, M.L., Y. Huang, H.F.O., M.M., M.J.M., M.Y., Y. Ito, M.F., Y. Iribe, K.N., D.C.H., enlarged polycystic kidneys with elevated mechanistic target of and L.S.S. performed research; H.H., M.B., E.S., L.T., W.M.L., and L.S.S. contributed new rapamycin complex 1 (mTORC1) activity (5) and increased mito- reagents/analytic tools; H.H., M.B., W.M.L., and L.S.S. analyzed data; M.J.M. and D.C.H. chondrial biogenesis through up-regulated peroxisome proliferator- provided pathological diagnosis; and H.H., M.B., W.M.L., and L.S.S. wrote the paper. activated receptor gamma coactivator 1-alpha (Ppargc1a) (6). The authors declare no conflict of interest. Muscle-targeted Flcn knockout mice displayed both red-colored This article is a PNAS Direct Submission. muscle with increased mitochondrial biogenesis caused by elevated Freely available online through the PNAS open access option. Ppargc1, and cardiac hypertrophy with up-regulated mTORC1, 1To whom correspondence should be addressed. Email: [email protected]. which were ameliorated by Ppargc1a inactivation, suggesting that This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. Flcn might serve as a critical link connecting mTORC1 with 1073/pnas.1419502112/-/DCSupplemental.

E1624–E1631 | PNAS | Published online March 16, 2015 www.pnas.org/cgi/doi/10.1073/pnas.1419502112 Downloaded by guest on October 2, 2021 (14, 15). Interestingly, a Flcn knockout mouse model using the (MEFs) suppressed Ppargc1a mRNA and ATP production PNAS PLUS tamoxifen-inducible ER (mutated form of the ligand-binding (Fig. 2 G–I). Because of the potential functional redundancies domain of the receptor)-Cre system also displayed between Fnip1 and Fnip2, we postulated that Fnip1 and Fnip2 defects in B-cell development (14), suggesting Fnip1 knockout expression might differ from organ to organ and that this vari- mice might develop phenotypes similar to those that develop as able expression may determine the specific roles for Fnip1 and a consequence of Flcn deficiency. Fnip2 in those tissues. The recent technology of ddPCR enabled Furthermore, FLCN has been shown to have a variety of func- us to measure absolute mRNA copy number. Using this tech- tions that might potentially link AMPK, mTOR, and Ppargc1a nology, we compared the absolute copy number of Fnip1 and with other important pathways. Crystallographic studies have Fnip2 mRNA in wild-type mouse tissues. Interestingly, we ob- shown that the C terminus of FLCN may be distantly related to served dominant expression of Fnip1 in heart, skeletal muscle, Differentially Expressed in Normal Cells and Neoplasia (DENN) and bone marrow, the tissues in which we observed phenotypes domain proteins and may possess guanine exchange in Fnip1 knockout mice, whereas there was no significant dif- factor activity toward RAB35 (16). FLCN modulates TFE3 lo- ference between Fnip1 and Fnip2 mRNA copy number in kidney calization (17), which may play an important role in the exit of (Fig. 2J), in agreement with the absence of kidney phenotypes in stem cells from pluripotency (18), and interacts with other sig- Fnip1 and Fnip2 knockout mice. These data support the idea that – naling pathways including the von Hippel-Lindau hypoxia in kidney, the Fnip2 expression level, which is commensurate – – inducible factor vascular endothelial growth factor axis (19 with that of Fnip1, might maintain Fnip function in Fnip1- 21), the TGF-beta pathway (22, 23), Rho A signaling (24, 25), deficient kidneys, and therefore, double inactivation of Fnip1 cell cycle regulation (26, 27), Rag-mediated and Fnip2 would be necessary to develop a kidney-specific sensing (28, 29), and autophagy (30, 31). These findings un- phenotype. derscore FLCN as an important molecule, inactivation of which affects multiple pathways. Kidney-Targeted Fnip1/Fnip2 Double-Knockout Mice Develop Enlarged To clarify the function of FLCN-interacting proteins Fnip1 Polycystic Kidneys. Strikingly, kidney-specific inactivation of both and Fnip2, we inactivated Fnip1 and/or Fnip2 in mouse kidneys, Fnip1 and Fnip2 resulted in enlarged kidneys (Fig. 3A). MRI muscle, and heart and investigated the effect on the mTOR imaging revealed multiple round-shaped structures in the Fnip1/ pathway and mitochondrial metabolism. The absolute mRNA Fnip2 double-deficient kidneys (Fig. 3B), which were confirmed copy number of Fnip1 and Fnip2 was measured using droplet by H&E staining to be polycystic lesions (Fig. 3C) displaying digital PCR (ddPCR) technology. To evaluate functional syn- hyperplastic cells that protruded into the cystic lumen (Fig. 3D). ergy of Fnip1 and Fnip2 with Flcn, we also inactivated Flcn, Kidney-specific Fnip1/Fnip2 double-knockout mice showed a Fnip1,andFnip2 simultaneously in mouse kidneys. Finally, we significantly increased kidney/body weight ratio (n = 12; per- searched for latent tumor development in Fnip1 and Fnip2 centage kidney/body weight ratio mean = 11.04%; P < 0.001) knockout mice. (Fig. 3E). The survival time of kidney-targeted Fnip1/Fnip2 = = Results double-knockout mice was about 3 wk (n 14; median survival 20 d; P < 0.0001) (Fig. 3F). The enlarged polycystic kidneys of Neither Kidney-Targeted Fnip1 nor Fnip2 Knockout Mice Develop kidney-targeted Fnip1/Fnip2 double-deficient mice were remi- a Kidney Phenotype. To investigate Fnip1 function in mouse niscent of the kidney phenotype observed in Flcn-deficient kid- kidney, we crossbred mice carrying loxP-flanked Fnip1 alleles neys (5), underscoring the phenotypic similarities between these (floxed, f) (14) with cadherin 16 (CDH16)-Cre transgenic mice, two genotypes. which express Cre recombinase driven by the CDH16 promoter, thereby deleting Fnip1 gene sequences specifically in kidney. We Fnip1/Fnip2 Double-Deficient Kidneys Are Identical to Flcn-Deficient observed no significant phenotype in the Fnip1-deficient kidneys Kidneys. Previously we observed increased mTOR activity (5) except occasional tiny cysts (Fig. 1A). Therefore, we decided to and Ppargc1a-driven mitochondrial biogenesis in Flcn-deficient analyze Fnip2 function in kidney by generating a Fnip2 condi- kidneys (6). In fact, Fnip1/Fnip2 double-deficient kidneys showed

tional mouse carrying loxP-flanked Fnip2 alleles (floxed, f)(Fig.1 MEDICAL SCIENCES increased protein expression of Ppargc1a and signaling mole- B–D) and crossbreeding with CDH16-Cre transgenic mice. How- cules in the mTOR pathway, including the downstream target of ever, kidney-targeted Fnip2 inactivation also did not cause any mTORC1, phospho-Ulk1 at Ser757, that suppresses autophagy, phenotype in mouse kidney (Fig. 1E). Indeed, we could not find which was confirmed by the accumulation of sequestosome-1 any phenotype in whole-body Fnip2 knockout mice that affected life span. (SQSTM1)/p62 (Fig. 4 A and B). Increased respiratory capacity (n = 4each;P < 0.001) (Fig. 4C) and increased mitochondrial < The Relative Expression Levels of Fnip1 and Fnip2 Differ from Organ surface area (13 cells each; P 0.001) (Fig. 4D)werealsoob- to Organ. Previously, we reported that Fnip1 knockout mice served in the Fnip1/Fnip2 double-deficient kidneys. These data showed B-cell developmental defects, which were also observed further support the concept that double inactivation of Fnip1/ in Flcn knockout mice using the tamoxifen-inducible ER-Cre Fnip2 in kidney cells mirrors the same phenotype as that as- system (14), suggesting Fnip1 knockout mice might show phe- sociated with kidney-targeted Flcn inactivation. We next asked notypes similar to those resulting from Flcn deficiency. In addi- whether the enlarged polycystic kidney phenotype in kidney- tion to the B-cell phenotype, we found similar Flcn-deficient and targeted Fnip1/Fnip2 double-knockout mice developed through Fnip1-deficient phenotypes in skeletal muscle and heart. Muscle- the same pathway that had produced the identical phenotype in targeted Fnip1 knockout mice showed red-colored muscle with kidney-specific Flcn knockout mice. To answer this question, we increased mitochondrial biogenesis (myoglobin and cox4 read- crossbred kidney-specific Fnip1/Fnip2 double-knockout mice outs; Fig. 2 A and B), as well as cardiac hypertrophy with ele- with kidney-specific Flcn knockout mice to see whether Flcn vated mTORC1 activity (Fig. 2 C–F), which we had previously inactivation might have a synergistic effect on Fnip1/Fnip2 observed in muscle-targeted Flcn knockout mice (6, 32). Se- doubly inactivated kidneys. In fact, Flcn inactivation did not quence similarity between FNIP1 and FNIP2 and the shared enhance the size (n = 8; P = 0.554) (Fig. 4E)oralterthe interaction of FNIP1 and FNIP2 with FLCN and AMPK (12) histology (Fig. 4F)ofFnip1/Fnip2 double-deficient kidneys, implied that FNIP1 and FNIP2 might be functionally redundant. suggesting that the enlarged polycystic kidney phenotypes of In support of this, we observed that expression of either FNIP1 kidney-specific Fnip1/Fnip2 double-knockout mice and kidney- or FNIP2 in Fnip1/Fnip2 null mouse embryonic fibroblasts specific Flcn knockout mice developed through the same

Hasumi et al. PNAS | Published online March 16, 2015 | E1625 Downloaded by guest on October 2, 2021 pathway, possibly triggered by loss of an important functional A 500m 500m interaction among Flcn, Fnip1, and Fnip2. *** N.S. Fnip1/Fnip2 Double-Knockout Mice Develop Kidney Cancer. Our pre- 1.5 1.0 vious finding that heterozygous Flcn knockout mice developed 1.0 kidney tumors by 24 mo of age after the loss of the remaining 0.5 Flcn allele, thereby mimicking human BHD renal tumorigenesis,

mRNA (Relative value) strongly supports a tumor-suppressive role for FLCN in kidney

0.0 % Kidney/ body weight 0.0 Fnip1 Fnip1 Fni Fnip1 Fnip1f/+, Fnip1f/f, +/+ f/f p1 and underscores FLCN as a classical tumor suppressor (8, 9). Fnip1 +/+ , CDH1 f/f CDH16-Cre CDH16-Cre , CDH16 Because Fnip1/Fnip2 double deficiency in kidney resulted in the 6-Cr -Cr e e identical enlarged polycystic kidney phenotype seen under Flcn deficiency, and knockdown of FNIP1/FNIP2 was permissive for proliferative cell growth in a FLCN-restored human BHD-asso- B 10 11 12 13 Fnip2 targeting vector Neor TK =loxP =frt ciated kidney cancer cell line (Fig. 5A), we postulated that double inactivation of Fnip1 and Fnip2 might lead to tumor development 9 10 11 12 13 1415 Fnip2 Wt allele in the kidney. Double homozygous inactivation of Fnip1 and 5'probe 3'probe Fnip2 in whole body resulted in embryonic lethality (Table 1), 22.6kb(BamH1) 19.1kb(EcoRV) further supporting the essential nature of the Fnips and the similarity between Flcn and Fnip1/Fnip2 double-deficient pheno- 9110 11 12 13 415types. Homozygous Fnip1 knockout mice did not survive long Fnip2 Targeted allele Neor 5'probe 3'probe enough to observe latent tumorigenesis (median survival = 292 d) 16.9kb(BamH1) because of severe immunodeficiency resulting from B-cell de- 8.2kb(EcoRV) velopmental defects (14). Therefore, we decided to achieve dou- +Flp ble inactivation of Fnip1 and Fnip2 by heterozygous knockdown of 9110 11 12 13 415 Fnip2f allele Fnip1 together with homozygous knockdown of Fnip2.Strikingly, these mice developed kidney cancer without developing any par- +Cre ticular extrarenal phenotype, whereas neither heterozygous Fnip1 9 10 11 1415 Fnip2d allele knockout mice nor homozygous Fnip2 knockout mice displayed a kidney tumor phenotype (n = 42; median kidney tumor-free survival = 796 d; P < 0.0001) (Fig. 5B). Kidney cancer in human C 2 ll 1 2 ll 1 l e e c cell cel BHD syndrome presents as multiple tumors with a variety of S S S c E E l E ES d el d d histologies including hybrid oncocytic tumors, chromophobe renal c e te S cell S et e rgeted rg rg argete t E a a cell carcinoma, oncocytoma, papillary , and Wt ET Ta W T T clear cell renal cell carcinoma (4). Indeed, we observed multiple

22.6kb: Wt allele n io

t 19.1kb: Wt allele kidney tumor lesions with a variety of histologies in heterozy- 16.9kb: Targeted allele s e

ig gous Fnip1/homozygous Fnip2 double-knockout mice, most of d

V which were hybrid oncocytic tumors (Fig. 5C). We observed

coR 8.2kb: Targeted allele increased protein expression of mTORC1/2 pathway members E BamH1 digestion and Ppargc1a in these tumors by Western blot analysis (Fig. 5D)

D /+ f/+ d d/d 2 P 12 13 deficient kidneys did not show differences except for infrequent tiny cysts + FNIP2 FNIP2 FNI FNIP2 (Right). (B) Fnip2 gene-targeting vector was constructed by recombineering =loxP 146bp d 12 13 =frt 457bp: FNIP2 methodology using homologous recombination. A neomycin resistance (Neo r) FNIP2f =Primer 1 cassette flanked by Frt (bar) and loxP (triangle) sequences was inserted =Primer 2 =Primer 3 238bp: FNIP2f into intron 11 for positive selection, and the thymidine kinase gene was 238bp FNIP2d 146bp: FNIP2+ included for negative selection. A second loxP sequence was inserted into intron 13. Correctly targeted embryonic stem cells were identified by 457bp Southern blot analysis and injected into blastocysts to produce chimeras. Backcrossing to C57BL/6 mice produced heterozygous F1 offspring with E germline transmission of the Fnip2 floxed (f)-Neo allele. The Neo cassette flanked by Frt sites was excised in vivo by crossing with mice expressing the ) e *** 500m 500m Flp recombinase transgene under the β-actin promoter. To produce the

valu Fnip2 deleted (d) allele, Fnip2 f/+ mice were crossed with mice expressing

e N.S.

iv 1.0 t the Cre recombinase transgene under the ubiquitous β-actin promoter. a l 0.8 e Deletion of 12 and 13 resulted in a frameshift and premature termi- / body weight (R 0.6

y A 0.4 nation codon in exon 14, which was predicted to cause mRNA degradation N

R 0.2 by the nonsense-mediated decay mRNA surveillance system. (C) The tar- m

2 0.0

0.0 % Kidne F Fnip2 F Fnip2 f/+ f/f geted embryonic stem cells were screened by Southern blotting of BamH1- nip2 nip2 Fnip2 , Fnip2 , +/ f/ + f/f and EcoRV-digested DNA, using two different external probes located out- Fnip + / CDH16-Cre CDH16-Cre f, C + , CDH16- DH16-Cre side the targeting sequence, as shown in B.(D) PCR-based genotyping was Cre performed using DNA extracted from mouse tails for routine monitoring of inheritance in offspring. Locations of PCR primers are indicated by arrows. Fig. 1. Neither kidney-targeted Fnip1 nor Fnip2 knockout mice show a (E) Kidney-specific inactivation of Fnip2 was achieved by crossing with kidney phenotype. (A) Conditional Fnip1 knockout mice were crossbred with CDH16-Cre transgenic mice. Inactivation of Fnip2 mRNA was confirmed by CDH16-Cre transgenic mice. Inactivation of Fnip1 mRNA was confirmed by real-time PCR. n = 6 each at 3 wk of age. Mean ± SD. Student t test (Left). real-time PCR. n = 6 each at 3 wk of age. Mean ± SD. Student t test (Left). Size of Fnip2-deficient kidney was not significantly different from that of The size of the Fnip1-deficient kidney was not significantly different from control kidney. n = 6 each at 3 wk of age. Mean ± SD. Student t test (Middle). that of the control kidney. n = 11 each at 3 wk of age. Mean ± SD. Student Representative H&E staining of 3-wk old control and Fnip2-deficient kidneys t test (Middle). Representative H&E staining of 3-wk old control and Fnip1- did not show any difference in histology (Right).

E1626 | www.pnas.org/cgi/doi/10.1073/pnas.1419502112 Hasumi et al. Downloaded by guest on October 2, 2021 and immunostaining (Fig. 5E), further supporting the idea that PNAS PLUS Fnip1f/f, Fnip1f/f these tumors might develop through the same pathway as tumors A B CKM-Cre+ that develop in the heterozygous Flcn knockout mouse model. Fnip1 Ppargc1a Discussion Myoglobin Here we report kidney tumor-suppressive roles for Fnip1 and Fnip2 in cooperation with Flcn. Fnip1/Fnip2 double inactivation Cox4 Fnip1f/f, Fnip1f/f targeted to mouse kidney resulted in an enlarged multicystic CKM-Cre+ p-AMPK(T172) kidney phenotype shortly after birth, which was identical to the Total AMPK phenotype observed in Flcn-deficient kidneys (5). The ratio of -tubulin absolute Fnip1 to Fnip2 mRNA copy number was high in the organs that demonstrated a phenotype after Fnip1 inactivation, whereas in kidney where no Fnip1-deficient or Fnip2-deficient CDFnip1f/f, phenotype was observed, the absolute Fnip2 mRNA copy num- CKM-Cre+ f/f ber was comparable to that of Fnip1. Therefore, these data Fnip1 suggest that the ratio of absolute Fnip1 to Fnip2 mRNA copy number may determine the function of each Fnip in a particular organ. Moreover, whole-body heterozygous Fnip1/homozygous Fnip2 double-knockout mice developed kidney tumors at 24 mo Fnip1f/f, of age, implying that loss of interaction between Flcn, Fnip1, and Fnip1f/f CKM-Cre+ Fnip2 may trigger kidney cancer development. According to protein sequence, FNIP1 and FNIP2 show 49% identity and 74% similarity (12), suggesting possible functional redundancy. Double homozygous inactivation of Fnip1 and Fnip2 E F Fnip1f/f, Fnip1f/f specifically in mouse kidney resulted in an enlarged polycystic CKM-Cre+ 50 kidney phenotype. However, expression of one allele of either **** Fnip1 Fnip1 or Fnip2 in kidney-targeted Fnip1/Fnip2 knockout mice was m) 40 p-AMPK(T172) sufficient to rescue this phenotype, suggesting Fnip1 and Fnip2

er (  Total AMPK 30 may have functional redundancy. Because worms and flies have p-mTOR(S2448)

amet only one Fnip protein (12), it is possible that a second Fnip with 20 Total mTOR overlapping functions evolved from the primary Fnip to ensure 10 p-S6R(S235/236) redundancy and conserve its critical role in regulating kidney cell

Muscle di Total S6R proliferation through its interaction with Flcn. 0 Fnip1f/f, In contrast, the differences between FNIP1 and FNIP2 amino Fnip1f/f Gapdh CKM-Cre+ acid sequences imply potentially distinct functions for each FNIP. Therefore, the nonoverlapped functions of the FNIPs, as GH well as additional overlapping functions, will need to be eluci- **** dated in future experiments. 1 *** IP IP2 1.2 N Crystallographic studies revealed that the C terminus of FLCN F FN RNA 1 + O m 0.8 shows distant homology to DENN domain proteins, a family a 0.6 DKO DKO+ DK – c1 0.4 of GDP GTP exchange factors that activate Rab GTPases in- g FNIP1 r 0.2 a volved in membrane trafficking in eukaryotes (16). Notably, a p

FNIP2 0 MEDICAL SCIENCES P D DK DKO+ KO subsequent bioinformatics study reported that FNIP1 and FNIP2 -tubulin O + F F NI N also have novel DENN modules (33), raising the possibility that P1 IP2 complex assembly of FLCN-FNIP1 or FLCN-FNIP2 might form a noncanonical DENN module critical for GDP–GTP exchange IJ**** that would suffer functional arrest if any of the components **** *** N.S. Fnip1 r 900 were absent. Disruption of a noncanonical DENN module, e Fnip2 30 b 800 whose conformation may be controlled by Flcn/Fnip1 or Flcn/ m ) 700 u g ***

n 600 Fnip2 interactions and may be essential for regulation of proper ole/cell 20 *** 500 kidney cell proliferation rates, might result in development

NA ( 400

10 R 300 l of the polycystic kidney phenotype in our kidney-targeted in Acopyn ATPfm N ta 200

R vivo models. 0 100 m D D D /to KO K KO+ 0 O+FNI B H KidneyQ one ma eart ua FNIP2 drice r P ro ps 1 w Gapdh served as a loading control. n = 3 each at 6 wk of age. (G) Western blotting shows the restoration of either FNIP1 or FNIP2 in Fnip1/Fnip2 null Fig. 2. Fnip1 and Fnip2 expression differs from organ to organ. (A)Muscle- MEFs (DKO). α-tubulin served as a loading control. (H) Restoration of either targeted Fnip1 knockout mice show red-colored muscle relative to control FNIP1 or FNIP2 suppressed Ppargc1a expression in Fnip1/Fnip2 null MEFs muscle (6 wk of age). (B) Western blotting shows increased mitochondrial bio- (DKO). n = 3 each. Mean ± SD. P < 0.001 for DKO+FNIP1, P < 0.0001 for genesis and decreased pAMPK in Fnip1-deficient muscle. α-tubulin served as DKO+FNIP2; Student t test. (I) Restoration of either FNIP1 or FNIP2 sup- a loading control. n = 3 each at 6 wk of age. (C) Fnip1-deficient heart is enlarged pressed ATP levels in Fnip1/Fnip2 null MEFs (DKO). n = 6 each. Mean ± SD. relative to control heart (8 wk of age). (Scale bar: 5 mm.) (D) Histology of Fnip1- P < 0.0001. Student t test. (J) ddPCR showed Fnip1 expression was signifi- deficient heart shows enlarged cardiac fibers (8 wk of age). (Scale bars: 3 mm and cantly higher than Fnip2 expression in bone marrow, heart, and quadriceps 60 μm.) (E)MusclediameterofFnip1-deficient hearts was increased. n = 20 each of C57BL/6 mice but was not significantly higher than Fnip2 expression in at 8 wk of age. P < 0.0001, Student t test. (F) Western blotting of Fnip1-deficient kidney of C57BL/6 mice. n = 6 each at 6 wk of age. Mean ± SD. P < 0.001 for hearts showed decreased pAMPK and activated mTORC1 signaling molecules. bone marrow, heart, quadriceps. N.S., not significant. Student t test.

Hasumi et al. PNAS | Published online March 16, 2015 | E1627 Downloaded by guest on October 2, 2021 After FLCN was identified as a two-hit A for BHD-associated chromophobe, hybrid oncocytic, and clear cell kidney cancers, a search for FLCN in a broad spectrum of sporadic kidney tumors was conducted, but genetic analyses of these samples only rarely identified mutations in the FLCN gene (34, 35). Somatic FNIP1 or FNIP2 mutations have been detected in several cancers by whole-exome sequencing as part of The Cancer Genome Atlas (TCGA) project. Sequencing Fnip1f/+ Fnip1f/f Fnip1f/+ Fnip1f/f efforts by TCGA project detected infrequent FNIP1/FNIP2 Fnip2f/+ Fnip2f/+ Fnip2f/f Fnip2f/f mutations in urologic cancers, including clear cell renal carci- CDH16-Cre CDH16-Cre CDH16-Cre CDH16-Cre nomas (7/424; 1.7%) and bladder cancer (7/130; 5.4%), and in all B but one of the urologic cancer samples, the tumors had a single of either FNIP1 or FNIP2 (36). In comparison, uterine corpus endometrioid cancer displayed the highest percentage of tumors with FNIP1 and FNIP2 alterations (17/240; 7.1%), of which 5 (2.1%) had mutations in both (37). Infrequent somatic mutations of FLCN, FNIP1,andFNIP2 genes in sporadic C kidney tumors indicate that genetic alteration of these genes is not the direct cause of sporadic kidney tumorigenesis, but rather, that the status of the FLCN/FNIP1 or FLCN/FNIP2 interactions might be critical for sporadic kidney tumor suppression. Protein– protein interaction studies of sporadic kidney cancers would be important to elucidate the functional status of the FLCN/FNIP1/ FNIP2 complex in sporadic kidney cancer, especially in sporadic Fnip1f/+ Fnip1f/f Fnip1f/+ Fnip1f/f chromophobe renal cell carcinoma and oncocytoma, which are Fnip2f/+ Fnip2f/+ Fnip2f/f Fnip2f/f the most frequent histologic subtypes observed in human BHD CDH16-Cre CDH16-Cre CDH16-Cre CDH16-Cre syndrome (Fig. S1). The findings of this study, which characterize the phenotype of D Fnip1, Fnip2,andFnip1/Fnip2 knockout mouse models in multiple organs, further our understanding of the Flcn tumor suppressor pathway and underscore important roles for Flcn/Fnip1/Fnip2 interactions in inhibiting kidney cancer development. These data may lead to the development of novel diagnostics and therapeutics for kidney cancer that target the FLCN-FNIP pathway. Materials and Methods Animals. Mice carrying Flcn alleles and Fnip1 alleles flanked by loxP sites (floxed, f) and mice carrying a Fnip1 deleted (d) allele were generated as previously described (5, 14). Mice carrying Fnip2 alleles flanked by loxP sites (floxed, f) and mice carrying a Fnip2 deleted (d) allele were generated using the same strategy (5, 14). Briefly, the Fnip2 targeting vector was generated by inserting a neomycin resistance cassette flanked by Frt and loxP se- quences into intron 11 of Fnip2 and inserting a second loxP sequence into + intron 13. Deletion of exon 12 and exon 13 in the Fnip2d/ mice resulted in EF20 *** a reading frameshift and premature termination codon in exon 14. CDH16- *** f/f f/f *** Fnip1 , Fnip2 , CDH16-Cre (n=14) ht Cre transgenic mice, which express Cre recombinase under the cadherin 16 Fnip1f/+, Fnip2f/+, CDH16-Cre (n=14) eig 15 (CDH16) promoter specifically in adult renal tubules and developing geni- tourinary tract (38), were crossed with mice carrying floxed (f) alleles of Flcn, 100 Fnip1, and Fnip2 to inactivate those genes. Because we did not observe any Log rank test: f/f f/f f/d

10 ing / Body w N.S. phenotypic difference between Fnip1 , Fnip2 , CDH16-Cre and Fnip1 , iv p<0.0001 v Fnip2f/d, CDH16-Cre mice, we used these two genotypes interchangeably * sur idney 50 throughout these studies. Muscle-targeted inactivation of Fnip1 was done K * 5 n f/f

% by crossing Fnip1 mice and CKM-Cre transgenic mice (Jackson Laborato- ries). Mice were housed in National Cancer Institute animal facilities and oportio 0 r killed by CO2 asphyxiation for analyses according to the National Cancer F Fn Fnip1 P n Fn 0 ip1 ip Institute–Frederick Animal Care and Use Committee guidelines. Animal care 1 ip 0102030 f/+ f/f 1 f f/ , /+ f,F Survival time (day) , Fnip2 Fn ,Fn nip ip2 f/+ ip 2 f f/+ 2 f/ /f f , , ,CD , CDH1 C CD (C) H&E staining shows enlarged polycystic kidneys in 3-wk-old kidney-spe- DH16- H16-C H 1 6 cific Fnip1/Fnip2 double-knockout mice. (Scale bars: 500 μm.) (D) H&E 6 -Cre C -C re re staining reveals detailed histology of kidneys from 3-wk-old kidney-targeted re ( (n n (n=10 =15) (n=11) =12 Fnip1/Fnip2 double-knockout mice displaying hyperplastic cells protruding ) ) into the lumen (arrow) within the medulla (Upper). Normal glomeruli (G) and proximal renal tubules (P) were observed in the cortex (Lower). Fig. 3. Kidney-targeted Fnip1/Fnip2 double-knockout mice develop en- (Scale bars: 50 μm and 20 μm.) (E) Kidney-specific Fnip1/Fnip2 double- larged polycystic kidneys. (A) Fnip1 and Fnip2 alleles were deleted specifi- knockout mice show an increased kidney/body weight ratio. Homozygous cally in kidney using CDH16-Cre transgenic mice. Double inactivation of Fnip1/heterozygous Fnip2 double-knockout mice show a slightly increased Fnip1 and Fnip2 targeted to the kidney resulted in enlarged kidneys relative kidney/body weight ratio as a result of occasionally observed tiny cysts. to the controls (3 wk of age). (B) T2 weighted images (T2WI) of MRI show Mean ± SD. Two-sided Student t test. Three weeks of age. (F) Survival curve multiple round-shaped structures in Fnip1/Fnip2 double-knockout kidneys at of kidney-specific Fnip1/Fnip2 double-knockout mice. Proportion surviving ± 3 wk of age (Left). Striations of medulla and renal pelvis are seen (Right). SD. Log rank test. n = 14 each at 3 wk of age.

E1628 | www.pnas.org/cgi/doi/10.1073/pnas.1419502112 Hasumi et al. Downloaded by guest on October 2, 2021 procedures followed the National Cancer Institute Animal Care and Use PNAS PLUS ABFnip1f/+, Fnip1f/f, Committee guidelines. Fnip1f/+, Fnip1f/f, Fnip2f/+, Fnip2f/f, f/+ f/f CDH16-Cre CDH16-Cre Fnip2 , Fnip2 , Southern Blot Analysis of Embryonic Stem Cells and PCR Genotyping of Fnip2 CDH16-Cre CDH16-Cre p-mTOR Knockout Mice. KOD Hot DNA polymerase (Novagen) was used for generating Fnip1 (S2448) probes and routine PCR genotyping. A 5′ external probe for Southern blot Flcn analysis of targeted embryonic stem cells was generated by PCR with the p-mTOR(S2448) p-S6R following primers: forward, 5′-GAACCAAGCGGGGAGAATCGAA-3′; reverse, (S240/244) ′ ′ ′ p-mTOR(S2481) 5 -GCAAAGGCTGAACCCGTTACCA-3 .A3 external probe was also gener- ated by PCR with the following primers: forward, 5′-TCAGGCCAAACCA- Total mTOR TAGCCTCA-3′; reverse, 5′-TCAGACTCAGCTGGGGAACGAA-3′. Genomic DNA p-S6R(S235/236) was isolated from tail samples of mice using DirectPCR Reagent (Viagen Total S6R C Fnip1f/+, Fnip2f/+, CDH16-Cre Biotech, Inc.) according to manufacturer’s protocols. Nonradioactive South- 4EBP1 Fnip1f/f, Fnip2f/f, CDH16-Cre ern blotting was performed with DIG OMNI System for PCR probes (Roche ’ p-AKT(T308) o Molecular Biochemicals) according to the manufacturer s protocol. PCR ti 250 ** genotyping was performed with three primer sets to amplify wild-type

p-AKT(S473) otein) r 200 (146-bp PCR product), floxed (238-bp PCR product), and deleted (457-bp PCR Total AKT product) Fnip2 alleles: P1, 5′-ATGCTAGTGAGGAGGAGCCATTG-3′,P2,5′-AGGA- ial p 150 ′ ′ ′

dr CAGAGAAAGCACGTGCTAG-3 and P3, 5 -AGTGTGCCACTTCCTTCTGGTCG-3 . p-Ulk1(S757)

nsumption ra 100 We confirmed that the CDH16-Cre transgenes had no detectable effect on *** co

Total Ulk1 es/min mouse phenotypes. l n 50 ***

SQSTM1/p62 ge 0 /g mitochon (  Mo ddPCR and Real-Time PCR. mRNA was extracted from bone marrow, heart, I Oxy x II IV Ppargc1a plex lex kidney, and quadriceps of C57BL/6, using TRIzol (Invitrogen), and 500 ng mp Gapdh ComCompleCo mRNA was transcribed into cDNA, using SuperScript III reverse transcriptase (Invitrogen) with 10 μL scale. A microliter of cDNA was used to make droplets *** DE*** using the QX100 ddPCR system (Bio-Rad), and droplets were amplified and 25 *** Fnip1f/+, Fnip2f/+, Fnip1f/f, Fnip2f/f, *** N.S. analyzed with the Taqman assays for Fnip1 (Mm00620486_m1) labeled with CDH16-Cre CDH16-Cre N.S. FAM and for Fnip2 (Mm01220192_m1) labeled with VIC. Real-time PCR was ght

i 20 N.S. done as previously described (6). Primer sequences are as follows: mouse Fnip1-forward, 5′-cacagttagtaatgggctgcttgg-3′;mouseFnip1-reverse, 5′-ctgc-

dy we 15 o aaagaaagaggcactcctg-3′;mouseFnip2-forward, 5′-tttgctgccttactgactgcggtg-3′; ′ ′ 500x 10 mouse Fnip2-reverse, 5 -cgagaaggctttaatgggagggtg-3 ; mouse Ppargc1a-

ney / B forward, 5′-atgaccctcctcacaccaaacccacag-3′;mousePpargc1a-reverse, 5′-cttgag- ′ β ′ ′ Kid catgttgcgactgcggttgtg-3 ;mouse -actin forward, 5 -gacatggagaagatctggca-3 ; 5 N.S. % and mouse β-actin reverse, 5′-ggtctcaaacatgatctgggt-3′. 0 re re re re re x -C C MRI Imaging. T2-weighted images were obtained using a fast spin echo se- 0 6 6- 6-C 6C 1 1 1 quence (rapid acquisition of relaxation enhancement) with an echo time of H- H- 50 0 D DH-1 DH- DH16-C CD C C 13msandarepetitiontimeof2,500msbya7TMRIscannercontrolled , ,C , /f ,C /f , /+ f/f f f f/+ f 2 with ParaVision 5.0 (Bruker BioSpin MRI GmbH). n n p lc ip2 lc F F ,Fni , Fnip2 Fnip1f/+, Fnip2f/+, CDH16-Cre ,Fn f/f f/f Western Blotting, Immunofluorescence Staining, and Antibodies. Frozen kidney f/+ 1 *** p ip1 · Fnip1f/f, Fnip2f/f, CDH16-Cre ni n samples were homogenized in RIPA buffer (20 mM Tris HCl at pH 7.5, 150 mM ip1 F n f ,F 01020 F f/ NaCl, 1 mM EDTA, 1.0% Triton X-100, 0.5% deoxycholate, 0.1% SDS) sup- % mitochondria area n lc plemented with Complete protease inhibitor mixture and PhosStop phos- F phatase inhibitor mixture (Roche). For immunoblotting, 20 μg protein was

F loaded in each well. Immunofluorescence staining of kidney samples was MEDICAL SCIENCES Flcn f/f, Fnip1 f/f, Fnip2 f/f, Flcn f/f,Fnip1f/f,Fnip2f/f, done as previously reported (5, 9). Antibodies used for Western blotting CDH-16 Cre CDH16-Cre CDH16-Cre included phospho-mTOR (Ser2448), phospho-mTOR (Ser2481), total mTOR, total AKT, phospho-AKT (Ser473), phospho-AKT (Thr308), phospho-S6K

nofluorescence shows increased staining of p-mTOR (S2448) and pS6R (S240/ 244) in Fnip1/Fnip2 double-deficient kidney relative to control kidney. Nuclei were stained with DAPI (Blue). Representative of three mice at 3 wk of age. (C) Respiratory capacity of isolated mitochondria is increased in Fnip1/ Fnip2 double-deficient kidney relative to control kidney. State 3 respiration of complex I and complex II, and complex IV-dependent respiration, were measured by Seahorse XF96 analyzer. Mean ± SD. n = 4 at 3 wk of age. Student t test (two-sided). (D) Electron microscope images show increased mitochondrial mass in Fnip1/Fnip2 double-deficient kidney compared with control kidney. Arrows indicate mitochondria. (Scale bars: 10 μmand 500 nm.) Percentage of mitochondrial area per cell was quantified for the indicated genotypes. Thirteen cells were evaluated for each genotype. Mean ± SD. Student t test (two-sided). Three weeks of age. (E) The kidney/ Fig. 4. Fnip1/Fnip2 double-deficient kidneys are identical to Flcn-deficient body weight ratios of 3-wk-old mice with Flcn-deficient, Fnip1/Fnip2 kidneys. (A) Western blotting shows increased Ppargc1 and phospho-pro- double-deficient, and Flcn/Fnip1/Fnip2 triple-deficient kidneys show no teins of the AKT-mTOR pathway in Fnip1/Fnip2 double-deficient kidney. significant differences. Mean ± SD, n = 8 each. Student t test (two- Increased phospho-Ulk1 at Ser757, one of the readouts of mTORC1 activity, sided). (F) The histologies of kidneys from 3-wk-old Flcn-deficient, Fnip1/ correlates with the accumulated SQSTM1/p62, normally degraded by Fnip2 double-deficient, and Flcn/Fnip1/Fnip2 triple-deficient mice show autophagy. Mouse Fnip2 protein is not shown because of technical difficulty no differences. Arrows indicate hyperplastic cells protruding into the cyst developing a unique Fnip2 antibody that does not cross-react with mouse lumens that were observed in all of the genotypes. (Scale bars: 1 mm, Fnip1. Gapdh served as a loading control. n = 3 at 3 wk of age. (B) Immu- 50 μm, and 20 μm.)

Hasumi et al. PNAS | Published online March 16, 2015 | E1629 Downloaded by guest on October 2, 2021 d/+ d/d Table 1. Neonates from matings of Fnip1 Fnip2 males Doxy - + N.S. A FLCN and females 7 ** -actin ** 6 Doxy(-)+SiCT Genotype Neonates Percentage neonates with genotype 5 N.S. Doxy + + + + Doxy(+)+SiCT SiRNA CT F1 F2 F1/2 4 * wt/wt d/d 3 * Doxy(+)+SiF1/2 Fnip1 Fnip2 47 36 FLCN d/+ d/d FNIP1 2 Fnip1 Fnip2 85 64 1 d/d d/d FNIP2 Cell proliferation Fnip1 Fnip2 00 0 -actin Day1 Day3 Day5 Total 132 100 B Fnip1d/+ Fnip2d/d males and Fnip1d/+ Fnip2d/d females were mated, and d/d d/d 100 Fnip1 d/+, Fnip2 d/d (n=42) 132 neonates were obtained from 53 litters. No Fnip1 Fnip2 neonates Fnip1 d/+ (n=25) were observed. Fnip2 d/d (n=15)

50 Log-rank test: p<0.0001 (Thr389), total S6K, 4EBP1, phospho-S6R(Ser235/236), total S6R, phospho- Ulk1(Ser757), total Ulk1, SQSTM1/p62, phospho-AMPK (Thr172), total AMPK, FLCN, GAPDH, β-actin (all from Cell Signaling), FNIP1 (11), FNIP2 (12) at 1:1,000 0 Kidney tumor free survival 0 500 1000 1500 dilution, and PPARGC1A (H300), myoglobin (Santa Cruz Biotechnology) at days 1:200 dilution. Phospho-mTOR (Ser2448), pshopho-S6R (Ser235/236), and COX4 antibodies (Cell Signaling) were used for immunofluorescence at 1:100 di- C lution. Antibody–protein complexes were detected using Odyssey imager (LI-COR biotechnology). Because of high sequence similarity between mouse Fnip1 and Fnip2, developing an antibody that detects mouse Fnip2 but does not cross-react with mouse Fnip1 was technically difficult and unsuccessful.

Electron Microscopy. Mouse kidney samples were immediately immersed in 4% (vol/vol) formaldehyde/2% (vol/vol) glutaraldehyde (Electron Microscopy Sciences)/PBS. Small blocks were then cut, osmicated, and dehydrated before embedding. The blocks were sectioned and observed in a Hitachi H7600 (Tokyo, Japan) transmission electron microscope equipped with an XR41B CCD camera (Advanced Microscopy Techniques Corporation). Percentage mitochondrial area was analyzed with Image J (National Institutes of Health). The ratio of mitochondrial area/cell area was measured in 13 cells and rep- resented as means and 95% confidence intervals.

Respiratory Capacity of Isolated Mitochondria from Kidney Tissue. For respiratory capacity measurements of isolated mitochondria from kidney, an XF96 V3 PET plate (Seahorse Bioscience) was coated overnight with 1:15,000 polyethylenimine

solution/assay buffer (137 mM KCl, 2 mM KH2PO4, 2.5 mM MgCl2, 20 mM Hepes, 0.5 mM EGTA, 0.2% fatty acid-free BSA). Mitochondria isolated from kidney using the standard Nagarse method was attached to the plate bottom at 936 g for 10 min. The plate was warmed at 37 °C for 10 min and transferred to the Seahorse XF96 analyzer. State III respiration (maximum ADP-stimulated oxygen consumption ratio under sufficient substrate for mitochondrial complex) of complex I was measured with 16 μg mitochondria immediately after the addi- tion of 5 mM glutamate, 5 mM malate, and 0.5 mM adenosine diphosphate. State III respiration of complex II was measured with 8 μg mitochondria imme- diately after the addition of 5 mM succinate, 0.28 μM rotenone, and 0.5 mM adenosine diphosphate. Complex IV-dependent respiration was measured with DE2 μg mitochondria immediately after the addition of 0.5 mM tetramethyl- Normal kidney Kidney tumor phenylenediamine and 3 mM ascorbic acid. Normal kidney Kidney tumor

p-mTOR (S2448) )8

p-mTOR (S2481) 442S(R Total mTOR p-S6K (T389) (B) Kidney tumor-free survival demonstrates that heterozygous Fnip1/

Total S6K OT homozygous Fnip2 double-knockout mice developed renal tumors at a median p-S6R (S235/236)

m-p age of 796 d. Neither heterozygous Fnip1 knockout mice nor homozygous Total S6R Fnip2 knockout mice developed kidney cancer. Kidney tumor-free survival ± Total 4EBP1 d/+ d/d d/+ d/d p-AKT (T308) )632 SD. Log rank test. n = 25, 15, and 42 for Fnip1 ,Fnip2 ,andFnip1 /Fnip2 ,

p-AKT (S473) /532 respectively. (C) H&E staining shows kidney tumor development in hetero- Total AKT zygous Fnip1/homozygous Fnip2 double-knockout mice. Tumor developed

Fnip1 S(R6S-p from cyst wall (Tumor1 of Mouse1, 692 d old) or within the kidney without Flcn prominent infiltration (Tumor2 of Mouse1). Cells lining the cyst walls (Cyst1 Ppargc1a Gapdh of Mouse2) were occasionally piled up (Cyst2 of Mouse2, 670 d old). The most frequent histology was the hybrid oncocytic tumor (Tumor1–Tumor3 of Fig. 5. Fnip1/Fnip2 double-knockout mice develop kidney cancer. (A) Double Mouse2). Papillary projections from the cyst wall were occasionally observed knockdown of FNIP1 and FNIP2 in human kidney cancer cell line caused in- (Mouse3, 699 d old). (Scale bars: 1 mm, 100 μm, and 20 μm.) (D) Western creased cell proliferation. UOK257 FLCN-null kidney cancer cell line was blotting shows increased Ppargc1a and phospho-proteins of the AKT-mTOR reconstituted with wild-type FLCN in a doxycycline-inducible manner (Upper pathway in kidney tumors from Fnip1/Fnip2 double-knockout mice. Gapdh Left). FNIP1 and FNIP2 expression was knocked down with siRNA (Lower Left). served as a loading control. n = 7 each, 628–699 d old. (E) Immunostaining Cell proliferation was evaluated using MTT assay at days 1, 3, and 5 (Right). of kidney tumors from Fnip1/Fnip2 double-knockout mice demonstrate in- CT, Control siRNA; F1, FNIP1 siRNA; F2, FNIP2 siRNA; F1/2, FNIP1 siRNA + FNIP2 creased phospho-mTOR and pS6R (downstream readout of mTOR) compared siRNA. Mean ± SD. Student t-test (two-sided). n = 5 each. N.S., no significance. with adjacent normal kidney.

E1630 | www.pnas.org/cgi/doi/10.1073/pnas.1419502112 Hasumi et al. Downloaded by guest on October 2, 2021 Cell Lines. Fnip1/Fnip2 null MEFs (DKO) were generated by deleting all significant at a value of P < 0.05. Survival curves were obtained using PNAS PLUS alleles of Fnip1 and Fnip2 in MEFs extracted from Fnip1f/f, Fnip2 f/f mice, GraphPad Prism version 6.01. using adenoviral Cre recombinase (17). These MEFs were restored with either wild-type FNIP1 (DKO+FNIP1) or wild-type FNIP2 (DKO+FNIP2), Animal Care. National Cancer Institute–Frederick is accredited by the Asso- using Tet3G system (Clontech) (32). To eliminate a possible artifact from ciation for Assessment and Accreditation of Laboratory Animal Care In- doxycycline, 0.5 ng/μL doxycycline was added to all lines. Doxycycline- ternational and follows the Public Health Service Policy for the Care and Use inducible wild-type FLCN-expressing UOK257 (6) was transfected with Silencer of Laboratory Animals. Animal care was provided in accordance with the Select Predesigned SiRNA (Ambion) for FNIP1:5′- GGCAUAUAAUCGAAUAG- procedures outlined in the National Research Council’s “Guide for Care and UUtt-3′ and FNIP2:5′-CCACAACUGAUGAUUAGUAtt-3′, using Lipofectamine Use of Laboratory Animals.” RNAi MAX (Invitrogen). Cell proliferation was analyzed using TetraColorOne (Seikagakukogyo) according to manufacturer’s protocol. ACKNOWLEDGMENTS. We thank Dr. Peter Igarashi for CDH16-Cre trans- genic mice, Dr. Daniel Crooks for helpful discussions, and Louise Cromwell ATP Measurement. ATP was measured using ATPlite 1step (Perkin-Elmer) for excellent technical support with the mouse studies. M.L. was supported according manufacture’s protocol. by an annual outbound fellowship from Fondazione Italiana Ricerca sul Cancro, “Fellowship For Abroad 2011.” This work was supported by the Intramural Research Program of the NIH, National Cancer Institute, Center Statistical Analysis. Experimental data are summarized as the mean values for Cancer Research. This project has been funded in part with federal funds with SD. Statistical analyses were performed using a two-tailed Student t test from the Frederick National Laboratory for Cancer Research, NIH, under (SPSS Statistics version 20), and differences were considered to be statistically Contract HHSN261200800001E.

1. Birt AR, Hogg GR, Dubé WJ (1977) Hereditary multiple fibrofolliculomas with tricho- 21. Yan M, et al. (2014) The tumor suppressor folliculin regulates AMPK-dependent discomas and acrochordons. Arch Dermatol 113(12):1674–1677. metabolic transformation. J Clin Invest 124(6):2640–2650. 2. Zbar B, et al. (2002) Risk of renal and colonic and spontaneous pneumo- 22. Cash TP, Gruber JJ, Hartman TR, Henske EP, Simon MC (2011) Loss of the Birt-Hogg- thorax in the Birt-Hogg-Dubé syndrome. Cancer Epidemiol Biomarkers Prev 11(4): Dubé tumor suppressor results in apoptotic resistance due to aberrant TGFβ-mediated – 393 400. . Oncogene 30(22):2534–2546. 3. Nickerson ML, et al. (2002) Mutations in a novel gene lead to kidney tumors, lung wall 23. Hong SB, et al. (2010) Tumor suppressor FLCN inhibits tumorigenesis of a FLCN-null defects, and benign tumors of the hair follicle in patients with the Birt-Hogg-Dubé renal cancer cell line and regulates expression of key molecules in TGF-beta signaling. – syndrome. Cancer Cell 2(2):157 164. Mol Cancer 9:160. 4. Vocke CD, et al. (2005) High frequency of somatic frameshift BHD gene mutations in 24. Medvetz DA, et al. (2012) Folliculin, the product of the Birt-Hogg-Dube tumor sup- Birt-Hogg-Dubé-associated renal tumors. J Natl Cancer Inst 97(12):931–935. pressor gene, interacts with the adherens junction protein p0071 to regulate cell-cell 5. Baba M, et al. (2008) Kidney-targeted Birt-Hogg-Dube gene inactivation in a mouse adhesion. PLoS ONE 7(11):e47842. model: Erk1/2 and Akt-mTOR activation, cell hyperproliferation, and polycystic kid- 25. Nahorski MS, et al. (2012) Folliculin interacts with p0071 (plakophilin-4) and de- neys. J Natl Cancer Inst 100(2):140–154. ficiency is associated with disordered RhoA signalling, epithelial polarization and 6. Hasumi H, et al. (2012) Regulation of mitochondrial oxidative metabolism by tumor – suppressor FLCN. J Natl Cancer Inst 104(22):1750–1764. cytokinesis. Hum Mol Genet 21(24):5268 5279. 7. Chen J, et al. (2008) Deficiency of FLCN in mouse kidney led to development of 26. Kawai A, Kobayashi T, Hino O (2013) Folliculin regulates cyclin D1 expression through ′ polycystic kidneys and renal neoplasia. PLoS ONE 3(10):e3581. cis-acting elements in the 3 untranslated region of cyclin D1 mRNA. Int J Oncol 42(5): 8. Hartman TR, et al. (2009) The role of the Birt-Hogg-Dubé protein in mTOR activation 1597–1604. and renal tumorigenesis. Oncogene 28(13):1594–1604. 27. Laviolette LA, et al. (2013) Human folliculin delays cell cycle progression through late 9. Hasumi Y, et al. (2009) Homozygous loss of BHD causes early embryonic lethality and S and G2/M-phases: Effect of and tumor associated mutations. PLoS kidney tumor development with activation of mTORC1 and mTORC2. Proc Natl Acad ONE 8(7):e66775. Sci USA 106(44):18722–18727. 28. Petit CS, Roczniak-Ferguson A, Ferguson SM (2013) Recruitment of folliculin to lyso- 10. Klomp JA, et al. (2010) Birt-Hogg-Dubé renal tumors are genetically distinct from somes supports the amino acid-dependent activation of Rag GTPases. J Cell Biol other renal neoplasias and are associated with up-regulation of mitochondrial gene 202(7):1107–1122. expression. BMC Med Genomics 3:59. 29. Tsun ZY, et al. (2013) The folliculin tumor suppressor is a GAP for the RagC/D GTPases 11. Baba M, et al. (2006) Folliculin encoded by the BHD gene interacts with a binding that signal amino acid levels to mTORC1. Mol Cell 52(4):495–505. protein, FNIP1, and AMPK, and is involved in AMPK and mTOR signaling. Proc Natl 30. Possik E, et al. (2014) Folliculin regulates ampk-dependent autophagy and metabolic – Acad Sci USA 103(42):15552 15557. stress survival. PLoS Genet 10(4):e1004273. 12. Hasumi H, et al. (2008) Identification and characterization of a novel folliculin- 31. Dunlop EA, et al. (2014) FLCN, a novel autophagy component, interacts with GABARAP – interacting protein FNIP2. Gene 415(1-2):60 67. and is regulated by ULK1 phosphorylation. Autophagy 10(10):1749–1760. 13. Takagi Y, et al. (2008) Interaction of folliculin (Birt-Hogg-Dubé gene product) with 32. Hasumi Y, et al. (2014) Folliculin (Flcn) inactivation leads to murine cardiac hyper- – a novel Fnip1-like (FnipL/Fnip2) protein. Oncogene 27(40):5339 5347. –

trophy through mTORC1 deregulation. Hum Mol Genet 23(21):5706 5719. MEDICAL SCIENCES 14. Baba M, et al. (2012) The folliculin-FNIP1 pathway deleted in human Birt-Hogg-Dubé 33. Zhang D, Iyer LM, He F, Aravind L (2012) Discovery of Novel DENN Proteins: Im- syndrome is required for murine B-cell development. Blood 120(6):1254–1261. plications for the Evolution of Eukaryotic Intracellular Membrane Structures and 15. Park H, et al. (2012) Disruption of Fnip1 reveals a metabolic checkpoint controlling B Human Disease. Front Genet 3:283. lymphocyte development. Immunity 36(5):769–781. 34. da Silva NF, et al. (2003) Analysis of the Birt-Hogg-Dubé (BHD) tumour suppressor 16. Nookala RK, et al. (2012) Crystal structure of folliculin reveals a hidDENN function in genetically inherited renal cancer. Open Biol 2(8):120071. gene in sporadic renal cell carcinoma and colorectal cancer. J Med Genet 40(11): – 17. Hong SB, et al. (2010) Inactivation of the FLCN tumor suppressor gene induces TFE3 820 824. transcriptional activity by increasing its nuclear localization. PLoS ONE 5(12):e15793. 35. Khoo SK, et al. (2003) Inactivation of BHD in sporadic renal tumors. Cancer Res 63(15): – 18. Betschinger J, et al. (2013) Exit from pluripotency is gated by intracellular re- 4583 4587. distribution of the bHLH transcription factor Tfe3. Cell 153(2):335–347. 36. Cancer Genome Atlas Research Network (2013) Comprehensive molecular charac- 19. Preston RS, et al. (2011) Absence of the Birt-Hogg-Dubé gene product is associated terization of clear cell renal cell carcinoma. Nature 499(7456):43–49. with increased hypoxia-inducible factor transcriptional activity and a loss of metabolic 37. Kandoth C, et al.; Cancer Genome Atlas Research Network (2013) Integrated genomic flexibility. Oncogene 30(10):1159–1173. characterization of endometrial carcinoma. Nature 497(7447):67–73. 20. Bastola P, et al. (2013) Folliculin contributes to VHL tumor suppressing activity in renal 38. Shao X, Somlo S, Igarashi P (2002) Epithelial-specific Cre/lox recombination in the cancer through regulation of autophagy. PLoS ONE 8(7):e70030. developing kidney and genitourinary tract. J Am Soc Nephrol 13(7):1837–1846.

Hasumi et al. PNAS | Published online March 16, 2015 | E1631 Downloaded by guest on October 2, 2021