BertolottoC Nature 2011 MITF SupplementaryInformation

w w w. n a t u r e . c o m / n a t u r e | 1

SuPPLementarY InFormatIondoi:10.1038/nature10539

4

Table S1. Frequency of the germline Mi-E318K (c.952G>A) substitution in controls and in

patients with melanoma only, according to melanoma type

Subjects Number

of

non-carriers

Number

of

carriers*

Total Frequency

of Mi-

E318K

FET

p-value

OR

[95%CI]

Controls 1649 10 1659 0.003 - Reference

Patients with

melanoma only

586 17 603 0.014 7.8 x 10-5

4.78 [2.05-

11.75]

With family

history

363 8 371

0.011† 0.009

3.63[1.24-

10.29]

Without family

history

223 9 232

0.019† 1.7x 10

-4

6.65 [2.36-

18.41 ]

Patients with

melanoma only

586 17 603 0.014 7.8 x 10-5

4.78 [2.05-

11.75]

Single melanoma 311 4 315 0.006††

0.26 2.12 [0.48-

7.40]

Multiple

melanomas

275 13 288 0.023††

3.91x10-6

7.79 [3.12-

20.04]

*All carriers are heterozygotes for Mi-E318K.

FET is the Fisher’s exact test for the difference in Mi-E318K allele frequency between each group of

patients and controls.

OR (95% CI) is the odds-ratio (with 95% confidence interval) associated with the Mi-E318K carrier

status.

* Family history of melanoma and/or pancreatic cancer, FHM in Table S2;

† The Mi-E318K allele frequency was not significantly different (p = 0.22; Fisher’s exact test) between

patients with melanoma that did/did not have a family history.

†† The Mi-E318K allele frequency was significantly different (p=0.02; Fisher’s exact test)

between

patients with a single melanoma and those with multiple primary melanomas, irrespective of family

history.

SUPPLEMENTARY INFORMATION

2 | w w w. n a t u r e . c o m / n a t u r e

RESEARCH

5

Table S2. Clinical and Pathological features of the 27 Mi-E318K carriers

ID Series† Sex Type of

1st

cancer‡

Age at 1st

cancer

diagnosis,

years

Type of

other

cancers‡

Age at other

cancers

diagnosis,

years

Family history of

cancers§

19525 M+RCC M ccRCC 70 NM, LMM

in situ

75, 76 None

20389 M+RCC M SSM 37 ccRCC 55 Maternal great uncle, CRC

10276 M+RCC M ccRCC 51 NM 62 Mother, UADT; maternal

1st cousin, lung

19034 M+RCC M ccRCC 69 NM 69 NA

24976 M+RCC M ccRCC 52 SSM 52 Mother, breast; maternal

uncle, CRC; paternal

uncle, leukaemia

10254 FHM F SSM 41 2 SSM +

Lymphoma

53, 54, 55 Brother, CM; mother,

bladder; brother prostate

15168¥ FHM M NM 54 SSM 54 Father, CM; sister, CM;

nephew, CM; first cousin

OM (MITF WT)

23817 FHM F MM 60 - - Brother, CM

15012 FHM F SSM 33 - - Maternal first cousin, CM;

Maternal aunt, NST+

breast; maternal uncle,

CRC + BCC

22822 FHM F SSM 56 - - Brother, pancreatic

cancer*; mother, uterine

21255¥ FHM F SSM 36 - - Sister: in situ CM

11311¥ FHM M SSM 40 2 SSM 54, 57 Maternal aunt, bladder;

maternal first cousin: in

situ CM; maternal grand-

father, pancreatic

19063 FHM F LMM 53 LM 58 Maternal uncle, 2CM;

maternal first cousin, CM

21000 MPM M NM 65 4 SSM 65, 65, 65,

66

Brother UADT; father

UADT

23777 MPM F SSM 44 2 SSM,

Breast

45, 45, 46 Father, RCC + seminoma

22112 MPM F SSM 39 In situ CM 40 None

22772 MPM M SSM 66 SSM 67 Maternal aunt, breast

22746 MPM M SSM 26 SSM 27 NA

21822 MPM M SSM 51 SSM 53 None

21281 MPM M SSM 59 SSM 59 Mother, leukaemia

20801 MPM F AM 36 CM 36 Father, CRC; mother,

ovarian

25372 MPM M NM 62 SSM 63 Father, Lung; Paternal

aunt, Leukemia

11473 RCC F jRCC 17 - - Mother, breast + melanoma

(MITF WT)

21309 RCC F PRCC II 33 - - Mother, CM

21939 RCC M PRCC II 36 - - NA

25220 RCC F ccRCC 79 - - NA

26534 RCC F PRCC I 62 - - NA


SUPPLEMENTARY INFORMATION RESEARCH

6

¥ Pedigrees are shown in Figure S1a

† M, melanoma; RCC, renal cell carcinoma; FHM, melanoma with family history of melanoma and/or

pancreatic cancer (see Methods); MPM, multiple primary melanomas.

‡ ccRCC, clear cell RCC; SSM, superficial spreading melanoma; NM, nodular melanoma; LMM,

lentigo malignant melanoma; MM, mucosal melanoma; LM, lentigo maligna; jRCC, juvenile

carcinoma t(X;17)(p11;q25); AM, acral melanoma; PRCC, papillary renal cell carcinoma.

§

CRC, colorectal cancer; UADT, upper aero-digestive tumour; NST, nervous system tumour; BCC,

basal Cell Carcinoma; RCC, renal cell carcinoma; OM, ocular melanoma; NA, not available.

7

Table S3. Association analysis of Mi-E318K among patients with melanoma, renal cell

carcinoma (RCC), or both cancers, and controls, with and without adjusting for

principal components (PCs).

†All carriers were heterozygotes for Mi-E318K.

‡The Wald test is a test of the effect of Mi-E318K carrier status on cancer risk using

logistic regression, with and without adjusting for principal components (PCs). The PCs were

estimated by a principal components analysis of genome-wide SNP data.

§FET is the Fisher’s exact test for assessing whether the Mi-E318K allele frequency was

significantly different between each group of patients and controls (this test does not adjust

for principal components).

‼OR (95% CI) is the odds-ratio (with 95% confidence interval) associated with the Mi-

E318K carrier status. The odds-ratios were estimated by adjusting for principal components.

Subjects Number

of non-

carriers

Number

of

carriers†

Total Frequency

of

p.E.318K

Wald

test

p-value

(adjusti

ng for

PCs)‡

Wald test

p-value

(not

adjusting

for PCs)‡

FET

p-value§

OR [95 % CI]

‼

Controls 1380 9 1389 0.003 - - - Reference

Melanoma

and/or RCC

patients

517 19 536 0.018 4.3×10-5

2.2×10-5

1.2×10-5

5.41 [2.41-

12.18]

Melanoma

+ RCC

36 5 41 0.061 7.1×10-7

1.5×10-7

2.8×10-5

20.34 [6.19-

66.92]

Melanoma

only

424 11 435 0.013 4.1×10-3

2.3×10-3

2.6×10-3

3.75 [1.52-

9.24]

RCC only 57 3 60 0.025 2.6×10-3

2.1×10-3

0.01 8.15 [2.08-

31.96]



RESEARCH

Table S4: The Mi-E318K signature in RCC4 renal carcinoma cells. Changes in gene expression (Top

22 up-regulated and top 10 down-regulated genes) induced by Mi-E318K compared to those induced

by Mi-WT in RCC4 renal cell carcinoma cells.

Fold Change p-value ID Entrez Gene Name 3.41 3.17E-02 CCR7 chemokine (C-C motif) receptor 7

2.649 1.02E-03 CLDN14 claudin 14 2.596 1.03E-04 ADCY1 adenylate cyclase 1 (brain) 2.493 3.05E-03 SLAMF7 SLAM family member 7 2.446 2.03E-02 TRIM63 tripartite motif containing 63 2.262 1.69E-02 HRK harakiri. BCL2 interacting protein (contains only BH3 domain)

2.09 8.88E-04 ABCB5 ATP-binding cassette. sub-family B (MDR/TAP). member 5 1.954 3.21E-05 ACSM3 acyl-CoA synthetase medium-chain family member 3 1.919 1.17E-04 ITGBL1 integrin. beta-like 1 (with EGF-like repeat domains) 1.903 6.36E-03 GADD45G growth arrest and DNA-damage-inducible. gamma 1.891 7.49E-04 AICDA activation-induced cytidine deaminase 1.864 4.95E-03 DMKN dermokine 1.852 2.33E-03 BMP7 bone morphogenetic protein 7 1.831 1.03E-03 SPINK1 serine peptidase inhibitor. Kazal type 1

1.78 3.73E-03 AQP11 aquaporin 11 1.764 3.37E-05 TRPV2 transient receptor potential cation channel. subfamily V. member 2

1.74 8.59E-03 LY96 lymphocyte antigen 96 1.715 1.11E-05 PPY pancreatic polypeptide

1.705 6.92E-06 BAAT bile acid CoA: amino acid N-acyltransferase (glycine N-choloyltransferase)

1.687 1.26E-03 PLA1A phospholipase A1 member A 1.659 6.35E-03 PNLIPRP3 pancreatic lipase-related protein 3 1.583 2.96E-04 HSD17B6 hydroxysteroid (17-beta) dehydrogenase 6 homolog (mouse)

-1.590 1.06E-09 IRAK2 interleukin-1 receptor-associated kinase 2 -1.608 1.69E-03 VCAM1 vascular cell adhesion molecule 1 -1.681 3.69E-03 WNT5A wingless-type MMTV integration site family. member 5A -1.698 3.67E-03 EDN2 endothelin 2 -1.712 2.11E-03 IL6 interleukin 6 (interferon. beta 2) -1.933 1.68E-03 AFAP1L2 actin filament associated protein 1-like 2 -1.972 2.52E-05 HIST1H1D histone cluster 1. H1d

-2.479 1.85E-02 CXCL1 chemokine (C-X-C motif) ligand 1 (melanoma growth stimulating activity. alpha)

-2.700 2.29E-03 SHISA2 shisa homolog 2 (Xenopus laevis) -3.721 1.32E-02 EGR1 early growth response 1



Table S5: Genome-wide analysis of MITF binding with chromatin immunoprecipitation (top 40

gene loci). The table gives the ratio of the read number determined for Mi-E318K/Mi-WT at the

corresponding genes in 501mel melanoma cells and the location of the peak relative to the

transcriptional start site (TSS).

Chromosome gene_name ratio e318k/wt TSS_distance chr17 RNASEK 3.20 -33chr13 FNDC3A 2.87 -55chr17 GAA 2.77 -69chr2 CYP20A1 2.73 932chr12 STAT6 2.72 -636chr19 ILVBL 2.70 44chr11 PICALM 2.61 -138chr16 RSPRY1 2.60 -88chr1 SPOCD1 2.45 -57chr12 VPS33A 2.32 117chr19 DHPS 2.32 101chr12 ERP29 2.27 -35chr4 FAM13A 2.24 -22chr16 CLN3 2.21 -45chr17 LRRC48 2.20 50chr8 XKR9 2.16 -32chr12 BLOC1S1 2.15 17chr22 HMOX1 2.14 -193chr6 TNFAIP3 2.05 -375chr19 BAX 1.98 -105chr3 DIRC2 1.95 -1178chr12 EIF4B 1.95 105chr5 DUSP1 1.93 -90chr7 VGF 1.92 -320chr22 TOM1 1.89 579chr19 PIAS4 1.87 -103chr5 BTF3 1.85 -1347chr16 CLCN7 1.84 74chr9 UBAP2 1.84 -214chr7 NUDCD3 1.81 -50chr15 COMMD4 1.80 -57chr11 USP28 1.80 -235chr1 KLHL12 1.79 -17chr17 AATF 1.79 -101chr7 GET4 1.77 -124chrX ZNF630 1.77 -764chr11 RPUSD4 1.75 -50chr1 ZMYND12 1.75 203chr1 MTHFR 1.72 -122chr3 CAPN7 1.72 -2



RESEARCH

Co

mp

on

en

t 2

Component 1

Controls Melanoma only

Melanoma+RCC RCC only

Figure S1Family 1

ID15179, 31 yrsMi-E318K

Mi-E318K

ID151682 CM 54

ID15318Mi-WT

Mi-WT

ID15858OM 51

ID27708CM 54

CM < 37

CM

Mi-E318K

ID21256CM 44

Mi-E318K

Family 2

ID11311

CM 40,

54, 57

Family 3

Mi-E318KMi-E318K

ID11131

CM 57

Pancreaticcancer

Bladdercancer, 72

a

Mi-E318K

b

ID21255CM 36

1

Supplemental information

Figure legends

Figure S1: Mi-E318K mutation segregation within 3 melanoma pedigrees and principal

components analysis (PCA). a, All family members with DNA available for genotyping are

shown. Those with Mi-WT or Mi-E318K are indicated. Filled symbols indicate individuals

with cancer and the age at diagnosis is shown; CM, cutaneous melanoma; OM, ocular

melanoma. Lines through the symbols indicate deceased. b, PCA for the genome-wide SNP

data in 1,389 cancer-free controls and 536 cases that passed quality control. Plot of first two

principal components. The case-control samples are identified by colour: controls are black

circles, melanoma + RCC cases are blue squares, melanoma-only cases are magenta

diamonds, RCC-only cases are green triangles.

Figure S2: Melanomas and kidney cancers immunostained with antibodies to MITF

(×400). a, Mi-WT melanomas displayed average to strong nuclear staining in melanocytic

cells (ID#03H3466). b, Mi-E318K melanomas displayed intense nuclear staining

(ID#10R1515; ID#22112 in Table S2). c and d, Two Mi-E318K carriers displayed nuclear

staining in different kidney cancer subtypes, papillary type I (ID#10R1348; #26534 in Table

S2) and clear cell (ID10R834; #25220 in Table S2) renal cell carcinoma, respectively.

Figure S3: Mi-E318K alters a MITF SUMO binding site. a, Protein sequence alignment

from Human to Tetraodon shows that the ΨKXE sumoylation motif is highly conserved. b,

Schematization of the main structure-function domains and post-translational modification

sites that affect MITF activity. AD1, transactivation domain; bHLH-LZ, basic helix loop helix



ba

dc

Figure S2

1


Figure legends






















RESEARCH

KXEE318K

a

S73 S298 S307 S409P P P P

b

1 419

K182 K201

AD1 bHLH-LZ

K316S U S

Fi S3Figure S3

1


Figure legends




















2

leucine zipper domain, characteristic of the Myc super family of transcription factors; P,

phosphorylated site; S, sumoylated site; U, ubiquitinated site.

Figure S4: Lack of sumoylation did not alter MITF nuclear localization. a, HEK293 cells

were co-transfected with wild type Mi- or mutant (p.K182R; Mi-E318K; p.K182R/p.E318K)

myc-tagged MITF and either a control pSG5 vector (EV) or pSG5 that carried the HA-

SUMO2 sequence. Cells were assessed by Western blot for MITF, SUMO2, and ERK2. b,

Immunofluorescence analysis of HEK293 cells transfected with myc-tagged wild type Mi or

mutant (p.K182R, Mi-E318K and p.K182R+p.E318K) MITF and labelled with an anti-myc

epitope tag, followed by Alexa Fluor 594 anti-mouse secondary antibody. Cell nuclei were

counterstained with DAPI. The bar represents 10 µm. c, Corresponding cell lysates were

analysed by Western blotting with MITF and ERK2 antibodies.

Figure S5: Mi-E318K did not affect MITF transcriptional activity on TYR and CDKN2A

promoters. Luciferase reporter gene containing either TYR or CDKN2A promoter were

cotransfected in 501mel human melanoma cells with pCDNA3 empty (EV) or encoding wild

type Mi or Mi-E318K. The results were expressed as the fold stimulation over basal

luciferase activity. Data represent the mean + SD of three independent experiments.

Figure S6: Gene profiling analysis of Mi-WT relative to Mi-E318K in melanoma and

renal carcinoma cells. a, The Venn diagram generated by the list comparison algorithm

showed an intersection of 165 genes when comparing wild type Mi and Mi-E318K effects in

A375 melanoma cells from 3 biological replicates. b, Ingenuity pathway analysis (IPA®) of

the genes in common to both genotypes in A375 cells. c, The Venn diagram generated by the

list comparison algorithm showed an intersection of 240 genes when comparing wild type Mi



MITF

a

72

95

kDa

120

R K

ERK2 43

17SUMO2

K182R

+E31

8K

HA‐SUMO2

b

‐myc

DAPI

Mi‐E318KMi‐WT K182R+E318K

K182R

c

MITF

ERK2

Figure S4

2
























1 0 | w w w. n a t u r e . c o m / n a t u r e

RESEARCH

3

y

TYR

1

1.2

y

CDKN2A/p16INK4A

0 5

1

1.5

2

2.5

Luci

fera

se A

ctiv

ity

0 2

0.4

0.6

0.8

1

Luci

fera

se A

ctiv

it

0

0.5

0

0.2

EV Mi-WT Mi-E318KEV Mi-WT Mi-E318K

Figure S5

2























w w w. n a t u r e . c o m / n a t u r e | 1 1


a Mi‐E318K b

41Functions/ disorders Nb of molecules p- value

Cancer 69 7.8E-13 – 2.9-3

Genetic disorder 99 2.6E-11 – 2.3E-3

A375 cells

165

28Dermatological Conditions 35 6.2E-10 – 4.1E-4

Cell growth & proliferation 61 3.4E-11 – 2.6E-3

Cell Death 53 5.0E-10 – 2.9E-3

Mi‐WT Cell Movement 45 1.3E-9 – 1.4E-3

c dMi‐E318K

i / di d b f l l l

259

Functions/ disorders Nb of molecules p- valueCancer 80 9.1E-12 – 1.9-3

Genetic disorder 121 3.4E-10 – 1.6E-3

D l t l Di d 33 2 0E-8 1 6E-3

RCC4 cells

240

9

Developmental Disorder 33 2.0E-8 – 1.6E-3

Cell Development 58 9.1E-10 – 1.9E-3

Lipid Metabolism 40 1.9E-8 – 1.9E-3

Molecular transport 35 1 9E-8 1 9E-3

Mi‐WTMolecular transport 35 1.9E 8 – 1.9E 3

Figure S6

2























3

and Mi-E318K effects in RCC4 cells from 2 biological replicates. d, IPA® of the genes in

common to both genotypes in RCC4 cells. Note that although the Venn diagrams show a

different transcriptional profile in A375 and RCC4 infected by wild type MI or Mi-E318K,

no gene is statistically differentially regulated by the two constructs in A375 cells. In contrast,

32 genes are differentially regulated by wild type Mi and Mi-E318K in RCC4 cells (see table

S3).

Figure S7: Mi-E318K signature in renal carcinoma cells. Ingenuity pathway analysis

(IPA®) was performed on the 32 genes specifically regulated by Mi-E318K in RCC4 cells.

Figure S8: Effect of Mi-E318K on migration and invasion of human A375 melanoma

cells. a, Migration and b, invasion of A375 melanoma cells. Bar shows the mean +SD of

three independent experiments. t-test compared Mi-WT to Mi-E318K, *, p<0.05,


1 2 | w w w. n a t u r e . c o m / n a t u r e

RESEARCH

Functions/ disorders Nb of molecules p- valueCell Growth and proliferation 14 3.8E-5 – 8.5-3

Inflammatory response 12 3.8E-6 – 8.5-3

Cell to cell signaling and 12 3.7E-6 – 8.5-3

interactionCancer 12 2.0E-4 – 8.5-3

Gene expression 10 1.8E-5 – 8.5-3

5 3Cell movement 9 2.8E-5 – 8.5-3

Figure S7Figure S7

3






S3).






80 80* *

20

40

60

20

40

60

cells

/fiel

d

cells

/fiel

d

0 0EV Mi- WT Mi-E318K EV Mi-WT Mi-E318K

Figure S8

3






S3).






BertolottoC Nature 2011 MITF SupplementaryInformation

Documents