Figure S1. DNMT1 and DNMT3A expression levels are upregulated in high CIMP ACC and are associated with poor clinical outcome

Q

Expression (log2 normalized signal intensity)

DNMT1

DNMT3A

DNMT3B

co


*

4.5


**

7

6

0

++

3.5

6

5

m

o

intermediate

low

high

intermediate

intermediate

low

high

low

high

Expression (log2 normalized signal intensity)

TET1

TET2

TET3


0


1

o

4.5

1

o

6

o

+

0

5

6

3.5

+

8

5

O

low

intermediate

high

low

intermediate

high

low

intermediate

high

b

DNMT1

DNMT3A

1.0

1.0

High expression

Overall survival (%)

Moderate expression

0.8

0.8

Low expression

0.6

0.6

0.4

0.4

++

0.2

0.2

0.0

p-value = 1.22e-5

0.0

p-value = 0.146

0

50

100

150

0

50

100

150

Time (months)

Time (months)

(a) Expression levels of DNMT1, DNMT3A, DNMT3B, TET1, TET2 and TET3 in the ENSAT patients exhibiting ICIMP, iCIMP and hCIMP (two-sided t-test, *p< 0.05, ** p< 0.01, *** p <0.005).

(b) Kaplan-Meier estimates of overall survival for ACC patients, as a function of DNMT1 or DNMT3A.

Figure S2. DNMT1 and DNMT3A expression is are associated with high proliferation Correlation between DNMT1 and DNMT3A expression and cell proliferation in the ENSAT patients. hCIMP samples are represented in red, iCIMP in orange and lCIMP in blue. Patients for whom the CIMP status was not defined are represented in gray.

DNMT1

DNMT3A

Expression (log2 normalized signal intensity

8.0

spearman rho = 0.718

normalized signal intensity

p-value < 1.6e-16

CIMP Status

7.5

5.5

spearman rho = 0.227

p-value < 0.17

7.0

Expression (log2

· High CIMP ☒

5.0

6.5

. Intermediate CIMP

6.0

4.5

. Low CIMP

5.5

4.0

5.0

3.5

4.5

3.0

-0.2

-0.1

0.0

0.1

0.2

0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

Proliferation score

Proliferation score

Figure S3. Gene Set Enrichment Analysis on TCGA datasets comparing hCIMP and ICIMP samples

G2M checkpoint

b Allograft rejection

Enrichment score (ES)

0.7

NES = 3.182

0.00

0.6

-0.05

0.5

p.value < 0

-0.10

0.4

-0.15

-0.20

0.3

-0.25

0.2

-0.30

NES = - 2.21

0.1

-0.35

-0.40

p.value < 0

0.0

-0.45

Cholesterol homeostasis

Fatty acid metabolism

Enrichment score (ES)

0.45

NES = 1.88

0.10

0.40

p.value < 4e-4

0.05

0.35

0.00

0.30

-0.05

0.25

-0.10

0.20

-0.15

0.15

-0.20

0.10

-0.25

NES = - 1.73

0.05

-0.30

p.value < 4e-3

0.00

-0.35

DNA repair

Protein secretion

Enrichment score (ES)

0.40

NES = 1.841

0.10

0.35

0.05

p.value < 3e-4

0.30

0.00

-0.05

0.25

-0.10

0.20

-0.15

0.15

-0.20

0.10

-0.25

NES = - 1.66

0.05

-0.30

p.value < 6e-3

0.00

-0.35

-0.40

(a) Representative example of gene sets enriched in hCIMP.

(b) Representative example of gene sets enriched in ICIMP.

Figure S4. CIMP in ACC is characterized by lower abundance of tumor-infiltrating immune cells Relative abundance of tumor-infiltrating immune and non-immune stromal cell populations, computed using MCP-counter, in lCIMP, iCIMP and lCIMP samples from the ENSAT dataset. Comparison of relative abundances of each population using Kruskal-Wallis test followed by Dunn's test with Benjamini-Hochberg corrections.

T cells

CD8 T cells

Cytotoxic lymphocytes

Relative abundance

3.1

3.2

0

2.9

0

3.0

2.9

2.8

2.7

2.8

·

2.7

2.4

2.5

1

Low

Medium

High

Low

Medium

High

Low

Medium

High

NK cells

B lineage

Monocytic lineage

Relative abundance

3.0

3.6

8

0

5.0

0

3.4

2.8

3.2

4.5

2.6

3.0

4.0

2.4

2.8

3.5

Low

Medium High

Low Medium Hi

High

Low

Medium

High

Myeloid dendritic cells

Neutrophils

Fibroblasts

Relative abundance

4.2

3.4

7.0

3.8

3.3

6.0

·

3.4

3.2

5.0

3.0

3.1

4.0

Low

Medium

High

Low

Medium

High

Low

Medium

High

Cell growth rate 6 days



1.2


*

1

* T

T

T

* T

T

*

% of control

0.8

T

0.6


T

0.4


T

0.2

0

H295R

MUC1

HUVEC

☒ Control

☒ ΑΖΑ 1μM

☐ ΑΖΑ 5 μΜ

☐ ΑΖΑ 10μΜ

Figure S5. Impact of demethylating agent on cell proliferation in ACC cell lines

Effect of a 6-day treatment with the DNA methylation inhibitor 5-azacytidine (AZA) at 1, 5, or 10uM in the H295R (n=4), MUC1 (n=5) cell lines and the non-tumor HUVEC cell line (n=5). Cell growth was estimated by neutral red proliferation assay. P-values of the Kruskal-Wallis test and Dunn’s test for stochastic dominance are reported: * p-value< 0.05; ** p-value < 0.01; *** p-value < 0.005

adjusted p-value (FDR)

Pathway in Cancer

MAPK Signaling Pathway

Focal Adhesion

Endocytosis

Neuroactive Lig. Receptor Interaction

Axon Guidance

Regulation of Actin Cytoskeleton

Leukocyte Transendothelial Migration

Hypertrophic Cardiomyopathy

Vascular Smouth Muscle Contraction

Calcium Signaling Pathway Viral Myocarditis

Dilated Cardiomyopathy

Hematopoietic Cell Lineage

Adherens Junction

Complement & Coagul. Cascade Junc.

Adipocytokine Signaling Pathway

Hedgehog Signaling Pathway

Glutathione Metabolism

Bladder Cancer

Amino Sugar & Nucleotide Sugar Met.

Renin Angiotensin System

Glycosphingolipid Biosynth. Ganglio

Glycosphingolipid Biosynthesis Globo

Glycosaminoglycan Biosynthesis

KEGG pathways

Figure S6. Gene pathways differentially active in H295R compared to MUC-1 cells due to DNA hypermethylation

1.3e-2

4.8e-3

4.07e-3

2.9e-3

3.03e-2

5.76e-4

3.07e-2

4.82e-3

4.07e-3

1.4e-2

9.62e-2

4.82e-3

3.6e-2

3.42e-2

1.75e-2

3.6e-2

9.62e-2

5.63e-2

3.84e-2

9.32e-2

9.63e-2

4.77e-2

3.66e-2

3.6e-2

9.04e-2

0

5

10

15

Number of genes

Pathway enrichment analysis on genes that exhibit a significantly hypermethylated DMR in their promoter and significantly lower expression in H295R than in MUC-1 cells. Pathways related to the immune response are highlighted in red.