Accepted Manuscript

Title: Toxicogenomic and bioinformatics platforms to identify

key molecular mechanisms of a curcumin-analogue DM-1
toxicity in melanoma cells

Authors: Érica Aparecida de Oliveira, Diogenes Saulo de

Lima, Lucas Esteves Cardozo, Garcia Ferreira de Souza,
Nayane de Souza, Debora Kristina Alves-Fernandes,
Fernanda Faião-Flores, José Agustı́n Pablo Quincoces, Silvia
Berlanga de Moraes Barros, Helder I. Nakaya, Gisele
Monteiro, Silvya Stuchi Maria-Engler

PII: S1043-6618(17)30566-2
DOI: http://dx.doi.org/10.1016/j.phrs.2017.08.018
Reference: YPHRS 3675

To appear in: Pharmacological Research

Received date: 10-5-2017

Revised date: 31-7-2017
Accepted date: 30-8-2017

Please cite this article as: de Oliveira Érica Aparecida, de Lima Diogenes
Saulo, Cardozo Lucas Esteves, de Souza Garcia Ferreira, de Souza Nayane,
Alves-Fernandes Debora Kristina, Faião-Flores Fernanda, Pablo Quincoces José
Agustı́n, de Moraes Barros Silvia Berlanga, Nakaya Helder I, Monteiro Gisele,
Maria-Engler Silvya Stuchi.Toxicogenomic and bioinformatics platforms to identify
key molecular mechanisms of a curcumin-analogue DM-1 toxicity in melanoma
cells.Pharmacological Research http://dx.doi.org/10.1016/j.phrs.2017.08.018

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
 1

Toxicogenomic and bioinformatics platforms to identify key molecular

mechanisms of a curcumin-analogue DM-1 toxicity in melanoma cells

Érica Aparecida de Oliveira1, Diogenes Saulo de Lima2, Lucas Esteves Cardozo2,

Garcia Ferreira de Souza3, Nayane de Souza1, Debora Kristina Alves-Fernandes1,
Fernanda Faião-Flores1, José Agustín Pablo Quincoces3, Silvia Berlanga de Moraes
Barros1, Helder I Nakaya2, Gisele Monteiro4, Silvya Stuchi Maria-Engler1

1
Skin Biology Group, Clinical Chemistry and Toxicology Department, School of
Pharmaceutical Sciences, University of Sao Paulo, FCF/USP, Sao Paulo, Brazil.
2
Computational Systems Biology Laboratory, School of Pharmaceutical Sciences,
University of Sao Paulo, FCF/USP, Sao Paulo, Brazil.
3
Laboratory of Organic Synthesis, Anhanguera University of São Paulo, UNIAN, Sao
Paulo, Brazil.
4
Biochemical Pharmaceutical Technology Department, School of Pharmaceutical
Sciences, University of Sao Paulo, FCF/USP, Sao Paulo, Brazil.

*Corresponding author:
Silvya Stuchi Maria-Engler, Ph.D.
Department of Clinical Chemistry & Toxicology
Faculty of Pharmaceutical Sciences
University of São Paulo
580 Professor Lineu Prestes Avenue
São Paulo, Brazil
Zip Code: 05508-000
Tel: +55-11-3091-3631
Fax: +55-11-3813-2197
Email: silvya@usp.br

Graphical abstract
 2

Abstract

Melanoma is a highly invasive and metastatic cancer with high mortality rates and
chemoresistance. Around 50% of melanomas are driven by activating mutations in
BRAF that has led to the development of potent anti-BRAF inhibitors. However
resistance to anti-BRAF therapy usually develops within a few months and
consequently there is a need to identify alternative therapies that will bypass BRAF
inhibitor resistance. The curcumin analogue DM-1 (sodium 4-[5-(4-hydroxy-3-methoxy-
phenyl)-3-oxo-penta-1,4-dienyl]-2-methoxy-phenolate) has substantial anti-tumor
activity in melanoma, but its mechanism of action remains unclear. Here we use a
synthetic lethal genetic screen in Saccharomyces cerevisiae to identify 211 genes
implicated in sensitivity to DM-1 toxicity. From these 211 genes, 74 had close human
orthologues implicated in oxidative phosphorylation, insulin signaling and iron and RNA
metabolism. Further analysis identified 7 target genes (ADK, ATP6V0B, PEMT, TOP1,
ZFP36, ZFP36L1, ZFP36L2) with differential expression during melanoma progression
implicated in regulation of tumor progression, cell differentiation, and epithelial-
mesenchymal transition. Of these TOP1 and ADK were regulated by DM-1 in
treatment-naïve and vemurafenib-resistant melanoma cells respectively. These data
reveal that the anticancer effect of curcumin analogues is likely to be mediated via
multiple targets and identify several genes that represent candidates for combinatorial
targeting in melanoma.

Keywords: curcumin-analogue, DM-1, toxicogenomic, bioinformatics, melanoma,

TOP1 and ADK

1. Introduction

Metastatic melanoma is highly resistant to conventional therapy, being

responsible for a great number of deaths over the past decades[1]. Mitogen-Activated

Protein Kinase (MAPK) pathway plays a critical role in melanoma development, and it

is the main target of highly effective drugs, such as Vemurafenib and Dabrafenib

(BRAF inhibitors) and Trametinib and Cometinib (MEK inhibitors) [2, 3]. However, due

to the high degree of intra-tumor heterogeneity, resistance to B-RAF inhibitors is often

 3

observed after about 7 months of treatment [4]. Although combining therapies clearly

benefit patients, overcoming resistance still represents a major challenge for melanoma

cure [5].

Phytochemicals are gaining considerable attention because of their low toxicity,

low cost, and public acceptance as dietary supplements [6]. Curcumin

(diferuloylmethane) is a component isolated from the rhizome of Curcuma longa plants

[7], and has potent anti-carcinogenic, anti-microbial and anti-inflammatory activities [8-

11]. There is a large extent of studies all over the world describing the medicinal

properties of curcumin. These properties include benefits in many treatments as

chronic pain, inflammatory dermatoses, wound closure, skin infections and

inflammations [12].

In the cancer field, curcumin has been described possessing anti-proliferative

effects in many tumors, besides the inhibition of the transcription factor NF-κB and

downstream genes, for example c-myc, Bcl-2, COX-2, NOS, Cyclin D1, TNF-α,

interleukins and MMP-9)[9, 12-14]. The involvement of curcumin in tumor progression

occurs through modifications in growth factor receptors and cell adhesion molecules.

Despite all these aspects, the low oral bioavailability of stimulated the development of

curcumin analogues, such as EF24, with greater anti-tumor efficacy and metabolic

stability [12, 15].

Among these analogues, the DM-1 compound (4-[5-(4-hydroxy-3-methoxy-

phenyl)-3-oxo-penta-1,4-dienyl] -2-methoxy-2-phenolate sodium) can effectively inhibit

tumor growth by stimulating the production of free radicals, and by activating intrinsic

and extrinsic apoptosis and cell cycle arrest [16, 17], besides provides a promising

improvement in vivo melanoma treatment with a reduction of side effects after

combinatory therapy with dacarbazine [18]. However, it is still unclear which molecular

pathways are targeted by DM-1.

 4

By systematic deleting genes from cells or organisms and then measuring its

response to a compound, functional toxicology identifies the genetic requirements for

chemical tolerance [19, 20]. In addition, gene expression analyses, as well as cell

viability assays can be valuable tools in assessing cell response to toxic agents [21-

24]. The Yeast Saccharomyces cerevisiae Deletion Collection (YKO) represents the

only complete, systematically-constructed deletion collection available [25]. The great

advantage in using this toxicogenomic yeast platform relies on the large amount of

functional information available for each gene in the platform [20]. Although many

cytotoxic compounds act via physiological mechanisms that are absent in yeast, the

majority of basic mechanisms of toxicity, adaptation and resistance to drugs are

apparently conserved between yeast and more complex organisms [26].

Here, we highlight toxicogenomics, aiming to gain insights into the molecular

aspects of DM-1 toxicity in melanoma utilizing a genome deletion set of individual

genes in Saccharomyces cerevisiae YKO library. We show that DM-1 may have

multiple targets that may each contribute to its toxicity towards sensitive and resistant

melanoma cells, and also, the same genes were relevant in melanoma-patients

database pointing their meaning in this scenario.

2. Methods

2.1 DM-1 synthesis

DM-1 compound (sodium 4-[5-(4-hydroxy-3-methoxyphenyl)-3-oxo-penta-1,4-

dienyl]-2-methoxy-phenolate) was obtained by dried sodium ethanolate (0.01 mol)

mixed with 1,5-bis (4-hydroxy-3-methoxyphenyl)-1,4-pentadien-3-one (0.01 mol; 3.26

g) at a 1:1 molar ratio and stirred at room temperature under anhydrous reaction

conditions followed by solvent rotoevaporation until solidification. The compound

C19H17O5Na has a molecular weight of 348 g. Results of the structural characterization

 5

of the isolated compound were confirmed by Nuclear Magnetic Resonance

Spectroscopy 1H NMR (300.13 MHz) and 13C NMR (75.5 MHz) spectra and were

consistent with those described previously [27].

2.2 Source and maintenance of yeast cells

Wild-type Saccharomyces cerevisiae strain BY4741 (MATa his3Δ1 leu2Δ0

met15Δ0 ura3Δ0) and the yeast Knockout (YKO) deletion collection (Invitrogen,

Carlsbad, CA, USA) of nonessential gene, which is isogenic to BY4741, were used in

this study as described before [23]. Cells were grown in YPD medium (2% w/v glucose,

2% w/v peptone, 1% w/v yeast extract, 2% w/v agar added in the case of solid media)

or SD complete synthetic medium (0.67% w/v yeast nitrogen base, 0.12% w/v dropout

mixture of amino acids, excluding leucine, tryptophan and histidine, 2% w/v glucose, 60

μg/ml leucine, 20 μg/ml histidine, 40 μg/ml tryptophan, 20 μg/ml uracil, pH 6.5).

2.3 Treatment of yeast cells with DM-1 compound

The wild-type strain was grown in YPD and SD complete synthetic medium at

30°C for 18 hours with 200 rpm shaking. Cells were diluted to an initial optic density of

OD600nm = 0.2 and serial dilutions were plated in solid medium containing 1 to 20 mM of

DM-1 or vehicle (DMSO) and incubated for 30ºC for 48 hours to evaluation of viability

by colony formation. The yeast mutant cells were grown in 96-well plates at 30ºC with

200rpm shaking for 48 hours and then plated with a 96-pins stamping tool in a solid

medium containing DM-1 or vehicle, used as control, for more 48 hours. The cells

viability (increase or decrease due to DM-1 exposure) were assessed comparing the

growth ratio in absence of DM-1 treatment. The cut-off for selection was total ausence

of growth after treatment.

2.4 Bioinformatics analyses

 6

The selected genes were analyzed using the Saccharomyces Genome

Database (SGD) tool, Yeast Mine – Batch Analysis

(http://yeastmine.yeastgenome.org/yeastmine/) [28]. Using the “Analyse” tab, the list of

all genes whose deletion caused some response to DM-1 was inputted.

Through the “Homologues” tab, it was possible to retrieve human homologue(s)

of yeast gene (s) and any of their associated Gene Ontology Enrichment. Pathway

enrichment analysis was performed for the human homologues using Enrichr tool [29]

with Panther and BioCarta pathways (Adjusted P-value < 0.05). A gene network was

constructed with the human homologues using interaction data (“physical interactions”,

“genetic interactions” and “pathway”) from GeneMania tool [30] and cytoscape [31].

Finally, gene expression was determined by using datasets GSE3189, GSE8401 and

GSE46517 from Gene Expression Omnibus (GEO). Additionaly, 643 samples obtained

from a meta-analysis published by Kaushik et al. [32] were used. Expression of the

genes from the list of human homologue involved in sensitivity to DM-1 were compared

among skin, benign nevus, primary melanoma and metastatic melanoma. Bioconductor

package limma was used to assess differential expression [33]. Genes with log2 fold

change higher than 1.5 and adjusted p-value < 0.05 were considered differentially

expressed. Genes with altered expression were submitted to The Cancer Genome

Atlas Study (TCGA) online tool to detect genomic alterations.

2.5 Melanoma tumor cell line challenge with DM-1

The melanoma cell lines SKMEL-19, SKMEL-28, SKMEL-29 and A375 which

carries the BRAFV600E mutation, was kindly donated by Dr. Marisol Soengas (Centro

Nacional de Investigaciones Oncológicas Madrid, Spain) and Dr. Keiran Smalley

(Moffitt Cancer Center, Tampa, USA). The cell lines were sequenced using the

IonAmpliSeqTMCancer Hotspot Panel v2 (Life TechnologiesTM), which analyzed the

presence of mutations in fifty cancer-related genes was performed at Molecular

 7

Oncology Research Center, of Barretos Cancer Hospital, Barretos, SP, Brazil [34],

following the protocol in accordance with manufacturer’s instructions (Life Technologies

TM). These cell lines were grown in DMEM supplemented with 10% FBS or RPMI

supplemented with 5% FBS and antibiotics (100 U/mL penicillin and 100 µg/ml

streptomycin) at 37ºC in a humidified, 5% CO2 containing atmosphere. To derive a

vemurafenib-resistant cell line, cells were treated with 0.5–6 µM vemurafenib every 3

days for 4–6 weeks; then clonal colonies were isolated as well described in the

literature [34, 35]. These cells were named SKMEL-19R, SKMEL-28R, SKMEL-29R

and A375R. SKMEL-19R were replenished with 3.0 µM, SKMEL–28R and SKMEL-

29Rcells with 6.0 µM and A375R with 4.5 µM of vemurafenib every 2–3 days. Trypan

blue exclusion was used to monitor the inhibition concentration of the cell lines upon

exposure to DM-1 compound for 48h. The cells were plated in 24-well plates (1 ×

104/well). At the indicated time point, cells were trypsinized to detach from the plates

and diluted with 0.4% Trypan Blue staining solution. Cells were counted using a

Neubauer chamber. Each experiment was performed in triplicate.

2.6 Quantitative real-time PCR for relative RNA levels

Total RNA was extracted using the RNeasy Plus Mini Kit (Qiagen, Hilden,

Germany) according to the manufacturer’s recommendations. The reverse transcription

reactions were performed using the High-Capacity cDNA Reverse Transcription Kit

(Applied Biosystems, New Jersey, USA). Real-time PCR analyses were performed

using the ABI Prism®7500 Sequence Detection System (Applied Biosystems, New

Jersey, USA). TaqMan® Probe-Based Gene Expression (Life Technologies) kits were

used to perform the assays. The target genes evaluated were: ADK

(Hs00417073_m1), ATP6V0B (Hs01072388_m1), PEMT (Hs00540979_m1), TOP1

(Hs00243257_m1) and ZFP36 (Hs00185658_m1). Data were normalized to beta-actin

 8

(Hs01060665_g1) levels in the samples in triplicates. Relative expression was

calculated using the ∆∆Ct method [36].

2.7 Western Blotting for protein levels

Cell lysates were prepared in RIPA buffer with protease inhibitor cocktail

(Roche, Penzberg, Upper Bavaria,Germany) and phosphatase inhibitor cocktails I and

II (Santa CruzBiotechnology, Santa Cruz, CA, USA). After that, 40 μg of total protein

were subjected to electrophoresis in 10-12% gradient SDS gels under reducing

conditions, and subsequently transferred topolyvinylidene fluoride (PVDF) membranes

(Hybond-P, Amersham Pharmacia Biotech, Piscataway, NJ, USA). The membranes

were blocked in 3% BSA diluted in TBS-Tween 20 (50 mMTris-HCl, pH 7.5, 150 mM

NaCl, 0.1% Tween-20) for 1h and then were probed with TOP1 antibody (Abcam

ab109374 1:1000), ADK antibody (Abcam ab133665 1:1000) and Vinculin antibody

(Sigma V9131 1:3000). Protein bands were detected by an enhanced

chemiluminescence system (ECL; Amersham Pharmacia Biotech, Piscataway, NJ,

USA).

2.8 Statistical analysis

Statistical analyses and graphical representations were generated using

GraphPad Version 7.0 (GraphPad Software, La Jolla, CA, USA). All data were

expressed as mean ± standard deviation (SD). One-way analysis of variance (ANOVA)

was used for comparison and significant differences in the means were determined

using multiple comparisons with the Tukey–Kramer test at a significance level of p <

0.05.

3. Results
 9

3.1 Genome-wide screen identifies deletion mutants with altered sensitivity to

DM-1 compound

S. cerevisiae wild type cells were sensitive to DM-1 when cultivated on the YPD

(FIG. 1A). The 1mM concentration was too toxic to the mutants yeast (Supplementary

FIG. 1). Then, we chose the concentration of 0.7mM of DM-1 at which the mutant

strains were still viable.

To identify mutations modulating sensitivity to DM-1 we screened a collection of

over 6,000 gene-disruption mutants of S. cerevisiae on YPD medium containing either

0.7 mM DM-1 or vehicle (5% DMSO). From the primary data collected, we identified

211 mutants that showed enhanced sensitivity when exposed to DM-1. Examples of

strains exhibiting different sensitivity to DM-1 are presented in Figure 1B. In case of

sensitive strains, we observed 116 mutants with no growth and 95 with formation of

microcolonies. The complete list of these affected genes are presented in

Supplementary Table 1.

3.2 Gene Ontology analysis of the genes involved in altered DM-1 sensitivity

From the 116 highly sensitive yeast mutants (no colony formation), we selected

42 candidate yeast strains that represent 79 human homologues (Table 1). To analyze

possible relationships between the functions of genes whose deletion results in

changes in sensitivity to DM-1, we performed Gene Ontology enrichment. The over-

represented categories included several pathways, such as insulin, iron and RNA

metabolism, oxidative phosphorylation and MAPK signaling (FIG. 2A). Gene network of

human homologues sensitive to DM-1 are shown in Figure 2B. (Table 2).
 10

3.3 DM-1 treatment induces modulation of human homologues genes that are

frequently deregulated in melanoma

To investigate the signaling pathways involved in melanoma progression

targeted by DM-1, we cross-referenced the 79 human homologues found in the

toxicogenomic screening with available melanoma datasets obtained from GEO. The

relative expression of those genes in normal skin, benign nevi, primary and metastatic

melanomas are shown in Figure 3. Several of them presented altered expression

during melanoma progression. We next focused on those genes frequently altered in

all datasets, especially in the transition between nevi to primary melanoma or primary

melanoma to metastatic melanoma. We found 7 target genes that were highly

differentially expressed in these situations: ADK, ATP6V0B, PEMT, TOP1, ZFP36,

ZFP36L1, ZFP36L2. These genes not only exhibit differential expression in melanoma

samples from the analyzed databases but the deletion of their yeast homologues

reduced fitness when treated with DM-1. Analysis of the TCGA melanoma cohort

confirmed modulation of these genes during the transition to malignancy in melanoma

(FIG. 4A-C), and genetic alterations from patients (FIG. 4D). Overall, expression of

ADK gene is upregulated during progression from nevus to melanoma in situ, but its

expression decreases in metastatic melanoma. The ATP6V0B and TOP1 genes

present high expression only in metastatic tumors. By contrast, PEMT and ZFP36 were

downregulated in invasive melanoma. All seven genes present some form of genetic

alteration in patient samples, most notably TOP1, which was altered in 27% of patients.

3.4 DM-1 compound affects multiple targets to promote toxicity in treatment-

naïve and vemurafenib-resistant melanoma cells

 11

Given the differential expression in melanoma progression and the sensitizing

effects of mutation of these genes to DM-1 in yeast, we next treated human melanoma

cell lines with an IC25 concentration of DM-1 for 48 h (Supplementary Figure 2) and

examined the effects on their gene expression profiles using RT qPCR. We included

BRAF mutated cell lines that were either vemurafenib sensitive or derivatives that had

previously been selected for resistance. The cell lines used were SKMEL 19 and

SKMEL 19R (3µM), SKMEL 28 (6µM), SKMEL 28R (5µM), SKMEL 29 (1.1µM), SKMEL

29R (1.4µM), A375 (1.6µM) and A375R (2.3µM). The results revealed those genes

whose expression was significantly altered on exposure to DM-1, most notably

downregulation of TOP1 in treatment-naïve SKMEL29 and SKMEL28 cells and up-

regulation of ADK in all three vemurafenib-resistant cells (FIG. 5A). These main genes

(ADK and TOP1) were further investigated at the protein level (FIG. 5B), where was

confirmed the TOP1 regulation after DM-1 treatment and in some cell lines, the ADK

regulation. We also present the relative basal expression between naïve-treatment

and vemurafenib-resistant cells (Supplementary Figure 3).

4. Discussion

In melanoma, multiple molecular targets are deregulated [38]. While most

synthetic drug discovery efforts are focused on identification of single biological targets,

current clinical therapeutic approaches have moved on to introduce a combination of

drugs that target multiple pathways [13, 39]. Heterogeneity is imperative for explain the

genetic divergence between nevus, primary and metastatic lesions. Although BRAF

mutations drive the initial events in malign transformation, the melanoma harbors a

varied population of different mutations that undergo a selection process with only

certain clones being able to grow at distant sites [40-43]. In that way, metastasis

frequently hold a genomic profile distinct from the primary tumor, a process described

as parallel tumor progression [44].

 12

In this sense, curcumin and its analogues are known to hit multiple targets

owing to its capacity to regulate the expression and activity of a wide variety of

molecules that play central roles in many diseases. For example curcumin can inhibit

cyclooxygenase-2, HER2 (human epidermal growth factor receptor 2), tumor necrosis

factor, EGFR (Epidermal Growth Factor Receptor), the proteasome, topoisomerase I

and II and vascular endothelial cell growth [13, 39, 45]. Curcumin also induces

apoptosis and up-regulation of p53, p21, and p27 in cancer cells [13, 14, 39, 45]. The

analogues developed by Yue et al. [46] regulated proteasome-dependent degradation

and effectively induced cancer cell apoptosis through ER stress involving ATF-6

pathway, caspase-12 activation, and caspase-9/3 cascade reaction. CA15 exhibited

anticancer effects on laryngeal cancer cells via targeting of NF-κB [47]. In melanoma,

the D6 analogue promotes melanoma cytotoxicity partially driven by up-regulation of

the p53 signalling pathways as well as by down-regulation of the PI3K/Akt and NF-kB

pathways [48].

Taking these into account, an analog like DM-1 that could maintain the

multitarget action of curcumin, but with better bioavailability, would be promising for

therapies. DM-1 has been shown previously to exhibit toxicity to melanoma cells [16,

17]. However, understanding better its mode of action and its effect in the context of

melanoma sensitivity or resistance to vemurafenib will be crucial if it is to progress as

an effective therapeutic option.

Here, by using a yeast synthetic lethality assay, we have identified a number of

genes that may be implicated in the response to curcumin. As we can see, most of

them have a similar mRNA expression when compared naïve and resistant cells,

although this expression can vary in a different cell line. TOP1 gene has lower mRNA

expression in the most of resistant cells, probably for this reason, we are unable to see

a modulation after DM-1 treatment at gene expression. However, at protein level, we

clearly can see the positive modulation in both treatment-naïve and resistant cells. It is
 13

arguable that would be expected that a downregulation in the genes after DM-1

treatment to correlate to cell death. However, we used subtoxic concentrations of DM-1

in an attempt to induce the mechanism and still maintain viable cells. We also

performed additional tests with higher concentrations where all target genes were then

downregulated at mRNA level. Under low concentrations of toxic treatments, the tumor

cells can overcome the regulation and stimulate a compensatory response. Yet, when

the doses are increased the damage is efficient.

Several genes also exhibited statistically differential expression in melanoma in

response to DM-1. Most notably, TOP1, an important topoisomerase implicated in DNA

transcription, regulation and frequently mutated in melanoma. TOP1 is regulated by

DM-1 in the naive cell lines treated with BRAF-inhibitor even at the protein level.

Importantly, up regulation of TOP1 has been already associated with worse prognosis

but never investigated as a target in melanomas [46]. On the other hand, there are also

works that described no correlation between TOP1 expression and progression of

tumors [50,51]. Therefore, the literature about TOP1 expression levels frequently report

inconclusive results [52-55]. Nevertheless, TOP1 inhibitors have been widely studied in

cancer therapy, including some FDA-approved compounds as reviewed by Xu and Her

[56] and based on the results presented here TOP1 seems to be a potential target in

melanoma.

By contrast in vemurafenib-resistant cells, the main affected gene was ADK, an

encoding adenosine kinase that has a key role in generating AMP from adenosine and

consequently is a key regulator of energy homeostasis [57-59]. Although increased

adenosine has been linked to cytotoxic and apoptotic effects by interaction with the

mTOR pathway in several cancer types, it also inhibits immune and inflammatory

responses and stimulates angiogenesis, effects that might benefit tumor growth [60-

62]. Targeting ADK would modulate an entire network based on activation of multiple
 14

adenosine receptor-dependent pathways and on adenosine receptor-independent

biochemical, bioenergetic, and epigenetic mechanisms [63].

PEMT encoding phosphatidylethanolamine N-methyltransferase was

significantly increased in the resistant SKMEL-19 cell line, but not in the treatment-

naïve cells. Increased PEMT activity is essential for maintaining membrane integrity

and for preventing cell death. Higher PEMT activity may be associated with tumor

aggressiveness and has been already investigated in lung [64], breast [65] and liver

[66] tumors, due to its role in lipid metabolism that is necessary for highly proliferating

cancer cells. Although PEMT expression was higher in SKMEL19 cells treated with low

doses of DM-1, when treated with an IC50 concentration its expression decreases (data

not shown).

In SKMEL-28 and SKMEL-29 cells ATP6V0B expression was decreased after

DM-1 treatment. Increased expression of this gene, a transcriptional target of E2F1

(E2F Transcription Factor 1) can increase v-ATPase and mTORC1 activity, consistent

with ATP6V0B being responsible for mediating the effects of E2F1 on these responses

[67]. As such ATP6V0B plays a key role in cell growth, apoptosis, invasion and

metastasis.

A375 cells exhibited an increase in expression of ZFP36 when treated with DM-

1. ZFP36, ZFP36L1 and ZFP36L2 are members of tristetraprolin family. TTP

expression is significantly decreased in various cancers [68], regulating MMP1 and

MMP13 expression [69], the epithelial-mesenchymal transition [70], repression of

inflammatory gene transcription by NFkB [71], Cyclin D1 and c-myc [72].

BRAF mutated cells contribute to cancer phenotypes through metabolic

reprogramming by altering glycolysis, serine biosynthesis and glutamine utilization

among others [73-77]. BRAF inhibitors can reverse these effects, however resistance

has already been associated with enhanced expression of metabolic markers, such as

ABC transporters mediated by ATP binding [78-80].

 15

The main target genes selected from the toxicogenomic platform were

significantly modulated in at least one cell line after treatment with DM-1 compound.

Importantly, all these genes have previously been identified as being involved in the

regulation of tumor progression, cell differentiation and epithelial-mesenchymal

transition and were differentially regulated during phenotypic transitions among tumors

in the TCGA database.

We can suggest a possible mechanism of action of DM-1 clearly dependent on

TOP1 signaling in treatment-naïve melanomas and partially dependent mechanism of

ADK signaling in resistant melanomas, once we looked to a great variety of sensitive

genes and found a consisted pattern in these scenarios. We would like to remember

that this study is the first that investigates the DM-1 mode of action, and so far, further

studies are warranted to deeply understand the mechanisms involved with this analog

molecule that is widely known as a multi-target compound.

In summary, the results suggest that DM-1 may have multiple targets that may

each contribute to its toxicity towards cancer cells. In this work, we reveal by which

molecular pathways DM-1 toxicity appears to act in sensitive and resistant melanoma

cells. Together, we can notice that the main target genes selected from the

toxicogenomic platform were significantly modulated in at least one cell line after

treatment with DM-1 compound. Importantly, all these genes are involved somehow in

the regulation of tumor progression, cell differentiation and epithelial-mesenchymal

transition. Also, the same genes were relevant in melanoma-patients database pointing

their meaning in this scenario.

5. Conclusion

DM-1 compound, as curcumin analogues, is effective on multiple targets in

melanoma. Molecular mechanisms are cell line-dependent: TOP1 is relevant in naive

 16

cells treated with BRAF inhibitor, while ADK gene plays a role in BRAF inhibitor-

resistant cells. Taken together, these findings provide meaningful details into the core

understanding of the toxicity of this curcumin-analogue DM-1, and encourage further

studies in the development of combinatorial treatments for melanomas.

Funding support: Pos-Doc Fellowship grant PNPD/CAPES and FAPESP grant

(2014/24400-0).

Conflict of interest: The authors declare no conflict of interest.

The manuscript does not contain clinical studies or patient data.

Acknowledgements: The authors are grateful for a Pos-Doc Fellowship grant

PNPD/CAPES and FAPESP 2014/24400-0 Project. We also thank Professor Colin
Goding of Ludwig Institute for Cancer Research for his critical revision of the
manuscript.
 17

References

1. Tsao, H., et al., Melanoma: from mutations to medicine. Genes Dev, 2012. 26(11): p.
1131-55.
2. Flaherty, K.T., F.S. Hodi, and D.E. Fisher, From genes to drugs: targeted strategies for
melanoma. Nat Rev Cancer, 2012. 12(5): p. 349-61.
3. Grob, J.J., et al., Comparison of dabrafenib and trametinib combination therapy with
vemurafenib monotherapy on health-related quality of life in patients with
unresectable or metastatic cutaneous BRAF Val600-mutation-positive melanoma
(COMBI-v): results of a phase 3, open-label, randomised trial. Lancet Oncol, 2015.
16(13): p. 1389-98.
4. Paraiso, K.H. and K.S. Smalley, Fibroblast-mediated drug resistance in cancer. Biochem
Pharmacol, 2013. 85(8): p. 1033-41.
5. Rizos, H., et al., BRAF inhibitor resistance mechanisms in metastatic melanoma:
spectrum and clinical impact. Clin Cancer Res, 2014. 20(7): p. 1965-77.
6. Pal, H.C., et al., Phytochemicals for the Management of Melanoma. Mini Rev Med
Chem, 2016. 16(12): p. 953-79.
7. Sharma, R.A., A.J. Gescher, and W.P. Steward, Curcumin: the story so far. Eur J Cancer,
2005. 41(13): p. 1955-68.
8. Rai, D., et al., Curcumin inhibits FtsZ assembly: an attractive mechanism for its
antibacterial activity. Biochem J, 2008. 410(1): p. 147-55.
9. Goel, A., A.B. Kunnumakkara, and B.B. Aggarwal, Curcumin as "Curecumin": from
kitchen to clinic. Biochem Pharmacol, 2008. 75(4): p. 787-809.
10. Seo, J.H., et al., Lysophosphatidic acid induces STAT3 phosphorylation and ovarian
cancer cell motility: their inhibition by curcumin. Cancer Lett, 2010. 288(1): p. 50-6.
11. Yu, T., J. Ji, and Y.L. Guo, MST1 activation by curcumin mediates JNK activation, Foxo3a
nuclear translocation and apoptosis in melanoma cells. Biochem Biophys Res Commun,
2013. 441(1): p. 53-8.
12. Nguyen, T.A. and A.J. Friedman, Curcumin: a novel treatment for skin-related disorders.
J Drugs Dermatol, 2013. 12(10): p. 1131-7.
13. Goel, A., S. Jhurani, and B.B. Aggarwal, Multi-targeted therapy by curcumin: how spicy
is it? Mol Nutr Food Res, 2008. 52(9): p. 1010-30.
14. Siwak, D.R., et al., Curcumin-induced antiproliferative and proapoptotic effects in
melanoma cells are associated with suppression of IkappaB kinase and nuclear factor
kappaB activity and are independent of the B-Raf/mitogen-activated/extracellular
signal-regulated protein kinase pathway and the Akt pathway. Cancer, 2005. 104(4): p.
879-90.
15. Lelli, D., C. Pedone, and A. Sahebkar, Curcumin and treatment of melanoma: The
potential role of microRNAs. Biomed Pharmacother, 2017. 88: p. 832-834.
16. Faiao-Flores, F., et al., The curcumin analog DM-1 induces apoptotic cell death in
melanoma. Tumor Biology, 2013. 34(2): p. 1119-1129.
17. Faiao-Flores, F., et al., Bcl-2 family proteins and cytoskeleton changes involved in DM-1
cytotoxic effect on melanoma cells. Tumour Biol, 2013. 34(2): p. 1235-43.
18. Faiao-Flores, F., et al., Curcumin analog DM-1 in monotherapy or combinatory
treatment with dacarbazine as a strategy to inhibit in vivo melanoma progression.
PLoS One, 2015. 10(3): p. e0118702.
19. Afshari, C.A., H.K. Hamadeh, and P.R. Bushel, The evolution of bioinformatics in
toxicology: advancing toxicogenomics. Toxicol Sci, 2011. 120 Suppl 1: p. S225-37.
20. Yasokawa, D. and H. Iwahashi, Toxicogenomics using yeast DNA microarrays. J Biosci
Bioeng, 2010. 110(5): p. 511-22.
 18

21. Navarro, F.J., L. Weston, and P. Nurse, Global control of cell growth in fission yeast and
its coordination with the cell cycle. Curr Opin Cell Biol, 2012. 24(6): p. 833-7.
22. Costanzo, M., et al., The genetic landscape of a cell. Science, 2010. 327(5964): p. 425-
31.
23. de Sousa, G.F., et al., Chemogenomic Study of Carboplatin in Saccharomyces cerevisiae:
Inhibition of the NEDDylation Process Overcomes Cellular Resistance Mediated by HuR
and Cullin Proteins. PLoS One, 2015. 10(12): p. e0145377.
24. Vanacloig-Pedros, E., M. Proft, and A. Pascual-Ahuir, Different Toxicity Mechanisms for
Citrinin and Ochratoxin A Revealed by Transcriptomic Analysis in Yeast. Toxins (Basel),
2016. 8(10).
25. Giaever, G. and C. Nislow, The yeast deletion collection: a decade of functional
genomics. Genetics, 2014. 197(2): p. 451-65.
26. Mager, W.H. and J. Winderickx, Yeast as a model for medical and medicinal research.
Trends Pharmacol Sci, 2005. 26(5): p. 265-73.
27. SUÁREZ, J.A.Q., et al., Methods to prepare penta-1,4-dien-3-ones and substituted
cyclohexanones and derivatives with antitumoral and antiparasitic properties, the
compounds and their uses. 2009, Google Patents.
28. Cherry, J.M., et al., Saccharomyces Genome Database: the genomics resource of
budding yeast. Nucleic Acids Res, 2012. 40(Database issue): p. D700-5.
29. Kuleshov, M.V., et al., Enrichr: a comprehensive gene set enrichment analysis web
server 2016 update. Nucleic Acids Res, 2016. 44(W1): p. W90-7.
30. Warde-Farley, D., et al., The GeneMANIA prediction server: biological network
integration for gene prioritization and predicting gene function. Nucleic Acids Res,
2010. 38(Web Server issue): p. W214-20.
31. Shannon, P., et al., Cytoscape: a software environment for integrated models of
biomolecular interaction networks. Genome Res, 2003. 13(11): p. 2498-504.
32. Kaushik, A., et al., Gene Network Rewiring to Study Melanoma Stage Progression and
Elements Essential for Driving Melanoma. PLoS One, 2015. 10(11): p. e0142443.
33. Ritchie, M.E., et al., limma powers differential expression analyses for RNA-sequencing
and microarray studies. Nucleic Acids Res, 2015. 43(7): p. e47.
34. Sandri, S., et al., Vemurafenib resistance increases melanoma invasiveness and
modulates the tumor microenvironment by MMP-2 upregulation. Pharmacol Res, 2016.
111: p. 523-33.
35. Nazarian, R., et al., Melanomas acquire resistance to B-RAF(V600E) inhibition by RTK or
N-RAS upregulation. Nature, 2010. 468(7326): p. 973-7.
36. Livak, K.J. and T.D. Schmittgen, Analysis of relative gene expression data using real-
time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods, 2001. 25(4): p.
402-8.
37. Kabbarah, O., et al., Integrative genome comparison of primary and metastatic
melanomas. PLoS One, 2010. 5(5): p. e10770.
38. Read, J., K.A. Wadt, and N.K. Hayward, Melanoma genetics. J Med Genet, 2016. 53(1):
p. 1-14.
39. Keith, C.T., A.A. Borisy, and B.R. Stockwell, Multicomponent therapeutics for networked
systems. Nat Rev Drug Discov, 2005. 4(1): p. 71-8.
40. Lin, J., et al., Polyclonality of BRAF mutations in acquired melanocytic nevi. J Natl
Cancer Inst, 2009. 101(20): p. 1423-7.
41. Loewe, R., et al., BRAF kinase gene V599E mutation in growing melanocytic lesions. J
Invest Dermatol, 2004. 123(4): p. 733-6.
42. Husemann, Y., et al., Systemic spread is an early step in breast cancer. Cancer Cell,
2008. 13(1): p. 58-68.
 19

43. Stoecklein, N.H., et al., Direct genetic analysis of single disseminated cancer cells for
prediction of outcome and therapy selection in esophageal cancer. Cancer Cell, 2008.
13(5): p. 441-53.
44. Klein, C.A., Parallel progression of primary tumours and metastases. Nat Rev Cancer,
2009. 9(4): p. 302-12.
45. Roth, G.N., A. Chandra, and M.G. Nair, Novel bioactivities of Curcuma longa
constituents. J Nat Prod, 1998. 61(4): p. 542-5.
46. Yue, X., et al., Identification of 4-arylidene curcumin analogues as novel proteasome
inhibitors for potential anticancer agents targeting 19S regulatory particle associated
deubiquitinase. Biochem Pharmacol. 2017. 1(137):p.29-50.
47. Chen, J., Curcumin Analogue CA15 Exhibits Anticancer Effects on HEp-2 Cells via
Targeting NF-κB. Biomed Res Int. 2017. 2017:p. 4751260.
48. Rozzo, C., Molecular changes induced by the curcumin analogue D6 in human
melanoma cells. Mol Cancer. 2013. 4:p. 12:37.
49. Ryan, D., et al., Topoisomerase I amplification in melanoma is associated with more
advanced tumours and poor prognosis. Pigment Cell Melanoma Res, 2010. 23(4): p.
542-53.
50. Zhao, M. and R.A. Gjerset, Topoisomerase-I PS506 as a Dual Function Cancer
Biomarker. PLoS One, 2015. 10(8): p. e0134929.
51. Naniwa, J., et al., Genetic diagnosis for chemosensitivity with drug-resistance genes in
epithelial ovarian cancer. Int J Gynecol Cancer, 2007. 17(1): p. 76-82.
52. Guerreiro, A., et al., alpha-Synuclein aggregation in the saliva of familial transthyretin
amyloidosis: a potential biomarker. Amyloid, 2012. 19(2): p. 74-80.
53, Goldwasser, F., et al., Topoisomerase I-related parameters and camptothecin activity in
the colon carcinoma cell lines from the National Cancer Institute anticancer screen.
Cancer Res, 1995. 55(10): p. 2116-21.
54. Nieves-Neira, W. and Y. Pommier, Apoptotic response to camptothecin and 7-
hydroxystaurosporine (UCN-01) in the 8 human breast cancer cell lines of the NCI
Anticancer Drug Screen: multifactorial relationships with topoisomerase I, protein
kinase C, Bcl-2, p53, MDM-2 and caspase pathways. Int J Cancer, 1999. 82(3): p. 396-
404.
55. Rao, S., et al., Quantification and expert evaluation of evidence for chemopredictive
biomarkers to personalize cancer treatment. Oncotarget, 2016.
56. Xu, Y. and C. Her, Inhibition of Topoisomerase (DNA) I (TOP1): DNA Damage Repair and
Anticancer Therapy. Biomolecules, 2015. 5(3): p. 1652-70.
57. Bontemps, F., G. Van den Berghe, and H.G. Hers, Evidence for a substrate cycle
between AMP and adenosine in isolated hepatocytes. Proc Natl Acad Sci U S A, 1983.
80(10): p. 2829-33.
58. Bontemps, F., M. Mimouni, and G. Van den Berghe, Phosphorylation of adenosine in
anoxic hepatocytes by an exchange reaction catalysed by adenosine kinase. Biochem J,
1993. 290 ( Pt 3): p. 679-84.
59. Bontemps, F., M.F. Vincent, and G. Van den Berghe, Mechanisms of elevation of
adenosine levels in anoxic hepatocytes. Biochem J, 1993. 290 ( Pt 3): p. 671-7.
60. Giglioni, S., et al., Adenosine kinase gene expression in human colorectal cancer.
Nucleosides Nucleotides Nucleic Acids, 2008. 27(6): p. 750-4.
61. Vannoni, D., et al., Enzyme activities controlling adenosine levels in normal and
neoplastic tissues. Med Oncol, 2004. 21(2): p. 187-95.
62. Vannoni, D., et al., Metabolism of adenosine in human colorectal tumour. Nucleosides
Nucleotides Nucleic Acids, 2004. 23(8-9): p. 1455-7.
63. Boison, D., Adenosine kinase: exploitation for therapeutic gain. Pharmacol Rev, 2013.
65(3): p. 906-43.
 20

64. Zinrajh, D., et al., Increased phosphatidylethanolamine N-methyltransferase gene

expression in non-small-cell lung cancer tissue predicts shorter patient survival. Oncol
Lett, 2014. 7(6): p. 2175-2179.
65. Li, D., et al., Epigenetic repression of phosphatidylethanolamine N-methyltransferase
(PEMT) in BRCA1-mutated breast cancer. Oncotarget, 2014. 5(5): p. 1315-25.
66. Tessitore, L., et al., Expression of phosphatidylethanolamine N-methyltransferase in
human hepatocellular carcinomas. Oncology, 2003. 65(2): p. 152-8.
67. Meo-Evoli, N., et al., V-ATPase: a master effector of E2F1-mediated lysosomal
trafficking, mTORC1 activation and autophagy. Oncotarget, 2015. 6(29): p. 28057-70.
68. Ross, C.R., S.E. Brennan-Laun, and G.M. Wilson, Tristetraprolin: roles in cancer and
senescence. Ageing Res Rev, 2012. 11(4): p. 473-84.
69. Ryu, J., et al., Tristetraprolin inhibits the growth of human glioma cells through
downregulation of urokinase plasminogen activator/urokinase plasminogen activator
receptor mRNAs. Mol Cells, 2015. 38(2): p. 156-62.
70. Yoon, N.A., et al., Tristetraprolin suppresses the EMT through the down-regulation of
Twist1 and Snail1 in cancer cells. Oncotarget, 2016. 7(8): p. 8931-43.
71. Chen, Y.L., et al., Transcriptional regulation of tristetraprolin by NF-kappaB signaling in
LPS-stimulated macrophages. Mol Biol Rep, 2013. 40(4): p. 2867-77.
72. Marderosian, M., et al., Tristetraprolin regulates Cyclin D1 and c-Myc mRNA stability in
response to rapamycin in an Akt-dependent manner via p38 MAPK signaling.
Oncogene, 2006. 25(47): p. 6277-90.
73. Hutton, J.E., et al., Oncogenic KRAS and BRAF Drive Metabolic Reprogramming in
Colorectal Cancer. Mol Cell Proteomics, 2016. 15(9): p. 2924-38.
74. Haq, R., et al., Oncogenic BRAF regulates oxidative metabolism via PGC1alpha and
MITF. Cancer Cell, 2013. 23(3): p. 302-15.
75. Baenke, F., et al., Resistance to BRAF inhibitors induces glutamine dependency in
melanoma cells. Mol Oncol, 2016. 10(1): p. 73-84.
76. Kang, H.B., et al., Metabolic Rewiring by Oncogenic BRAF V600E Links Ketogenesis
Pathway to BRAF-MEK1 Signaling. Mol Cell, 2015. 59(3): p. 345-58.
77. Nagarajah, J., et al., Correlation of BRAFV600E Mutation and Glucose Metabolism in
Thyroid Cancer Patients: An (1)(8)F-FDG PET Study. J Nucl Med, 2015. 56(5): p. 662-7.
78. Michaelis, M., et al., Association between acquired resistance to PLX4032
(vemurafenib) and ATP-binding cassette transporter expression. BMC Res Notes, 2014.
7: p. 710.
79. Tang, C., et al., Tyrosine kinase inhibitor resistance in chronic myeloid leukemia cell
lines: investigating resistance pathways. Leuk Lymphoma, 2011. 52(11): p. 2139-47.
80. Balabanov, S., et al., Abcg2 overexpression represents a novel mechanism for acquired
resistance to the multi-kinase inhibitor Danusertib in BCR-ABL-positive cells in vitro.
PLoS One, 2011. 6(4): p. e19164.
 21

Fig. 1 Cell viability in wild-type yeast and Plate 6 as an example of mutant strains
exhibiting different sensitivity to DM-1 compound compared with the vehicle
after 48h treatment. (A) Wild-type exposure in YPD medium; (B) Plate 6 control with
DMSO; (C) Plate 6 with DM-1 0,7mM.
 22

Fig. 2 Functional toxicology of DM-1. (A) Pathway enrichment analyses of 79 human

homologues sensitive to DM-1. Brown and green bars represent the adjusted P-value
of enrichment for BioCarta and Panther pathways, respectively. Only pathways with
Adjusted P-value < 0.05 were selected. (B) Gene network of human homologues
sensitive to DM-1. The size and the color of nodes are proportional to the Closeness
Centrality and degree, respectively.
 23

Fig. 3 Expression analysis in Melanoma tumors of human homologues predicted

to be involved in DM-1 response. (A) Expression patterns of human homologues
during melanoma progression (normal skin, nevus, primary and metastatic
melanomas). Expression data of each gene (rows) was z-score normalized across all
samples and the mean of each group is represented (columns). Data was obtained
from Kabbarah et al. study (GSE46517)[37]; (B) Differential expression analysis of
human homologues in the melanoma meta-analysis performed by Kaushik et al. [32].
The graphs display the log2 fold-change values of genes whose expression is
significantly different (adjusted P-value < 0.05) in 3 different comparisons. The size of
the circles is proportional to the log2 fold-change value.
 24

Fig. 4 Bioinformatic analysis of the genes differentially expressed during

melanoma progression and implicated in the response to DM-1. (A) Volcano maps
of up and down regulated genes between nevus and melanoma as well, primary and
metastatic melanomas; (B) Graphic comparison of the mainly genes based in the
volcano maps; (C) Individual expression of the genes mostly altered after DM-1
treatment during melanoma progression; (D) Genetic alterations by TCGA database of
the mainly genes selected by GSE databases.
 25

Fig. 5 Relative mRNA expression and protein expression in sensitive and

resistant-melanoma cell lines treated for 48h with IC25 of DM-1 compound (A)
Relative mRNA expression of ADK, ATP6V0B, PEMT, TOP1 and ZFP36 genes by
Real time PCR. Beta-actin was used as housekeeping gene and all samples were
normalized by melanocytes expression. The error bars correspond to technical and
experimental triplicates. One way ANOVA ** p<0,01 e *** p<0,001and (B) protein
expression of ADK and TOP1 by Western Blotting. Vinculin was used as endogenous
control.
 26

Table 1. Identification of the yeast set selected

Systematic Standard Human Systematic Standard Human

Name Name Homolog Name Name Homolog
YOR002W ALG6 ALG6 YJL141C YAK1 HIPK3
YOR008C SLG1 MUC15 HIPK4
YOR067C ALG8 ALG8 DYRK1A
YOL004W SIN3 SIN3B HIPK1
SIN3A HIPK2
YOL006C TOP1 TOP1 DYRK3
TOP1MT DYRK4
YOL081W IRA2 NF1 PRPF4B
YDR126W SWF1 ZDHHC4 DYRK1B
YDR129C SAC6 LCP1 YOR270C VPH1 TCIRG1
PLS1 ATP6V02A
PLS3 ATP6V04A
YDR392W SPT3 SUPT3H ATP6V01A
YJL176C SWI3 SMARCC1 YJL128C PBS2 MAP2K2
SMARCC2 MAP2K3
YDR151C CTH1 ZFP36L1 MAP2K5
ZFP36L2 MAP2K7
ZFP36 MAP2K4
YDR226W ADK1 AK2 YBR127C VMA2 ATP5B
YNL229C URE2 GSTZ1 ATP6V1B1
GDAP1 ATP6V1B2
GDAP1L1 YDR448W ADA2 TADA2A
YNL219C ALG9 ALG9 TADA2B
YIL007C NAS2 PSMD9 YDR529C QCR7 UQCRB
YIL038C NOT3 CNOT3 YER131W RPS26B RPS26
YIL042C PKP1 BCKDK YDL047W SIT4 PPP6C
YFL025C BST1 PGAP1 YHR021C RPS27B RPS27
YIR042C NAT9 YPR131C NAT3 NAA20
YMR142C RPL13B RPL13 YNL220W ADE12 ADSSL1
YMR116C ASC1 GNB2L1 ADSS
APAF1 YNR031C SSK2 MAP3K4
YMR125W STO1 NCBP1 YNL064C YDJ1 DNAJA2
YHR026W VMA16 ATP6V0B DNAJB11
YJR073C OPI3 PEMT DNAJA4
YNL040W PTGES3L YGR063C SPT4 SUPTH4H1
PTGES3L – YLR447C VMA6 ATP6V0D2
AARSD1
PTGES3 ATP6V0D1
AARS YDL116W NUP84 NUP107
AARS2 YJR105W ADO1 ADK
AARSD1

Highly sensitive yeast set selected involved in the cellular response to

DM-1, according to the fitness defect showed in the presence of DM-1.
 27

Total number of genes analyzed = 116; genes that possess human

homologs = 42 (48,7%).

Table 2. Enrichment of human homolog. Set size: total of genes in the set; Genes in
input list: selected human genes presented in the set.
Genes
Set in
Pathway P-value Bonferroni FDR Datasets
size input
list

Insulin Receptor Recycling 23 6 2,44384E-09 3,2503E-06 2,82647E-06 Reactome

Transferrin Endocytosis and Recycling 25 6 4,25034E-09 5,65295E-06 2,82647E-06 Reactome

Latent Infection of Homo Sapiens with

33 6 2,56732E-08 3,41453E-05 1,13818E-05 Reactome
Mycobacterium Tuberculosis

Iron Uptake and Transport 36 6 4,45646E-08 5,9271E-05 1,48177E-05 Reactome

Epithelial cell Signaling in Helicobacter

68 7 9,80635E-08 0,000130424 2,60849E-05 KEGG
Pylori infection

Vibrio Cholerae infection 56 6 6,81031E-07 0,000905772 0,000150962 KEGG

Oxidative Phosphorylation 135 8 8,34401E-07 0,001109753 0,000158536 KEGG

Anthrax Pathway 17 4 2,15088E-06 0,00286067 0,000350292 PID

Signaling by insulin receptor 108 7 2,3704E-06 0,003152632 0,000350292 Reactome

MAPK Pathway 87 6 9,25716E-06 0,012312026 0,001231203 Biocarta

JNK MAPK Pathway 40 4 7,49439E-05 0,099675362 0,009061397 ST

Metabolism of RNA 330 9 9,09024E-05 0,120900172 0,010075014 Reactome

MAPK Activation in TLR Cascade 50 4 0,000181055 0,240803076 0,017807189 Reactome

Metabolism of mRNA 284 8 0,000187444 0,249300648 0,017807189 Reactome

GNRH Signaling Pathway 101 5 0,000264062 0,35120181 0,023413454 KEGG

Biosynthesis of the N Glycan Precursor

Dolichol lipid linked oligosaccharide 29 3 0,000582424 0,774624217 0,048414014 Reactome
LLO and transfer to a nascent protein