1755-8794-1-29 1755-8794 Research article <p>Characterization of global transcription profile of normal and HPV-immortalized keratinocytes and their response to TNF treatment</p> Termini Lara ltermini@ludwig.org.br Boccardo Enrique eboccardo@ludwig.org.br Esteves H Gustavo gesteves@uepb.edu.br Hirata Roberto Jr rhirata@gmail.com Martins K Waleska waleska.martins@funed.mg.gov.br Colo Estela L Anna annacolo@gmail.com Neves E Jordão neves@ime.usp.br Villa Lina Luisa llvilla@ludwig.org.br Reis FL Luiz luiz.reis@hcancer.org.br

Ludwig Institute for Cancer Research, São Paulo, Brazil

Hospital do Câncer A. C. Camargo, São Paulo, Brazil

Instituto de Matemática e Estatística da Universidade de São Paulo, São Paulo, Brazil

 BMC Medical Genomics 1755-8794 2008 1 1 29 http://www.biomedcentral.com/1755-8794/1/29 18588690 10.1186/1755-8794-1-29 01 2 2008 27 6 2008 27 6 2008 2008 Termini et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Background

Persistent infection by high risk HPV types (e.g. HPV-16, -18, -31, and -45) is the main risk factor for development of cervical intraepithelial neoplasia and cervical cancer. Tumor necrosis factor (TNF) is a key mediator of epithelial cell inflammatory response and exerts a potent cytostatic effect on normal or HPV16, but not on HPV18 immortalized keratinocytes. Moreover, several cervical carcinoma-derived cell lines are resistant to TNF anti-proliferative effect suggesting that the acquisition of TNF-resistance may constitute an important step in HPV-mediated carcinogenesis. In the present study, we compared the gene expression profiles of normal and HPV16 or 18 immortalized human keratinocytes before and after treatment with TNF for 3 or 60 hours.

Methods

In this study, we determined the transcriptional changes 3 and 60 hours after TNF treatment of normal, HPV16 and HPV18 immortalized keratinocytes by microarray analysis. The expression pattern of two genes observed by microarray was confirmed by Northern Blot. NF-κB activation was also determined by electrophoretic mobility shift assay (EMSA) using specific oligonucleotides and nuclear protein extracts.

Results

We observed the differential expression of a common set of genes in two TNF-sensitive cell lines that differs from those modulated in TNF-resistant ones. This information was used to define genes whose differential expression could be associated with the differential response to TNF, such as: KLK7 (kallikrein 7), SOD2 (superoxide dismutase 2), 100P (S100 calcium binding protein P), PI3 (protease inhibitor 3, skin-derived), CSTA (cystatin A), RARRES1 (retinoic acid receptor responder 1), and LXN (latexin). The differential expression of the KLK7 and SOD2 transcripts was confirmed by Northern blot. Moreover, we observed that SOD2 expression correlates with the differential NF-κB activation exhibited by TNF-sensitive and TNF-resistant cells.

Conclusion

This is the first in depth analysis of the differential effect of TNF on normal and HPV16 or HPV18 immortalized keratinocytes. Our findings may be useful for the identification of genes involved in TNF resistance acquisition and candidate genes which deregulated expression may be associated with cervical disease establishment and/or progression.

Background

Human papillomaviruses (HPVs) are double-stranded DNA tumor viruses that infect keratinocytes of the anogenital tract epithelium 1. Persistent infection by high risk HPV types (e.g., HPV-16, -18, -31, and -45) is the main risk factor for the development of cervical intraepithelial neoplasia and cervical cancer 23. High-risk HPV DNA is detected in more than 90% of cervical carcinomas worldwide 4 and it has been shown that HPV types 16 and 18 can immortalize normal cells in culture, a function that is attributed to E6 and E7 oncogenes 5. These are the only HPV genes consistently retained and expressed in cervical carcinomas. Besides, their continued expression is required to maintain the malignant phenotype 678. The proteins encoded by these genes disturb cell proliferation and differentiation by physical and functional interaction with several cellular factors involved in cell cycle regulation 9. E6 is best known for its ability to bind to p53 and induce its ubiquitin-dependent degradation 1011, whereas E7 was initially recognized by its ability to interact with members of the retinoblastoma protein family, namely pRb, p107 and p130 12 and its capacity of enhancing their degradation 13.

Persistence of HPV infections and development of neoplasia is influenced by local cell-mediated immune response 14. Tumor necrosis factor-alpha (TNF) is one of the main mediators of skin and mucosa inflammation and has a potent antiproliferative effect on normal primary human keratinocytes (PHKs). This cytokine is a key regulator of diverse inflammatory and immune processes in human epithelia and its expression by keratinocytes is enhanced in response to tissue injury, inflammation, viral infection, and UV radiation 151617. Furthermore, TNF has been identified as a key mediator for the regression of HPV-induced lesions 18192021. Previous studies from our group had shown that TNF exerts a potent cytostatic effect on normal and HPV16 immortalized keratinocytes. On the other hand, keratinocytes immortalized by HPV18 or SV40, as well as HPV16 or HPV18-positive cervical tumor-derived cell lines continue to proliferate normally in the presence of this cytokine 2223. In addition, it has been observed that continuous HPV18-gene expression in malignant HeLa-fibroblasts hybrids, as well as increased tumorigenicity of HPV16-transformed human keratinocytes is associated with TNF resistance 2425. These observations underscore the importance of TNF-resistance acquisition in HPV mediated pathogenesis and suggest that this event could be an important factor in HPV-associated neoplasia outcome. However, the molecular basis of HPV-mediated TNF resistance has not been elucidated.

The aim of the present study was to characterize and compare the global transcription profile of normal and HPV-immortalized keratinocytes. Furthermore, we sought to analyze their response to TNF in order to identify differences that contribute to explain their divergent response to this cytokine. For this purpose, we used microarray analysis to determine transcriptional changes upon 3 and 60 hours after TNF exposure. The 3 hours treatment would favor the identification of immediate early TNF regulated genes. On the other hand, the 60 hours treatment was used because the cytostatic effect exerted by this cytokine on normal and HPV16-immortalized keratinocytes reaches its maximum at this time-point 2223. Our experimental setting allowed us to: 1) identify genes that are differentially expressed between TNF-sensitive and TNF-resistant cells; 2) identify genes that are differentially modulated by TNF at two-time points (3 and 60 hours); 3) analyze the effect of HPV-induced immortalization on TNF-regulated genes and, 4) find genes that are differentially expressed between cells immortalized by two different high-risk HPV types. Using this approach, we identified differentially expressed genes that are involved in different cell processes such as immune and inflammatory responses, cell differentiation, cell death, proliferation, extracellular matrix remodeling and DNA repair. The implications of these results are discussed.

Methods

Cell Culture and TNF treatment

Cultures of primary human keratinocytes (PHK), recovered from newborn foreskin (Cambrex, Walkersville, MD, USA), were maintained in keratinocyte serum-free medium – KSFM (Life Technology, Inc., Gaithersburg, MD, USA) for 3 to seven passages 26. HF698 and HF18Nco are cell lines obtained from human keratinocytes immortalized by HPV16 and HPV18 whole genome, respectively. These cell lines (from now on referred as HPV16 and HPV18, respectively) were kindly provided by R. Schlegel, Georgetown University Medical Center, Washington, DC 27, were grown in 3+1 medium, consisting of a mixture of 3 parts KSFM and 1 part DMEM, supplemented with 10% fetal calf serum. Cells were grown in 100-mm tissue culture dishes to 30% confluence and treated with 2 nM of human recombinant TNF (Boehringer Mannheim, Germany), for 3 or 60 hours. Cells were then trypsinized, washed 3 times with PBS and frozen until RNA extraction. For all time points, RNA was obtained from two independent experiments, including the control plates.

RNA extraction, amplification, labeling, and hybridizations

For each sample, total RNA was extracted using TRIzol Reagent (Life Technologies, Inc., Grand Island, NY, USA) following the procedure recommended by the manufacturer. Three micrograms of target and reference (a pool of RNA from all control conditions) total RNA were linearly amplified using T7-based protocol, converted to cyanine-modified cDNA, and labeled as described previously 28.

Hybridizations were performed in duplicate, using dye-swap, on a cDNA platform of ORESTES representing 4,600 unique genes with known full-length sequence selected from the clone collection derived from the Human Cancer Genome Project 29. cDNA amplification, purification, identity verification and printing were performed as previously described 28. A detailed description of the cDNA microarray platform used and the raw data of this study are available at the GEO website under the accession numbers GPL1930 and GSE4524, respectively 30. Slides were scanned on a confocal laser scanner (Arrayexpress; Packard Bioscience, USA) and, for each spot, signal and background intensities were measured using histogram method of Quantarray software (version 3.0, Packard BioScience, BioChip Technologies LLC, USA).

Statistical Analysis

Data analysis was performed with R project for statistical computing 31 and tools of the associated project, Bioconductor 32. Prior to analysis, signal intensity was corrected by background subtraction, and data normalized by loess method, using span = 0.4 and degree = 2. For the identification of differentially expressed genes, we used ANOVA model when just one variable was considered. For the identification of differentially expressed genes in a pair-wise manner, we used t-test and determined the nominal p-value for each individual gene. Those nominal p-values can be conservatively adjusted for multiple testing with the Bonferroni correction by multiplying them by the number of genes in our chip. For clustering samples on the basis of their expression profile, we applied hierarchical clustering based on correlation distance and complete linkage.

Northern Blotting

For Northern blot analysis, 15 μg of total RNA was fractionated through a 1% denaturing agarose gel and transferred by capillarity onto Hybond N filters (GE Healthcare BioSciences, NJ, USA). Prehybridization, hybridization, and washes were performed as described by Church and Gilbert 33. The KLK7 and SOD2 cDNA probes were the same used for immobilization in the array. The human GAPDH cDNA probe was used as control for ensuring equal RNA loading. Probes were labeled by random priming, using Ready-To-Go Labeling Beads (GE Healthcare Bio-Sciences, NJ, USA) and [α-32P]dCTP (3000 Ci/mmol). Nylon filters were exposed to Kodak Hyperfilm (GE Healthcare BioSciences, NJ, USA) with intensifying screen.

Electrophoretic mobility shift assay (EMSA)

Nuclear extracts were obtained from monolayer cultures of PHK, and from cell lines HPV16 and HPV18 treated with 2 nM of human recombinant TNF for 1 h. Briefly, cell plates were washed with ice-cold PBS and cells were scraped in 5 ml of PBS. Cells were transferred to a 15 ml Falcon tube and centrifuged at 3000 rpm for 3 min. Cell pellets were ressuspended in 4 ml of lysis buffer (10 mM HEPES pH 7,9, 10 mM KCl, 0,2 mM EDTA, 1 mM DTT), incubated on ice for 5 min, centrifuged and ressuspended in 4 ml of lysis buffer. Nuclei obtained were centrifuged at 2000 rpm for 2 min, ressuspended in 100 μl of extraction buffer (20 mM HEPES pH7,9, 0,42 M NaCl, 2 mM EDTA, 1 mM DTT, 1 mM PMSF, 2 μM pepstatin, 0,6 μM leupeptin, 25 mU/ml aprotinin) and incubated on ice for 30 min. Finally, the samples were centrifuged at 12000 rpm for 15 min at 4°C. The supernatants were stored at -80°C. The protein concentration was determined by the Bradford method (Bio-Rad, CA, USA).

For gel retardation the following double-stranded oligonucleotide, corresponding to the NF-κB binding sequence, was used: forward-5'-GCCTGGGAAAGTCCCCTCAACT-3' (Invitrogen, CA, USA) was used. The annealed oligonucleotide was labeled with [γ-32]ATP (Amersham, Buckinghamshire, UK; 3,000 Ci/mmol) using TK polynucleotide kinase according to the manufacturer instructions (Biolabs, MA, USA) and purified using Sephadex G50 columns followed by phenol:chloroform extraction and precipitation using 10 μg of salmon sperm DNA as a carrier (Invitrogen, CA, USA). DNA pellets were ressuspended in binding buffer (20 mM HEPES pH 7,9, 20% glycerol, 0,1 M KCl, 2 mM EDTA, 1 mM PMSF, 2 μM pepstatin, 0,6 μM leupeptin, 25 mU/ml aprotinin) to a final concentration of 2,5 fmol/μl. The incorporated radioactivity was quantitated using a LS6500 scintillation counter (Beckman Coulter, CA, USA).

The binding of NF-κB was performed in a reaction containing 5 μg of protein extract, 5 μg of BSA, 5 μg of salmon sperm DNA and binding buffer to a final volume of 32 μl on ice. After 10 min, 8 μl of the [γ-32]ATP 5'-end-labeled double-stranded oligonucleotide probe was added, and the incubation was continued for an additional 15 min at 30°C. The DNA-protein complexes were resolved on 4% nondenaturing polyacrylamide gels (29:1 cross-linking ratio), dried, and exposed overnight to X-ray films (Amersham, Buckinghamshire, UK).

Results

Glass arrays containing 4.800 cDNA sequences were used in order to determine the effects of HPV infection in human keratinocytes as well as the impact of TNF treatment on global gene expression, in HPV negative or positive cells.

In order to identify differentially expressed genes as a function of a unique variable (cell type or TNF-treatment) our dataset was first analyzed by one way ANOVA. The comparisons performed allowed us to determine 1) genes that are differentially expressed between TNF-sensitive and TNF-resistant cells; 2) identify genes that are differentially modulated by TNF at two-time points (3 and 60 hours); 3) analyze the effect of HPV-induced immortalization on TNF-regulated genes and, 4) find genes that are differentially expressed between cells immortalized by two different high-risk HPV types (Figure 1).

<p>Figure 1</p>

Experimental setup for the analysis of HPV and TNF effects on keratinocytes gene expression

Experimental setup for the analysis of HPV and TNF effects on keratinocytes gene expression. In order to characterize and compare the global transcription profile of normal and HPV-immortalized keratinocytes and to analyze their response to TNF we used an experimental setting that allowed us to: 1) identify differential expressed genes between normal PHK, HPV16 and HPV18-immortalized keratinocytes (comparisons represented by dashed arrows); 2) identify genes modulated by TNF upon treatment for three and sixty hours (comparisons represented by solid arrows) and; 3) compare the effect of TNF between normal PHK and cells immortalized by two different high-risk HPV types (comparisons represented by round dot arrows).

Differentially expressed genes as a function of cell type or TNF treatment

The identification of differentially expressed genes, with statistical significance, as a function of cell type was performed by ANOVA. Samples and differentially expressed genes (cutoff p-value <10-10) were grouped hierarchically, using correlation distance and complete linkage (Figure 2). As it can be observed, normal (PHK) and HPV16-immortalized keratinocytes (HPV16), which are sensitive to TNF cytostatic effect, grouped together while TNF-resistant HPV18-immortalized cell line (HPV18) formed an independent branch. This indicates that TNF-sensitive cell lines share a group of genes which are regulated in a way that clearly differentiate them from the TNF-resistant one. Samples were further clusterized by the time in culture after the last medium change (3 or 60 hs) and finally separated as a function of TNF treatment. This clusterization pattern may reflect differences in cell density and other cultures variables such as nutrients availability or medium conditioning. Initially, all treatments were performed using 30% cell density cultures. As expected, due to TNF cytostatic effect on normal and HPV16-immortalized keratinocytes, cell density at the end of the 60 hours period was different between treated (40–50%) and control cells (70–80%) for these cell lines. On the other hand, both cytokine treated and control HPV18-immortalized cells reached 80–90% cell density by the end of the 60 hours period. Flow-cytometry analysis revealed that the TNF effect on sensitive cells was characterized by the accumulation of cells in the G1-phase of the cell cycle. Conversely, TNF-induced G1-arrest was not observed in HPV18-immortalized keratinocytes [2223 and data not shown]. Finally, no differences in cell density were observed for cultures corresponding to 3 hours-treatment group.

<p>Figure 2</p>

Hierarchical grouping based on differentially expressed genes as a function of cell type

Hierarchical grouping based on differentially expressed genes as a function of cell type. These genes where identified by the ANOVA method and the samples where grouped considering the correlation distance and complete linkage. After sample grouping the genes (p values <10-10) were hierarchically grouped by their correlation distances. High gene expression is shown in red, low gene expression is shown in green and black indicates non-differential gene expression. Samples: Primary human keratinocytes: controls and treated for 3 or 60 hours with TNF, respectively (PHK_3H, PHK_60H, PHK_3H.TNF, PHK_60H.TNF); HPV16-immortalized keratinocytes: controls and treated for 3 or 60 hours with TNF, respectively (HPV16_3H, HPV16_60H, HPV16_3H.TNF, HPV16_60H.TNF); HPV18-immortalized keratinocytes: controls and treated with 3 or 60 hours for TNF, respectively (HPV18_3H, HPV18_60H, HPV18_3H.TNF, HPV18_60H.TNF).

Among the differentially regulated genes we found some related with inflammatory response (SOD2, TGFB1, CD44, INHBA, OAS1, SIMP), epidermal development, differentiation and proliferation (ADAMST1, RARRES1, CREG, HBP17, MCM2, PRSS1, S100P, CREG1), proteolysis regulation (KLK7, PI3, LXN), and cell adhesion (CD44, PARVA, PROS1). The name and function of the genes described are listed in Table 1. We next determined the global changes in gene expression as a function of TNF treatment. The name and annotated function of the identified genes that best distinguish samples based on TNF treatment (cutoff p-value <10-2,9) are listed in Table 2. As expected, many of these genes are involved in the inflammatory response and/or are direct targets of TNF e.g. CCL20, CD44, HLA-F, IL1F9, NFKBIA, INHBA, SOD2, MARCKS, RFX5. Samples and genes were hierarchically clusterized on the basis of their correlation distance using complete linkage (Figure 3). Samples from TNF-treated normal keratinocytes (PHK) grouped together and apart from the others. HPV-positive samples exhibited a complex clusterization pattern suggesting that the presence of either HPV16 or 18 has an impact on TNF-regulated gene expression. Furthermore, the grouping of treated PHKs apart from the other samples could reflect the fact that PHKs are the only normal cells used in this study and, as such, the only cell type expected to have an unaltered TNF-signaling network. This could contribute to explain the differences in gene expression upon TNF treatment observed between normal and HPV-immortalized keratinocytes.

<p>Table 1</p>

Name and function of the differentially expressed genes that best distinguish samples by cell type variable

GENE

UniGene ID

GENE NAME

FUNCTION

ADAMTS1

Hs.534115

disintegrin-like and metalloprotease (reprolysin type)

negative regulation of cell proliferation

ARHGEF10

Hs.443460

Rho guanine nucleotide exchange factor (GEF) 10

GTPase activator activity

CD44

Hs.502328

CD44 antigen

cell adhesion

CITED1

Hs.40403

Cbp/p300-interacting transactivator

transcription regulator activity

CREG

Hs.5710

cellular repressor of E1A-stimulated genes 1

cell proliferation

CSTA

Hs.518198

cystatin A (stefin A)

cysteine protease inhibitor activity

D4S234E

Hs.518595

DNA segment on chromosome 4 (unique) 234 expressed sequence

dopamine receptor signaling pathway

DHRS3

Hs.289347

dehydrogenase/reductase (SDR family) member 3

fatty acid metabolism

EPB41L1

Hs.437422

erythrocyte membrane protein band 4.1-like 1

structural molecule activity

FLJ20105

Hs.47558

FLJ20105

regulation of transcription

FLJ21511

Hs.479703

FLJ21511

unknown function

FLJ21616

Hs.591836

FLJ21616

regulation of transcription

FLJ30525

Hs.7962

FLJ30525

unknown function

GALNT11

Hs.647109

UDP-N-acetyl-alpha-D-galactosamine

transferase activity, transferring glycosyl groups

GC20

Hs.315230

translation factor sui1 homolog

regulation of translational initiation

GLDC

Hs.584238

glycine dehydrogenase

glycine metabolism

GSR

Hs.271510

glutathione reductase

glutathion metabolism

HBP17

Hs.1690

fibroblast growth factor binding protein 1

regulation of cell proliferation

INHBA

Hs.28792

inhibin, beta A

cell cycle arrest, negative regulation of immune cell differentiation

JPH3

Hs.592068

junctophilin 3

unknown function

KIAA0368

Hs.368255

KIAA0368

ER-associated protein catabolism

KLK7

Hs.151254

kallikrein 7 (chymotryptic, stratum corneum)

epidermis development, proteolysis and peptidolysis, chymotrypsin activity

LCN2

Hs.204238

lipocalin 2 (oncogene 24p3)

transporter activity

LXN

Hs.478067

latexin

enzyme inhibitor activity

MAL2

Hs.201083

T-cell differentiation protein 2

Unknown function

MCM2

Hs.477481

minichromosome maintenance deficient 2

cell cycle

MGC45400

Hs.389734

transcription elongation factor A (SII)-like 8

translation elongation factor activity

MGEA5

Hs.500842

meningioma expressed antigen 5 (hyaluronidase)

glycoprotein catabolism

MYO5B

Hs.200136

acetyl-Coenzyme A acyltransferase 2

fatty acid metabolism

NMES1

Hs.112242

normal mucosa of esophagus specific 1

unknown function

NMU

Hs.418367

neuromedin U

neuropeptide signaling pathway, digestion

NT5E

Hs.153952

5'-nucleotidase, ecto (CD73)

DNA metabolism

OAS1

Hs.524760

2',5'-oligoadenylate synthetase 1

immune response to viral infections

ODC1

Hs.467701

ornithine decarboxylase 1

polyamine biosynthesis

PARVA

Hs.607144

parvin, alpha

cell adhesion, actin binding

PEX3

Hs.7277

peroxisomal biogenesis factor 3

peroxisome organization

PI3

Hs.112341

protease inhibitor 3, skin-derived (SKALP)

elastase-specific inhibitor

PLAU

Hs.77274

plasminogen activator

chemotaxis

PPGB

Hs.517076

protective protein for beta-galactosidase

intracellular protein transport

PROS1

Hs.64016

protein S (alpha)

cell adhesion, endopeptidase inhibitor activity

PRSS11

Hs.501280

protease, serine, 11 (IGF binding)

insulin-like growth facto binding, regulation of cell growth

RARRES1

Hs.131269

retinoic acid receptor responder

negative regulation of cell proliferation

RDH-E2

Hs.170673

epidermal retinal dehydrogenase 2

oxidoreductase activity

RPL15

Hs.381219

ribosomal protein L15

protein biosynthesis

RUTBC3

Hs.474914

RUN and TBC1 domain containing 3

unknown function

S100P

Hs.440880

S100 calcium binding protein P

cell cycle progression and differentiation

SEPT10

Hs.469615

septin 10

cell cycle

SF3B4

Hs.516160

myotubularin related protein 11

inositol or phosphatidylinositol phosphatase activity

SIMP

Hs.475812

immunodominant MHC-associated peptides

protein amino acid glycosylation

SMOC2

Hs.487200

SPARC related modular calcium binding

calcium ion binding

SOD2

Hs.487046

superoxide dismutase 2

age-dependent response to reactive oxygen species, cellular defense response

STAF65 (gamma)

Hs.6232

SPTF-associated factor 65 gamma

regulation of transcription, DNA- dependent

SYTL3

Hs.436977

synaptotagmin-like 3

intracellular protein transport

TFRC

Hs.529618

transferrin receptor (p90, CD71)

endocytosis

TGFB1

Hs.645227

transforming growth factor, beta 1

cell proloferation

TPD52

Hs.368433

tumor protein D52

morphogenesis

TPX2

Hs.244580

microtubule-associated protein homolog

cell proliferation

TRIM31

Hs.493275

tripartite motif-containing 31

protein ubiquitination, ubiquitin ligase activity

YME1L1

Hs.499145

YME1-like 1 (S. cerevisiae)

protein catabolism

ZNF198

Hs.644041

zinc finger protein 198

regulation of transcription, DNA- dependent

*Genes are listed in alphabetical order. The cutoff p-value was set as <10-10.

 <p>Table 2</p>

Name and function of the differentially expressed genes that best distinguish samples by TNF treatment variable

GENE

UniGene ID

GENE NAME

FUNCTION

ADORA2b

Hs.167046

adenosine A2b receptor

activation of MAPK

AKAP1

Hs.463506

A kinase (PRKA) anchor protein 1

RNA binding

BTG2

Hs.519162

BTG family, member 2

negative regulation of cell proliferation

C3

Hs.529053

complement component 3

inflammatory response

CCL20

Hs.75498

chemokine (C-C motif) ligand 20

inflammatory response

CD44

Hs.502328

CD44 antigen

cell adhesion

cig5

Hs.17518

radical S-adenosyl methionine domain containing 2

Catalytic activity

CLCA4

Hs.546343

chloride channel, calcium activated, family member 4

chloride transport

DC-UbP

Hs.179852

dendritic cell-derived ubiquitin-like protein

Protein modification

FAD104

Hs.159430

fibronectin type III domain containing 3B

cell differentiation

FLJ21511

Hs.479703

FLJ21511

Unknown function

FMNL3

Hs.179838

formin-like 3

cell organization and biogenesis

GFPT2

Hs.30332

glutamine-fructose-6-phosphate transaminase 2

carbohydrate biosynthesis

HLA-F

Hs.519972

major histocompatibility complex, class I, F

antigen presentation, endogenous antigen

IL1F9

Hs.211238

interleukin 1 family, member 9

inflammatory response

INHBA

Hs.28792

inhibin, beta A

cell cycle arrest, negative regulation of immune cell differentiation

KIAA0303

Hs.133539

microtubule associated serine/threonine kinase family member 4

protein kinase activity

KIAA1279

Hs.279580

KIAA1279

Unknown function

LAP3

Hs.479264

leucine aminopeptidase 3

Protein metabolism

MARCKS

Hs.75061

MARCKS-like 1

calmodulin binding, macrophage activation

MGAT4B

Hs.437277

mannosyl (alpha-1,3-)-glycoprotein beta-1,4-N-acetylglucosaminyltransferase, isoenzyme B

cytokine activity

MGC45400

Hs.389734

transcription elongation factor A (SII)-like 8

translation elongation factor activity

MMP9

Hs.297413

matrix metalloproteinase 9

proteolysis and peptidolysis

NFKBIA

Hs.81328

nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor, alpha

cytoplasmic sequestering of NF-kappaB

NMES1

Hs.112242

normal mucosa of esophagus specific 1

Unknown function

OAS1

Hs.524760

2',5'-oligoadenylate synthetase 1

immune response to viral infections

PLAU

Hs.77274

plasminogen activator

chemotaxis

RDH-E2

Hs.170673

epidermal retinal dehydrogenase 2

oxidoreductase activity

RFX5

Hs.166891

regulatory factor X, 5

inflammatory response, HLA class II expression

RIG-1

Hs.17466

retinoic acid receptor responder (tazarotene induced) 3

negative regulation of cell proliferation

RIPK2

Hs.103755

receptor-interacting serine-threonine kinase 2

inflammatory response

SASH1

Hs.193133

SAM and SH3 domain containing 1

Negative regulation of cell cycle

SDCBP

Hs.200804

syndecan binding protein (syntenin)

intracellular signaling cascade, interleukin-5 receptor binding

SEC24A

Hs.211612

SEC24 related gene family, member A

intracellular protein transport

SERPINB2

Hs.514913

encoding serine (or cysteine) proteinase inhibitor, clade B (ovalbumin), member 2

anti-apoptosis

SF3B4

Hs.412818

myotubularin related protein 11

RNA splicing

SOD2

Hs.487046

superoxide dismutase 2

age-dependent response to reactive oxygen species, cellular defense response

TMSB4

Hs.522584

thymosin, beta 4, X-linked

cytoskeleton organization and biogenesis

VMP1

Hs.444569

transmembrane protein 49

Unknown function

*Genes are listed in alphabetical order. The cutoff p-value was set as <10-2,9.

 <p>Figure 3</p>

Hierarchical grouping based on differentially expressed genes as a function of TNF treatment

Hierarchical grouping based on differentially expressed genes as a function of TNF treatment. These genes where identified by the ANOVA method and the samples where grouped considering the correlation distance and complete linkage. After sample grouping the genes (p values <10-2,9) were hierarchically grouped by their correlation distances. High gene expression is shown in red, low gene expression is shown in green and black indicates non-differential gene expression. Samples: Primary human keratinocytes: controls and treated for 3 or 60 hours with TNF, respectively (PHK_3H, PHK_60H, PHK_3H.TNF, PHK_60H.TNF); HPV16-immortalized keratinocytes: controls and treated for 3 or 60 hours with TNF, respectively (HPV16_3H, HPV16_60H, HPV16_3H.TNF, HPV16_60H.TNF); HPV18-immortalized keratinocytes: controls and treated for 3 or 60 hours with TNF, respectively (HPV18_3H, HPV18_60H, HPV18_3H.TNF, HPV18_60H.TNF).

Since our experimental setting included the comparative analysis of global gene expression at two time points (3 or 60 hours), we searched for genes that best differentiate our samples as a function of time. We found 48 genes that clearly differentiate samples from the analyzed time points. The name and annotated function of the identified genes (cutoff p-value <10-9) are listed in additional file 1. Hierarchical clusterization divided samples in two main branches (additional file 2). Each branch was exclusively composed of samples from the same time point, namely, 3 or 60 hours. Samples from the 3 hours-time point formed a secondary branch that divided normal from HPV-immortalized keratinocytes. On the other hand, samples from the 60 hours-time point formed a secondary branch that divided normal and HPV16-immortalized keratinocytes (TNF-sensitive samples) from HPV18-immortalized keratinocytes (TNF-resistant samples).

<p>Additional file 1</p>

Table with name and function of the differentially expressed genes that best distinguish samples by time variable. The cutoff p-value was set as <10-9.

Click here for file

 <p>Additional file 2</p>

Hierarchical grouping based on differentially expressed genes as a function of time. These genes where identified by the ANOVA method and the samples where grouped considering the correlation distance and complete linkage. After sample grouping the genes were hierarchically grouped by their correlation distances. High gene expression is shown in red, low gene expression is shown in green and black indicates non-differential gene expression, p values <10-9. Samples: Primary human keratinocytes: controls and treated for 3 or 60 hours with TNF, respectively (PHK_3H, PHK_60H, PHK_3H.TNF, PHK_60H.TNF); HPV16-immortalized keratinocytes: controls and treated for 3 or 60 hours with TNF, respectively (HPV16_3H, HPV16_60H, HPV16_3H.TNF, HPV16_60H.TNF); HPV18-immortalized keratinocytes: controls and treated for 3 or 60 hours with TNF, respectively (HPV18_3H, HPV18_60H, HPV18_3H.TNF, HPV18_60H.TNF).

Click here for file

Differentially expressed genes between TNF-sensitive and TNF-resistant cells

In order to identify differentially expressed genes between specific samples we performed a series of pair-wise comparisons. For each pair-wise comparison, we generated a list of differentially expressed genes with p-value lower than 0,01. The complete list of all pair-wise comparisons performed is presented in additional file 3. We next aimed to characterize genes that were differentially expressed between TNF-sensitive (PHK and HPV16) and TNF-resistant cells (HPV18). To achieve this goal we selected the thirty genes with the lowest p-values that best distinguish both PHK and HPV16 from HPV18 and the thirty genes with lowest p-values that best distinguish both PHK+TNF and HPV16+TNF from HPV18+TNF (considering treatment with TNF for 3 h). Twelve genes were common to both lists giving a total of 48 different genes identified (Table 3). Using the expression profile of these 48 genes, samples were grouped hierarchically, based on their correlation distance and complete linkage (Figure 4).

<p>Additional file 3</p>

Table of differentially expressed genes obtained by pair-wise comparison of all the samples and their respective expression ratio ordered by p-value. PHK_3H (primary human keratinocytes – control/3 hours), PHK_60H (primary human keratinocytes – control/60 hours), HPV16_3H (HPV16-immortalized keratinocytes – control/3 hours), HPV16_60H (HPV16-immortalized keratinocytes – control/60 hours), HPV18_3H (HPV18-immortalized keratinocytes – control/3 hours), HPV18_60H (HPV18-immortalized keratinocytes – control/60 hours), PHK_TNF3H (primary human keratinocytes – TNF/3 hours), PHK_TNF60H (primary human keratinocytes – TNF/60 hours), HPV16_TNF3H (HPV16-immortalized keratinocytes – TNF/3 hours), HPV16_TNF60H (HPV16-immortalized keratinocytes – TNF/60 hours), HPV18_TNF3H (HPV18-immortalized keratinocytes – TNF/3 hours), HPV18_TNF60H (HPV18-immortalized keratinocytes – TNF/60 hours).

Click here for file

 <p>Table 3</p>

List of differentially expressed genes that best distinguish TNF-resistant cells (HPV18) from TNF-sensitive cells (PHK and HPV16), in normal culture conditions or upon treatment with TNF for 3 hours

(PHK and HPV16) vs HPV18

(PHK_TNF and HPV16_TNF) vs HPV18_TNF

GENE

GENE NAME

FOLD

p VALUE

FOLD

p VALUE

ABCE1

ATP-binding cassette, sub-family E (OABP), member 1

0.592

0.001384

----

----

ACBD5

acyl-Coenzyme A binding domain containing 5

----

----

0.378

0.00086

ALDH3A2

encoding aldehyde dehydrogenase 3 family, member A2

1.623

0.00122

----

----

APG12L

APG12 autophagy 12-like (S. cerevisiae)

1.739

0.000924

----

----

APPBP1

amyloid beta precursor protein binding protein 1

0.540

0.000283

----

----

ARF4L

ADP-ribosylation factor 4-like

----

----

0.578

5.70E-05

BCLAF1

BCL2-associated transcription factor 1

0.611

0.000876

----

----

BOC

brother of CDO

----

----

2.337

6.00E-06

CCNA2

cyclin A2

0.577

0.000604

----

----

CDCA2

cell division cycle associated 2

----

----

0.539

7.20E-05

CDK2AP1

CDK2-associated protein 1

----

----

1.523

0.000148

CPSF3

cleavage and polyadenylation specific factor 3

0.586

0.000466

----

----

CYP1B1

cytochrome P450, family 1, subfamily B, polypeptide 1

0.499

0.001386

----

----

DEK

DEK oncogene

0.480

0.000278

----

----

FAM31C

family with sequence similarity 31, member C

----

----

2.548

0.000663

    FLJ20105

hypothetical protein LOC54821

0.026

5.00E-06

0.029

2.00E-06

    GALNAC4S-6ST

B cell RAG associated protein

1.827

0.000145

1.673

0.000215

H105E3

encoding NAD(P) dependent steroid dehydrogenase-like

----

----

0.569

6.80E-05

HLCS

holocarboxylase synthetase

----

----

1.551

0.00011

JPH3

junctophilin 3

0.154

0.000192

----

----

KIAA0795

kelch-like 18 (Drosophila)

0.857

0.001325

----

----

KIAA1023

IQ motif containing E

1.575

0.000723

----

----

KIF1B

kinesin family member 1B

----

----

0.570

0.000632

    KLK7

encoding kallikrein 7 (chymotryptic, stratum corneum)

0.421

0.000416

0.374

2.30E-05

LCN2

lipocalin 2 (oncogene 24p3)

----

----

0.216

0.000686

LOC151242

protein phosphatase 1, regulatory (inhibitor)

1.928

0.000268

----

----

Lrp2bp

low density lipoprotein receptor-related protein binding protein

0.629

0.000574

----

----