Hypoxia induces up-regulation of progranulin in neuroblastoma cell lines

Neurochemistry International 57 (2010) 893–898

Hypoxia induces up-regulation of progranulin in neuroblastoma cell lines

Paola Piscopo, Roberto Rivabene, Alice Adduci, Cinzia Mallozzi, Lorenzo Malvezzi-Campeggi,Alessio Crestini, Annamaria Confaloni *

Department of Cell Biology and Neurosciences, Istituto Superiore di Sanita, Viale Regina Elena 299, 00161 Rome, Italy

A R T I C L E I N F O

Article history:

Received 6 April 2010

Received in revised form 31 August 2010

Accepted 26 September 2010

Available online 7 October 2010

Keywords:

Progranulin

Gene expression

Hypoxia

Oxidative stress

Neuroblastoma cell lines

A B S T R A C T

Progranulin (PGRN) is a widely expressed multifunctional protein, involved in regulation of cell growth

and cell cycle progression with a possible involvement in neurodegeneration. We looked for PGRN

regulation in three different human neuroblastoma cell lines, following exposure to two different

stimuli commonly associated to neurodegeneration: hypoxia and oxidative stress. For gene and protein

expression analysis we carried out a quantitative RT-PCR and western blotting analysis. We show that

PGRN is strongly up-regulated by hypoxia, through the mitogen-actived protein kinase (MAPK)/

extracellular signal-regulated kinase (MEK) signaling cascade. PGRN is not up-regulated by H2O2-

induced oxidative stress. These results suggest that PGRN in the brain could exert a protective role

against hypoxic stress, one of principal risk factors involved in frontotemporal dementia pathogenesis.

� 2010 Elsevier Ltd. All rights reserved.

Contents lists available at ScienceDirect

Neurochemistry International

journa l homepage: www.e lsev ier .com/ locate /neuint

1. Introduction

Progranulin (PGRN) is a widely expressed multifunctionalprotein involved in the regulation of cell growth and cell cycleprogression (Bateman et al., 1990; He and Bateman, 2003). Innormal peripheral tissues, the PGRN has a complex function, withthe full-length form of the protein, which exerts trophic and anti-inflammatory activity. Proteolytic cleavage of PGRN generatesgranulin peptides that promote inflammatory activity. Little isknown about the role of PGRN in the central nervous system (CNS).

The protein is widely expressed during early neural develop-ment (Daniel et al., 2003), but with maturation its expressionbecomes restricted to defined neuronal populations, such ascortical and hippocampal pyramidal neurons and Purkinje cells(Daniel et al., 2000). PGRN has also been implicated in the sexualdifferentiation of the brain (Suzuki and Nishiahara, 2002). Theprotein is up-regulated in activated microglial cells (Baker andManuelidis, 2003) but not in astrocytes or oligodendrocytes.Recently, Van Damme and colleagues observed that PGRN and GRNE (one of the proteolytic fragments of PGRN) promote neuronalsurvival and enhance neurite outgrowth in cultured neurons (VanDamme et al., 2008). Several nonsense and frameshift mutations in

Abbreviations: FTD, frontotemporal dementia; GLUT-1, glucose transporter 1; GSH,

reduced glutathione; GSSG, oxidized glutathione; H2O2, hydrogen peroxide;

MAPKs, mitogen-activated protein kinases; MEK, (MAPK)/extracellular signal-

regulated kinase kinase; PGRN, progranulin.

* Corresponding author. Tel.: +39 06 49902930; fax: +39 06 49387143.

E-mail address: [email protected] (A. Confaloni).

0197-0186/$ – see front matter � 2010 Elsevier Ltd. All rights reserved.

doi:10.1016/j.neuint.2010.09.008

the PGRN gene are associated with frontotemporal lobar degener-ation (FTLD) (Baker et al., 2006), the third most common cause ofneurodegenerative dementia after AD and dementia with Lewybodies (Bird et al., 2003). FTLD is caused by both genetic andenvironmental factors, but the exact cause are still unknown. It hasbeen hypothesized that both ischemia/hypoxia and oxidativestress are involved in the pathogenesis of several neurodegenera-tive diseases, including FTD (Martin et al., 2001; Gerst et al., 1999).In fact, the CNS is particularly susceptible to changes in local O2

levels, which can affect neuronal activity (Pena and Ramirez, 2005)and promote the development of disorders, including dementia(Bazan et al., 2002).

Interestingly, hypoxia increases progranulin expression infibroblast cultures (Guerra et al., 2007), but its effect on geneexpression in neuronal cells is still unknown. Several signalingpathways have been identified to regulate gene expression duringhypoxia (Seta et al., 2002). Among these, the hypoxia-inducedactivation of the MEK-ERK MAPK cascade could be an importantsignal transduction pathway of PGRN due to the presence ofrepeated consensus sequence for the transcription factor activatorprotein-1 (AP-1) in the promoter region of the gene (Bhandari et al.,1996).

On the other side, a possible role for oxygen free radicals inneurodegeneration has been also extensively examined (Mattson,2002; Fatokun et al., 2008).

Here, we investigate the in vitro effect on PGRN regulation oftwo different stimuli commonly associated with the progression ofneurodegenerative processes: hypoxia and oxidative stress. Our

P. Piscopo et al. / Neurochemistry International 57 (2010) 893–898894

results demonstrate that chronic hypoxic exposure up-regulatesPGRN expression in neuroblastoma cell lines via the mitogen-actived protein kinase (MAPK)/extracellular signal-regulatedkinase (MEK) signaling cascade. In contrast, PGRN is not modulatedby H2O2 treatment. These data, together with those of others(Larade, 2008; Guerra et al., 2007) suggest a possible protectiverole of PGRN against hypoxic/anoxic insults in the brain.

2. Methods

2.1. Cell lines

Human neuroblastoma cell lines SK-N-BE and SK-N-SH (kind gift of Prof. G.

Poiana, University ‘‘La Sapienza’’, Rome, Italy) were cultivated in RPMI-1640

medium (Euroclone) and Dulbecco’s modified Eagle’s medium (DMEM, Euroclone)

respectively. The human neuroblastoma cell line SH-SY5Y (kindly supplied by Dr. G.

Pani, Catholic University, Rome, Italy) was cultivated in Dulbecco’s modified Eagle’s

medium with 4.5 g glucose added per liter (DMEM high glucose, Euroclone). All

growth media were supplemented with 10% heat-inactivated (v/v) Foetal Bovine

Serum (FBS), 5 mM L-glutamine, penicillin (100 IU/ml) and streptomycin (100 mg/

ml). Cell cultures were maintained at 37 8C in a humidified atmosphere of 5% CO2.

For experiments, cells were seeded on 60 mm plastic culture dishes at a density of

1 � 104 cm�2 and were grown to 80% confluence, at which point, the medium was

changed. The general morphology of cell monolayers, before and after treatments,

was monitored by light microscopy. Cell proliferation was estimated by counting

the total number of cells in each dish with use of a hematocytometer. Cell viability

was determined by Trypan blue dye exclusion test (Sigma–Aldrich). We also

evaluated the number of cells that detached from the substrate and were found to

be freely floating in cell medium.

2.2. Cell treatments

For stimulating hypoxia, we used a hypoxic/anaerobic chamber (BBLTM

GasPakTM, USA). The system was set up at 37 8C in 5% CO2, 95% N2. Cells were

transferred into the humidified chamber and incubated with the appropriate media

for up to 24 h then, lysed for RNA isolation (below). Control cells were maintained in

the incubator under normoxic conditions. For stimulating intracellular oxidative

stress, sub-confluent cells were exposed to a final concentration of 100 mM H2O2 for

24 h under normoxic conditions, then, washed with PBS, harvested by trypsiniza-

tion, and processed as described below.

For stimulating inhibition of MEK cascades, sub-confluent cells were exposed to

a 40 mM 2-(2-amino-3-methoxyphenyl)-4H-1-benzopyran-4-one (PD98059; Sig-

ma–Aldrich) inhibitor for 24 h under hypoxic conditions; PD98059-treated cells

grown under normoxic conditions were used as controls. Cells were successively

washed with PBS, harvested by trypsinization and mRNA was recovered.

2.3. RNA isolation and quantitative real-time PCR

Total RNA was extracted from SK-N-BE, SK-N-SH and SH-SY5Y cells using the

Invisorb SpinCell RNA kit (Invitek). cDNA was made by retrotranscription using

SuperScript III first-strand cDNA synthesis kit (Invitrogen Inc., Carlsbad, CA) with

random primers, according to the manufacturer’s protocol. Reverse transcription

quantitative real-time PCR (RT-qPCR) was performed using the ABI PRISM 7000

Sequence Detection System (Applied Biosystems, Foster City, USA) with TaqMan

Universal PCR Master Mix and TaqMan Gene expression assays (Applied

Biosystems). The parameters for PCR amplification were: 50 8C for 2 min, 95 8Cfor 10 min followed by 40 cycles of 95 8C for 15 s and 60 8C for 1 min. PCR was

performed in triplicate for each sample; 18S rRNA was chosen as reference gene.

The relative expression of mRNA was calculated using the comparative Ct method.

2.4. RNA interference

RNA interference was performed using Thermo Scientific Dharmacon1 Accell1

siRNA Dharmacon (Euroclone S.p.A, Italy), according to manufacturer’s protocol.

This protocol is especially modified for delivering small-interfering RNA (siRNA)

into cells without use of a transfection reagent; the methodology works at a higher

probe concentration than conventional siRNA technology, with a minimal

disruption of cell expression profiles. For genetic knockdown experiments, cells

were plated at the appropriate cell density and grown overnight, then incubated

with 2 ml of fresh complete medium containing 1 mM Accell Smartpool siRNA

human PGRN (cat. no. E-009285-00-0050), in which a mixture of four siRNAs

targeting human PGRN gene was added. A negative control was used (cat. no. D-

001910-10-20), with a mixture of four siRNAs that had no significant homology to

any known human gene sequence; whereas a positive control (cat. no. D-001930-

10-20) had a mixture of four siRNAs that targeted the human GAPDH human gene.

After 24 h of incubation, cells were transferred into the hypoxic chamber for

additional 24 h incubation under hypoxic conditions, as described above. Cells were

then harvested, and processed for RNA extraction. The RNA silencing was not toxic

to cells: no statistically significant reduction in cellular viability was observed

relative to untreated hypoxic cell cultures (data not shown).

2.5. Evaluation of oxidative stress

The total intracellular glutathione content was determined, in three different

neuroblastoma cell lines, with use of the enzymatic recycling assay with

glutathione reductase (type IV, Sigma–Aldrich) and 5,50-dithiobis-2-nitrobenzoic

acid (DTNB, Sigma–Aldrich) (see Anderson, 1985). For the measurement of oxidized

glutathione (GSSG), the acidified homogenates were submitted to derivatization

with undiluted 2-vinylpyridine (Aldrich, Milwaukee, WI, USA) in the presence of

triethanolamine (Sigma–Aldrich), for 1 h at room temperature. Samples were then

assayed for total glutathione measurement. The amount of reduced glutathione

(GSH) present in the samples was calculated as the difference between total

glutathione and GSSG levels. Data were expressed as nmoles of GSH or GSSG per mg

of cell protein.

2.6. Protein content

Protein concentration was measured by the commercially available Coomassie

brilliant blue dye-binding assay (Bio-Rad, Hercules, CA, USA), following the

manufacturer’s instructions. Bovine g-globulin was used as a standard.

2.7. Western blot analysis

Cells were harvested in RIPA buffer (25 mM Tris–HCl pH 7.4, 150 mM NaCl, 1.0%

NP-40, 0.1% SDS, 1% Na-deoxycholate), containing a protease inhibitor cocktail

(Sigma–Aldrich), 1 mM Na3VO4, 5 mM NaF, 1 for 1 h in ice. The cellular extracts

were solubilized in 4X Laemmli sample buffer and boiled for 5 min. Proteins (20 mg)

were separated on 10% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and

transferred to polyvinylidene difluoride membranes (Westran, Schleicher and

Schuell). Blots were first blocked with 5% non-fat powdered milk in TBS/Tween

0.1%, then probed overnight at 4 8C with mouse monoclonal anti-Granulin (Abcam

no. ab 52557, 1:4000), mouse monoclonal anti-b-actin (Sigma–Aldrich, 1:80,000),

polyclonal anti-pERK1/2 (Cell Signalling, 1:1000), polyclonal anti-ERK1/2 (Cell

Signalling, 1:1000). Membranes were washed, incubated with horseradish

peroxidase-conjugated anti-mouse or anti-rabbit secondary antibody (Chemicon,

1:20,000) and developed by ECL (BIORAD) according to manufacturer’s instructions.

Densitometric analysis was performed with FX-Imager densitometer and the

Quantity One software (Bio-Rad Laboratories).

2.8. Statistical analysis

Data are expressed as mean � SEM. Comparisons among groups were made using

Student’s t-test or using one way ANOVA followed by Bonferroni’s multiple

comparison test, with significance set at P < 0.05.

3. Results

3.1. Preliminary assessment of hypoxia

We analysed the expression of glucose transporter GLUT1,which is regulated under hypoxic condition, as a marker ofhypoxia (Fisk et al., 2006). As shown in Fig. 1A, GLUT1 levelswere up-regulated in response to hypoxia in all cell lines tested,with a dramatical increase at 24 h of incubation. Moreover, 24 hof hypoxic treatment was not cytotoxic for the cells, at leastunder the assay conditions we have used here. In fact, thenumber of cells treated with hypoxia did not differ bymore than 5% from the number of control cells, without anydecrease in cell viability (96 � 2%). Twenty-four hours of H2O2

treatment did not alter GLUT1 gene expression in any of the celllines.

3.2. Preliminary assessment of oxidative stress

As a marker of oxidative stress we used the GSH/GSSG ratio,considered to be an indicator of redox potential of the cells (Esteveet al., 1999). Twenty-four hours of H2O2 treatment induced aconsistent oxidative stress in SK-N-BE cells, as indicated by asustained decrease in GSH and a parallel decrease in GSSG, with areduction of the GSH/GSSG ratio, (Table 1). However, under ourexperimental conditions, 24 h of chronic hypoxia did not alterglutathione levels. A significant reduction in cell number(�42 � 4.5%) accompanied by a moderate decrease in cell viability(88 � 4%), with respect to controls, was observed in the H2O2 treated

[(Fig._1)TD$FIG]

Fig. 1. Twenty-four hours of hypoxia treatment, affect GLUT-1 and PGRN gene expression in neuroblastoma cell lines. (A and B) The histograms show mRNA levels of GLUT-1

(A) and PGRN (B) after hypoxia administration. Data, compared to 18S rRNA, are expressed as fold changes relative to the control at each treatment. Mean � SEM of six

experiments is shown. (C) Representative Western blot analysis of PGRN protein. Densitometric analysis data are presented as ratios normalized by b-actin. The histogram shows

the mean � SEM of at least four preparations. **P < 0.01 and *P < 0.05 vs. untreated normoxic cells.

[(Fig._2)TD$FIG]

P. Piscopo et al. / Neurochemistry International 57 (2010) 893–898 895

cells. In parallel the number of cells detached from the substrate andfreely floating in the medium, increased. Similar results wereobtained in both SK-N-SH and SH-5YSY cells (data not shown).

3.3. Hypoxia, but not oxidative stress, induces PGRN up-regulation in

neuroblastoma cell lines

To test the effect of hypoxia and hydrogen peroxidetreatment s on the regulation of the PGRN gene, we analysedthe PGRN transcripts of multiple neuroblastoma cell lines byquantitative real-time PCR, under normoxic and hypoxicconditions. Fig. 1B shows a modulation in PGRN transcriptionafter 24 h of hypoxic compared to normoxic conditions. Inparticular, we found that PGRN mRNA was increased2.44 � 0.15-fold in SK-N-BE, 1.39 � 0.1-fold in SK-N-SH and1.96 � 0.19-fold in SH-SY5Y cells. Since the best hypoxic/normoxicPGRN expression ratio was displayed by SK-N-BE cells, we decidedto perform all subsequent analyses in this cell line. To test anotherdegenerative stimulus, we treated SK-N-BE cells for 24 h with100 mM H2O2. No effects on PGRN expression were observed afterexposure to hydrogen peroxide (Fig. 2). Moreover, to investigatewhether the differences noted in mRNA levels were also reflectedat the protein level, a specific immunoblot analysis was performedin SK-N-BE cells, revealing a consistent and significant increase ofPGRN proteins after 24 h of hypoxic treatment, relative tonormoxic conditions (Fig. 1C).

Table 1Twenty-four hours of 100 mM H2O2 treatment, but not hypoxia, induce oxidative

stress in SK-N-BE cells.

Parameter Control H2O2 Hypoxia

GSH (nmol/mg protein) 11.7� 0.99 6.88�0.72** 13.5�1.47

GSSG(nmol/mg protein) 0.37� 0.09 1.23�0.18** 0.45�0.07

GSH/GSSG 32.66�2.0 5.59�0.49*** 30�2.19

The means� SEM of six separate experiments, each performed in duplicate are shown.** P<0.01.*** P<0.001 vs. untreated normoxic cells.

3.4. siRNA-mediated down regulation of PGRN expression is

associated with altered cell morphology

In order to verify if and how much, PGRN protectedneuroblastoma cells from hypoxic damage we knocked downPGRN with use of small-interfering RNA (siRNA) technology. Theefficacy of silencing treatment was monitored by both RT-PCR andwestern blot analysis. Fig. 3A shows that, PGRN siRNA exposuredepressed PGRN expression in hypoxic SK-N-BE cells by 50%, withrespect to untreated hypoxic cells. Gene silencing was alsoconfirmed by the reduced expression of down-stream PGRNprotein, as revealed by western blot analysis (Fig. 3B). A parallelanalysis of cell monolayers performed by phase contrast micros-copy showed that cells under normoxic conditions (CTRL),displayed a flat and polygonal aspect with very short fillopodia;no significantly modifications were visible after 24 h of hypoxicconditions (hyp); when siRNA for PGRN was added for 24 h (siRNAPGRN), the cells displayed greater tendency to form clumps; anumber of cells displayed elongated cell bodies (Fig. 3C, arrows). Inaddition, a moderate decrease in the number of adherent cells wasappreciablely lower (�25 � 3%) compared to untreated cells, with aparallel increase (+12 � 2%) in the number of detached cells observedfloating freely in the medium. Finally, a significant reduction in cell

Fig. 2. Twenty-four hours of H2O2 did not affect PGRN and GLUT-1 gene expression

in SK-N-BE cells. Data, compared to 18S rRNA, are expressed as fold changes relative

to the control at each treatment. Mean � SEM of six experiments is shown.

[(Fig._3)TD$FIG]

Fig. 3. PGRN silencing determines cell number and morphology alterations. (A) siRNA PGRN transcripts were evaluated by quantitative Real time PCR. Data, compared to 18S

rRNA, are expressed as fold changes relative to the normoxic control at each treatment. Mean � SEM of three experiments is shown. *P < 0.05 vs. untreated normoxic cells. (B)

Western blot analysis in SK-N-BE cells silenced or not for PGRN protein for 48 h in hypoxic and normoxic conditions. (C) Phase contrast images of SK-N-BE showed cells in normoxic,

hypoxic and silencing conditions. Pictures are representative of three independent experiments and at least 10 fields of view.

P. Piscopo et al. / Neurochemistry International 57 (2010) 893–898896

viability was detected in siRNA-treated cells under hypoxic condi-tions, with respect to untreated hypoxic cells (90 � 3% vs. 96 � 2%;P < 0.05). Interestingly, the progranulin decrease induced by siRNAunder normoxic conditions did not significantly reduce cell viability(94 � 3%), nor cause any cellular shape modification.

3.5. PD98059 inhibitor reverts hypoxia-dependent

PGRN up-regulation

To evaluate if the MEK pathway is involved in hypoxia-inducedPGRN up-regulation, cells were treated with the specific MEKinhibitor PD98059. As showed in Fig. 4B the treatment with theinhibitor induces a �60% decrease of ERK 1/2 phosphorylation.PD98059 treatment reverted the PGRN mRNA transcripts, which

[(Fig._4)TD$FIG]

Fig. 4. (A) MEK inhibitor PD98059 reverts the PGRN hypoxia-induced up-regulation

in SK-N-BE neuron-like cells. Data, compared to 18S rRNA, are expressed as fold

changes relative to the normoxic control at each treatment. Mean � SEM of almost

three experiments is shown; **P < 0.01. (B) Activation of MEK/ERK pathway in SK-N-BE

cells after 24 h of hypoxic treatment. The specific MEK1 inhibitor PD98059 abolishes

ERK activation (representative Western blot). The histogram shows the data obtained

from densitometric analysis of the p-ERK/ERK ratio (n = 4). Bars represent the

mean � SEM; *P < 0.05.

increased after 24 h of hypoxia, to levels similar to those of controls(Fig. 4A). Moreover, the use of the inhibitor did not modify PGRNtranscription in SK-N-BE cells that were maintained undernormoxic conditions (Fig. 4A).

To evaluate if the MEK pathway is involved in hypoxia-inducedPGRN up-regulation, cells were treated with the specific MEKinhibitor PD98059. This treatment reverted the PGRN mRNAtranscripts, which were increased after 24 h of hypoxia, to levelssimilar to those of controls (Fig. 4A). Moreover, the use of theinhibitor did not modify PGRN transcription in SK-N-BE cells thatwere maintained under normoxic conditions (Fig. 4A). To establishthe strength of the inhibition we also carried out an analysis onactivation of MEK/ERK pathway in SK-N-BE cells after 24 h ofhypoxic treatment. As showed in Fig. 4B the inhibitor induces a�60% decrease of ERK 1/2 phosphorylation.

4. Discussion

Ischemia/hypoxia and oxidative stress play a role in thepathogenesis of several neurodegenerative diseases, includingFTD (Martin et al., 2001; Gerst et al., 1999). To evaluate a possibleinvolvement for PGRN in neurodegeneration, we evaluated themodulation of the PGRN gene following exposure to these twoneurodegenerative stimuli. Here, we report that levels of bothPGRN gene and protein expression were significantly elevatedfollowing chronic hypoxia. In contrast, neither gene expression northe protein, were altered following hydroperoxide-inducedoxidative stress. Thus, each of the two stress conditions, associatedwith neurodegeneration, has distinct effects on gene expression.Interestingly, Yin et al. described that PGRN-deficient hippocampalslices were hypersusceptible to deprivation of oxygen (Yin et al.,2010). Moreover, a study on a marine gastropod reportsaccumulation of l-grn transcripts as part of the response to anoxia(Larade, 2008). Guerra and colleagues also describe an up-regulation of PGRN transcripts in fibroblasts under condition ofhypoxia (Guerra et al., 2007). These findings are in agreement withthe present data, wherein we demonstrate a modulation not onlyof PGRN mRNA level, but also of its protein expression. Moreover,PGRN silencing was associated with changes in cell morphologyand a fewer adherent cells, suggesting a possible protective role forthis gene under hypoxic conditions. This neuroprotective action isalso confirmed by a study on serum-deprivation in motor neuronsdescribing that progranulin is cytoprotective over prolonged

P. Piscopo et al. / Neurochemistry International 57 (2010) 893–898 897

periods when over-expressed in a neuronal cell line (Ryan et al.,2009).

Oxygen-deprived stress induces a number of spatially andtemporally regulated intracellular responses, ranging fromreduced channel activity to altered gene expression. In fact,depending on the duration and severity of oxygen deprivation,cellular oxygen-sensor responses activate a variety of short- andlong-term energy-saving and cellular protection mechanisms.Hypoxic adaptation encompasses an immediate depolarizationblock by changing potassium, sodium and chloride ion fluxesacross the cellular membrane. Moreover, the physiologic responseto hypoxia modulates gene expression and protein synthesis, witha HIF-mediated up-regulation of enzymes or growth factorsinducing angiogenesis, anaerobic glycolysis, cell survival orneural stem cell growth (Van Elzen et al., 2008; Acker and Acker,2004).

Many potential regulatory elements associated with thefunctional activity of promoters of other eukaryotic genes arescattered throughout the 50 flanking region of the human PRGN.However, among these, the consensus sequences for the family ofhypoxia-inducible factors are not included. Notably, 25 sequencesshow a 7 out of 8 bp match with the AP-1 consensus sequenceTGA(C/G)TCAG (Bhandari et al., 1996). Members of the AP-1transcription factor family are well-known targets of the MEKsignaling cascade. Moreover, the MAPKs respond to a variety ofenvironmental stresses (Gaitanaki et al., 2004) including hypoxia(Haddad, 2004) and have a primary role in mediating stress-induced gene expression. Our data, for the first time, providestrong evidence that the highly selective inhibition of the MEKsignal transduction pathway abolishes the up-regulation of PGRNin SK-N-BE human neuron-like cell culture, as stimulated byhypoxia treatment. These findings demonstrate a direct involve-ment of MAPKs in the induction of mRNA transcription. Moreover,the inhibition of MEK activity does not alter PGRN mRNAexpression under normoxic conditions, suggesting that the PGRNgene transcription is specifically triggered by the hypoxic stimulus.As PGRN expression also promotes neuronal survival, andenhances neurite outgrowth in vitro, (Van Damme et al., 2008)it is reasonable that the observed cellular response to hypoxiacould be adaptative; in this case our data agree with the reportedPGRN blocking effect on tamoxifen-induced apoptosis in breasttumors (Tangkeangsirisin et al., 2004). Interestingly, PGRNactivates the extracellular regulated kinases, and the phosphatidylinositol-3 kinase signal cascades and also increases expression ofcyclins D and B (Lu and Serrero, 2001). So, regulation of PGRNexpression, exerted by the MAPKs under hypoxia, could beindicative of a survival loop which is switched on by cells toblunt pro-apoptotic environmental conditions, preserving neuro-nal viability. Interestingly, a similar hypothesis regarding thereciprocal functional interaction of PGRN and MEK pathway wasformulated by Kamrava et al. in their study on lysophosphatidicacid and endothelin-induced proliferation of ovarian cancer celllines suggesting a diffused proliferation and survival loop sharedby different cell types and unrelated environmental stimuli(Kamrava et al., 2005). In conclusion, our results suggest thatPGRN is highly susceptible to changes in local O2 tension and, as aspecific hypoxic sensor, could have an adaptative/protective roleagainst one of principal risk factors involved in AD and FTDpathogenesis as hypoxic stress.

Acknowledgments

The authors would address particular thanks to Dr. SonalJhaveri for reviewing the manuscript. This study was supported bya grant from the Italian Ministry of Health: ‘‘Genetic determinants

and modulator factors in neurodegenerative diseases: clinical andexperimental model’’ to A.C.

References

Acker, T., Acker, H., 2004. Cellular oxygen sensing need in CNS function: physiolog-ical and pathological implications. J. Exp. Biol. 207, 3171–3188.

Anderson, M.E., 1985. Determination of glutathione and glutathione disulfide inbiological samples. Methods Enzymol. 113, 548–555.

Baker, C.A., Manuelidis, L., 2003. Unique inflammatory RNA profiles of microglia inCreutzfeldt–Jakob disease. Proc. Natl. Acad. Sci. U.S.A. 100, 675–679.

Baker, M., Mackenzie, I.R., Pickering-Brown, S.M., Gass, J., Rademakers, R., Lindholm,C., Snowden, J., Adamson, J., Sadovnick, A.D., Rollinson, S., Cannon, A., Dwosh, E.,Neary, D., Melquist, S., Richardson, A., Dickson, D., Berger, Z., Eriksen, J.,Robinson, T., Zehr, C., Dickey, C.A., Crook, R., McGowan, E., Mann, D., Boeve,B., Feldman, H., Hutton, M., 2006. Mutations in progranulin cause tau-negativefrontotemporal dementia linked to chromosome 17. Nature 442, 916–919.

Bateman, A., Belcourt, D., Bennett, H., Lazure, C., Solomon, S., 1990. Granulins, anovel class of peptide from leukocytes. Biochem. Biophys. Res. Commun. 173(3), 1161–1168.

Bazan, N.G., Palacios-Pelaez, R., Lukiw, W.J., 2002. Hypoxia signaling to genes:significance in Alzheimer’s disease. Mol. Neurobiol. 26 (2–3), 283–298.

Bhandari, V., Daniel, R., Lim, P.S., Bateman, A., 1996. Structural and functionalanalysis of a promoter of the human granulin/epithelin gene. Biochem. J. 15,441–447.

Bird, T., Knopman, D., VanSwieten, J., Rosso, S., Feldman, H., Tanabe, H., Graff-Raford,N., Geschwind, D., Verpillat, P., Hutton, M., 2003. Epidemiology and genetics offrontotemporal dementia/Pick’s disease. Ann. Neurol. 54 (Suppl. 5), S29–31.

Daniel, R., Daniels, E., He, Z., Bateman, A., 2003. Progranulin (acrogranin/PC cell-derived growth factor/granulin–epithelin precursor) is expressed in the pla-centa, epidermis, microvasculature, and brain during murine development.Dev. Dyn. 227, 593–599.

Daniel, R., He, Z., Carmichael, K.P., Halper, J., Bateman, A., 2000. Cellular localizationof gene expression for progranulin. J. Histochem. Cytochem. 48, 999–1009.

Esteve, J.M., Mompo, J., Garcia de la Asuncion, J., Sastre, J., Asensi, M., Boix, J., Vina,J.R., Vina, J., Pallardo, F.V., 1999. Oxidative damage to mitochondrial DNA andglutathione oxidation in apoptosis: studies in vivo and in vitro. FASEB J. 13,1055–1064.

Fatokun, A.A., Stone, T.W., Smith, R.A., 2008. Oxidative stress in neurodegenerationand available means of protection. Front Biosci. 13, 3288–3311.

Fisk, L., Nalivaeva, N.N., Turner, A.J., 2006. Regulation of endothelin-convertingenzyme-1 expression in human neuroblastoma cells. Exp. Biol. Med. (May-wood) 231 (6), 1048–1053.

Gaitanaki, C., Kefaloyianni, E., Marmari, A., Beis, I., 2004. Various stressors rapidlyactivate the p38-MAPK signaling pathway in Mytilus galloprovincialis (Lam.).Mol. Cell. Biochem. 260, 119–127.

Gerst, J.L., Siedlak, S.L., Nunomura, A., Castellani, R., Perry, G., Smith, M.A., 1999. Roleof oxidative stress in frontotemporal dementia. Dement. Geriatr. Cogn. Disord.10 (Suppl. 1), 85–87.

Guerra, R.R., Kriazhev, L., Hernandez-Blazquez, F.J., Bateman, A., 2007. Progranulin isa stress–response factor in fibroblasts subjected to hypoxia and acidosis.Growth Factors 25, 280–285.

Haddad, J.J., 2004. Hypoxia and the regulation of mitogen-activated protein kinases:gene transcription and the assessment of potential pharmacologic therapeuticinterventions. Int. Immunopharmacol. 4, 1249–1285.

He, Z., Bateman, A., 2003. Progranulin (granulin–epithelin precursor, PC-cell-de-rived growth factor, acrogranin) mediates tissue repair and tumorigenesis. J.Mol. Med. 81, 600–612.

Kamrava, M., Simpkins, F., Alejandro, E., Michener, C., Meltzer, E., Kohn, E.C., 2005.Lysophosphatidic acid and endothelin-induced proliferation of ovarian cancercell lines is mitigated by neutralization of granulin–epithelin precursor (GEP), aprosurvival factor for ovarian cancer. Oncogene 24 (47), 7084–7093.

Larade, K., Storey, K.B., 2008. Identification of a granulin-like transcript expressedduring anoxic exposure and translated during aerobic recovery in a marinegastropod. Gene 410 (1), 37–43.

Lu, R., Serrero, G., 2001. Mediation of estrogen mitogenic effect in human breastcancer MCF-7 cells by PC-cell-derived growth factor (PCDGF/granulin precur-sor). Proc. Natl. Acad. Sci. U.S.A. 98, 142–147.

Martin, J.A., Craft, D.K., Su, J.H., Kim, R.C., Cotman, C.W., 2001. Astrocytes degeneratein frontotemporal dementia: possible relation to hypoperfusion. Neurobiol.Aging 22, 195–207.

Mattson, M.P., 2002. Oxidative stress, perturbed calcium homeostasis, and immunedysfunction in Alzheimer’s disease. J. Neurovirol. 8 (6), 539–550.

Pena, F., Ramirez, J.M., 2005. Hypoxia-induced changes in neuronal network prop-erties. Mol. Neurobiol. 32, 251–283.

Ryan, C.L., Baranowski, D.C., Chitramuthu, B.P., Malik, S., Li, Z., Cao, M., Minotti, S.,Durham, H.D., Kay, D.G., Shaw, C.A., Bennett, H.P., Bateman, A., 2009. Progra-nulin is expressed within motor neurons and promotes neuronal cell survival.BMC Neurosci. 10, 130.

Seta, K.A., Spicer, Z., Yuan, Y., Lu, G., Millhorn, D.E., 2002. Responding to hypoxia:lessons from a model cell line. Sci. STKE (146), re11.

Suzuki, M., Nishiahara, M., 2002. Granulin precursor gene: a sex steroid-induciblegene involved in sexual differentiation of the rat brain. Mol. Genet. Metab. 75,31–37.

P. Piscopo et al. / Neurochemistry International 57 (2010) 893–898898

Tangkeangsirisin, W., Hayashi, J., Serrero, G., 2004. PC cell-derived growth factormediates tamoxifen resistance and promotes tumor growth of human breastcancer cells. Cancer Res. 64 (5), 1737–1743.

Van Damme, P., Van Hoecke, A., Lambrechts, D., Vanacker, P., Bogaert, E., vanSwieten, J., Carmeliet, P., Van Den Bosch, L., Robberecht, W., 2008. Progranulinfunctions as a neurotrophic factor to regulate neurite outgrowth and enhanceneuronal survival. J. Cell. Biol. 81, 37–41.

Van Elzen, R., Moens, L., Dewilde, S., 2008. Expression profiling of thecerebral ischemic and hypoxic response. Expert Rev. Proteomics 5, 263–282.

Yin, F., Banerjee, R., Thomas, B., Zhou, P., Qian, L., Jia, T., Ma, X., Ma, Y., Iadecola, C.,Beal, M.F., Nathan, C., Ding, A., 2010. Exaggerated inflammation, impaired hostdefense, and neuropathology in progranulin-deficient mice. Exp. Med. 207 (1),117–128.