Valproic acid downregulates heme oxygenase-1 independently of Nrf2 by increasing ubiquitination and proteasomal degradation

Accepted Manuscript Valproic acid downregulates heme oxygenase-1 independently of Nrf2 by increasing ubiquitination and proteasomal degradation Mateusz Jez, Maciej Ciesla, Jacek Stępniewski, Agnieszka Langrzyk, Lucie Muchova, Libor Vitek, Alicja Jozkowicz, Jozef Dulak PII:

S0006-291X(17)30309-1

DOI:

10.1016/j.bbrc.2017.02.041

Reference:

YBBRC 37289

To appear in:

Biochemical and Biophysical Research Communications

Received Date: 2 February 2017 Accepted Date: 7 February 2017

Please cite this article as: M. Jez, M. Ciesla, J. Stępniewski, A. Langrzyk, L. Muchova, L. Vitek, A. Jozkowicz, J. Dulak, Valproic acid downregulates heme oxygenase-1 independently of Nrf2 by increasing ubiquitination and proteasomal degradation, Biochemical and Biophysical Research Communications (2017), doi: 10.1016/j.bbrc.2017.02.041. 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.

ACCEPTED MANUSCRIPT Jez et al.

Valproic acid downregulates heme oxygenase-1 independently of Nrf2 by increasing

RI PT

ubiquitination and proteasomal degradation

Mateusz Jez1,¶, Maciej Ciesla1,¶, Jacek Stępniewski1, Agnieszka Langrzyk2, Lucie Muchova3,

Department of Medical Biotechnology, Faculty of Biochemistry, Biophysics and

M AN U

1

SC

Libor Vitek3, Alicja Jozkowicz1, Jozef Dulak1,2*

Biotechnology, Jagiellonian University, Krakow, Poland 2

Kardio-Med Silesia, M. Curie-Skłodowskiej 10c, Zabrze, Poland

3

Fourth Department of Internal Medicine and Institute of Medical Biochemistry and

D

Laboratory Medicine, First Faculty of Medicine, Charles University in Prague, Prague, Czech

EP

TE

Republic.

AC C

* Corresponding author:

e-mail: [email protected] ¶

1

contributed equally

ACCEPTED MANUSCRIPT Jez et al.

Abstract

RI PT

Aims. Heme oxygenase-1 (HO-1; HMOX1 in human, Hmox1 in mice) is an antioxidative enzyme affecting wide range of sub-cellular processes. It was shown to modulate tumor growth or vascular-related diseases, thus being putative molecular target for tailored therapies. Therefore it is of importance to elucidate novel compounds regulating HO-1

SC

activity/expression and to delineate mechanisms of their action. In the present study we aimed to understand mode of action of valproic acid (VA), an antiepileptic drug, on HO-1

M AN U

expression.

Results. We demonstrated that HO-1 expression is decreased by VA at protein but not mRNA level in human alveolar rhabdomyosarcoma cell line CW9019. Nrf2 transcription factor, the activator of HO-1 expression through ARE sequence, was excluded as a mediator of HO-1

D

decrease, as VA downregulated Bach1, a Nrf2 repressor, concomitantly upregulating ARE

TE

activation. Also miRNA-dependent inhibition was excluded as a mechanism of HMOX1 regulation. However, co-immunoprecipitation assay showed a higher level of ubiquitinated

EP

HO-1 after VA treatment. Accordingly, MG132, an inhibitor of proteasomal degradation, reversed the effect of VA on HO-1 suggesting that decrease in HO-1 expression by VA is

AC C

through protein stability.

The inhibitory effect of VA on HO-1 was also observed in murine cells including embryonic fibroblasts isolated from Nrf2-deficient mice, what confirms Nrf2-independent effect of the compound. Importantly, VA decreased also HO-1 expression and activity in murine skeletal muscles in vivo. Conclusion. Our data indicate that VA downregulates HO-1 by acting through ubiquitinproteasomal pathway leading to decrease in protein level. 2

ACCEPTED MANUSCRIPT Jez et al. Keywords: Heme Oxygenase-1, Valproic acid, Ubiquitination, Rhabdomyosarcoma, Nrf2 Introduction

RI PT

Heme oxygenase-1 (HO-1; HMOX1 in human, Hmox1 in mice), a cytoprotective enzyme degrading heme to carbon monoxide, ferrous ions and biliverdin, modulates a number of biochemical and physiological processes, including angiogenesis[1], cell differentiation[2] and carcinogenesis[3]. Importantly, due to its anti-oxidative and pro-angiogenic properties,

SC

increased expression of HO-1 influences development and progression of many tumors and

thus it is regarded as a promising target for anti-cancer therapy[3,4]. Our recent data indicate

M AN U

also the crucial role of HO-1 in development of rhabdomyosarcoma (RMS), the tumor arising from aberrant myogenesis. Overexpression of HO-1 promoted tumorigenic transformation of murine myoblasts[5] and higher HMOX1 expression was observed in human RMS cell lines and clinical primary tumors of more aggressive, alveolar phenotype. Significance of HO-1 for

D

tumor growth, progression and resistance to chemotherapy justifies inhibition of this protein

TE

for anti-cancer treatment[3]. The major shortcoming of designing anti-cancer HO-1-based therapies, however, is the limited number of so far established compounds inhibiting

EP

transcription and/or activity of this enzyme[3]. Vast majority of HO-1-targeting molecules belong to the family of metalloporphyrin-

AC C

based compounds which are insoluble in hydrophilic solvents. Mostly studied tin and zinc protoporphyrins and their modifications initially brought promising results in in vitro experiments[6,7]. However due to their toxicity and wide range of off-target effects, the clinical utility of these compounds is limited. Alternatively, class of imidazole-based molecules with better biocompatibility and less adverse effects than metalloporphyrins were shown to efficiently inhibit HO-1 in vivo[8–10]. Both groups, however, decrease only the enzymatic activity of HO-1 which is followed by increased expression of HO-1 whereas

3

ACCEPTED MANUSCRIPT Jez et al. recent findings indicate the non-enzymatic functions of this protein in various biologic processes, including carcinogenesis[11,12]. This suggests the need to inhibit both expression and enzymatic activity of HO-1 for therapeutic purposes.

RI PT

HO-1 level is upregulated upon different external stimuli usually linked with stressful conditions such as inflammation or high level of prooxidants[13]. HMOX1/Hmox1 expression is mainly regulated at the transcriptional level by Nrf2 (nuclear factor E2-related factor 2)

SC

being the major transcription factor implicated in HO-1 induction[14]. It binds to antioxidant response element (ARE), present in the promoters of numerous genes involved in

M AN U

antioxidative response signaling, including HO-1[15]. In basal conditions Nrf2 is sequestrated in the cytoplasm and further degraded, however, upon stressful stimuli it translocates to the nucleus, binds to small MAFs proteins and activates ARE. This signaling pathway is inhibited by Bach1, which competes with Nrf2 in binding to MAFs and ARE sequence and lacks

D

activator domain[16].

TE

Valproic acid (VA) is commonly used anti-convulsant and anti-epileptic drug[17], that was recently described to decrease HO-1 production by influencing proteasome activity in the

EP

rat brain[18]. VA is mainly known as histone deacetylases (HDACs) inhibitor leading to formation of open chromatin structure. This results in increased expression of many genes and

AC C

seems to be an upstream effector of other altered pathways upon VA treatment[19]. Importantly, it was shown to be a promising compound against various cancers, as it enhanced differentiation, inhibited proliferation and increased apoptosis of cancer cells[20– 22]. However, mechanisms governing the VA-mediated action is not fully understood. Interestingly, in our recent study the effect of pharmacologic inhibition of HO-1 by tin protoporphyrin IX (SnPPIX) on human RMS growth and vascularization in xenograft model was mimicked by VA[23]. Taking into consideration the reported inhibition of HO-1 by

4

ACCEPTED MANUSCRIPT Jez et al. VA[18] we investigated the mechanism of action of VA in the process of HO-1 suppression in RMS cells and muscle tissue.

RI PT

Materials and methods Animals

C57BL/6 Nfe2l2-/-(Nrf2 knockout) mice from our colony were originally kindly gifted

SC

by dr. Masayuki Yamamoto (Tohoku University, Sendai, Japan[24]). C57/B16 mice were purchased from Charles River Laboratories (Massachusetts, USA). All experiments on

M AN U

animals were performed in accordance with national and European regulations upon approval by the First Local Ethical Committee on Animal Testing at the Jagiellonian University in Krakow. Cell cultures

D

Murine embryonic fibroblasts (MEFs) were isolated from control and Nrf2 knockout mice as previously described[25]. CW9019 (alveolar) human rhabdomyosarcoma cell line

TE

was kindly provided by Dr. Peter Houghton (St Jude Children’s Research Hospital, Memphis, TN). NIH3T3 murine fibroblast cell line was purchased from ATCC.

EP

Cells were cultured under normal conditions (37oC, 5% CO2 and 90% humidity) in growth

AC C

medium (GM): Dulbecco’s Modified Eagles Medium with 4.5g/L of glucose (DMEM HG, Lonza) supplemented with 10% fetal bovine serum (FBS, Biowest), penicillin (100 U/ml) and streptomycin (100 μg/ml Sigma-Aldrich). To evaluate the effect of VA on HO-1 at mRNA and protein level, CW9019 cells were

treated with 1-2 mM VA (Sigma-Aldrich) for 2 or 5 days in GM. Medium was changed every second day. To evaluate the effect of VA on HO-1 at the protein level in murine NIH3T3 fibroblasts, cells were stimulated with 0.1, 0.5, 1 and 2 mM VA for 24h in GM. Subsequently,

5

ACCEPTED MANUSCRIPT Jez et al. 1 mM and 2 mM VA was used to assess its effect on HO-1 mRNA in both human and murine cell lines. Additionally, control and Nfe2l2-/- mouse embryonic fibroblasts were stimulated with 2 mM VA as the most effective dose for 24h in GM. To investigate a possible

RI PT

mechanism of VA action on HO-1 at protein level, CW9019 cell line was pre-treated with 10 M proteasomal inhibitor MG132 (Bioshop) for 24h and then 2 mM VA was applied for next 24h. Protein was isolated after 48h. Additionally, RNA was isolated after 24h treatment with

SC

10 μM MG132. Doses of VA were chosen based on other studies evaluating the anti-cancer properties of this compound[19,20,22].

M AN U

In vivo administration of VA

VA was dissolved in phosphate buffered saline and administered intraperitoneally into C57/Bl6 mice (both sexes) every other day for 1 week at the dose of 500 mg/kg body weight (bw) within the range of standard doses of this compound utilized in vivo studies[18].

D

Animals were sacrificed with ketamine (90mg/kg bw) mixed with xylazine (10 mg/kg bw).

TE

Muscles and liver were collected for RNA and protein isolations and for assessment of HO activity in cell lysates. For HO activity measurement, 2 mg of muscle tissue was incubated for

EP

15 minutes at 37°C in CO-free septum-sealed vials containing identical volumes of 150 μM heme and 4.5 mM NADPH. Reaction was terminated by adding 5μl of 30% (w/v)

AC C

sulfosalicylic acid. The amount of CO generated by the reaction was quantified by gas chromatography with a Reduction Gas Analyzer (Peak Laboratories) as described elsewhere[26].

microRNAs binding to 3’ UTR of HO-1 luciferase assay To assess binding of miRNAs to HO-1 3’UTR CW9019 cells were stimulated with 2 mM VA and after 3h were transfected with plasmid containing luciferase reporter gene and 3’UTR of human HO-1 (pLightSwitch_HMOX1_3UTR, Switchgear Genomics) using

6

ACCEPTED MANUSCRIPT Jez et al. jetPRIME transfection kit (Polyplus Transfection). After next 24h cells were washed with PBS and measurement of luciferase activity was performed with LuciferaseAssay System (Promega) according to the vendor’s protocol. To normalize signal for protein content, protein

RI PT

concentration was measured using BCA assay (Bioshop). ARE-luciferase assay

To evaluate the effect of VA on Nrf2 activity, plasmid containing luciferase coding

SC

sequence under the control of ARE was utilized (pARE-Luc, kindly gifted by Dr J.A.

Johnson, University of Wisconsin, Madison, USA[27]). CW9019 cells were treated with 2

M AN U

mM VA and after 3h transfection with pARE-Luc using JetPRIME reagent was performed. After 24h luciferase activity was measured with Luciferase Assay System. To normalize signal for protein content, protein concentration was measured using BCA assay. RNA isolation and qRT-PCR

D

For in vitro studies RNA was isolated using Fenozol reagent (A&A Biotechnology)

previously[23].

TE

according to manufacturer’s protocol. From murine muscles RNA was isolated as described

EP

To check the mRNA expression of HMOX1 and muscle-specific miRNAs qRT-PCR was performed using SybrGreen Mix (Sigma-Aldrich) and specific primers (Oligo.pl) (listed

AC C

in table 1). Eef2 (encoding eukaryotic translation elongation factor 2, EF-2) or ACTB (encoding β actin) were used as housekeeping gene for the reference for RNA content normalization. For miRNAs small nuclear RNA U6 was used as reference. Universal primers for human HMOX1 and mouse Hmox1 were used. For quantification of miRNAs universal reverse primer supplied with NCode VILO miRNA cDNA Synthesis kit was used.

7

ACCEPTED MANUSCRIPT

Table 1. List of primers.

RI PT

Jez et al.

Primer

Sequence

HMOX1/Hmox

5’CTT TCA GAA GGG TCA GGT GTC C 3’

SC

1 forward

5’GTG GAG ACG CTT TAC GTA GTG C 3’

HMOX1/Hmox 1 reverse

5’GAC ATC ACC AAG GGT GTG CA 3’

Eef2 reverse

5’TCA GCA CAC TGG CAT AGA GG 3’

ACTB forward

5' AAG GGA CTT CCT GTA ACA ACG CA 3'

ACTB reverse

5' CTG GAA CGG TGA AGG TGA CA 3'

miR-1

5`TGG AAT GTA AAG AAG TAT GTA T 3`

miR-133a

5’TTG GTC CCC TTC AAC CAG CTG T 3’

miR-133b

5’TTG GTC CCC TTC AAC CAG CTA 3’

miR-206

5’TGG AAT GTA AGG AAG TGT GTG G 3’

Western Blotting

TE

D

M AN U

Eef2 forward

Whole cell lysate was obtained using RIPA buffer whereas cytoplasmic and nuclear

EP

fractions were obtained using Nuclear Extraction kit (Active Motif), according to vendor

AC C

protocol. Level of HO-1, Nrf2 and Bach1proteins was analyzed by Western Blot. Primary antibodies recognizing: HO-1 (Enzo rabbit-ADI-SPA-894 1:500), Nrf2 (Santa Cruz rabbit-H300 1:500), Bach1 (Santa Cruz mouse-F9 1:200), Lamin A (Santa Cruz rabbit-H102 1:500), α tubulin (Calbiochem mouse-CPO6 1:1000). Secondary antibodies: HRP-Goat anti rabbit (BD Biosciences 1:5000) and HRP-goat anti mouse (Cell Signaling 1:10000).

8

ACCEPTED MANUSCRIPT Jez et al. Co-immunoprecipitation assay To check the possible interaction between HO-1 and ubiquitin upon VA treatment, coimmunoprecipitation assay was performed. For that CW9019 cells were stimulated with 2mM

RI PT

VA for 48h. After isolation, 100 µg of proteins were incubated with 1μg of anti-ubiquitin antibody (P4D1, Santa Cruz Biotechnology) overnight at 4oC on rotating wheel. On the next

day, 4.2 μl of Protein G DYNAbeads (Thermo Fischer Scientific) were added for 1h at 4oC on rotating wheel. Then, beads were washed 3x with 300 μl of RIPA buffer and 2x with PBS

SC

using magnets. In order to release conjugated protein, beads were dissolved in 24 μl of 0.1 M

M AN U

glycine pH 2.6 and incubated on ice for 10 minutes. Then 6 μl of loading buffer was added and Western Blot for HO-1 was performed. Immunocytochemistry

To check intracellular interaction of HO-1 with ubiquitin, immunocytochemistry staining was performed. After stimulation of CW9019 cells with 2 mM VA for 48h, cells

D

were stained as described previously[23]. Primary antibodies used for immunocytochemistry:

TE

Ubiquitin (Enzo mouse-FK2 1:200) and HO-1 (Enzo rabbit-ADI-SPA-894 1:200). Secondary: HRP-Goat anti mouse and HRP-goat anti rabbit (both Life Technologies 1:1000). Pictures

EP

were taken on confocal microscope Leica TCS SP5 II.

AC C

Statistical analysis

All experiments are presented as mean ± SD of three independent repetitions unless

described differently. Statistical analysis was performed using two-tailed unpaired Student's ttest or one-way ANOVA.

Results 9

ACCEPTED MANUSCRIPT Jez et al. VA decreases HO-1 at the protein but not transcript level in human RMS cell lines independently of Nrf2 Our recent findings indicate that inhibition or silencing of HO-1 decreases the growth

RI PT

of RMS cells in vitro and in vivo[23]. Therefore, in this study the effect of VA on HO-1 was investigated in malignant CW9019 cell line. Interestingly, although VA decreased HO-1 protein (Fig. 1A), it did not affect HMOX1 mRNA (Fig. 1B) in CW9019 cells.

SC

To delineate the mechanism of VA-mediated downregulation of HO-1 protein, three

main pathways known to regulate HO-1 were evaluated: i) Nrf2-dependent signaling as Nrf2

M AN U

is a key transcription factor inducing HO-1 expression[14], ii) activation of Bach1 as it can inhibit Nrf2 activity[16] and, iii) activity of HMOX1 3`UTR as it reflects level of miRNAs targeting 3`UTR of HO-1 mRNA.

To check whether Nrf2 transcription factor is involved in VA-driven downregulation

D

of HO-1 total Nrf2 protein was analyzed after 48h of VA stimulation (Fig. 1A,C). As no

TE

changes were observed in Nrf2 protein abundance in the whole cell lysates despite concomitant downregulation of HO-1 (Fig. 1A), Nrf2 levels were evaluated in cytoplasmic

EP

and nuclear fractions. Again, no changes in Nrf2 levels were found (Fig. 1C) indicating that Nrf2 translocation is not affected by VA. Interestingly Bach1, repressor of Nrf2, was

AC C

downregulated by stimulation with VA (Fig. 1D). In line with these results activity of ARE was increased (Fig. 1E). Interestingly again, also activity of HMOX1 3`UTR was augmented, which suggested decreased pool of miRNAs targeting HMOX1. Nevertheless, level of candidate miRNAs (basing on screening in microRNA base www.microrna.org) -328, -3385p, -429 and -485-5p remained unchanged (data not shown) which indicates a lower content of other miRNAs targeting 3`UTR of HMOX1.

10

ACCEPTED MANUSCRIPT Jez et al. VA increases ubiquitination of HO-1, whereas inhibition of proteasomal degradation reverses effect of VA on HO-1 expression The obtained results suggested that VA may influence protein stability of HO-1

RI PT

instead of affecting its transcription or translation. To test this hypothesis Western Blot

analysis of HO-1 expression after stimulation with MG132, the inhibitor of proteasomal

degradation[28] was performed. Results revealed the reversal of effects of VA on HO-1 level

SC

after inhibition of proteasome-ubiquitination system (Fig. 2A). However, MG132 was also

reported to exhibit pro-oxidative properties[29] which may per se increase HO-1 expression.

M AN U

Therefore, the effect of MG132 on HO-1 transcript was evaluated. No statistically significant increase in HO-1 mRNA expression was observed in CW9019 cells upon VA treatment (data not shown).

To further verify the hypothesis that VA may affect HO-1 protein stability via proteasome-ubiquitination system, the co-immunoprecipitation of HO-1 and ubiquitin was

D

performed. Results of pull down of all ubiquitinated proteins and comparison of ubiquitinated

TE

HO-1 level in control and VA-treated samples indeed revealed the higher levels of ubiquitinated HO-1 after VA treatment (Fig. 2B).

EP

Finally, co-localization of HO-1 and ubiquitin was assessed by confocal imaging of immunocytochemical staining of these two proteins. Again, the HO-1 in the cells was

AC C

decreased (Fig. 2C). Additionally, in line with previously described results, slightly higher, statistically significant co-localization of HO-1 with ubiquitin was observed after VA treatment (Fig. 2C).

VA decreases HO-1 levels in vivo in muscles and upregulates levels of pro-myogenic microRNAs

11

ACCEPTED MANUSCRIPT Jez et al. To evaluate whether VA can similarly influence HO-1 level in vivo, mice were injected every second day with VA (i.p., 500 mg/kg bw) for one week. This resulted in markedly decreased HO-1 protein in muscle lysates of VA-treated animals in comparison to

RI PT

their control counterparts (Fig 3A). Additionally, activity of this protein determined by the measurement of CO production was significantly decreased in muscles (Fig. 3B) but not in

liver (not shown) upon VA treatment. This indicates certain tissue specificity of VA action in

SC

vivo, the mechanism of which needs further studies. Importantly, and in agreement with the

inhibitory effect of VA on HO-1, the content of the class of muscle specific microRNAs, i.e.

M AN U

miR-1, 133a/b and 206, that are downregulated by HO-1[5,23], was increased in VA-treated mice (Fig. 3C).

VA decreases HO-1 level in murine fibroblasts independently of Nrf2 To further assess the effect of VA on HO-1 expression in murine cells, the NIH3T3

D

fibroblast cells were treated with this compound. Importantly, 2 mM VA decreased HO-1

TE

protein level (Fig. 4A) whereas, similarly as in case of RMS cells, the HO-1 expression at mRNA level was not affected (Fig. 4B). Importantly, downregulation of HO-1 protein was

EP

observed after 2 mM VA treatment in primary fibroblasts isolated from Nfe2l2-/- mice which do not express Nrf2 (Fig. 4C). Also further analysis revealed no effect of VA on Nrf2 nuclear

AC C

translocation in wild type cells (Fig. 4D). This indicates that VA can influence HO-1 level in both human and mouse cells independent of Nrf2.

Discussion

We report in this study that VA decreases HO-1 protein level in human RMS cells by increasing HO-1 ubiquitination. This effect is independent of Nrf2 transcription factor

12

ACCEPTED MANUSCRIPT Jez et al. signaling pathway or regulation of miRNA targeting HO-1 mRNA. Ubiquitin-dependent mechanisms of HO-1 downregulation upon VA treatment was previously reported in rat brain[18]. Our study extends this observation to human rhabdomyosarcoma cells and murine

RI PT

fibroblasts showing additionally for the first time that it is Nrf2-independent. Expression of HO-1 is mainly regulated at the transcriptional level and Nrf2 transcription factor is considered to be

SC

one of the major drivers of HO-1 expression[30]. Noteworhy, we have not observed changes in Nrf2 nuclear translocation. Nevertheless, downregulation of Bach1, the main competitor of

M AN U

Nrf2 transcriptional activity, facilitated activation of ARE.

Given the lack of effect of VA on HMOX1 transcript, it could be hypothesized that these phenomena are linked to changed translation rather than to decreased gene transcription. Interestingly, the levels of several miRNAs targeting HMOX1 3’UTR were unchanged by VA

D

stimulation. However, as biocomputational prediction of miRNA binding to 3’UTR is not a

TE

perfect tool and does not preclude that some other miRNAs may be involved, the functional effect of miRNAs was verified. The results showed activation of HMOX1 3’UTR-mediated

EP

luciferase activity upon VA treatment, what suggests decreased activity of miRNAs, indicating that this regulatory pathway is not directly responsible for decreased HO-1 levels in

AC C

VA stimulated cells.

The lack of transcription- or miRNAs-dependent effect of VA on HMOX1 mRNA

prompted us to investigate the effect of VA on protein stability. Indeed, coimmunoprecipitation experiments demonstrated enrichment in ubiquitinated HO-1 upon VA treatment. Accordingly, MG132 (an inhibitor of proteasomal degradation[28]) attenuated effect of VA on HO-1 expression. Finally, VA increased interaction between ubiquitin and

13

ACCEPTED MANUSCRIPT Jez et al. HO-1 in co-localization assay. This indicates that VA downregulates HO-1 protein due to a higher proteasomal degradation. Additionally, inhibitory effect of VA was demonstrated in murine cells in vivo and in

RI PT

vitro. Importantly, inhibition of HO-1 in skeletal muscles in vivo correlated with increased

expression of muscle specific microRNAs (myomirs), confirming our previous[5] and current data[23] on reverse relationship between HO-1 and microRNAs. Moreover, also in murine

SC

cells VA inhibits HO-1 apparently in Nrf2-independent way, as evidenced by VA-exerted downregulation of murine HO-1 in Nrf2-deficient fibroblasts. Those results indicate for

M AN U

generality of the mechanisms of VA action.

In summary, we propose a mechanism of VA-dependent decrease in HO-1 protein level in human and murine cells. We suppose that downregulation of HO-1 is mediated by increased proteasomal degradation of this protein and not by miRNA action or reduced gene

EP

Figures and legends

TE

D

transcription.

Fig. 1. Effect of VA on HMOX1 regulation. A. VA downregulates HO-1 but does not influence expression of Nrf2 As a loading control lamin A (for nuclear fraction) and tubulin (cytoplasmatic fraction) were used. B. Lack

AC C

of differences between VA-stimulated (2mM) and control samples at HMOX1 mRNA level in qRT-PCR analysis. HMOX1 expression was normalized to Eef2 level. C. VA (2 mM) does not change Nrf2 sub-cellular localization in CW9019 cell line D. VA (2 mM) downregulates Bach1 protein. E. VA (2 mM) increases HO-1 3`UTR activity in 3`UTR-Luc activity assay (left panel) and ARE in ARE-luc assay. Bars represents mean ± SD of N=3 experiments. * - p<0.05 vs. control. Fig. 2. Effect of VA on HO-1 ubiquitination and co-localization of HO-1 and ubiquitin. A. CW9019 cells pretreated with 10 μM MG132 exhibit higher level of HO-1, also in the presence of VA (2 mM). B. VA increases interaction between HO-1 and ubiquitin in co-immunoprecipitation assay. C. VA stimulation leads to

14

ACCEPTED MANUSCRIPT Jez et al. slight increase in co-localization of HO-1 and ubiquitin according to confocal microscopy-based analysis. Bars represent mean ± SD from six independent pictures; western blots – representative experiment out of 2

RI PT

performed **** - p<0.001.

Fig. 3. Effect of VA on HO-1 in vivo. A. VA downregulates HO-1 protein in skeletal muscles . Tubulin was used as a loading control. B. VA treatment inhibited HO enzymatic activity in the muscle tissue. C. VA upregulated muscle-specific miRNAs: miR-1, -133a/b and -206 in qRT-PCR analysis. U6 snRNA was used as a

M AN U

SC

reference gene for normalization. N=4 animals were analyzed.* - p<0.05.

Fig. 4. Effect of VA on HO-1 in murine cells. A. VA downregulates HO-1 protein level in NIH3T3 fibroblasts. Densitometric analysis was performed with Western Blots from 3 independent experiments with tubulin as a reference protein. B. VA does not affect the Hmox1 mRNA in NIH3T3 fibroblasts, normalized to Eef2 levels. C. VA (2mM) decreases HO-1 in Nfe2l2-/-(Nrf2 KO) MEFs. D. VA (2 mM) does not affect nuclear translocation of

EP

Acknowledgments

TE

Additional Information

D

Nrf2. For each experiment N=3. Bars represent mean ± SD.* - p<0.05, *** - p<0.005, **** - p<0.0001.

This study was supported by grant of the NCN: Maestro (2012/06/A/NZ1/0004), grant

AC C

of the Foundation for Polish Science: Ventures (2011-7/2 and 2011-8/8) and the grant from the NCBiR (PBS2/B7/0/2013 and STRATEGMED (2/269415/11/NCBR/2015). Faculty of Biochemistry, Biophysics and Biotechnology of Jagiellonian University is a partner of the Leading National Research Center (KNOW) supported by the Ministry of Science and Higher Education. We thank prof. Jerzy Dobrucki and Agnieszka Hoang from the Department of Cell Biophysics for confocal microscopy analysis.

15

ACCEPTED MANUSCRIPT Jez et al. Competing financial interests: The authors declare no competing financial interests. Bibliography

[6] [7]

[8]

[9]

[10] [11]

RI PT

AC C

[12]

SC

[5]

M AN U

[4]

D

[3]

TE

[2]

J. Dulak, J. Deshane, A. Jozkowicz, A. Agarwal, Heme oxygenase-1 and carbon monoxide in vascular pathobiology: focus on angiogenesis., Circulation. 117 (2008) 231–41. M. Kozakowska, K. Szade, J. Dulak, A. Jozkowicz, Role of heme oxygenase-1 in postnatal differentiation of stem cells: a possible cross-talk with microRNAs., Antioxid. Redox Signal. 20 (2014). A. Loboda, A. Jozkowicz, J. Dulak, HO-1/CO system in tumor growth, angiogenesis and metabolism - Targeting HO-1 as an anti-tumor therapy., Vascul. Pharmacol. 74 (2015) 11–22. G.F. Vile, S. Basu-Modak, C. Waltner, R.M. Tyrrell, Heme oxygenase 1 mediates an adaptive response to oxidative stress in human skin fibroblasts., Proc. Natl. Acad. Sci. U. S. A. 91 (1994) 2607–10. M. Kozakowska, M. Ciesla, A. Stefanska, et al., Heme Oxygenase-1 Inhibits Myoblast Differentiation by Targeting Myomirs, Antioxid. Redox Signal. 16 (2012) 113–127. R.E. Böni, R.A. Huch Böni, et al., Tin-mesoporphyrin inhibits heme oxygenase activity and heme-iron absorption in the intestine., Pharmacology. 47 (1993) 318–29. S.K. Sahoo, T. Sawa, J. Fang, et al., Pegylated zinc protoporphyrin: a water-soluble heme oxygenase inhibitor with tumor-targeting capacity., Bioconjug. Chem. 13 1031– 8. J.Z. Vlahakis, C. Lazar, G. Roman, et al., Heme oxygenase inhibition by α-(1Himidazol-1-yl)-ω-phenylalkanes: effect of introduction of heteroatoms in the alkyl linker., ChemMedChem. 7 (2012) 897–902. G. Roman, J.Z. Vlahakis, D. Vukomanovic,et al., Heme oxygenase inhibition by 1aryl-2-(1h-imidazol-1-yl/1h-1,2,4-triazol-1-yl)ethanones and their derivatives., ChemMedChem. 5 (2010) 1541–55. V. Pittalà, L. Salerno, G. Romeo, et al., A focus on heme oxygenase-1 (HO-1) inhibitors., Curr. Med. Chem. 20 (2013) 3711–32. R. Hori, M. Kashiba, T. Toma, et al., Gene transfection of H25A mutant heme oxygenase-1 protects cells against hydroperoxide-induced cytotoxicity., J. Biol. Chem. Q.S. Lin, S. Weis, G. Yang, T. Zhuang, A. Abate, P.A. Dennery, Catalytic inactive heme oxygenase-1 protein regulates its own expression in oxidative stress., Free Radic. Biol. Med. 44 (2008) 847–55. A.M. Choi, J. Alam, Heme oxygenase-1: function, regulation, and implication of a novel stress-inducible protein in oxidant-induced lung injury., Am. J. Respir. Cell Mol. Biol. 15 (1996) 9–19. T. Ishii, K. Itoh, S. Takahashi et al., Transcription Factor Nrf2 Coordinately Regulates a Group of Oxidative Stress-inducible Genes in Macrophages, J. Biol. Chem. 275 (2000) 16023–16029. T. Prestera, P. Talalay, J. Alam, et al., Parallel induction of heme oxygenase-1 and chemoprotective phase 2 enzymes by electrophiles and antioxidants: regulation by upstream antioxidant-responsive elements (ARE)., Mol. Med. 1 (1995) 827–837. S. Dhakshinamoorthy, A.K. Jain, D.A. Bloom, A.K. Jaiswal, Bach1 competes with Nrf2 leading to negative regulation of the antioxidant response element (ARE)mediated NAD(P)H:quinone oxidoreductase 1 gene expression and induction in response to antioxidants., J. Biol. Chem. 280 (2005) 16891–900.

EP

[1]

[13]

[14]

[15]

[16]

16

ACCEPTED MANUSCRIPT Jez et al.

AC C

EP

TE

D

M AN U

SC

RI PT

[17] A.J. Williams, F.C. Tortella, X.M. Lu, et al., Antiepileptic drug treatment of nonconvulsive seizures induced by experimental focal brain ischemia., J. Pharmacol. Exp. Ther. 311 (2004) 220–7. [18] K.J. Kwon, J.N. Kim, M.K. Kim, et al., Neuroprotective effects of valproic acid against hemin toxicity: possible involvement of the down-regulation of heme oxygenase-1 by regulating ubiquitin-proteasomal pathway., Neurochem. Int. 62 (2013) 240–50. [19] M. Göttlicher, S. Minucci, P. Zhu, et al., , Valproic acid defines a novel class of HDAC inhibitors inducing differentiation of transformed cells., EMBO J. 20 (2001) 6969–78. [20] M. Kaiser, I. Zavrski, J. Sterz, C. Jakob, C. Fleissner, P.-M. Kloetzel, O. Sezer, U. Heider, The effects of the histone deacetylase inhibitor valproic acid on cell cycle, growth suppression and apoptosis in multiple myeloma., Haematologica. 91 (2006) 248–51. [21] L. Altucci, N. Clarke, A. Nebbioso, et al., Acute myeloid leukemia: therapeutic impact of epigenetic drugs., Int. J. Biochem. Cell Biol. 37 (2005) 1752–62. [22] A. Angelucci, A. Valentini, D. Millimaggi, et al., Valproic acid induces apoptosis in prostate carcinoma cell lines by activation of multiple death pathways., Anticancer. Drugs. 17 (2006) 1141–50. [23] M. Ciesla, P. Marona, M. Kozakowska,et al., Heme oxygenase-1 controls an HDAC4MIR-206 pathway of oxidative stress in rhabdomyosarcoma, Cancer Res. 76 (2016) 5707–5718. [24] K. Itoh, T. Chiba, S. Takahashi, et al., An Nrf2/small Maf heterodimer mediates the induction of phase II detoxifying enzyme genes through antioxidant response elements., Biochem. Biophys. Res. Commun. 236 (1997) 313–22. [25] J. Stepniewski, N. Kachamakova-Trojanowska, D. Ogrocki, et al., Induced pluripotent stem cells as a model for diabetes investigation., Sci. Rep. 5 (2015) 8597. [26] H.J. Vreman, R.J. Wong, T. Kadotani, D.K. Stevenson, Determination of carbon monoxide (CO) in rodent tissue: effect of heme administration and environmental CO exposure., Anal. Biochem. 341 (2005) 280–9. [27] M.J. Calkins, R.J. Jakel, D.A. Johnson, et al., Protection from mitochondrial complex II inhibition in vitro and in vivo by Nrf2-mediated transcription., Proc. Natl. Acad. Sci. U. S. A. 102 (2005) 244–9. [28] D.H. Lee, A.L. Goldberg, Proteasome inhibitors: valuable new tools for cell biologists., Trends Cell Biol. 8 (1998) 397–403. [29] M. Lauricella, A. D’Anneo, M. Giuliano, et al., Induction of apoptosis in human osteosarcoma Saos-2 cells by the proteasome inhibitor MG132 and the protective effect of pRb., Cell Death Differ. 10 (2003) 930–2. [30] A. Loboda, M. Damulewicz, E. Pyza, et al., Role of Nrf2/HO-1 system in development, oxidative stress response and diseases: an evolutionarily conserved mechanism, Cell. Mol. Life Sci. (2016).

17

AC C

EP TE D M AN U

SC

RI PT

AC C

EP TE D M AN U

SC

RI PT

AC C

EP TE D M AN U

SC

RI PT

AC C

EP TE D M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

RI PT SC M AN U D TE EP

 

Valproic acid decreases HO-1 level at protein but not mRNA level in rhabdomyosarcoma and murine cells Valproic acid decreases HO-1 through increased proteasomal degradation Downregulation of HO-1 protein by valproic acid results in increased expression of myomiRs in vivo

AC C

