Summer Research Fellowship Programme of India's Science Academies

Effect of valproic acid treatment on hepatocytes in vitro

Saiprakash Rout

M.Sc in Chemistry, Pondicherry University, Puducherry 605014

Dr. Soumen Kanti Manna

Associate Professor, Biophysics and Structural Genomics Division, Saha Institute of Nuclear Physics, Kolkata, West Bengal 700064


Valproic acid (VPA) is an anticonvulsant drug, which is used in the treatment of conditions like epilepsy, bipolar disorder. VPA has been reported to cause three clinically distinguishable forms of hepatotoxicity such as hyperammonemia, acute hepatocellular injury with jaundice and Reye-like syndrome. However no method to enable prediction of onset, type and severity of hepatotoxicity exists till date. In this study, we aim to analyze metabolic rearrangement associated with valproic acid exposure to hepatocytes in vitro using mass spectrometry and find out correlation between hepatotoxicity and metabolic signatures. We hope it would eventually help to elucidate the mechanism underlying hepatotoxic effects of valproate and identify putative targets to minimize the same. We designed a dose-response experiment on VPA-induced hepatotoxicity using human hepatoma (HepG2). Cells were treated with vehicle or 0.5 mM to 8mM sodium valproate and effect on cell morphology, growth, viability and death were examined. Metabolites were extracted from vehicle or 4 mM sodium valproate-treated cells using organic solvent mixture. Metabolic signatures were analyzed using gas chromatography coupled with mass spectrometry (GCMS) following trimethylsilylation using MSTFA. Results indicated significant changes in metabolome associated with valproate treatment.

Keywords: valproic acid, hepatotoxicity, metabolic signature, mass spectrometry


VPA  Valproic Acid 
 GABA Gamma Amino Butyrate 
HDAC  Histone Deacetylase
HCC Hepatocellular carcinoma 
PCOS Polysystic ovary syndrome 
GC-MS  Gas Chromatography-Mass Spectrometry 
DMEM  Dulbecco modified Eagle Medium 
FBS  Fetal Bovine Serum 
MSTFA  N-Methyl-N-(trimethylsilyl)trofluoroacetamide 
rcf retative centrifugal force 
 DMSODimethyl sulphoxide 
CPS Carbamoylphosphate synthetase 



Valproic acid (2-propylvaleric acid, 2-propylpentanoic acid or n-dipropylacetic acid; see Figure 1), is a branched short chain fatty acid. VPA was first prepared in 1881 and came into medicinal use in 1962. It is widely used and included the World Health Organization List of Essential Medicines, most warranted and effective medicines for healthcare system.

vpa diagram.png
    Valproic Acid

    VPA as a therapeutic agent is available under various brand names such as Depacon, Mylproin, Depakote, Myproic acid and Convulex.

    Uses of VPA

    VPA is primarily used to treat epilepsy and bipolar disorder. It also used to prevent migraine headaches. Due to it’s anti-convulsant activity, it’s also used in treatment of tonic-clonic seizures, absence seizures and myoclonic seizures [1-4].

    In human brain, VPA alters the activity of the neurotransmitter gamma amino butyric acid (GABA) by potentiating the inhibitory activity of GABA through several mechanisms, including inhibition of GABA degradation, increased GABA synthesis [5] and decrease turnover ( Figure 2).

    Valproic acid

    ↓ Inhibit

    GABA transaminase

      Incresased conc. Of GABA → Inhibit Presynaptic dischage

    Post synaptic discharge

    In brain

     Fig2:Effect of VPA on GABA

    VPA attenuates N-Methyl-D-Aspartate-medicated excitation and blocks voltage gated channels (Na, K, Ca) [6]. VPA has also been described as an HDAC inhibitor [7, 8], which increased interest for its use in cancer therapy. VPA as a HDAC inhibitor is able to alter expression of many genes. Corresponding proteins were described to play important role in cellular activity and influence several pathways such as cell cycle control, differentiation and apoptosis.

    Recently it has been shown that VPA is also able to induced mono, di-, tri-methylation of histone 3, particularly at lysine 9 (H3K4). Methylation of histones at this lysine is associated with increased transcriptional activity [9-12].


    Although there are existing and promising therapeutic uses of VPA, it is often associated with side effects like nausea, vomiting, low blood pressure, etc. VPA has also been reported to cause three clinically distinguishable forms of hepatotoxicity such as hyperammonemia, acute hepatocellular injury with jaundice and Reye-like syndrome [13]. Hyperammonemia is a condition characterized by elevation in the level of ammonia in serum above 40 mmols/L. It is commonly associated with liver disease or inborn errors of metabolism. Valproic acid and its derivatives are thought to inhibit the enzyme, carbamoylphosphate synthetase I (CPS I), thereby preventing the incorporation of ammonia into the urea cycle [14]. In addition, it may inhibit the action of N-acetylglutamate, an activator of CPS I and also increase the concentration of pyruvate, a potent inhibitor of CPS I. All these may reduce incorporation of ammonia in the urea cycle and increase serum ammonia level (Figure 3).

      Schematic representation of urea cycle and the inhibition effect of valproic acid.


      Although cause of hyperammonemia is understandable, the mechanism underlying other types of hepatotoxic effects of valproic acid is not clearly understood. It is worth noting that hepatotoxicity occurs in only a fraction of patients treated with valproate and the nature and severity varies widely. Thus, identifying the determinants of valproate-induced hepatotoxicity and signatures that can monitor and predict the same would be of great help towards safer and effective use of the drug.

      Rationale for using hepatocytes

      Cultures of primary hepatocytes have been the gold standard for in vitro testing as they can maintain functional activities for 24-72 hours, can be used for enzyme induction and inhibition studies, allow for medium-throughput screening of compounds, and are ideal for examining interspecies and inter-individual differences in metabolism [15,16]. In vitro systems also allow to study chemical metabolism, evaluate the mechanisms of toxicity, measure enzyme kinetics, and examine dose-response relationships [17]. The HepG2 line was generated in the 1970s and expresses many liver-specific genes. In fact, it has successfully used to study metabolism and toxicity of several drugs including acetaminophen, tamoxifen, etc [18].

      Rationale for using metabolomics

      The majority of valproate metabolism occurs in the liver. VPA is also known to be metabolized by the cytochrome P450 enzymes like CYPA6, CYP2B6, CYP2C9 and CYP3A5. In adult patient, 30-50% of the administered dose is eventually excreted in urine as glucuronide conjugate. It is interesting note that the activity of cytochrome P450s and glucuronosyltransferase are essentially linked to metabolic processes including central carbon metabolism. In addition, cytochrome P450s often leads to production of ROS and ROS scavenging is connected to sulphur metabolism [19,20]. Thus it seems, in addition to its effect on urea cycle valproic acid may have a wider effect on cellular metabolism, which in turn affects the fate of hepatocytes. Metabolomics, which refers to global profiling of exogenous and endogenous biochemicals (< 1200 Da), may thus not only be helpful in understanding the mechanism underlying VPA toxicity but also identify signatures that could help to monitor and predict VPA toxicity.


      1. Analysis of the effect of various VPA concentrations and treatment durations on hepatocytes in vitro.

      2. Analysis of the effect of VPA treatment on hepatocyte metabolome.


      The HepG2 cell line was kindly gifted by the laboratory of Prof. D Mukhopadhyay from Saha institute of Nuclear Physics, Kolkata. All the chemicals, solvents, MTT reagent purchased from Sigma Aldrich, valproic acid drug brought from TCI chemicals Pvt. Ltd., Chennai. The media and FBS are purchased from HiMedia Laboratories Pvt. Ldt. Mumbai.

      Materials required for cell culture

      1. DMEM

      DMEM contains all the ingredients that help in normal cell growth in vitro. The composition of DMEM is given in table 1.


      Table 1: Composition of DMEM

      Ingredients mg/L
      Calcium chloride dehydrate 265.00
      Ferric nitratenonahydrate 0.10
      Magnesium sulphate anhydrous 97.72
      Potassium chloride 400.00
      Sodium chloride 6400.00
      Glycine 30.00
      L-cystine dihydrochloride 62.57
      L-glutamine 584.00
      L-histidine hydrochloride monohydrate 42.00
      L-isoleucine 105.00
      L-leucine 105.00
      L-lysine hydrochloride 146.00
      L- arginine hydrochloride 84.00
      L-methionine 30.00
      L-phenylalanine 66.00
      L-serine 42.00
      L-threonine 95.00
      L-tryptophan 16.00
      L-valine 94.00
      L-tyrosine disodium salt 103.79
      Choline chloride 4.00
      Nicotinamide 4.00
      D-Ca-Pantothenate 4.00
      Folic acid 4.00
      Pyridoxal hydrochloride 4.00
      Rivoflavin 0.4
      Thiamine hydrochloride 4.00
      i-Inositol 7.2
      D-Glucose 4500.00
      Phenol red sodium salt 15.90

      2. FBS

      FBS provides growth supplements for the normal growth of the cell.It is associated with paracrine, endocrine and autocrine growth factors.FBS with DMEM act as complete media for the cell culture.

      3. PBS

      PBS contains phosphate buffer which maintain osmotic balance between internal and external environment of the cell and also maintain physiological pH of the cell. The composition of PBS is given in table 2.

      Table 2: Composition of PBS

      Compound g/L
      NaCl 8
      KCl 0.2
      37% HCl --
      NaH2PO4 1.14
      KH2PO4 0.27

      4.Trypsin-EDTA solution

      Trypsin is a proteolytic enzyme. It acts by cutting amino acids specifically lysine, arginine on their C- terminals. Trypsin-EDTA solution is used to detach the cells from each other as well as from the plate.


      Cell Culcuture

      HepG2 cells (stored at -80°C in DMSO) were cultured in DMEM (incomplete media), supplemented with 10% FBS. The cells were seeded in a 100mm plate with 10 ml media. Plate was incubated in humidified atmosphere at 37°C and 5% CO2 for 24 hrs. After 24 hrs media was replaced and kept for another 24 hrs to get 80% of confluency. Cells were transferred to second plate for suitable growth, by splitting them via trypsinization method. In Trypsinisation method media was removed from plate and washed thrice with PBSto remove leftover media, 330 µl of trypsin was added to the plate, swirled and kept in incubator for 3 mins to detach cells from the plate. Trypsin was then removed and cells were transferred to a plate containing new media (FBS+ DMEM). After 48 hrs cells were harvested again by trypsinization and transferred to 96 well plate for final treatment of drug.

      cell culture plate.jpg
        cell culture plate 

        MTT Assay

        Cells were incubated for 24 hours after seeding in 96 well plates. Then 100 µL of VPA solutions with final VPA concentrations 0.5,1, 2, 4 and 8mM were added to wells (n = 4). Four wells each were also treated with DMSO (3.2%, equivalent to 32µl of DMSO content in 4 mM VPA). All these eight were incubated for 24 hours. Another similar set of eight wells were incubated for 48 hours after treatment. 10µl of MTT reagent was added to each well at the end of the incubation period followed by 5µl of solubilizing agent was added to monitor the formation of purple formazen dye (Figure 4) according to reaction shown in Figure 5.

          96-well plate MTT assays

          Absorption was measured at 570 nm. Graph was plotted as % of viability normalized to control against concentration of drug.

          mtt assay reaction_1.jpg
            Formazan formation

            Extraction of metabolite

            Cells harvested after 24 hrs were scraped off using a scraper and collected cell suspension was set to centrifuge at 4500 rcf at 4°C , the pellet obtained on removal of supernatant was washed with 150mM NaCl. 200µl of this solution was again centrifuged under same condition as before. Then to the pellet 200µl of extracting solvent (water/acetonitrile/isopropanol = 2:3:3 containing 10µM homovanillic acid as internal standard) was added. Cells were lysed to release metabolites by repeated freeze-thaw cycles using liquid nitrogen. Lysate was centrifuged at 14000 rcf for 25 mins at 4°C and the supernatant containing cellular metabolites were collected carefully. 150µL of this supernatant was taken in glass vial and evaporated to dryness under vacuum, then derivatized by adding 50µl of MSTFA and finally heated at 65°C for 60 mins. The prepared sample was loaded on the autosampler of for GC-MS.

              GC-MS Instrument

              GC-MS analysis

              Samples were analyzed using an Agilent 30m HP-5MS column with 0.25ᶙm film thickness and 0.25mm inner diameter. Column was conditioned by injecting pooled samples 5 times before actual sample injection. Samples were injected in a randomized fashion and pooled samples were injected intermittently for quality control. The GC oven temperature were held at 70°C for 5 minutes and increased linearly to 280°C at 5°C/min followed by a final clean-up at 295°C. The column was equilibrated at 70°C for 3 minutes before each injection. The mass spectra were acquired using a single-quad EI-MS. The spectrum was acquired by scanning the mass in the range of 40-600Da.

              Statistical analysis

              We calculated mean, standard deviation (±), standard error of the mean (SEM) by using Microsoft Excel. The statistical changes between different conditions were analyzed by two tailed student t-test and p<0.05 was considered as significant. The p<0.05, 0.01 and 0.001 are indicated by *, ** and ***, respectively.


              Cell morphology

              We treated VPA drug concentration of 4000µM in 6-well plates for 24 hours. After that we observed under microscope to visualize the morphological changes in cell. As can be seen in Figure 7, there was no notable change in cell morphology upon valproate treatment.

              vpa cell_1.jpg
                (A) Control without any DMSO, (B) Vehcle control with same amount of DMSO as in drug, (C) and (D) Drug-treated cells.

                MTT assay

                After 24 hours

                The viability of HepG2 cells was measured by MTT assay and plotted as normalized viability with respect to DMSO-treated control. It was observed that after 24 hours of treatment (Figure 8), cells treated with 0.5(p <0.04) or 1(p <0.02) mM VPA showed more viability compared to DMSO-treated cells whereas cell treated with 8 mM (p <0.05) VPA showed less viability compared to DMSO-treated control.

                vpa 24 hr.jpg
                  Effect of VPA on the viability of HepG2 cell on 24hours (p<0.05 and 0.01 and 0.001 are indicated by * and **respectively.)

                  After 48 hours

                  Similar results were also obtained after 48 hours of treatment showing higher viability of 0.5 (p <0.002), 1 (p <0.01) and 2 (p <0.02) mM VPA-treated cells compared to the DMSO-treated control (Figure 9). 4mM VPA-treated showed similar viability while 8 mM VPA treated showed a significant (p <0.0005) decrease in cell viability.

                  vpa 48 hour_1.jpg
                    EffectofVPAontheviabilityofHepG2cellon48hours.(p<0.05, 0.01 and 0.001 are indicated by *, ** and *** respectively.)

                    This apparently anomalous result might be a result of the fact that 0.5, 1 and 2 mM VPA treatment exposed cells to less DMSO compared to DMSO-treated control and DMSO is known to have cytotoxic effects. This also raises the question whether the lower viability seen in 8mM VPA-treated cells is actually due to VPA or higher DMSO content. Essentially this calls for a more careful experiment with exact DMSO control for each concentration as well as an untreated control to account for the effect of DMSO. This was adhered to in subsequent experiments for cell counting and metabolite extraction.

                    GCMS Analysis

                    The cellular metabolome was analyzed by GCMS. Approximately 250 features (Figure 10) representing metabolites including organic acids related to central carbon metabolism, amino acids, sugars, fatty acids and metabolites involved in nucleic acid metabolism was observed. A comparison of total ion chormatograms indicate that the cellular metabolome of untreated and DMSO-treated HepG2 cells were similar whereas that of the VPA treated cells were different. A detailed analysis metabolomic signature and pathways is currently underway to identify changes in metabolome associated with VPA treatment and their potential implication in hepatocellular injury.

                    VPA GC-MS.jpg
                      The total ion chromatograms for GCMS-based metabolic profiling of (A) untreated, (B) DMSO-treated and (C) 4mM VPA-treated HepG2 cells


                      The effect of different concentrations of VPA on HepG2 cells was examined and results do not conclusively indicate any significant cytotoxicity after 48 hours of treatment with concentrations up to 4mM of sodium valproate. However, GC-MS analysis indicates changes in endogeneous metabolome that might be linked to log-term hepatocellular injury. This warrants further detailed and careful investigation.


                      I would like to acknowledge all those people who have made this research work possible, as experience that I will cherish forever.

                      First of all I owe my sincere thanks and the deepest sense of gratitude to my guide Dr. Soumen Kanti Manna, who has been the source of inspiration and support throughout the period of project, for his guidance and constant encouragement. I have been extremely fortunate to have worked under his supervision.

                      I am indebted to the Indian Academy of Sciences, Bangaore for giving me this opportunity to work at Saha Institute of N uclear Physics, Kolkata, one of the top institute for scientific research and higher education in India.

                      I would like to extend my heartest thanks to Mr. SK Ramiz Islam ( PhD scholar), Mr. Saran Chattopadhyaya (PhD scholar) and Mr. Raju Dutta (technical staff) without whom it would have been difficult for me to complete the work within the stipulated time, for their inspirational guidance, timely help, valuable suggestions and discussions throughout my project.

                      I would like to express my heartful thanks and gratitude to my friend Nishi who made these two months highly informative and enjoyable with valuable caring and support.

                      Its my previlege to express my gratitude to AuthorCafe which offers an amazing platform to expose my research experience to the public.

                      Most impotantly, I thank my family who gave always supported and gave me the life which I wanted to pursue.


                      1. Chateauvieux S, Morceau F, Dicato M, Diederich M. Molecular and therapeutic potential and toxicity of valproic acid. J Biomed Biotechnol. 2010 doi: 10.1155/2010/479364. Epub ahead of print. [PMC free article] [PubMed] [CrossRef] [Google Scholar]

                      2. G. Erenberg, A. D. Rothner, C. E. Henry, and R. P. Cruse, “Valproic acid in the treatment of intractable absence seizures in children. A single-blind clinical and quantitative EEG study,” American Journal of Diseases of Children, vol. 136, no. 6, pp. 526–529, 1982. View at Google Scholar · View at Scopus

                      3. G. Coppola, G. Auricchio, R. Federico, M. Carotenuto, and A. Pascotto, “Lamotrigine versus valproic acid as first-line monotherapy in newly diagnosed typical absence seizures: an open-label, randomized, parallel-group study,” Epilepsia, vol. 45, no. 9, pp. 1049–1053, 2004. View at Publisher · View at Google Scholar · View at PubMed · View at Scopus

                      4. J. C. Dean and J. K. Penry, “Valproate monotherapy in 30 patients with partial seizures,” Epilepsia, vol. 29, no. 2, pp. 140–144, 1988. View at Google Scholar · View at Scopus

                      5. Owens MJ, Nemeroff CB (2003). "Pharmacology of valproate". Psychopharmacol Bull. 37 Suppl 2: 17–24. PMID 14624230

                      6. Martin F, Ufodiama C, Watt I, Bland M, Brackenbury WJ. Therapeutic Value of Voltage-Gated Sodium Channel Inhibitors in Breast, Colorectal, and Prostate Cancer: A Systematic Review. Front Pharmacol. 2015;6:273. Published 2015 Nov 12. doi:10.3389/fphar.2015.00273

                      7. Ghodke-Puranik Y, Thorn CF, Lamba JK, Leeder JS, Song W, Birnbaum AK, Altman RB, Klein TE (April 2013). "Valproic acid pathway: pharmacokinetics and pharmacodynamics". Pharmacogenet. Genomics. 23 (4): 236–241. doi:10.1097/FPC.0b013e32835ea0b2PMC 3696515PMID 23407051.

                      8. ​"Valproic acid". DrugBank. University of Alberta. 29 July 2017. Archived from the original on 31 July 2017. Retrieved 30 July 2017

                      9. C. A. Bradbury, F. L. Khanim, and F. L. Khanim, “Histone deacetylases in acute myeloid leukaemia show a distinctive pattern of expression that changes selectively in response to deacetylase inhibitors,” Leukemia, vol. 19, no. 10, pp. 1751–1759, 2005. View at Publisher · View at Google Scholar · View at PubMed · View at Scopus

                      10. K. P. Nightingale, S. Gendreizig, D. A. White, C. Bradbury, F. Hollfelder, and B. M. Turner, “Cross-talk between histone modifications in response to histone deacetylase inhibitors: MLL4 links histone H3 acetylation and histone H3K4 methylation,” Journal of Biological Chemistry, vol. 282, no. 7, pp. 4408–4416, 2007. View at Publisher · View at Google Scholar · View at PubMed · View at Scopus

                      11. K. N. Harikrishnan, T. C. Karagiannis, M. Z. Chow, and A. El-Osta, “Effect of valproic acid on radiation-induced DNA damage in euchromatic and heterochromatic compartments,” Cell Cycle, vol. 7, no. 4, pp. 468–476, 2008. View at Google Scholar · View at Scopus

                      12. S. Lunke and A. El-Osta, “The emerging role of epigenetic modifications and chromatin remodeling in spinal muscular atrophy,” Journal of Neurochemistry, vol. 109, no. 6, pp. 1557–1569, 2009. View at Publisher · View at Google Scholar · View at PubMed · View at Scopus

                      13.  "DEPAKOTE- divalproex sodium tablet, delayed release"Archived from the original on 5 March 2016. Retrieved 10 November 2015

                      14. Mittal V, Muralee S, Tampi RR. Valproic Acid-induced hyperammonemia in the elderly: a review of the literature. Case Rep Med. 2009;2009:802121. doi:10.1155/2009/802121

                      15. LeCluyse EL. Human hepatocyte culture systems for the in vitro evaluation of cytochrome P450 expression and regulation. European journal of pharmaceutical sciences : official journal of the European Federation for Pharmaceutical Sciences. 2001;13:343–368.

                      16. Sivaraman A, Leach JK, Townsend S, Iida T, Hogan BJ, Stolz DB, Fry R, Samson LD, Tannenbaum SR, Griffith LG. A microscale in vitro physiological model of the liver: predictive screens for drug metabolism and enzyme induction. Current drug metabolism. 2005;6:569–591.

                      17. DelRaso NJ. In vitro methodologies for enhanced toxicity testing. Toxicol Lett. 1993;68:91–99.

                      18. Soldatow VY, Lecluyse EL, Griffith LG, Rusyn I. In vitro models for liver toxicity testing. Toxicol Res (Camb). 2013;2(1):23–39. doi:10.1039/C2TX20051A

                      19. "Valpro sodium valproate"TGA eBusiness Services. Alphapharm Pty Limited. 16 December 2013. Retrieved 14 February 2014

                      20. Sadeque AJ, Fisher MB, Korzekwa KR, Gonzalez FJ, Rettie AE. Human CYP2C9 and CYP2A6 mediate formation of the hepatotoxin 4-ene-valproic acid. J Pharmacol Exp Ther. 1997;283:698–703.[PubMed] [Google Scholar]


                      Written, reviewed, revised, proofed and published with