Denna sida på svenska

Our research aims at understanding the basis for differences between individuals in human drug metabolism, toxicity and drug response. The emphasis is on genetic polymorphism of the genes encoding drug transporters, drug metabolizing enzymes and drug targets.

The research performed in our laboratory examines the genetic basis for interindividual differences in drug response and adverse drug reactions (ADRs).

We start with patient outliers and examine mechanisms and genetic factors explaining the differences. Genetic biomarkers are identified and used in the clinics.

Among our previous discoveries is the stable gene duplication/amplification in the human genome, the identification of ultrarapid metabolizers of drugs and the identification of several different gene variants of importance for the efficacy of drug treatment.

In addition, we are characterizing the human hepatic epigenome responsible for interindividual differences in drug metabolism and transport, developing a new approach for treatment of colon cancer and utilizing novel mouse models and human genomics in order to elucidate new bases for treatment of depression.

We are also developing novel in vitro methods based on hollow fiber bioreactors and spheroids that can predict drug toxicity, thus finding new ways of producing hepatocytes and non parenchymal liver cells from stem cells in collaboration with Cellectis (Cellartis).

Group members

Tommy B AnderssonAdjunct professor
Isabel BarraganAssistant professor
Catherine BellPostdoc
Isa CavacoPostdoc
Linsey De GrootStudent
Laura EspinoPostdoc
Lisa FredrikssonPostdoc
Pedro GilSenior researcher
Delilah HendriksPhD student
Magnus Ingelman-SundbergProfessor
Maxim IvanovPostdoc
Guo JiaPhD student
Inger JohanssonSenior lecturer
Marin JukicPostdoc
Volker LauschkePostdoctoral researcher, Marie Curie
Tony LlorensStudent
Rosario LopezPostdoc
Maria Isabel Mendes VeigaAssociated
Frederic MichonAssociated
Souren MkrtchianSenior researcher
Åsa NordlingBiomedical scientist
Fatemah RezayeeStudent
Ina Schuppe KoistinenAssociated
Sarah SimSenior researcher
Michelle (Ming) TangPostdoc
Sabine VorrinkPostdoc
Tang XiaojingPhD student


1. Genomic and epigenomic regulation of human hepatocyte formation and function. Implications for drug development and hepatotoxicity

Summary. Studies of liver function, drug and viral induced hepatotoxicity have previously been impossible in in vitro systems due to the rapid de-differentiation of human hepatocytes in cultures. We have worked out conditions for formation of hepatic spheroids in normal and diseased states which are phenotypically stable for several weeks in vitro. Furthermore we have developed novel methods for analyses of the hepatic genome allowing single nucleotide analyses of 5-mC as well as 5-hmC. In this project 220 different livers and spheroid systems obtained from human liver donations or cryopreserved hepatocyte cultures, are used in combination with different types of non-parenchymal cells, in order to

  1. Evaluate the mechanistic basis for de-differentiation of hepatocytes to progenitor cells in 2D cultures,
  2. Study and improve the differentiation process of stem cells to fully differentiated hepatocytes,
  3. Evaluate short term and long term (epigenomic imprinting) epigenomic regulation of hepatic gene expression following de-differentiation or drug treatment taking into account both 5-mC and 5-hmC using novel protocols for genomic analyses,
  4. Evaluate the roles of microRNAs (miRNAs) and long non-coding RNAs (lncRNAs) in the processes of liver de- and re-differentiation and hepatic gene expression,
  5. Develop novel in vitro systems for prediction of drug hepatotoxicity including the identification of novel biomarkers, investigate epigenomic regulation in diseased livers and in diseased liver spheroid models.

This information will be of importance i) for increased possibilities to avoid the development of hepatotoxic drugs,  ii) to develop novel protocols for formation of hepatocytes from stem cells, iii) for understanding the bases for interindividual differences in drug metabolism and drug hepatotoxicity and iv) for understanding the genome alterations in different liver diseases.

The spheroid model

In our lab we have established and further developed novel spheroid based 3D systems for studies of liver function and properties in vitro (cf. Gunness et al., 2013). Starting with hepatocytes obtained from the cell transplantation unit at Huddinge University Hospital or with cryopreserved hepatocytes we form spheroids from primary human hepatocyte fractions (PHH spheroids) using the InSphero gravity plus system or the Corning ultra-low attachment plates. In the spheroids, the cells have relevant cell-cell interactions, a cell-extracellular matrix (ECM) contact, improved cell polarity and form functional bile canaliculi (see Figure 2). Our work has shown that the spheroids i) express albumin, urea etc  and the major forms of hepatic cytochrome P450s for at least 54 days, ii) retain to a great extent their phenotype from the liver resulting in mosaic formation (perivenous vs periportal)  in the spheroids, iii) can be infected with virus, iv) can be formed from different proportions of hepatocytes and different types of non-parenchymal cells. 

De-differentiation of human hepatocytes and re-differentiation in hepatocyte spheroid formation

In collaboration with Takara Bio Europe AB in Gothenburg and recently with EU EBiSC Stem Bank we study mechanisms of formation of hepatocytes form stem cells (Sivertsson et al., 2013) as well as mechanisms of de-differentiation of hepatocytes with the aim to understand critical factors needed for successful differentiation of stem cells into hepatocytes. 

To understand the dedifferentiation and the phenotypic difference of the hepatocytes in 3D spheroids as compared to 2D hepatocyte cultures, we follow >285,000 full-length transcripts, 245,000 coding transcripts, 40,000 non-coding transcripts, 1200 miRNAs and 12000 lncRNAs during de-differentiation and re-differentiation of human hepatocytes from three different liver donors.

Novel methods for analyses of the dynamics of the hepatic epigenome.

Using the sensitive and reliable LC-MS method for the detection of cytosine modifications, we recently discovered in collaboration with the Estonian Genome Project that genomic DNA from adult human livers contain a substantial level of 5hmC (Ivanov et al., 2013). In this study we also enriched genomic DNA samples from 8 fetal and 7 adult livers for the 5hmC-containing fraction, subjected it to next-generation sequencing (NGS) and generated the first genome-wide map of 5hmC peaks in the hepatic epigenome. We have now used a TAB-Methyl-SEQ protocol for the epigenomic profiling of ADME genes in 20 adult human livers. We can then find that the bisulfite methods shadows regions where 5hmC is heavily enriched as illustrated in Figure 4.

The new powerful tools for determining the 5-hmC and 5-mC distribution allow site-specific and inducible manipulation of methylation and hydroxymethylation patterns at candidate genomic loci and result in the functional validation of epigenetic polymorphisms. Issues that are in addition considered are the regulation of epigenomes by epidrugs, drug inducers and during hepatocyte differentiation.

Furthermore the epigenomic alterations seen during steatosis, hepatitis (drug or virus induced) as well as fibrosis are studied both in the liver specimens characterized for these conditions by a pathologist as well as in the diseased spheroid models where such conditions have been induced.


The project encompasses completely novel techniques for studying epigenome regulation taking both 5mC and 5hmC modifications into account which is important in view of our recent finding (Ivanov et al., 2013) that the activatory modification 5hmC accounts for about 30-40 % of all cytosine modifications in active genes in adult liver and based on our preliminary data here shown which shows on a single cytosine resolution the erroneous results obtained from sole bisulfite analyses missing elements where a high amount of 5hmC modification has occurred.

The newly developed liver in vitro systems allow long term studies of liver functions in vitro for 4-5 weeks at a high and stable level of differentiated hepatocyte function. This in vitro model also allows studies of diseased liver in vitro like drug- or virus induced hepatitis, steatosis, fibrosis etc. Furthermore, the phenotypic preservation of the hepatocyte phenotype in the spheroids and their de-differentiation and re-differentiation during spheroid formation, allows an investigation of the signal transduction systems and mechanisms of importance for developing new protocols for stem cell derived hepatocyte formation.

In conclusion, the projects allow to utilize novel technologies for understanding the basis behind hepatic gene regulation and hepatocyte differentiation, things that are of high importance for optimizing drug therapy and for development of primary cell and stem cell derived systems that can facilitate drug development in the future. 

2. CYP2W1, a novel target for colon cancer therapy

We have found and characterized novel specific forms of P450 in cancer cells and in extrahepatic tissues. One interesting enzyme is CYP2W1, which is expressed in fetal colon but not in adult tissue, except for colon tumors. We have found in collaboration with David Edler found that the CYP2W1 expression can serve as a prognostic marker for malignancy.

In collaboration with Laurence Pattersson in Bradford, we have also been able to identify chemicals that can be bioactivated by CYP2W1 in the cancer cells to effectively kill the cells at 300 nM concentration. We are making xenograft models for treatment of colon cancers in vivo using these substances and found very promising results with respect to killing of colon cancer tumors in vivo with these agents. We will in due time develop chemical structures that can possibly be used in vivo in human for treatment of colon cancer.

The CYP2W1 gene is evolutionary conserved; it is expressed in embryonic gastrointestinal tract, and remained for a short period in the neonatal tissues, followed by epigenetic silencing throughout the adult life. The Cyp2w1 null mouse model has been established in our lab which preliminary shows a phenotype and  we are currently studying its endogenous function in embryogenesis, particularly in the gut development.

3. Endogenous function of CYP2C19: Role in brain and liver phenotypes

The CYP2C19 enzyme metabolizes endogenous and synthetic psychoactive compounds including steroid hormones, cannabinoids, SSRIs, tricyclic antidepressants, and benzodiazepines. Two major variant alleles are known, CYP2C19*2 which is defective and CYP2C19*17, identified by our group, which causes increased enzyme expression.

We detected CYP2C19 expression in fetal human brain and in collaboration with Nancy Pedersen, we showed that the presence of two defective (CYP2C19*2) alleles causes a decrease in depressive symptoms. We found that transgenic mice containing human CYP2C19 gene also show expression of this gene in developing brain, as well as decreased hippocampal volume, anxious phenotype, and increased hippocampal activation after acute stress. Currently, we utilize this model to study the role of CYP2C19 in developing brain and to identify molecular mechanisms of its anxiety- and depression-like phenotype. In addition, we are conducting studies with the aim to translate present and prospective findings from this model to psychiatric patients.

Participation i EU-projects

Notox. Leader of work package 1.

Scr&Tox. Participant in work package 1 and work package 2.

MIP-DILI. Leader of work package 2.

SafeSciMET. Responsible for Course 6.3 on pharmacogenomics.

Financial support

Selected publications

Stem cell-derived systems in toxicology assessment.
Suter-Dick L, Alves P, Blaauboer B, Bremm K, Brito C, Coecke S, et al
Stem Cells Dev. 2015 Jun;24(11):1284-96

Polymorphic expression of CYP2C19 and CYP2D6 in the developing and adult human brain causing variability in cognition, risk for depression and suicide: the search for the endogenous substrates.
Ingelman-Sundberg M, Persson A, Jukic M
Pharmacogenomics 2014 ;15(15):1841-4

Genetic and epigenetic regulation of gene expression in fetal and adult human livers.
Bonder M, Kasela S, Kals M, Tamm R, Lokk K, Barragan I, et al
BMC Genomics 2014 ;15():860

Epigenetic mechanisms of importance for drug treatment.
Ivanov M, Barragan I, Ingelman-Sundberg M
Trends Pharmacol. Sci. 2014 Aug;35(8):384-96

Whole-exome sequencing reveals defective CYP3A4 variants predictive of paclitaxel dose-limiting neuropathy.
Apellániz-Ruiz M, Lee M, Sánchez-Barroso L, Gutiérrez-Gutiérrez G, Calvo I, García-Estévez L, et al
Clin. Cancer Res. 2015 Jan;21(2):322-8

Personalized medicine into the next generation.
Ingelman-Sundberg M.
J Intern Med. 2015 Feb;277(2):152-4.

High frequency and founder effect of the CYP3A4*20 loss-of-function allele in the Spanish population classifies CYP3A4 as a polymorphic enzyme.
Apellániz-Ruiz M, Inglada-Pérez L, Naranjo M, Sánchez L, Mancikova V, Currás-Freixes M, et al
Pharmacogenomics J. 2015 Jun;15(3):288-92

 In-solution hybrid capture of bisulfite-converted DNA for targeted bisulfite sequencing of 174 ADME genes.
Ivanov M, Kals M, Kacevska M, Metspalu A, Ingelman-Sundberg M, Milani L
Nucleic Acids Res. 2013 Apr;41(6):e72

 Colon cancer-specific cytochrome P450 2W1 converts duocarmycin analogues into potent tumor cytotoxins.
Travica S, Pors K, Loadman P, Shnyder S, Johansson I, Alandas M, et al
Clin. Cancer Res. 2013 Jun;19(11):2952-61

 High warfarin sensitivity in carriers of CYP2C9*35 is determined by the impaired interaction with P450 oxidoreductase.
Lee M, Borgiani P, Johansson I, Oteri F, Mkrtchian S, Falconi M, et al
Pharmacogenomics J. 2014 Aug;14(4):343-9

 Ontogeny, distribution and potential roles of 5-hydroxymethylcytosine in human liver function.
Ivanov M, Kals M, Kacevska M, Barragan I, Kasuga K, Rane A, et al
Genome Biol. 2013 ;14(8):R83

 Decreased hippocampal volume and increased anxiety in a transgenic mouse model expressing the human CYP2C19 gene.
Persson A, Sim S, Virding S, Onishchenko N, Schulte G, Ingelman-Sundberg M
Mol. Psychiatry 2014 Jun;19(6):733-41

 Amidoxime reductase system containing cytochrome b5 type B (CYB5B) and MOSC2 is of importance for lipid synthesis in adipocyte mitochondria.
Neve E, Nordling A, Andersson T, Hellman U, Diczfalusy U, Johansson I, et al
J. Biol. Chem. 2012 Feb;287(9):6307-17 

Vacant positions

Applications for postdoctoral fellows are welcome. If interested, please send a cover letter and an updated CV to Magnus Ingelman-Sundberg, research group leader.

Contact us


Magnus Ingelman-Sundberg

Phone: 08-524 877 35
Organizational unit: Ingelman-Sundberg Magnus group - Pharmacogenetics

A group within the section: 

Pharmacoepigenetics in Health and Disease

Our ultimate goal is to understand the pharmacological response and mechanisms of toxicity. Gaining knowledge about novel biomarkers that can be implemented in treatment management and safety assessment is essential. We are currently characterizing the epigenetic contribution to differential gene regulation relevant for drug action and metabolism. 

Unmasking the liver ADME genes associated with epigenetic variation: contribution of epigenetics to the modulation of their expression and activity.

There is an important interindividual variability in drug response that is not explained by genetic or environmental factors and contribute to unpredicted side effects and insufficient efficacy. Today we know that epigenetics takes part in the regulation of liver absorption, distribution, metabolism, and excretion (ADME) genes by DNA methylation and histone deacetylation. These are some of the most common epigenetic elements associated with gene regulation of more than 75 genes important for drug metabolism (Trends Pharmacol Sci, 2014. 35(8): p. 384-96) In addition, these same epigenetic features are related to changes in the genome function under exogenous influence. Epi-drugs, including inhibitors of DNA methylation (e.g. Azacytidine, Vidaza®), and histone deacetylation (e.g., Vorinostat, Zolinza®) are very well known modulators of the gene expression. On top of that, studies in mice have reported that other drugs such as phenobarbital, a prototype nuclear receptor-mediated inducer of ADME genes, is also associated with perturbation of the epigenome, including changes in the recently “re-discovered” cytosine modification 5-hydroxymethylcytosine (5-hmC). Interestingly, we have identified 5-hmC as a new epigenetic marker abundant in the liver and likely to be functionally opposite to methylation (Genome Biol, 2013. 14(8): p. R83)

We are employing a pharmacological unmasking approach to identify in which genes epigenetics is a regulatory component; We are using a bona fide liver organotypic system based on primary cells from human adult healthy donors. Our unique strength to this fundament is the state-of-the-art 3D culturing systems in addition to several in-house developed organ tailored epigenetic profiling such as hydroxymethylation mapping. 

The liver microtissues’ added value to the gold standard 2D culturing is that they cope better with the de-differentiation and the subsequent progressive loss of the hepatocyte-specific phenotype.

Capacity for phase I, phase II drug metabolism, as well as drug transport is retained. Bile production has also been detected. The microtissues are stable and functional for more than 4 weeks in culture, with higher drug sensitivity than 2D cultures and reflection of immune-mediated toxicity.

Picture: Immunohistochemical detection of the hepatic marker CK8 in Kupffer macrophagues, endothelial cells, bile canaliculi and hepatic transporters, respectively, found in human liver spheroids (InSphero AG). 

From left to right: CK8, CD68, CD31, MDR1. 

Among our methods to identify which genes are affected by the epi-drugs and the ADME genes inducers we are employing RNA-Seq, western blotting, and LC/MS for metabolite determination as a measurement of the enzymatic activity. In order to investigate the epigenetic modifications associated to the alterations in the ADME and related genes we are using a novel oxidative protocol adaptation for 450K Illumina arrays that allows the mapping in parallel of methylation and hydroxymethylation, and ChIP-seq for histone modifications profiling.

We also account on a panel of 200 human liver samples that has been characterized at the genomewide level for mRNA and genotype, as well as for the activity of the main liver metabolic enzymes.

Whole genome methylation/hydroxymethylation and miRNA analysis of the cohort, together with pathological featuring for fibrosis, cirrhosis, steatosis, cholestasis and inflammation is ongoing.

This panel will serve for the validation of the identified correlations in gene expression versus epigenetic elements and to detect the variability of those epigenetic marks that are altered due to the drug exposure. With this, we expect to identify the missing gap in the lack of drug efficacy and the unpredictable adverse reactions.

Group members

Antoni Llorens Internship student
Laura Espino Postdoctoral researcher
Luis Ferrández Master student
Tang (Michelle) Ming Postdoctoral researcher
Tang Xiaojing PhD Student
Isabel Barragan Assistant Professor

Financial support

  • European Union, Marie Curie
  • Karolinska Institutet
  • Department of Physiology and Pharmacology
  • Varia

Open calls 

Postdoctoral studies in epigenetic regulation of drug metabolism genes and drug response (scholarship) Netrecruiter referensnummer: 2-256/2015

Selected publications


Epigenetic mechanisms of importance for drug treatment.
Ivanov M, Barragan I, Ingelman-Sundberg M
Trends Pharmacol. Sci. 2014 Aug;35(8):384-96

Second most downloaded article. On the Top 5 of “Editors’ Select”

Original articles

Cytostatic effect of repeated exposure to simvastatin, a mechanism for chronic myotoxicity revealed by the use of mesodermal progenitors derived from human pluripotent stem cells. 
Peric D, BARRAGAN I, Giraud-Triboult K, Egesipe AL , Meyniel-Schicklin L , Cousin C, Lotteau V, Petit V , Touhami J, Battini JL , Sitbon M, Pinset C, Ingelman-Sundberg M, Laustriat D, Peschanski M. 
Article under revision, Journal Stem Cells 

Genetic and epigenetic regulation of gene expression in fetal and adult human livers.
Bonder M, Kasela S, Kals M, Tamm R, Lokk K, Barragan I, et al
BMC Genomics 2014 ;15():860

Long-term chronic toxicity testing using human pluripotent stem cell-derived hepatocytes.
Holmgren G, Sjögren A, Barragan I, Sabirsh A, Sartipy P, Synnergren J, et al
Drug Metab. Dispos. 2014 Sep;42(9):1401-6

Ontogeny, distribution and potential roles of 5-hydroxymethylcytosine in human liver function.
Ivanov M, Kals M, Kacevska M, Barragan I, Kasuga K, Rane A, et al
Genome Biol. 2013 ;14(8):R83

CYP2W1 polymorphism: functional aspects and relation to risk for colorectal cancer.
Stenstedt K, Travica S, Guo J, Barragan I, Pors K, Patterson L, et al
Pharmacogenomics 2013 Oct;14(13):1615-22

Strong effects of environmental factors on prevalence and course of major depressive disorder are not moderated by 5-HTTLPR polymorphisms in a large Dutch sample.
Peyrot W, Middeldorp C, Jansen R, Smit J, de Geus E, Hottenga J, et al
J Affect Disord 2013 Mar;146(1):91-9

Mutation screening of multiple genes in Spanish patients with autosomal recessive retinitis pigmentosa by targeted resequencing.
González-del Pozo M, Borrego S, Barragán I, Pieras J, Santoyo J, Matamala N, et al
PLoS ONE 2011 ;6(12):e27894

Copy-number variations in EYS: a significant event in the appearance of arRP.
Pieras J, Barragán I, Borrego S, Audo I, González-Del Pozo M, Bernal S, et al
Invest. Ophthalmol. Vis. Sci. 2011 Jul;52(8):5625-31

Mutation spectrum of EYS in Spanish patients with autosomal recessive retinitis pigmentosa.
Barragán I, Borrego S, Pieras J, González-del Pozo M, Santoyo J, Ayuso C, et al
Hum. Mutat. 2010 Nov;31(11):E1772-800

EYS, encoding an ortholog of Drosophila spacemaker, is mutated in autosomal recessive retinitis pigmentosa.
Abd El-Aziz M, Barragan I, O'Driscoll C, Goodstadt L, Prigmore E, Borrego S, et al
Nat. Genet. 2008 Nov;40(11):1285-7

Published commentaries

EYS, encoding an ortholog of Drosophila spacemaker, is mutated in autosomal recessive retinitis pigmentosa.
Abd El-Aziz M, Barragan I, O'Driscoll C, Goodstadt L, Prigmore E, Borrego S, et al
Nat. Genet. 2008 Nov;40(11):1285-7 

Contact us

Assistant professor

Isabel Barragan

Phone: 08-524 862 39
Organizational unit: Ingelman-Sundberg Magnus group - Pharmacogenetics