In Vivo Lithium Treatment Effects on Gene Expression Levels in Lymphoblastoid Cell Lines From Human Healthy Subjects

Background

One of the major limitations of psychiatric research is the necessity to use indirect methods to study neuronal functions. Among these, studies of transformed lymphocytes are an invaluable tool. Epstein Barr virus (EBV) transformed lymphoblasts can be used for a variety of purposes including in vitro studies of gene expression and investigations of cellular responses to pharmacological treatment. They can be kept and re-grown as needed and can often serve as a backup supply for DNA for genetic analyses. For these reasons, lymphoblasts cell lines (LCLs) have been collected and used by a number of research groups, including our own.

Most investigators work with the implicit assumption that such transformed cells represent trait characteristics of the person and their illness. In other words, the cell transformation and repeated passages are assumed to reduce the impact of any confounding factors present at the time of blood sampling and lymphocyte isolation. These factors may include, among others, the clinical state, the actual treatment, or time of the day. Surprisingly, we could not find any published data supporting or contradicting this assumption.

Here we propose to investigate the effects of one particular factor that may influence various cellular measures, namely in vivo drug treatment. Specifically, we are interested in assessing the effect of treatment with lithium, an ion that possesses mood-stabilizing properties. Indeed, lithium is the most effective treatment in bipolar disorder, with approximately 30% of patients achieving complete illness remission and prevention of episode recurrence (Baldessarini and Tondo, 2000; Garnham et al., 2007).

Rationale

Lithium has well established regulatory effects on the expression of specific target genes. Gene expression studies on LCLs from patients characterized for response to lithium treatment have identified a series of molecular target directly modulated by this drug. A microarray study (Sun et al., 2004) on LCLs of patients with BD characterized for full response to lithium demonstrated its effect in decreasing the expression of seven genes: somatostatin receptor type 2 (SSTR2), nuclear factor kappa-B DNA binding subunit (NF-kB), alpha1B-adrenoceptor (1B-AR), acetylcholine receptor protein alpha chain precursor (ACHR), cAMP-dependent 3', 5'-cyclic phosphodiesterase 4D (PDE4D), substance-P receptor (SPR), and ras-related protein (RAB7), the latter five being validated by Northern blotting analysis. Recently, using LCLs from three healthy subjects, Sugawara and coworkers (2010) identified 44 genes whose expression was regulated by lithium. Among the ten genes most down-regulated by lithium were Bax, zuotin related factor 1 (ZRF1) and thioredoxin domain containing 13 (TXNDC13), while platelet-activating factor acetylhydrolase, isoform Ib, beta subunit 30 kDa (PAFAH1B2), Synovial sarcoma translocation, chromosome 18 (SS18) and peroxisome biogenesis factor 1 (PEX1) were the most up-regulated. Moreover, Washizuka et al. (2009) showed that valproate, but not lithium, significantly increased the expression of the gene encoding for a subunit of mitochondrial complex I (NDUFV2) in LCLs from Japanese BD patients. In regard to other psychiatric phenotypes, unpublished data from LCLs of BD patients characterized for different risk of suicidal behaviour are showing that in vitro lithium treatment significantly perturbed the expression of the gene coding for the rate limiting enzyme spermidine/spermine N(1)-acethyltransferase (SAT1) (Squassina et al., in preparation).

Besides gene expression studies, LCLs have also been used to investigate the effect of lithium on protein levels in BD subjects. The study from Tseng et al. (2008) revealed that basal BDNF protein levels are decreased in LCLs from lithium responsive BD patients when compared with both their unaffected relatives and with healthy control participants. Interestingly, in vitro treatment with lithium of the LCLs decreased BDNF levels in all participants, but the difference between BD patients and healthy controls remained.

In addition, the pleiotropic effect of lithium on gene and protein expression has been deeply investigated in a series studies using animal (Bosetti et al., 2002; McQuillin et al., 2007; Chetcuti et al., 2008; Chen et al., 1999) and human (Sun et al., 2007; Seelan et al., 2008) cell tissues.

Specifically: 1) Lithium has been shown to increase the expression of the anti-apoptotic gene BCL2 with reduction of the expression of the pro-apoptotic genes p53 and Bax (Chen et al., 1999) clearly indicating a role in influencing the molecular cascade regulating the programmed cell death. This evidence acquires particularly interest in light of recent findings on BCL2. Two recent studies (Machado-Vieira et al., 2011, Uemura et al., 2011) demonstrated that, in individuals with BD, BCL2 gene expression regulated by the single nucleotide polymorphism (SNP) rs956572 directly impacted intracellular Ca2+ homeostasis dysregulation, a molecular signalling pathway proved to play a significant role in the pathogenesis of BD. 2) Using a genome wide gene expression approach (GWGE) on multiple prostate human cancer cell lines that were incubated with lithium, Sun and coworkers 11 showed a marked downregulation of genes involved in DNA replication. In another study, Seelan et al. (2008), in the attempt of profiling the lithium-modulated gene expression in human neuronal cells with microarray, identified peroxiredoxin 2 (PRDX2), an antioxidant enzyme, as the most upregulated gene, and tribbles homolog 3 (TRB3), a pro apoptotic protein, as the most downregulated, further suggesting a role of these pathways in the mood stabilization process.

In summary, lithium has been proven to significantly modulate the magnitude of the expression of a number of genes, among which the most robust and replicated changes were for BCL2, BAX, p53, and SAT1. Thus, we can take advantage of the well-established property of lithium of regulating the expression of these genes, and proteins, in a significant and stable manner. Conversely, these genes could be used as markers of the effect on lithium on the gene expression, providing a measure for the investigation of the effectiveness of EBV immortalization and repeated passages in eliminating the in vivo treatment effects.

Trial Objectives

This proposal aims to validate the assumption that LCLs, via EBV immortalization and repeated passages, are not influenced by environmental conditions, and especially drug treatment, at the time of sampling.

To do so, fresh lymphocytes and LCLs will be sampled in 20 healthy volunteers before (T0) and after (T1) four weeks of lithium treatment at a stable dose.

First, a set of molecular studies in fresh lymphocytes will examine the expression levels in target genes (namely BCL2, BAX, p53, SAT1) and protein (BDNF), and the activity of Complex I (all biological measures already known to be up-/down- regulated by lithium) at T0 and T1. These biological measures will serve as an assay sensitivity. We expect them to be regulated by in vivo lithium treatment and consequently to be significantly different between T0 and T1 in fresh lymphocytes.

Only the biological measures showing significant difference in fresh lymphocytes (not transformed with EBV) will be then analyzed in LCLs. Differential expression between LCLs sampled at T0 and T1 will indicate that the EBV transformation and repeated passages do not eliminate the environmental influences and, specifically, the effect of lithium treatment at sampling.

Finally, by studying healthy volunteers, we expect to decrease the confounding factors given by the presence of illness status.