, Volume 14, Issue 3, pp 582–587 | Cite as

Genomics of Lithium Action and Response

Open Access


Lithium is the most successful mood stabilizer treatment for bipolar disorder. However, unlike conventional drugs that are designed to interact with a specific molecular target, the actions of lithium are distributed across many biological processes and pathways. Treatment response is subject to genetic variation between individuals and similar genetic variation may dictate susceptibility to side effects. Transcriptomic, genomic, and cell-model research strategies have all been deployed in the search for the genetic factors and biological systems that mediate the interaction between genetics and the therapeutic actions of lithium. In this review, recent findings from genome-wide studies and patient cell lines will be summarized and discussed from a standpoint that genuine progress is being made to define clinically useful mechanisms of this treatment, to place it in the context of bipolar disorder pathology, and to move towards a time when the prescription of lithium is targeted to those individuals who will derive the greatest benefit.


Lithium Bipolar disorder Pharmacogenomics Therapeutic response Toxicity 

Introduction to Lithium, the Mood Stabilizer

The serendipitous discoveries of therapeutic drugs are a fascinating chapter in twentieth-century medicine. The application of lithium as a treatment for mood disorders is one of the key examples. In the late 1940s, John Cade, a psychiatrist working in Melbourne, Australia accidentally discovered lithium’s sedative properties during rodent experiments designed to explore the role of uric acid in psychiatric conditions (although Cade knew about lithium as a historical treatment for uric acid build up in gout) [1]. After a lengthy lag period caused by concerns over its safety, the clinical development of lithium for “psychotic excitement”, as Cade originally termed it, accelerated in the 1960s and 1970s. Fifty years on, lithium is still the “gold standard” treatment for bipolar disorder with cumulative and overwhelming evidence for its therapeutic effectiveness and consequent life-saving outcomes [2, 3].

In an era of rational high-throughput drug screening of complex libraries chosen to represent the full dimensionality of chemical space, it is somewhat bemusing to discover that the blunt tool of a metal ion is a top-performing mood stabilizer. This is a “small molecule” but without a single defined “target”. Lithium ions possess a single positive charge and have a radius of 76 pm [4]. We must assume that these 2 properties alone mediate its pharmacological action, most likely by mimicking and compromising the actions and targets of more physiologically relevant trace metals with similar physical properties such as Cu+ (77 pm), Mg2+ (72 pm), Zn2+ (74 pm), and the “high spin” form of Fe2+ (78 pm).

There is a stock view presented in numerous reviews of therapeutic lithium that we understand little about the biological mechanisms of its action. The sheer scale of lithium’s potential physical interactions perhaps explains why it has been so difficult to define its major biological modes of action—both good and bad—until the advent of more sophisticated and objective methodologies. A second confounding factor has been the relative invisibility of lithium atoms in X-ray crystallography. Consequently, any presumed functional interaction with a protein, such as glycogen synthase kinase 3β (GSK3β), has been very hard to confirm or characterize. The physical association of lithium and magnesium ions with the phosphate chain of energy intermediate molecule, adenosine triphosphate, has only recently been inferred by nuclear magnetic resonance, and demonstrated to alter its purinergic signaling capacity [5]. Likewise, lithium’s pharmacological action within the catalytic site of (bacterial) inositol monophosphatase has only been indirectly observed through magnesium displacement upon increasing lithium concentrations [6]. This review will acknowledge the technical challenges facing the field, but will also offer encouraging evidence for progress from recent studies of the mechanisms of lithium’s therapeutic action, the basis for its side effects, and the genetic determinants of patient response.

Side Effects of Lithium

Given the lack of specificity, it is surprising that lithium’s therapeutic action is so potent and the toxic side effects not more pronounced. It is, however, a close-run race—the therapeutic index of lithium is very narrow, indicating dosages with therapeutic action are close to dosages with toxic side effects. Patients undergoing lithium treatment require regular blood tests to ensure that safe and effective levels are maintained. They also undergo occasional assessment of kidney and parathyroid function: the 2 organs most susceptible to lithium toxicity. This potential for harm provides a strong motivation to deconvolute the biological action of lithium to identify targets that could be pharmacologically manipulated with more specific and less risky drugs and, in addition, to define harmful “off-target” effects.

Multiple side effects of lithium have been described since it was licensed for the treatment of acute mania and then bipolar disorder in the 1970s. These include hypothyroidism, nausea, diarrhea, and weight gain. However, it is the damage to the kidneys correlated with long-term use and excessive serum concentrations that is the most well documented and characterized [7, 8, 9]. This primarily manifests as a common (~50% of patients) increase in thirst resulting from polyuria—excessive urination—and, less commonly, as a progression to nephrogenic diabetes [10]. Both states indicate a failure to concentrate urine in the collecting duct of the nephron. The epithelial cells of the collecting duct permit the entry of lithium through amiloride-sensitive ENaC sodium channels. As an aside, a recent paper described a screen for single-nucleotide polymorphism (SNP) variants in selected candidate genes that predispose those prescribed with lithium to long-term deterioration of estimated glomerular filtration rates [11]. A CC genotype at SNP rs378448, located in ACCN1, was identified as the most significant predisposing factor. Additionally, a SNP in the first exon of this gene has been associated with lithium response in a small Sardinian cohort [12]. ACCN1 encodes a member of the ENaC superfamily, but it is not known if this plays a role in the collecting duct pathology. Once lithium enters the duct cells, there are many and, in some instances, hotly debated routes of pathological action, including increased cyclooxygenase-2 expression generating prostaglandin E2, inhibition of GSK3β/vasopressin action, decreased urea transporter expression, increased cell-cycle activity leading to cellular remodeling, and reduced inositol monophosphate/cyclic adenosine monophosphate signaling [13, 14, 15]. Ultimately, there is reduced induction and expression of the aquaporin-2 channels in the duct cell wall, preventing the passive reuptake of water from the duct and therefore leading to the production of excess, dilute urine. Much less common (~1%) in those prescribed lithium is a further chronic progression to nephrotoxicity and histologically visible damage to the kidney [16].

Adding to the complexity of the dosage balancing act is the variable lithium response exhibited by each patient. Approximately 30% of those prescribed lithium exhibit a good response, with the remainder falling into partial- and nonresponder groupings. This spectrum may be owing to heterogeneous genetic and environmental components—certainly there is good evidence for heritable response to lithium in bipolar disorder family clusters [17, 18]. Other factors that correlate with response have been comprehensively reviewed previously and include age–at–onset and periods of remission [19]. Patient response might also directly relate to genetically determined variation in lithium’s physiological availability, excretion and molecular targets. For example, in the rat, the major determinant of lithium reabsorption back into the bloodstream from the kidney glomerular filtrate was recently identified as a foscarnet- and parathyroid hormone-sensitive sodium–phosphate co-transporter activity. This raises the possibility that human genetic variation at this locus/loci could significantly influence the half-life of circulating lithium in the bloodstream [20].

Archetypal Mediators of Lithium’s Therapeutic Effects

In terms of lithium’s therapeutic action, there are a number of classically studied proteins and pathways that are definitive lithium targets. These have been reviewed extensively elsewhere [21, 22]; only 2 will be briefly summarized here. First is lithium’s known inhibition of the protein GSK3β [23, 24, 25]. This protein is a component of the WNT signaling pathway, inhibition of which allows the accumulation of β-catenin protein and consequent downstream gene regulation. GSK3β also regulates other proteins (e.g., arrestins) and is itself regulated by other pathways (e.g., Akt). Second, an excess of inositol is believed to be a biochemical pathology in bipolar disorder affecting multiple cellular systems, including mitochondrial function, autophagy, growth cone function, and calcium signaling. Lithium depletes this excess signaling by directly inhibiting inositol monophosphatase 1—diverting the metabolic pathway that generates inositol.

Transcriptomic Approaches Struggle to Reveal Lithium Action

In addition to these established modes of action, there are now multidisciplinary approaches to fully characterize lithium’s biology and the factors that influence a patient’s response. One approach is to employ transcriptomic methods to identify genes up- or downregulated in response to lithium treatment. The premise is that these expression changes will reflect the principal target genes and biological pathways that underlie therapeutic action. However, the results have been disappointingly inconsistent. For this review, an informal analysis was carried out on differentially expressed gene sets from 5 representative transcriptomics papers spanning human/rodent studies and in vitro/vivo sample types [26, 27, 28, 29, 30]. This identified only 2 genes, CD93 and SULT1A1, shared between any 2 studies. A more sophisticated analysis of this kind was carried out by Breen et al. [31], to act as validation of their own gene expression study. The authors identified a number of results shared across publications, including STC2, HADH, GAMT, MAT2A, HSP90AA1, CRIP1, CKB, FOS, LAX1, and RSAD2, but, further indicating the difficulties of the gene expression approach, these genes were not identified in the studies assessed in this review. Many of these inconsistencies may lie in the choice of cell type, treatment regimen, and even species.

Induced Pluripotent Stem Cells Offer a Means to Explore Lithium Response in Well-Controlled Human Models

The emergence of induced pluripotent stem (iPS) cell technologies has permitted the detailed investigation of mental illness in biologically relevant human cell types. The acute nature of the brain’s response to lithium has been convincingly modeled in an in vitro study of dentate gyrus-like neurons derived from 6 patients with bipolar disorder and 4 healthy control iPS cell lines [32]. Disease-associated phenotype differences observed in the patient neurons included altered mitochondrial function, abnormal calcium signaling, and, most noteworthy, a general hyperexcitability manifested as increased sodium ion currents and increased frequency of spontaneous action potentials. The authors selected 3 good lithium responders and 3 nonresponders when recruiting the cohort with bipolar disorder. The hyperexcitability phenotype was reversed by lithium treatment in the good responder subset alone: extraordinary evidence that the heterogeneous nature of complex genetic phenotypes can occasionally “collapse” into a simple and discriminating marker. The authors went on to identified 45 genes with altered expression in bipolar patient lines and biological categorization of these confirmed the involvement of mitochondrial processes. Importantly, the number of genes with significant expression changes in response to lithium treatment was an order of magnitude greater in the responding cells than in the nonresponding cells. Among these genes were a number that have been previously implicated in bipolar disorder (PDE11A, PRKCH, PTPRB, SCN11A, NKAIN, KCNA1, and KCNJ12), indicating that a good response to lithium correlates with a partial rescue of pathological gene expression profiles. One of the disease-associated, lithium-responding genes, KCNA1, was also identified in the study by McQuillin et al. [26]. This gene encodes a potassium channel, proposed to mediate a homeostatic response to the cellular hyperexcitability: expression increases in disease, but decreases as lithium reverses this electrophysiological phenotype. Similarly, the disease-associated, lithium-responding gene Pde11a, encoding a phosphodiesterase that catabolizes cyclic adenosine monophosphate and cyclic guanosine monophosphate, was shown to be expressed at a lower levels in the hippocampi of mouse strains that respond to lithium [33]. Pde11a knockout mouse strains also displayed an increased response to lithium compared with wild-type littermates.

The iPS work above has been neatly complemented by a recent publication that used lymphoblastoid cell-derived iPS cells from patients with bipolar disorder and healthy control cohorts of the same size and composition as above [34]. When differentiated into dentate gyrus-like granule cell neurons, those from patients with bipolar disorder also displayed an increased likelihood of the hyperexcitable phenotype. However, a more discerning electrophysiological characterization of the cells revealed that lithium responder and nonresponder lines achieved this hyperexcitability through different perturbations in their underlying sodium and potassium channel activity and action potential parameters, with distinct electrophysiological changes after lithium treatment. This enabled the construction of a lithium response prediction model based on electrophysiology alone.

A third study employing iPS cells generated from responsive and nonresponsive patients focused on the proteomic changes that follow lithium treatment [35]. Of the 15 proteins identified with altered expression or gel migration, collapsin response mediator protein 2 (CRMP2) was chosen as an attractive candidate for further study as it showed increased phosphorylation on threonine position 514 in lithium-responsive patient iPS-derived neurons. This site can be phosphorylated by GSK3β (among other kinases), and both lithium and a specific GSK3β inhibitor were shown to reduce its phosphorylation level. CRMP2 regulates the cellular response to semaphorin 3A (formerly known as “collapsin”), an extracellular signaling molecule that binds to neuropilin/plexin receptors. This pathway shapes dendritic spines and axonal growth cone morphology during development and in the adult—with implications for synapse formation and function. Spine morphologies were shown to be altered in vitro and in postmortem brain samples, suggesting this is a genuine bipolar disorder pathology, and one that can be ameliorated by lithium, most likely through action on the GSK3β/CRMP2/semaphorin 3A pathway. The antidepressant tianeptine, antiepileptic lacosamide, and neurotrophic brain metabolite lanthionine ketimine have all been shown to act on CRMP2 [36].

Together, these studies have created overlapping frameworks of disease pathology and drug action that significantly advance our understanding and provide discerning functional assays for drug development and pharmacogenomic profiling.

The Era of Genome-Wide Studies Into Lithium Response

Objective genome-wide approaches to study complex genetic disorders, such as SNP-based association and high-throughput sequencing, have also been instrumental in the search for novel facets of pathology and pharmacogenomics across the full range of complex genetic disorders. A key step in the assessment of disease-predisposing genetic risk factors in association studies is to compensate for false positives that will inevitably arise when genotyping hundreds of thousands of genetic variants in each individual—patient or healthy control. Rather than use the standard statistically significant p-value threshold of 0.05, the requirement is for a p-value of 10–8 to be met, or surpassed, before a finding can be considered “genome-wide significant”. For bipolar disorder, the emergence of genome-wide significant variants within calcium channel genes perhaps offers the most pharmacologically tractable finding from the multitude of genes discovered and validated [37, 38]. Direct assessment of lithium response factors via genome-wide association study has been reported in 4 studies with sufficient participants to be considered adequately powered (see Table 1 for summary). The phenotypic definition of lithium response at the level of the patient is most often achieved using the ‘Alda’ rating scale based on case notes and prescription history [17]. As with all psychiatric disorders, there is no other clear physical biomarker that can be employed for this purpose. Discounting the new evidence for in vitro electrophysiological changes, there is only one in vivo physical measure that could act as a quantifiable phenotype: the robust association of brain region size increases that correlate with extended lithium use [39]. A correlation with lithium response has not been tested (although perhaps inferable from long-term use) but would be an indication that cell proliferation/cell size are additional pathologies worthy of exploration. The first of the 4 genome-wide studies failed to reach statistical significance at any SNP locus [40]. However, the second study, despite being made up of only 264 patients with bipolar disorder, and then a further 100 patients in a validation study, achieved extremely significant p-values for SNP markers within GADL1 [41]. Depending on the parameters used in the definition of lithium response, the p-values for the most significant SNP, rs17026688, were between 9.19 × 10−18 and 5.50 × 10−37. This extraordinary level of significance, and associated effect size, should be practically predictive for patient lithium response. The gene encodes the glutamate decarboxylase-like protein 1 that, because of its sequence similarity to γ-aminobutryic acid-ergic metabolism enzymes glutamate decarboxylase 1 and 2, appears a plausible candidate. However, GADL1 is seemingly not expressed in the brain and is principally engaged in taurine metabolism [42]. While there is validation of this GWAS finding in a second study in the Han Chinese population [43], other population studies have failed to replicate association [44, 45], which suggests that the gene’s candidacy should be treated with a level of caution until a statistical consensus emerges. In the third study, a number of association comparisons were made across several collected cohorts in which lithium response was either subjectively (self-) reported or objectively assessed [46, 47]. The only case–control permutation which yielded genome-wide statistical significance was between 387 objectively assessed lithium responders and 6684 healthy controls. This pinpointed a SNP, rs146727601 (p-value:1.22 × 10−9), located within the gene, PTS, encoding 6-pyruvoyltetrahydropterin synthase, an enzyme that participates in the synthesis of the cofactor tetrahydrobiopterin (BH4). BH4 aids the production of nitric oxide and monoamine neurotransmitter synthesis. However, a mouse knockout of Pts alters metabolism and abdominal fat distribution without an overt central nervous system phenotype or perturbation of neurotransmitter levels [48]. There are other SNPs in this study that fall marginally below the genome-wide statistical threshold but which have intriguing links to established psychiatric biology, suggesting that larger surveys are required. Finally, a fourth study carried out by the Consortium on Lithium Genetics (ConLiGen) consisted of a genome-wide association study of 2563 individuals diagnosed with bipolar disorder and phenotypically assessed for lithium response using both continuous and categorical assessment tools [49]. Marginally sub-genome-wide significant SNPs were identified using the continuous phenotypic model in the genes CSMD2 [rs9662615 (p = 5·26×10–7); gene previously implicated in multiple neuropsychiatric diagnoses [50]] and HDAC9 [rs61549860 (p = 5·44×10–7); gene disrupted in one reported case of schizophrenia [51]]. However, a cluster of genome-wide significant SNPs was identified on chromosome 21 over a potential long noncoding mRNA associated with an expressed sequence tag with accession number AL157359 (most significant SNP rs74795342; p = 3·31×10–9). The biological significance of this transcript is currently unknown.
Table 1

Summary of the top findings from 3 genome-wide association studies of lithium response

SNP identifier


Implicated gene



Chen et al. [41]




5.50 × 10−37

Discovery cohort

Hou et al. [49]




1·37 × 10–8

European/Asian ancestry cohorts combined




1·31 × 10–8

European/Asian ancestry cohorts combined




3·31 × 10–9

European/Asian ancestry cohorts combined




3·50 × 10–9

European/Asian ancestry cohorts combined




5·26 × 10–7

European cohort alone




7·01 × 10–7

European cohort alone




5·44 × 10–7

European cohort alone

Song et al. [46, 47]



PTS,<space> PLET1

1.22 × 10−9

Corrigenda p-value




7.59 × 10−7

Corrigenda p-value

p-Values in bold indicate genome-wide significant findings


It is evident that research findings in the last few years have sharpened our resolution of the processes that are affected by lithium treatment in the context of bipolar pathology. Results from differentiated iPS cells have been the most perceptive—revealing disorder-associated protein and electrophysiological changes that are reversed by lithium and distinguish patient response profiles. Such cells will be vital tools to help bridge the biological gap between the known genetic factors controlling lithium response and quantifiable electrophysiology. With current advances in microfluidic devices that model neural networks and the emergence of organoid production, we may soon be able to examine lithium’s effects at higher functional levels: brain structural changes and the connectivity changes that might underlie its mood stabilizing activities. Genomic studies, however, are still on a cusp—mirroring the state of complex disease gene identification some 5 years ago, immediately before meta-analyses pushed case numbers into the high thousands and unleashed a torrent of significant SNP findings. Doubtless that will come, and in the meantime we should be cautious not to over-interpret. Ultimately, the hope is that the genomic Venn diagram of lithium response and bipolar disorder risk will overlap in a restricted but insightful manner and, in this way, give drug companies clear direction for the development of equally potent but safer alternatives to lithium.

Supplementary material

13311_2017_554_MOESM1_ESM.pdf (1.6 mb)
ESM 1 Required Author Forms Disclosure forms provided by the authors are available with the online version of this article. (PDF 1647 kb)


  1. 1.
    Strobusch AD, Jefferson JW. The checkered history of lithium in medicine. Pharm Hist 1980;22(2):72–76.PubMedGoogle Scholar
  2. 2.
    BALANCE investigators and collaborators, Geddes JR, Goodwin GM, et al. Lithium plus valproate combination therapy versus monotherapy for relapse prevention in bipolar I disorder (BALANCE): a randomised open-label trial. Lancet 2010;375(9712):385–395.CrossRefGoogle Scholar
  3. 3.
    Cipriani A, Hawton K, Stockton S, Geddes JR. Lithium in the prevention of suicide in mood disorders: updated systematic review and meta-analysis. BMJ 2013;346:f3646.CrossRefPubMedGoogle Scholar
  4. 4.
    Shannon RD. Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides. Acta Crystallogr A 1976;32:751–767.CrossRefGoogle Scholar
  5. 5.
    Briggs KT, Giulian GG, Li G, Kao JPY, Marino JP. A molecular model for lithium’s bioactive form. Biophys J 2016;111(2):294–300.CrossRefPubMedGoogle Scholar
  6. 6.
    Dutta A, Bhattacharyya S, Dutta D, Das AK. Structural elucidation of the binding site and mode of inhibition of Li(+) and Mg(2+) in inositol monophosphatase. FEBS J 2014;281(23):5309–5324.CrossRefPubMedGoogle Scholar
  7. 7.
    Zyoud SH, Waring WS, Sweileh WM, Al-Jabi SW. Global research trends in lithium toxicity from 1913 to 2015: a bibliometric analysis. Basic Clin Pharmacol Toxicol 2017 Jan 8 [Epub ahead of print].Google Scholar
  8. 8.
    McKnight RF, Adida M, Budge K, Stockton S, Goodwin GM, Geddes JR. Lithium toxicity profile: a systematic review and meta-analysis. Lancet 2012;379:721–728.CrossRefPubMedGoogle Scholar
  9. 9.
    Boton R, Gaviria M, Batlle DC. Prevalence, pathogenesis, and treatment of renal dysfunction associated with chronic lithium therapy. Am J Kidney Dis 1987;10(5):329–345.CrossRefPubMedGoogle Scholar
  10. 10.
    Gitlin M. Lithium side effects and toxicity: prevalence and management strategies. Int J bipolar Disord 2016;4(1):27.CrossRefPubMedPubMedCentralGoogle Scholar
  11. 11.
    Tsermpini EE, Zhang Y, Niola P, et al. Pharmacogenetics of lithium effects on glomerular function in bipolar disorder patients under chronic lithium treatment: a pilot study. Neurosci Lett 2017;638:1–4.CrossRefPubMedGoogle Scholar
  12. 12.
    Squassina A, Manchia M, Borg J, et al. Evidence for association of an ACCN1 gene variant with response to lithium treatment in Sardinian patients with bipolar disorder. Pharmacogenomics 2011;12(11):1559–1569.CrossRefPubMedGoogle Scholar
  13. 13.
    Rej S, Pira S, Marshe V, et al. Molecular mechanisms in lithium-associated renal disease: a systematic review. Int Urol Nephrol 2016;48(11):1843–1853.CrossRefPubMedGoogle Scholar
  14. 14.
    Li Y, Shaw S, Kamsteeg E-J, Vandewalle A, Deen PMT. Development of lithium-induced nephrogenic diabetes insipidus is dissociated from adenylyl cyclase activity. J Am Soc Nephrol 2006;17(4):1063–1072.CrossRefPubMedGoogle Scholar
  15. 15.
    Kortenoeven MLA, Schweer H, Cox R, Wetzels JFM, Deen PMT. Lithium reduces aquaporin-2 transcription independent of prostaglandins. Am J Physiol Cell Physiol 2012;302(1):C131-C140.CrossRefPubMedGoogle Scholar
  16. 16.
    Gong R, Wang P, Dworkin L. What we need to know about the effect of lithium on the kidney. Am J Physiol Renal Physiol 2016;311(6):F1168–F1171. AGoogle Scholar
  17. 17.
    Grof P, Duffy A, Cavazzoni P, et al. Is response to prophylactic lithium a familial trait? J Clin Psychiatry 2002;63(10):942–947.CrossRefPubMedGoogle Scholar
  18. 18.
    Duffy A, Alda M, Milin R, Grof P. A consecutive series of treated affected offspring of parents with bipolar disorder: is response associated with the clinical profile? Can J Psychiatry 2007;52(6):369–376.CrossRefPubMedGoogle Scholar
  19. 19.
    Tighe SK, Mahon PB, Potash JB. Predictors of lithium response in bipolar disorder. Ther Adv Chronic Dis 2011;2(3):209–226.CrossRefPubMedPubMedCentralGoogle Scholar
  20. 20.
    Uwai Y, Arima R, Takatsu C, Furuta R, Kawasaki T, Nabekura T. Sodium-phosphate cotransporter mediates reabsorption of lithium in rat kidney. Pharmacol Res 2014;87:94–98.CrossRefPubMedGoogle Scholar
  21. 21.
    Alda M. Lithium in the treatment of bipolar disorder: pharmacology and pharmacogenetics. Mol Psychiatry 2015;20:661–670.CrossRefPubMedPubMedCentralGoogle Scholar
  22. 22.
    Pisanu C, Melis C, Squassina A. Lithium pharmacogenetics: where do we stand? Drug Dev Res 2016;77(7):368–373.CrossRefPubMedGoogle Scholar
  23. 23.
    Williams R, Ryves WJ, Dalton EC, et al. A molecular cell biology of lithium. Biochem Soc Trans 2004;32(Pt 5):799–802.CrossRefPubMedGoogle Scholar
  24. 24.
    Freland L, Beaulieu J-M. Inhibition of GSK3 by lithium, from single molecules to signaling networks. Front Mol Neurosci 2012;5:14.CrossRefPubMedPubMedCentralGoogle Scholar
  25. 25.
    Malhi GS, Outhred T. Therapeutic mechanisms of lithium in bipolar disorder: recent advances and current understanding. CNS Drugs 2016;30(10):931–949.CrossRefPubMedGoogle Scholar
  26. 26.
    McQuillin A, Rizig M, Gurling HMD. A microarray gene expression study of the molecular pharmacology of lithium carbonate on mouse brain mRNA to understand the neurobiology of mood stabilization and treatment of bipolar affective disorder. Pharmacogenet Genomics 2007;17(8):605–617.CrossRefPubMedGoogle Scholar
  27. 27.
    Seelan RS, Khalyfa a., Lakshmanan J, Casanova MF, Parthasarathy RN. Deciphering the lithium transcriptome: Microarray profiling of lithium-modulated gene expression in human neuronal cells. Neuroscience 2008;151(4):1184–1197.CrossRefPubMedGoogle Scholar
  28. 28.
    Sugawara H, Iwamoto K, Bundo M, et al. Effect of mood stabilizers on gene expression in lymphoblastoid cells. J Neural Transm 2010;117(2):155–164.CrossRefPubMedGoogle Scholar
  29. 29.
    Lee RS, Pirooznia M, Guintivano J, et al. Search for common targets of lithium and valproic acid identifies novel epigenetic effects of lithium on the rat leptin receptor gene. Transl Psychiatry 2015;5:e600.CrossRefPubMedPubMedCentralGoogle Scholar
  30. 30.
    Anand A, McClintick JN, Murrell J, Karne H, Nurnberger JI, Edenberg HJ. Effects of lithium monotherapy for bipolar disorder on gene expression in peripheral lymphocytes. Mol Neuropsychiatry 2016;2(3):115–123.CrossRefPubMedGoogle Scholar
  31. 31.
    Breen MS, White CH, Shekhtman T, et al. Lithium-responsive genes and gene networks in bipolar disorder patient-derived lymphoblastoid cell lines. Pharmacogenomics J 2016;16(5):446–453.CrossRefPubMedGoogle Scholar
  32. 32.
    Mertens J, Wang Q-W, Kim Y, et al. Differential responses to lithium in hyperexcitable neurons from patients with bipolar disorder. Nature 2015;527(7576):95–99.CrossRefPubMedPubMedCentralGoogle Scholar
  33. 33.
    Pathak G, Agostino MJ, Bishara K, et al. PDE11A negatively regulates lithium responsivity. Mol Psychiatry 2016 Sep 20 [Epub ahead of print].Google Scholar
  34. 34.
    Stern S, Santos R, Marchetto MC, et al. Neurons derived from patients with bipolar disorder divide into intrinsically different sub-populations of neurons, predicting the patients’ responsiveness to lithium. Mol Psychiatry 2017 Feb 28 [Epub ahead of print].Google Scholar
  35. 35.
    Tobe BTD, Crain AM, Winquist AM, et al. Probing the lithium-response pathway in hiPSCs implicates the phosphoregulatory set-point for a cytoskeletal modulator in bipolar pathogenesis. Proc Natl Acad Sci U S A 2017;114:E4462–E4471.CrossRefPubMedPubMedCentralGoogle Scholar
  36. 36.
    Hensley K, Venkova K, Christov A, Gunning W, Park J. Collapsin response mediator protein-2: an emerging pathologic feature and therapeutic target for neurodisease indications. Mol Neurobiol 2011;43(3):180–191.CrossRefPubMedGoogle Scholar
  37. 37.
    Sklar P, Smoller JW, Fan J, Ferreira MAR, Perlis RH, Chambert K. Whole-genome association study of bipolar disorder. Mol Psychiatry 2008;13:558–569.CrossRefPubMedPubMedCentralGoogle Scholar
  38. 38.
    Ferreira MA, O’Donovan MC, Meng YA, et al. Collaborative genome-wide association analysis supports a role for ANK3 and CACNA1C in bipolar disorder. Nat Genet 2008;40(9):1056–1058.CrossRefPubMedPubMedCentralGoogle Scholar
  39. 39.
    López-Jaramillo C, Vargas C, Díaz-Zuluaga AM, et al. Increased hippocampal, thalamus and amygdala volume in long-term lithium-treated bipolar I disorder patients compared with unmedicated patients and healthy subjects. Bipolar Disord 2017;19:41-49.CrossRefPubMedGoogle Scholar
  40. 40.
    Perlis RH, Smoller JW, Ferreira MAR, et al. A genomewide association study of response to lithium for prevention of recurrence in bipolar disorder. Am J Psychiatry 2009;166(6):718–725.CrossRefPubMedPubMedCentralGoogle Scholar
  41. 41.
    Chen C-H, Lee C-S, Lee M-TM, et al. Variant GADL1 and response to lithium therapy in bipolar i disorder. N Engl J Med 2014;370(2):119–128.CrossRefPubMedGoogle Scholar
  42. 42.
    Liu P, Ge X, Ding H, Jiang H, Christensen BM, Li J. Role of glutamate decarboxylase-like protein 1 (GADL1) in taurine biosynthesis. J Biol Chem 2012;287(49):40898–40906.CrossRefPubMedPubMedCentralGoogle Scholar
  43. 43.
    Chen C-K, Lee C-S, Chen H-Y, et al. GADL1 variant and medication adherence in predicting response to lithium maintenance treatment in bipolar I disorder. BJPsych Open 2016;2(5):301–306.PubMedPubMedCentralGoogle Scholar
  44. 44.
    Kotambail A, Mathur A, Bhat SM, Rai PS, Sharma PS, Satyamoorthy K. GADL1 gene polymorphisms and lithium response in bipolar I disorder: lack of association from an Indian population. Psychiatr Genet 2015;25(1):39–40.CrossRefPubMedGoogle Scholar
  45. 45.
    Cruceanu C, Alda M, Dion PA, Turecki G, Rouleau GA. No evidence for GADL1 variation as a bipolar disorder susceptibility factor in a Caucasian lithium-responsive cohort. Am J Psychiatry 2015;172(1):94–95.CrossRefPubMedGoogle Scholar
  46. 46.
    Song J, Bergen SE, Di Florio A, et al. Genome-wide association study identifies SESTD1 as a novel risk gene for lithium-responsive bipolar disorder. Mol Psychiatry 2016;21(9):1290–1297.CrossRefPubMedGoogle Scholar
  47. 47.
    Song J, Bergen SE, Di Florio A, et al. Genome-wide association study identifies SESTD1 as a novel risk gene for lithium-responsive bipolar disorder. Mol Psychiatry 2017 Feb 14 [Epub ahead of print].Google Scholar
  48. 48.
    Korner G, Scherer T, Adamsen D, et al. Mildly compromised tetrahydrobiopterin cofactor biosynthesis due to Pts variants leads to unusual body fat distribution and abdominal obesity in mice. J Inherit Metab Dis 2016;39(2):309–319.CrossRefPubMedGoogle Scholar
  49. 49.
    Hou L, Heilbronner U, Degenhardt F, et al. Genetic variants associated with response to lithium treatment in bipolar disorder: a genome-wide association study. Lancet 2016;387(10023):1085–1093.CrossRefPubMedPubMedCentralGoogle Scholar
  50. 50.
    Håvik B, Le Hellard S, Rietschel M, et al. The complement control-related genes CSMD1 and CSMD2 associate to schizophrenia. Biol Psychiatry 2011;70(1):35–42.CrossRefPubMedGoogle Scholar
  51. 51.
    Tam GWC, van de Lagemaat LN, Redon R, et al. Confirmed rare copy number variants implicate novel genes in schizophrenia. Biochem Soc Trans 2010;38:445–451.CrossRefPubMedGoogle Scholar

Copyright information

© The Author(s) 2017

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Authors and Affiliations

  1. 1.Strathclyde Institute of Pharmacy and Biomedical SciencesUniversity of StrathclydeGlasgowUK

Personalised recommendations