Verapamil in treatment resistant depression: a role for the P-glycoprotein transporter?

Verapamil in treatment resistant depression: a role for theP-glycoprotein transporter?

Gerard Clarke1,2*, Siobhain M O’Mahony2,3, John F Cryan2,4 and Timothy G Dinan1,2

1Department of Psychiatry, University College Cork, Cork, Ireland2Alimentary Pharmabiotic Centre, University College Cork, Cork, Ireland3Department of Anatomy, University College Cork, Cork, Ireland4Department of Pharmacology & Therapeutics, University College Cork, Cork, Ireland

Introduction Verapamil is a calcium channel blocker that also inhibits the P-glycoprotein (Pgp) membrane transporter. We have found thatadministration of verapamil with a recognised antidepressant improves clinical outcome in previously treatment resistant cases despite thefact that verapamil does not possess inherent antidepressant activity. In this study we examined the hypothesis that the antidepressant-likeeffects of verapamil are mediated through its blockade of the Pgp transporter in the blood brain barrier (BBB).Methods Following pre-treatment with verapamil (20mg/kg) or a saline solution male Sprague Dawley rats were injected with imipramine(15mg/kg). Two hours later, the animals were sacrificed, trunk blood collected and brain regions dissected out. High performance liquidchromatography (HPLC) was used to quantitate antidepressant drug concentrations in all samples.Results Verapamil pre-treatment significantly elevated imipramine concentrations in all brain regions studied. The effect was mostpronounced in the brainstem and frontal cortex where we observed in excess of a doubling in the brain region: serum ratios.Conclusion Our results verify inhibition of Pgp as a potential mechanism of action for verapamil during treatment resistant depression. Theimplications of these findings are discussed in the context of novel treatment strategies in depression. Copyright# 2009 JohnWiley& Sons, Ltd.

key words—P-glycoprotein; verapamil; treatment resistant depression; antidepressant; mechanism of action

abbreviations—BBB, blood brain barrier; CNS, central nervous system; HPLC, high performance liquid chromatography; OPA,orthophosphoric acid; PGP, P-glycoprotein; SSRI, selective serotonin reuptake inhibitor; TCA, tricyclic antidepressant

INTRODUCTION

The prevalence of treatment resistance depression,with as many as 46% of patients failing to respond tothe available pharmacological interventions (Fava andDavidson 1996; Greden 2002), reflects poorly oncurrent therapeutic strategies. Interestingly we havefound that co-administration of verapamil, primarily acalcium channel blocker (Schinkel and Jonker 2003),with a recognised antidepressant improves the clinicaloutcome in previously resistant cases. For example, a46-year-old man with an 18 month history of majordepression and melancholic features failed to respondto courses of escitalopram, venlafaxine, buproprionand the tricyclic lofepramine, each in therapeutic dosesfor at least 6 weeks. Verapamil 240mg noctewas added

to lofepramine 140mg twice daily and within 3 daysthere was a dramatic improvement in symptoms. At theend of 1 week he was entirely symptom free and hasremained so for the past 6 months. However, at2 months he went on vacation forgetting to bringthe verapamil but remaining on lofepramine. Hebecame symptomatic within 1 week but respondedto the reintroduction of verapamil on his return.Although themechanism of action is unclear at present,verapamil itself does not seem to have antidepressantproperties per se (Adlersberg et al., 1994).One plausible explanation for the apparent efficacy

of verapamil in treatment resistant depression revolvesaround its competitive blockade of an important effluxtransporter, P-glycoprotein (Pgp) (Aszalos et al.,1999). Pgp is a membrane transporter that is expressedin a variety of tissues including the intestine, liver andkidney and its role as a drug efflux transporter has beenextensively reviewed elsewhere (Kusuhara andSugiyama, 2001a,b; Loscher and Potschka, 2005b;Schinkel 1999). Of particular interest to this study is its

human psychopharmacologyHum. Psychopharmacol Clin Exp 2009; 24: 217–223.

Published online 11 February 2009 in Wiley InterScience

(www.interscience.wiley.com) DOI: 10.1002/hup.1008

*Correspondence to: G. Clarke, Department of Psychiatry, Lab 1.27, Bio-sciences Institute, University College Cork, Cork, Ireland. Tel: 00353 214901415. Fax: 00353 21 4901722E-mail: [email protected]

Copyright # 2009 John Wiley & Sons, Ltd.

Received 3 October 2008

Accepted 2 January 2009

role in the blood brain barrier (BBB) where it limits thebioavailability of a range of drugs, including anti-depressants, to the brain (Uhr et al., 2000). Indeedthe presence of Pgp in the BBB has been identified as amajor impediment to the successful treatment of manyCNS disorders (Pardridge 2005).It has previously been proposed that Pgp might play

a role in treatment resistant CNS disorders (Loscherand Potschka, 2005a) and initial studies on Pgp knockout mice produced up to 100 fold alterations in drugconcentrations compared to wild type animals (Wanget al., 2004a,b). However, the follow up studies havebeen not been enthusiastically received with onlymodest increases in the brain levels of the antidepress-ant nortriptyline observed following prior treatmentwith cyclosporine A, another Pgp inhibition strategy(Ejsing and Linnet, 2005). The fact that the authorsused whole brain in their analysis as opposed to distinctCNS structures may have obscured important region-specific central increases in drug concentrations. It isknown for instance that antidepressants have a regiondependant distribution (Daniel et al., 1991). Further-more, Pgp expression itself is also unevenly distributed(Kwan et al., 2002, 2003) and the averaged results fromthe Ejsing and Linnet (2005) study do not take thesephenomena into account. In any case, many cliniciansrecognise that even modest increases in CNS druglevels can improve response and that the magnitude ofalteration in central drug levels seen in Pgp knock-outanimals need not be replicated in the clinical setting toobtain meaningful therapeutic improvements. While itis perplexing that the area has received so littleattention in subsequent years, the recent associationbetween polymorphisms in the transporter andantidepressant treatment response (Uhr et al., 2008)should serve to reinvigorate the field.In this study we hypothesised that the improvement

in clinical symptoms seen following verapamiladministration was due to its inhibition of the Pgpefflux transporter and we sought to demonstrate thatPgp blockade could elicit physiologically relevantincreases in CNS levels of the prototypical tricyclicantidepressant (TCA), imipramine.

MATERIALS AND METHODS

Drugs and chemicals

High performance liquid chromatography (HPLC)grade acetonitrile, potassium dihydrogen phosphateand orthophosphoric acid (OPA) were obtained fromAlkem/Reagecon (Cork, Ireland). Imipramine, desi-pramine, trimipramine and verapamil were obtained

from Sigma-Aldrich as were all other chemicals unlessotherwise stated.

Animals

The animals used in this study were male SpragueDawley rats that were bred in the Biological ServicesUnit in the University College Cork. The animal roomremained temperature controlled (20� 18C) and a 12hlight/dark cycle employed (lights on at 8 a.m.). Thecages were cleaned twice a week as part of the animalroom routine. European Communities Council Directiveof 24November 1986 (86/609/EEC) for the care and useof laboratory animals was followed in all instances.

Animal experiments

Rats were randomly assigned to two groups (n¼ 7) andthe study was powered to detect differences in analytelevels at the 0.05 level. Animals were administeredverapamil i.p. (20mg/kg) or a saline solution at timezero. Ninty minutes later, all rats were treated withimipramine i.p. (15mg/kg). Two hours after imipraminetreatment, the unanesthetised animals were sacrificed ina random order by decapitation using a small animalguillotine (Biological Instruments, Italy) and the trunkblood and tissue was processed as described below.

Sample preparation

The brain was rapidly removed from the cranium anddissected out on an ice-cold plate as previouslydescribed (Harkin et al., 2001). The hypothalamus,frontal cortex, brainstem and hippocampus wereremoved, weighed and placed in 1.5ml microcentri-fuge tubes (Sarstedt) which contained 1ml of chilledhomogenisation buffer (HPLC mobile phase spikedwith trimipramine as internal standard). Each samplewas sonicated for 5 s (Sonoplus, Bandelin), centrifugedat 14 000RPM (Mikro 22R, Hittich, Germany) for15min at 48C and the supernatant stored at�808C untilextraction for analysis by HPLC. On the day of analysisimipramine and its main metabolite desipramine wereextracted by liquid–liquid partitioning as describedpreviously (Ejsing and Linnet 2005) with somemodifications. Briefly 1ml of sodium hydroxide(2M) and 3ml of water were added to 0.5ml of thesupernatant. Extraction was carried out in 7.5ml of1.5% isopentyl alcohol in n-heptane by agitation on amechanical shaker for 15min followed by centrifu-gation at 6000RPM for 15min at 48C. The uppersolvent layer was transferred to a tube containing200 ul of 25mMOPA, agitated on a mechanical shakerfor 15min followed by centrifugation at 6000RPM for

Copyright # 2009 John Wiley & Sons, Ltd. Hum. Psychopharmacol Clin Exp 2009; 24: 217–223.DOI: 10.1002/hup

218 g. clarke ET AL.

15minutes at 48C. Twenty microlitres of the loweraqueous phase was injected onto the HPLC system foranalysis. Serum samples had 200 ul of homogenisationbuffer added per 500 ul of serum and extracted asdescribed for brain homogenate supernatants.

Preparation of standards

Stock standards (10mg/ml) of imipramine/desipra-mine/trimipramine were prepared in HPLC gradewater. Working dilutions were prepared in the HPLCmobile phase, aliquoted out and stored at �808C untilanalysis.

HPLC equipment

The HPLC detection (FLD) system consisted of aWaters 510 pump, 717plus cooled Autosampler, 996PDA detector, a busSAT/IN module and a Croco-Cilcolumn oven. System components were used inconjunction with Waters Empower software. Allsamples were injected onto a reversed phase Luna3mm C18(2) 150� 2mm column (Phenomenex),which was protected by Krudkatcher disposable pre-column filters (Phenomenex) and SecurityGuardcartridges (Phenomenex).

HPLC conditions

The HPLC method was adapted from a previouslydescribed method (Frahnert et al., 2003). Briefly, themobile phase which was used on the HPLC system wascomposed of a mixture of 25mM potassium dihydro-gen phosphate (25mM, pH 7 with 4N NaOH) andHPLC grade acetonitrile (55:45). Mobile phase wasfiltered through Millipore 0.45mm HV Duraporemembrane filters (AGB, Dublin) and vacuum degassedprior to use. Compounds were eluted isocratically overa 40min runtime at a flow rate of 0.2ml/min after a20ml injection. The column was maintained at atemperature of 308C and samples/standards were keptat 88C in the cooled autoinjector prior to analysis.

Analyte identification and quantitation.

Imipramine, desipramine and trimipramine (internalstandard) were identified by their characteristicretention times as determined by standard injectionswhich were run at regular intervals during sampleanalysis. Analyte:internal standard peak height ratioswere measured and compared with standard injectionsand results were expressed at nanograms of analyte pergram fresh weight of tissue or per ml of serum asappropriate.

Data analysis

All data are reported as mean� SEM. Student’s t-testswere employed to determine differences in analytelevels. Bonferroni corrections for multiple t-tests wereemployed as necessary.

RESULTS

Hippocampal drug concentrations

Imipramine concentrations were significantlyincreased in the hippocampus of the verapamil treatedanimals (1.92 fold increase, p< 0.001, t¼ 4.409,df¼ 12). An observed increased in desipramine levelsdid not reach statistical significance (1.46 foldincrease) (see Table 1 and Figure 1).

Hypothalamic drug concentrations

Verapamil pre-treatment was also found to signifi-cantly increase imipramine concentrations in thehypothalamus (1.48 fold increase, p< 0.05,t¼ 2.197, df¼ 12). There was no statistical alterationin desipramine concentrations (1.25 fold increase) (seeTable 1 and Figure 2).

Frontal cortex drug concentrations

Verapamil pre-treated animals showed a 2.87 foldincrease in imipramine concentrations compared to thecontrol group (p< 0.01, t¼ 3.887, df¼ 12). Theseanimals also showed a 2.23 fold increase in desipra-mine concentrations (p< 0.05, t¼ 3.783, df¼ 12) (seeTable 1).

Brainstem drug concentrations

A 2.67 fold increase in imipramine concentrations wasobserved in the verapamil pre-treated rodents(p< 0.0001, t¼ 5.986, df¼ 12). A 1.73 fold increasein desipramine concentrations was observed in thesame animals (p< 0.05, t¼ 2.283, df¼ 12) (seeTable 1).

Brain-serum ratios

Hippocampus:serum imipramine ratios were elevatedsignificantly (1.5 fold increase, p< 0.05, t¼ 2.612,df¼ 12) in the verapamil treated group. There was nodiscernable alteration in hippocampus:serum desipra-mine ratios between groups (see Table 1).We did not observe any alterations in either the

hypothalamic:serum imipramine ratio or hypothala-mus:serum desipramine ratio (see Table 1).


pgp inhibition in treatment resistant depression 219

There was a 2.38 fold increase in the frontalcortex:serum ratio for imipramine (p< 0.05, t¼ 2.682,df¼ 12) and this resulted in a 1.43 fold increase in thefrontal cortex:serum ratio for desipramine (p< 0.01,t¼ 3.451, df¼ 12) (see Table 1).Verapamil pre-treatment also resulted in a significant

elevation in the brainstem:serum ratio for imipramine(2.07 fold increase, p< 0.001, t¼ 4.353, df¼ 12)although an apparent increase in the brainstem:serumratio for desipramine did not reach significance (1.34fold) (see Table 1).

Serum drug concentrations

Verapamil pre-treatment significantly increased serumimipramine concentrations (1.34 fold increase,p< 0.05, t¼ 2.505, df¼ 12). This group also showedincreases in desipramine concentrations but thisincrease did not reach statistical significance (1.30fold increase) (see Table 1).

Figure 1. Concentration of imipramine (ng/g tissue) in the hippocampusof control and verapamil treated animals (n¼ 7, ��p< 0.001)

Figure 2. Concentration of imipramine (ng/g tissue) in the hypothalamusof control and verapamil treated animals (n¼ 7, �p< 0.05)T

able

1.

Serum

concentration,brain

tissueconcentrationandbrain

tissue:

serum

ratiosforim

ipramineanditsmainmetabolite,desipramin

Serum

Hypothalam

us

Hippocampus

Frontalcortex

Brainstem

(ng/m

l)(ng/g)

Ratio

(ng/g)

Ratio

(ng/g)

Ratio

(ng/g)

Ratio

Imipramine

Control

104.4�12.14

2155�237.5

21.51�2.65

1863�318.3

17.36�2.38

1593�3

27.3

11.94�1.53

1299�261.2

12.17�2.02

Verapam

iltreatm

ent

140.4�7.68

3189�406.5

23.62�3.75

3574�221.9

25.99�2.29

4577�694.5

28.37�5.93

3473�252.2

25.17�2.20

p-value(t-test)

0.0277�

0.0484�

0.6538

0.0009��

0.0227�

0.0022��

0.0199�

<0.0001��

0.0009��

Desipramine

Control

137.8�20.07

3949�802.2

31.45�7.11

1801�236.2

13.84�1.33

1588�365.1

13.22�0.92

1335�277.4

9.98�1.86

Verapam

iltreatm

ent

178.5�25.64

4953�854.9

32.32�8.25

2628�310.7

15.09�0.85

3539�364.4

18.95�1.38

2315�327.4

13.38�1.34

p-value(t-test)

0.2356

0.4081

0.9379

0.0556

0.4431

0.0026��

0.0048��

0.0415�

0.1638

n¼7.

� p<0.05.

��p<0.01.

�� p

<0.001.



DISCUSSION

In support of our hypothesis on the mechanism ofaction of verapamil in treatment resistant depression,our data demonstrates that inhibition of the Pgptransporter by verapamil can significantly increasebrain levels of the antidepressant imipramine in anumber of important CNS regions. Moreover thedegree of increase in imipramine concentrations variedbetween regions analysed. We observed over a 2.5 foldincrease in the concentration of imipramine in both thefrontal cortex and brainstem of verapamil pre-treatedanimals compared to controls. Even when taking intoaccount the more modest 34% increase in serumimipramine levels, the brain region:serum ratio forthese regions is still increased over 2 fold compared tocontrol animals. Interestingly desipramine concen-trations, the main metabolite of imipramine, alsoincreased significantly in these regions. Although boththe hypothalamus and hippocampus also showedsignificantly elevated imipramine concentrations andbrain region:serum ratios, the magnitude of thesechanges (1.1 fold in the hypothalamus and 1.5 fold inthe hippocampus) would not seem to be as physio-logically relevant as in the other regions. In support ofthis view, increases in actual desipramine concen-trations did not reach statistical significance in eitherbrain area. It should be noted the desipraminemeasured in this study is derived from the parentcompound and that had desipramine been administereddirectly we may have found similar increases to thosedemonstrated for imipramine.The magnitude of CNS increase seen here for

imipramine following Pgp inhibition is greater than thepreviously observed modest brain increases in nor-triptyline (Ejsing and Linnet 2005) and support theview of the Pgp transporter as an important ‘gate-keeper’ (Schinkel 1999) in the BBB. A number ofimportant issues arise from these results that meritfurther comment. Firstly, our clinical case studieshighlight the importance of verapamil as a co-therapyin treatment resistant depression but its efficacy has, todate, lacked a prescribed mechanism. Verapamil doesnot possess inherent antidepressant activity (Adlers-berg et al., 1994) although calcium channel blockershave been shown to produce an antidepressant-likeeffect in the mouse forced swimming test (Cohen et al.,1997). However this effect was mediated by agentsacting at the dihydropyridine binding site and not thephenyalanine binding site upon which verapamil acts.Interestingly, nimodipine, a calcium channel blockerthat does act at the dihydropyridine site, has repeatedlybeen shown to augment the antidepressant properties of

standard pharmacological agents in the clinical setting(Taragano et al., 2001, 2005). Although studiesindicate that, in contrast to verapamil, nimodipinedoes have independent antidepressant activity (Deet al., 1997), this remains to be conclusivelydemonstrated. It should also be noted that BBBjunction integrity can be altered by various calcium-dependant signal transduction cascades and thatcalcium ions can interact directly with junctionproteins to modulate BBB function (Brown and Davis2002). Nevertheless, it is not clear from the literaturewhether the verapamil dose administered in this studywould disrupt the BBB integrity sufficiently, indepen-dently of Pgp inhibition, to elicit the magnitude ofalteration in CNS drug levels we have reported.Our reasonable hypothesis that verapamil owed its

effects to inhibition of the Pgp transporter is at leastpartly validated by the CNS drug alterations describedhere. Investigators interested in treatment resistantepilepsy have advanced similar theories (Loscher2005; Summers et al., 2004) and have found similarvalidation of their theories in pre-clinical studies(Seegers et al., 2002; Volk et al., 2004). Of courseinhibition of Pgp as a therapeutic strategy has longbeen a source of interest in cancer therapy and hasspawned the development of novel second and thirdgeneration inhibitors (Thomas and Coley 2003). Inlight of the possibility of the effects in this studybeing mediated by calcium channel blockade andthe recent advances in drug development, futurestudies should seek to utilise these alternativepharmacological agents to further clarify the role ofthe Pgp transporter.Secondly, one needs to consider the current state of

antidepressant therapy and the impact these resultsmight have therein. It is generally recognised that thesuccess of the SSRIs centres largely around their moremanageable side effect profile compared to the TCAsand not necessarily their greater efficacy (Anderson1998; Steffens et al., 1997). The results of this studywould then seem to extend beyond the confines of ourinitial observations concerning treatment resistantdepression. It is worth speculating that the possibilityof lowering the TCA dose to limit their peripheral sideeffect profile while at the same time maintainingeffective CNS drug levels would be of great interest tomany clinicians (Gillman 2007). Finally recent claimsthat polymorphisms in the Pgp transporter gene areassociated with antidepressant response might pave theway for personalised treatment (Uhr et al., 2008) offerfurther expansion of the ideas expressed here. Surelythe same concept could be utilised to identify suitablecandidates for co-administration of verapamil and



reduce the need for stressful trial and error dosingregimes?It is appropriate at this stage to consider some of the

limitations of this study. We used imipramine as aninitial theory verification probe but further studies arerequired on the range of currently available anti-depressant therapies. Nevertheless, desipramine, themain metabolite of imipramine, is itself an anti-depressant and the results shown here would suggestthat Pgp inhibition would also be an effective strategyfor that particular antidepressant. Our study offers datagleaned from an acute paradigm only and one cannotdiscount the possibility that chronic inhibition of Pgpmight result in upregulation of both transporterfunction and abundance. While some studies havedemonstrated upregulation of the transporter followingprolonged treatment with one of its substrates (Baueret al., 2006; Hennessy et al., 2002) a definitiveexamination of both efflux pump activity andexpression following concurrent chronic treatmentwith verapamil and an antidepressant is now required.It is however worth pointing out the initial clinicalobservations on which this study was based were drawnfrom patients on prolonged treatment regimes.Additionally, although all the brain areas in this studyhave been implicated in depression (Connor andLeonard, 2004) the differential effects which Pgpinhibition seems to exert need to be examined further.Our results suggest that the brainstem and frontalcortex are afforded superior protection by the BBBthan the hippocampus or hypothalamus and con-sequently derive a greater ‘benefit’ from Pgp inhi-bition. This is backed up by studies showing that somenuclei of the hypothalamus lack a fully intact BBB(Norsted et al., 2008). Furthermore, it is interesting tospeculate that the mdr1b isoform of the Pgptransporter, which is expressed in the hippocampusand preferentially transports corticosterone (Parianteet al., 2004), might actually predominate over theantidepressant transporting mdr1a isoform. Whetherthis regional disparity in effect is relevant to theefficacy of verapamil requires further clarification.Attention should be drawn to the fact that this studywas conducted in ‘normal’ rats. Care should be takenwhen extrapolating these findings to cases of clinicaldepression. BBB efflux mechanisms can be affected byboth stress (Sukhai and Piquette-Miller 2000) andinflammation (Ho and Piquette-Miller 2006), both ofwhich have been implicated in the pathogenesis ofdepression (Connor and Leonard 2004; Dinan et al.,2008). However our clinical findings imply that themechanism we propose is relevant. Finally, drug–druginteractions between verapamil and imipramine could

in part account for the findings of increased serumimipramine concentrations in this study (Hermannet al., 1992; Zhou et al., 2004). Nevertheless the brainregion:serum ratio demonstrates that the increases inbrain tissue imipramine concentrations is over andabove that which could be contributed by increasedserum concentrations on their own.In conclusion, our results implicate inhibition of Pgp

as a potential mechanism of action for verapamilduring treatment resistant depression. Additionalstudies are urgently required to take advantage of thispotentially exciting therapeutic strategy in the field ofdepression.

ACKNOWLEDGEMENTS

The Alimentary Pharmabiotic Centre (APC) is principallyfunded by the Science Foundation of Ireland (SFI). It is alsoin receipt of funds from the Industrial DevelopmentAuthority (IDA) and Glaxosmithkline. Professor Dinan isadditionally in receipt of funds from the Wellcome trust.

REFERENCES

Adlersberg S, Toren P, Mester R, Rehavi M, Skolnick P, Weizman A. 1994.Verapamil is not an antidepressant in patients resistant to tricyclicantidepressants. Clin Neuropharmacol 17: 294–297.

Anderson IM. 1998. SSRIS versus tricyclic antidepressants in depressedinpatients: a meta-analysis of efficacy and tolerability. Depress Anxiety 7(Suppl 1): 11–17.

Aszalos A, Thompson K, Yin JJ, Ross DD. 1999. Combinations of P-glycoprotein blockers, verapamil, PSC833, and cremophor act differentlyon the multidrug resistance associated protein (MRP) and on P-glyco-protein (Pgp). Anticancer Res 19: 1053–1064.

Bauer B, Yang X, Hartz AM, et al. 2006. In vivo activation of humanpregnane X receptor tightens the blood-brain barrier to methadonethrough P-glycoprotein up-regulation. Mol Pharmacol 70: 1212–1219.

Brown RC, Davis TP. 2002. Calcium modulation of adherens and tightjunction function: a potential mechanism for blood-brain barrier disrup-tion after stroke. Stroke 33: 1706–1711.

Cohen C, Perrault G, Sanger DJ. 1997. Assessment of the antidepressant-like effects of L-type voltage-dependent channel modulators. BehavPharmacol 8: 629–638.

Connor TJ, Leonard BE. 2004.BiologicalMarkers of DepressionHandbookof Experimental Pharmacology. Springer Verlag: New York; 117–148.

Daniel W, Danek L, Janczar L, Nocon H, Melzacka M. 1991. Regionaldistribution of imipramine, desipramine and specific [3H] desipraminebinding sites in the rat brain after acute and chronic treatment withimipramine. J Pharm Pharmacol 43: 31–315.

De Vry J, Fritze J, Post RM. 1997. The management of coexistingdepression in patients with dementia: potential of calcium channelantagonists. Clin Neuropharmacol 20: 22–35.

Dinan T, Siggins L, Scully P, O’Brien S, Ross P, Stanton C. 2008.Investigating the inflammatory phenotype of major depression: Focuson cytokines and polyunsaturated fatty acids. J Psychiatr Res. 43: 471–476.

Ejsing TB, Linnet K. 2005. Influence of P-glycoprotein inhibition on thedistribution of the tricyclic antidepressant nortriptyline over the blood-brain barrier. Hum Psychopharmacol 20: 149–153.

Fava M, Davidson KG. 1996. Definition and epidemiology of treatment-resistant depression. Psychiatr Clin North Am 19: 179–200.

Frahnert C, Rao ML, Grasmader K. 2003. Analysis of eighteen antidepress-ants, four atypical antipsychotics and active metabolites in serum by



liquid chromatography: a simple tool for therapeutic drug monitoring.J Chromatogr B Analyt Technol Biomed Life Sci 794: 35–47.

Gillman PK. 2007. Tricyclic antidepressant pharmacology and therapeuticdrug interactions updated. Br J Pharmacol 151: 737–748.

Greden JF. 2002. Unmet need: what justifies the search for a new anti-depressant? J Clin Psychiatry 63 (Suppl 2): 3–7.

Harkin A, Connor TJ, Mulrooney J, Kelly JP, Leonard BE. 2001. Priorexposure to methylenedioxyamphetamine (MDA) induces serotonergicloss and changes in spontaneous exploratory and amphetamine-inducedbehaviors in rats. Life Sci 68: 1367–1382.

Hennessy M, Kelleher D, Spiers JP, et al. 2002. St Johns wort increasesexpression of P-glycoprotein: implications for drug interactions. Br J ClinPharmacol 53: 75–82.

Hermann DJ, Krol TF, Dukes GE, et al. 1992. Comparison of verapamil,diltiazem, and labetalol on the bioavailability and metabolism of imi-pramine. J Clin Pharmacol 32: 176–183.

Ho EA, Piquette-MillerM. 2006. Regulation of multidrug resistance by pro-inflammatory cytokines. Curr Cancer Drug Targets 6: 295–311.

Kusuhara H, Sugiyama Y. 2001a. Efflux transport systems for drugs at theblood-brain barrier and blood-cerebrospinal fluid barrier (Part 1). DrugDiscov Today 6: 150–156.

Kusuhara H, Sugiyama Y. 2001b. Efflux transport systems for drugs at theblood-brain barrier and blood-cerebrospinal fluid barrier (Part 2). DrugDiscov Today 6: 206–212.

Kwan P, Sills GJ, Butler E, Gant TW, Brodie MJ. 2003. Differentialexpression of multidrug resistance genes in naive rat brain. NeurosciLett 339: 33–36.

Kwan P, Sills GJ, Butler E, Gant TW, Meldrum BS, Brodie MJ. 2002.Regional expression of multidrug resistance genes in geneticallyepilepsy-prone rat brain after a single audiogenic seizure. Epilepsia43: 1318–1323.

Loscher W. 2005. Mechanisms of drug resistance. Epileptic Disord 7(Suppl 1): S3–9.

Loscher W, Potschka H. 2005a. Blood-brain barrier active efflux transpor-ters: ATP-binding cassette gene family. NeuroRx 2: 86–98.

Loscher W, Potschka H. 2005b. Role of drug efflux transporters in the brainfor drug disposition and treatment of brain diseases. Prog Neurobiol 76:22–76.

Norsted E, Gomuc B, Meister B. 2008. Protein components of the blood-brain barrier (BBB) in the mediobasal hypothalamus. J Chem Neuroanat.36: 107–121.

Pardridge WM. 2005. The blood-brain barrier: bottleneck in brain drugdevelopment. NeuroRx 2: 3–14.

Pariante CM, Thomas SA, Lovestone S, Makoff A, Kerwin RW. 2004.Do antidepressants regulate how cortisol affects the brain? Psychoneur-oendocrinology 29: 423–447.

Schinkel AH. 1999. P-Glycoprotein, a gatekeeper in the blood-brain barrier.Adv Drug Deliv Rev 36: 179–194.

Schinkel AH, Jonker JW. 2003. Mammalian drug efflux transporters of theATP binding cassette (ABC) family: an overview. AdvDrugDeliv Rev 55:3–29.

Seegers U, Potschka H, Loscher W. 2002. Expression of the multidrugtransporter P-glycoprotein in brain capillary endothelial cells and brainparenchyma of amygdala-kindled rats. Epilepsia 43: 675–684.

Steffens DC, Krishnan KR, Helms MJ. 1997. Are SSRIs better than TCAs?Comparison of SSRIs and TCAs: a meta-analysis.Depress Anxiety 6: 10–18.

Sukhai M, Piquette-Miller M. 2000. Regulation of the multidrug resistancegenes by stress signals. J Pharm Pharm Sci 3: 268–280.

Summers MA, Moore JL, McAuley JW. 2004. Use of verapamil as apotential P-glycoprotein inhibitor in a patient with refractory epilepsy.Ann Pharmacother 38: 1631–1634.

Taragano FE, Allegri R, Vicario A, Bagnatti P, Lyketsos CG. 2001. A doubleblind, randomized clinical trial assessing the efficacy and safety ofaugmenting standard antidepressant therapy with nimodipine in thetreatment of ’vascular depression’. Int J Geriatr Psychiatry 16: 254–260.

Taragano FE, Bagnatti P, Allegri RF. 2005. A double-blind, randomizedclinical trial to assess the augmentation with nimodipine of antidepress-ant therapy in the treatment of ‘vascular depression’. Int Psychogeriatr17: 487–498.

Thomas H, ColeyHM. 2003. Overcomingmultidrug resistance in cancer: anupdate on the clinical strategy of inhibiting p-glycoprotein. CancerControl 10: 159–165.

Uhr M, Steckler T, Yassouridis A, Holsboer F. 2000. Penetration ofamitriptyline, but not of fluoxetine, into brain is enhanced in mice withblood-brain barrier deficiency due to mdr1a P-glycoprotein gene disrup-tion. Neuropsychopharmacology 22: 380–387.

Uhr M, Tontsch A, Namendorf C, et al. 2008. Polymorphisms in the drugtransporter gene ABCB1 predict antidepressant treatment response indepression. Neuron 57: 203–209.

Volk HA, Potschka H, Loscher W. 2004. Increased expression of themultidrug transporter P-glycoprotein in limbic brain regions after amyg-dala-kindled seizures in rats. Epilepsy Res 58: 67–79.

Wang JS, Ruan Y, Taylor RM, Donovan JL, Markowitz JS, DeVane CL.2004a. The brain entry of risperidone and 9-hydroxyrisperidone isgreatly limited by P-glycoprotein. Int J Neuropsychopharmacol 7:415–419.

Wang JS, Ruan Y, Taylor RM, Donovan JL, Markowitz JS, DeVane CL.2004b. Brain penetration of methadone (R)- and (S)-enantiomers isgreatly increased by P-glycoprotein deficiency in the blood-brain barrierof Abcb1a gene knockout mice. Psychopharmacology (Berl) 173: 132–138.

Zhou S, Chan E, Lim LY, et al. 2004. Therapeutic drugs that behave asmechanism-based inhibitors of cytochrome P450 3A4. Curr Drug Metab5: 415–442.



Documents

Verapamil in treatment resistant depression: a role for the P-glycoprotein transporter?