O R I G I N A L
`A R T I C L E
`
`doi: 10.1111/j.1472-8206.2011.00957.x
`
`Sirolimus and everolimus intestinal
`absorption and interaction with calcineurin
`inhibitors: a differential effect between
`cyclosporine and tacrolimus
`
`Fabien Lamoureuxa,b, Nicolas Picarda,b,c,*, Belkacem Bousseraa,b,
`Franc¸ois-Ludovic Sauvagea,b,c and Pierre Marqueta,b,c
`aINSERM UMR-S850, 2 rue du Dr marcland, 87025 Limoges, France
`bFaculty of Medicine, University of Limoges, 2 rue du Dr marcland, 87025 Limoges, France
`cC.H.U. of Limoges, Department of Pharmacology – Toxicology, 2 avenue Martin Luther-King, 87042 Limoges, France
`
`Keywords
`bioavailability,
`Caco-2 cells,
`calcineurin inhibitors,
`intestinal transport,
`mTOR inhibitors
`
`Received 4 November 2010;
`revised 11 April 2011;
`accepted 20 April 2011
`
`*Correspondence and reprints:
`nicolas.picard@unilim.fr
`
`A B S T R A C T
`
`The mTOR inhibitors (ImTORs) sirolimus (SRL) and everolimus (EVR) have been
`increasingly used in renal transplantation as part of calcineurin inhibitor (CNI)
`sparing or avoidance regimens. Those drugs have low and variable oral bioavail-
`ability that is increased when combined with cyclosporine or tacrolimus (TAC). We
`investigated the mechanisms involved in ImTORs intestinal absorption in vitro and
`associated it with their drug–drug interactions with CNIs. The transport of ImTORs
`across Caco-2 cells was studied in the apical (A) to basolateral (B) and B to A
`directions, in the absence or presence of cyclosporine, TAC, and GF120918 (P-gp
`inhibitor). In Caco-2 cells, EVR and SRL displayed a polarized transport with 8.7- and
`5.9-fold higher Papp,BfiA than Papp,AfiB, respectively. P-gp inhibition by GF120918
`resulted in a 70 and 41% decrease in EVR and SRL efflux, respectively. Cyclosporine
`and TAC led to a comparable and significant decrease in the efflux ratio of ImTORs,
`suggesting inhibition of a P-gp-mediated efflux transport. Cyclosporine also exhibited
`a specific increase of Papp,BfiA, which may be attributed to the inhibition of other
`transporters and/or metabolizing enzymes. In conclusion, EVR and SRL are both
`subject to an apically directed efflux mediated by P-gp. TAC mainly inhibits this efflux
`mechanism, while the effect of cyclosporine appears to be more complex with
`mechanisms to be confirmed by further studies.
`
`I N T R O D U C T I O N
`
`formerly known as rapamycin, and
`Sirolimus (SRL),
`everolimus [EVR; 40-O-(2-hydroxyethyl)-rapamycin] are
`mTOR inhibitors (ImTORs) commonly used to prevent
`allograft rejection after solid organ transplantation.
`ImTORs are being used increasingly in combination
`with low doses of calcineurin inhibitors (CNIs), cyclo-
`sporine A (CsA), or tacrolimus (TAC).
`mTOR inhibitor s as well as CNIs present a low average
`oral bioavailability (approximately 25%) with wide
`interindividual variations (range: 4–89%) [1,2]. The
`causes of this remain unclear but are presumably linked
`
`with the variable activity of metabolic enzymes [cyto-
`chrome P450 (CYP) 3A] and of active efflux transporters
`such as P-glycoprotein (P-gp) or members of the multi-
`drug resistance-associated proteins (MRPs) family (espe-
`cially MRP1 and MRP2) in the small intestine [3]. CsA,
`TAC, SRL, and EVR are indeed extensively metabolized
`by CYP 3A4 and, to a lesser extent, by CYP3A5 [4–6].
`These drugs are also known substrates of P-gp [7,8]. In
`addition, we recently showed that ImTORS are not
`transported by uptake transporters of the organic anion-
`transporting polypeptides (OATPs) family [9].
`Although structurally similar, EVR and SRL undergo
`different metabolism. In contrast to SRL, the presence of
`
`ª 2011 The Authors Fundamental and Clinical Pharmacology ª 2011 Socie´ te´ Franc¸aise de Pharmacologie et de The´ rapeutique
`Fundamental & Clinical Pharmacology 26 (2012) 463–472
`
`463
`
`Fundamental & Clinical Pharmacology
`
`Ex. 1047-0001
`
`

`
`464
`
`F. Lamoureux et al.
`
`Table I Concentration-dependent effect of cyclosporine (a), tacroli-
`mus (b) and GF120918 (c) on the transport of digoxin (1 lM), a
`P-gp probe, across Caco-2 cell monolayers.
`
`(a)
`
`CsA (lM)
`
`Control
`
`1
`
`10
`
`50
`
`(b)
`
`TAC (lM)
`
`Control
`
`1
`
`10
`
`50
`
`(c)
`
`GF120918 (lM)
`
`Control
`
`2
`
`Papp,AfiB
`)6 cm/s)
`(10
`
`1.02 ± 0.28
`
`0.84 ± 0.14
`
`3.10 ± 0.23
`
`2.41 ± 0.60
`
`Papp,AfiB
`)6 cm/s)
`(10
`
`0.98 ± 0.25
`
`0.57 ± 0.14
`
`3.40 ± 0.72
`
`2.53 ± 0.59
`
`Papp,AfiB
`)6 cm/s)
`(10
`
`1.02 ± 0.28
`
`2.77 ± 0.64
`
`Papp,BfiA
`)6 cm/s)
`(10
`
`21.25 ± 2.21
`
`17.20 ± 1.82
`
`3.49 ± 0.10
`
`3.73 ± 0.32
`
`Papp,BfiA
`)6 cm/s)
`(10
`
`20.76 ± 2.32
`
`17.92 ± 1.52
`
`3.64 ± 0.36
`
`3.31 ± 0.43
`
`Papp,BfiA
`)6 cm/s)
`(10
`
`Efflux ratio
`Papp,BfiA/
`Papp,AfiB
`
`20.79
`
`20.43
`
`1.13**
`
`1.55**
`
`Efflux ratio
`Papp,BfiA/
`Papp,AfiB
`
`21.18
`
`31.37*
`
`1.07*
`
`1.31*
`
`Efflux ratio
`Papp,BfiA/
`Papp,AfiB
`
`21.25 ± 2.21
`
`6.22 ± 1.04
`
`20.79
`
`2.25*
`
`Data are expressed as mean ± standard deviation and are representative of
`
`three independent triplicate experiments.
`Papp,AfiB and Papp,B fi A, apparent permeability in the apical-to-basal and
`basal-to-apical directions, respectively, calculated according to the equation
`
`described in the 2.8 section.
`*P £ 0.05 and **P £ 0.01 (measured efflux ratio vs. control).
`
`a 40-O-2-hydroxyethyl group on EVR prevents its 39-O-
`demethylation and also decreases two major hydroxyl-
`ation pathways, which results in an overall decrease in
`metabolism as shown using human liver microsomes
`[10]. However, Crowe et al. suggested on the basis of
`experimental studies in rats and Caco-2 cells that EVR is
`metabolized by the intestine to a greater extent than SRL,
`which is balanced in terms of bioavailability by a higher
`intrinsic permeability [11,12].
`Concurrent administration of CsA significantly in-
`creases SRL area under curve (area under the plasma
`drug concentration vs. time curve), Cmax, and tmax [13],
`while TAC seems to have a lower effect [14,15]. Similar
`effects of CNIs on the pharmacokinetics of EVR have
`recently been described [16]. The mechanisms of these
`drug–drug interactions at the intestinal level are still not
`
`fully understood but are likely to be linked to the
`inhibitory effect of CNIs on ImTORs intestinal and/or
`hepatic metabolism and transport. CsA and to a lesser
`extent TAC are inhibitors of CYP3A4 and P-gp [17,18].
`The human colon adenocarcinoma cell line Caco-2 is
`widely used as a model to study drug transport in
`intestinal epithelium. When fully differentiated, polarized
`Caco-2 cells exhibit morphological and functional sim-
`ilarities with human intestinal enterocytes, expressing
`efflux transporters (e.g., P-gp, MRP1, and MRP2) [19] as
`well as metabolic enzymes at significant levels [20]. A
`predictive relationship between the permeability of
`Caco-2 monolayers and human in vivo intestinal
`absorption has been reported by several authors [21,22].
`The present study aimed to compare the transepithe-
`lial passage of ImTORs across Caco-2 cells and their
`interaction with CNIs at this level.
`
`M A T E R I A L S A N D M E T H O D S
`
`Materials and chemicals
`The human colon adenocarcinoma cell line Caco-2 was
`obtained from the American Type Culture Collection
`(Manassas, VA, USA). Dulbecco’s modified Eagle’s med-
`ium (DMEM), fetal calf serum (FCS), glutamine, nones-
`sential amino acids (NEAA), penicillin–streptomycin
`(10 000 units/mL and 10 mg/mL in 0.9% sodium
`chloride, respectively), 0.05% trypsin – 0.53 mM EDTA-
`4Na, Hank’s balanced salt solution (HBSS), HEPES
`solution, and Dulbecco’s phosphate-buffered saline
`(PBS) were purchased from GibcoBrl Life Technology
`(Cergy-Pontoise, France). Matrigel was purchased from
`BD biosciences Discovery Labware (Le Pont de Claix,
`France). EVR and CsA were kindly provided by Novartis
`Pharma AG (Basel, Switzerland), SRL by Wyeth-Lederle´
`(Paris, France), and TAC by Astellas Pharma (Levallois-
`Perret, France). The MDR1 chemical inhibitor GF120918
`(9,10-dihydro-5-methoxy-9-oxo-N-[4-[2(1,2,3,4-tetra-
`hydro-6,7-dimethoxy-2-isoquinolinyl)ethyl]-4-acridinecar-
`boxamide hydrochloride salt) was kindly donated by
`GlaxoSmithKline (Marly-le-Roi, France). Digoxin and
`atenolol were obtained from Sigma (Saint Quentin
`Fallavier, France). Organic solvents and chemicals used
`for drug analysis were of analytical grade.
`
`Caco-2 cells culture conditions
`Caco-2 cells were routinely cultured at a density of
`7 · 106 cells on 75-cm2 plastic culture flasks (BD Bio-
`sciences Discovery Labware), containing DMEM, 10%
`inactivated FCS, 2 mM glutamine, 1% NEAA, 100 units/
`
`ª 2011 The Authors Fundamental and Clinical Pharmacology ª 2011 Socie´ te´ Franc¸aise de Pharmacologie et de The´ rapeutique
`Fundamental & Clinical Pharmacology 26 (2012) 463–472
`
`Ex. 1047-0002
`
`

`
`mTOR inhibitors intestinal absorption and interaction with calcineurin inhibitors
`
`465
`
`mL of penicillin, and 100 lg/mL of streptomycin. All the
`cells used in this study were passaged 45–60 times.
`For the transepithelial transport study, Caco-2 cells
`were seeded on the apical side of matrigel-treated filter
`inserts (12 mm diameter and 0.4-lm pore size, Millipore,
`Molsheim, France) at a density of 6 · 105 cells per filter.
`The cells were grown until differentiation for at least
`21 days and used for experiments between 21 and
`28 days postseeding. All transport studies were con-
`ducted in transport buffer (HBSS supplemented with
`10 mM HEPES; pH 7.4) in both apical-to-basolateral (A
`to B) and basolateral-to-apical (B to A) directions under
`iso-pH conditions (pH 7.4 on both sides).
`
`Cell toxicity and viability assessment
`The viability of CNI- or ImTOR-treated cells was assessed
`by the trypan blue dye exclusion test and the MTT assay.
`Cytotoxicity experiments were conducted by applying
`1–100 lM CNI and 1–10 lM ImTOR for 2 h at 37 °C on
`differentiated Caco-2 cells. Cells were then incubated for
`4 h with 5 mg/mL MTT (Sigma) in PBS and lysed in 200
`lL SDS 10%/HCl 0.01 N for 4 h. Aliquots of the lysates
`were transferred in 96-well plates, and absorbance was
`recorded at 550 nm using a Multiskan EX (Labsystems,
`Milford, MA, USA) microplate spectrophotometer system.
`Trypan blue exclusion was also used to qualitatively
`assess cell viability. After exposure to increasing con-
`centrations of CNIs (1, 10, 50, or 100 lM) or ImTORs (1,
`5, or 10 lM), Caco-2 monolayers were washed with PBS,
`detached from the support with trypsin – EDTA, incu-
`bated with 0.4% trypan blue solution for 1 min, and
`counted in a hemocytometer using a light microscope.
`Exposure to CNIs or ImTORs for 2 h did not decrease
`mitochondrial activity or viability of Caco-2 cells in
`comparison with controls, as assessed during these assays.
`
`Assessment of cell monolayer integrity
`The integrity of cell monolayers was evaluated by two
`different methods. The transepithelial electrical resistance
`(TEER) was measured using a Millicell-ERS equipment
`(Millipore) as described previously [23]. Only monolayers
`with TEER > 600 W/cm2 were used for further experi-
`ments. Additionally, TEER was measured after each
`permeability experiment to confirm monolayer integrity.
`Exposure of the cell monolayers to ImTORs did not induce
`a significant decrease in TEER before and after transport
`studies (n = 120, 688 ± 35 W cm2 at day 28 postsee-
`ding vs. 600 ± 61 cm2 just after transport experiments).
`The integrity of the cell monolayers was also studied
`by determining the paracellular permeability (apparent
`
`permeability coefficient, Papp) of atenolol, as previously
`recommended [24,25]. The permeability of atenolol
`through untreated cells was the same for both transport
`
`directions (Papp,AfiB: 1.44 ± 0.09 · 10)6 cm/s, Papp,BfiA:
`)6 cm/s), consistent with the literature
`1.38 ± 0.11 · 10
`values [21,26] and unaffected after incubation of the
`monolayers with the immunosuppressive agents tested
`in comparison with the control.
`
`Transport studies
`Transport of ImTORs accross Caco-2 monolayers
`Fresh medium containing EVR or SRL at various
`concentrations (1, 5, and 10 lM, corresponding to levels
`expected after oral dosing) was added to either the apical
`(for absorption studies: A fi B direction) or basal (for
`secretory studies: B fi A direction) side. An equal
`volume of
`incubation medium without drugs was
`systematically added to the opposite side of the mono-
`layers. Monolayers were then incubated for 2 h at 37 °C
`in a humidified 5% CO2 atmosphere, in three indepen-
`dent and triplicate experiments. Incubation time was set
`at 2 h according to ImTORs pharmacokinetic data: the
`tmax reported are 1–1.75 h for EVR [16,27] and 1.3 + /
`0.5 h for SRL [28], and concomitant administration of
`CNIs with mTORs tends to delay the tmax [13,16,29].
`Samples were taken from the apical and basolateral
`compartments at the end of the 2-h period and kept at
`)22 °C prior to liquid chromatography - Tandem Mass
`Spectrometry (LC-MS/MS) analysis.
`
`Infuence of CsA, TAC, and GF120918 on the transport of
`ImTORs
`The transport of ImTORs (EVR and SRL at 1, 5, and
`10 lM) was also studied in both the A fi B and B fi A
`directions, in the absence or presence of CsA (10 lM),
`TAC (10 lM), and GF120918 (2 lM; a third-generation
`P-gp inhibitor with no effect on CYP3A) for 2 h at 37 °C,
`in triplicates. Samples from the A and B compartments
`were then taken and kept at )20 °C until analysis.
`As a positive inhibition control, the effect of CsA, TAC
`(1-10-50 lM), and GF120918 (2 lM) on the efflux of the
`P-gp substrate digoxin (1 lM) [30] across Caco-2 cell
`monolayers was investigated using similar experimental
`conditions.
`
`LC-MS/MS analysis
`The concentrations of EVR and SRL in compartments A
`and B were determined using turbulent flow chroma-
`tography–tandem mass spectrometry (TFC-MS/MS). The
`LC system used for all analyses was a high turbulence
`
`ª 2011 The Authors Fundamental and Clinical Pharmacology ª 2011 Socie´ te´ Franc¸aise de Pharmacologie et de The´ rapeutique
`Fundamental & Clinical Pharmacology 26 (2012) 463–472
`
`Ex. 1047-0003
`
`

`
`466
`
`F. Lamoureux et al.
`
`liquid chromatography 2300 turbulent-flow chromato-
`graphy system (Cohesive Technologies, Milton Keynes,
`UK) equipped with a CTC HTC Pal autosampler (CTC
`Analytics AG, Zwingen, Switerland) kept at 4 °C, two
`binary high-pressure Agilent 1100 pumps (Agilent
`Technologies, Palo Alto, CA, USA), and three-six-port
`switching valves controlled by the Aria OS software
`package (Cohesive Technologies, Franklin, MA, USA).
`Specific and sensitive detection and quantitation were
`performed using a triple stage quadrupole mass spectro-
`meter Quantum Discovery tandem mass spectrometry
`system (Thermo-Fischer Scientific, Les Ulis, France)
`equipped with an orthogonal electrospray ionization
`source and controlled by the Xcalibur computer pro-
`gram. The drugs were monitored in the positive ion,
`selected reaction monitoring mode, following two tran-
`sitions per compound. Quantitation limits were 10 lg/L
`for all drugs (10.4, 10.8, 8, 12, and 13 nM/L for EVR,
`SRL, CsA, TAC, and digoxin, respectively). Excellent
`calibration curves were obtained using quadratic regres-
`sion from the limit of quantitation up to 2000 lg/L
`(2.09, 2.19, 1.66, 2.49, and 2.56 lM/L for EVR, SRL,
`CsA, TAC, and digoxin, respectively).
`
`Calculation of transport data
`The flux (J) was calculated using the following equation:
`JX ¼ dQ=dt
`where Q (pM) is the amount of drug transported within a
`given time period dt (s).
`X denotes transport direction, either absorptive (A to
`B, A fi B) or secretory (B to A, B fi A).
`Permeability was estimated by calculating the appar-
`
`
`)6 cm/s) across
`ent permeability coefficient (Papp · 10
`Caco-2 monolayers in both the A ! B Papp;A!B
`
`
`and
`B ! A Papp;B!A
`directions, according to the following
`equation:
`Papp ¼ J=ðA C0Þ
`where A is the surface area of the monolayer exposed to
`the compound (0.6-cm2) and C0 (ng/ml) the initial con-
`centration of test compound in the donor compartment.
`The polarization (Efflux) ratio (ER) was defined as:
`Efflux ratio (ER) ¼ Papp;B!A=Papp;A!B
`where Papp,BfiA and Papp,AfiB represent the apparent
`permeability coefficient
`in the B to A and A to B
`directions, respectively.
`
`Statistical analysis
`Results are presented as mean values ± SD of triplicate
`A fi B or B fi A experiments. Comparisons of Papp and
`
`ER across experiments were performed using the non-
`parametric one-tailed Mann–Whitney test with Stat-
`viewÒ (SAS Institute, Cary, NC, USA), with a level of
`significance set at 0.05.
`
`R E S U L T S
`
`Transepithelial transport of mTOR inhibitors across
`Caco-2 cell monolayers
`As shown in Figure 1a,b, respectively, EVR and SRL
`displayed a polarized transport. At all the concentrations
`tested (1, 5, and 10 lM), their Papp,BfiA was significantly
`higher than their Papp,AfiB (P = 0.003 for both EVR
`1 lM and SRL 1 lM; P = 0.05 for EVR and SRL 5 lM and
`10 lM), suggesting an apically directed efflux of ImTORs.
`The Papp,AfiB of EVR and SRL increased up to
`)6 cm/
`)6 cm/s and 2.84 ± 1.40 · 10
`1.59 ± 0.45 · 10
`s at the highest concentration (10 lM), respectively,
`while their Papp,BfiA also varied over the concentration
`range, with a trend to decrease when ImTOR concen-
`trations increased (Figure 1a,b). These variations resulted
`in a steadily decreasing efflux ratio (ER) over the
`concentration range examined (P = 0.012),
`from 9.2
`to 2.3 and from 7.15 to 2 for EVR and SRL, respectively,
`suggesting a concentration-dependent and saturable
`transepithelial
`transport of
`ImTORs across Caco-2
`monolayers. As a result of these observations, all the
`subsequent transport studies were performed at a non-
`saturating level of 1 lM for EVR and SRL.
`At this concentration, we observe no difference in
`transepithelial transport between EVR and SRL, but a
`trend toward a higher ER for EVR (P = 0.139; Table II).
`GF120918, a known chemical P-gp inhibitor, was
`used to evaluate the P-gp-mediated efflux of EVR and
`SRL in Caco-2 cells, while digoxin was employed as a
`P-gp substrate (positive control for inhibition). In the
`presence of 2 lM GF120918, the mean ER of digoxin
`decreased from 20.79 to 2.25 (P = 0.05) as the result of
`both a decreased Papp,BfiA and an increased Papp,AfiB
`(Table Ic); the Papp,BfiA of EVR and SRL (1 lM) was also
`significantly decreased (P = 0.012), but their Papp,AfiB
`was not significantly modified (Table II), resulting never-
`theless in a 70% decrease in EVR ER (P = 0.012) and a
`41% decrease in SRL ER (P = 0.13) (Figure 2a,b).
`
`Effect of Cyclosporine A and Tacrolimus on the
`transepithelial transport of ImTORs
`To confirm the inhibitory effect of CsA and TAC on P-gp,
`incubations were first performed with the P-gp substrate
`digoxin. Addition of CsA or TAC resulted in a significant
`
`ª 2011 The Authors Fundamental and Clinical Pharmacology ª 2011 Socie´ te´ Franc¸aise de Pharmacologie et de The´ rapeutique
`Fundamental & Clinical Pharmacology 26 (2012) 463–472
`
`Ex. 1047-0004
`
`

`
`mTOR inhibitors intestinal absorption and interaction with calcineurin inhibitors
`
`467
`
`Papp,A→B
`Papp,B→A
`ER
`
`Efflux ratio
`
`Efflux ratio
`
`0
`
`0
`
`024681
`
`024681
`
`12
`
`(a)
`
`**
`
`**
`
`ns
`
`EVR 1 μM
`
`EVR 5 μM
`
`EVR 10 μM
`
`**
`
`*
`
`ns
`
`SRL 1 μM
`
`SRL 5 μM
`
`SRL 10 μM
`
`0369
`
`Papp (×10–6 cm/s)
`
`12
`
`(b)
`
`0369
`
`Papp (×10–6 cm/s)
`
`Figure 1 Everolimus (a) and sirolimus
`(b) Papp,AfiB, Papp,B fi A and efflux ratio
`across Caco-2 cell monolayers over a
`1–10 lM concentration range (data
`shown as mean ± SD; results are repre-
`sentative of three independent triplicate
`experiments). Efflux ratios were
`compared using the nonparametric
`Mann–Whitney test (*P £ 0.05 and
`**P £ 0.01).
`
`Table II Effects of GF120918 (a P-gp inhibitor) and calcineurin
`inhibitors on the transepithelial transport of everolimus 1 lM (a)
`and sirolimus 1 lM (b) across Caco-2 cell monolayers. Data are
`expressed as mean ± SD and are representative of three indepen-
`dent triplicate experiments.
`
`Papp,AfiB
`)6 cm/s)
`(10
`
`Papp,BfiA
`
`(10)6 cm/s)
`
`Efflux ratio
`Papp,BfiA/Papp,AfiB
`
`0.78 ± 0.20
`
`6.76 ± 1.98
`
`9.23 ± 3.20
`
`0.99 ± 0.04
`
`2.74 ± 0.36
`
`2.77 ± 0.47**
`
`12.94 ± 1.53
`
`15.66 ± 0.74
`
`1.22 ± 0.09**
`
`2.48 ± 0.19
`
`3.11 ± 0.28
`
`1.26 ± 0.07*
`
`1.03 ± 0.58
`
`6.11 ± 1.20
`
`7.15 ± 3.05
`
`0.84 ± 0.19
`
`3.27 ± 0.56
`
`4.16 ± 1.81
`
`9.89 ± 2.58
`
`10.24 ± 1.71
`
`1.09 ± 0.32**
`
`2.76 ± 0.21
`
`2.85 ± 0.14
`
`1.04 ± 0.11*
`
`(a)
`EVR 1 lM (control)
`+GF120918 2 lM
`
`+Cyclosporine A
`10 lM
`
`+Tacrolimus
`10 lM
`
`(b)
`SRL 1 lM (control)
`+GF120918 2 lM
`
`+Cyclosporine A
`10 lM
`+Tacrolimus 10 lM
`
`*P £ 0.05 and **P £ 0.01 (measured efflux ratio vs. control).
`Papp,AfiB and Papp,BfiA: apparent permeability in the apical-to-basal and basal-
`to-apical directions, respectively, calculated according to the equation
`
`described in the 2.8 section.
`
`in digoxin
`increase
`concentration-dependent
`and
`Papp,AfiB and decrease in Papp,BfiA (Table Ia,b). At the
`highest CsA or TAC concentrations tested, the mean
`digoxin ER was 1.55 and 1.31, which is consistent with
`
`an almost complete inhibition of active efflux (Table Ia,b,
`respectively), similar to that obtained with the P-gp
`inhibitor GF120918 (2 lM) (Table Ic).
`The effects of CsA and TAC on the apically directed
`efflux of EVR and SRL are presented in Figure 2 and
`In the presence of CsA (10 lM), both the
`Table II.
`Papp,BfiA and the Papp,AfiB of EVR and SRL (1 lM) were
`significantly and markedly increased (P = 0.012 and
`P = 0.014, respectively). TAC also increased signifi-
`in contrast to
`cantly their Papp,AfiB (P = 0.014) but,
`CsA, decreased their Papp,BfiA (P = 0.012) in a signifi-
`cant manner. In both cases, these variations resulted in a
`significant and important decrease in EVR and SRL mean
`ERs (Table II).
`
`D I S C U S S I O N
`
`Using the enterocyte-like Caco-2 cell line, we found that
`EVR and SRL exhibit a polarized transport, with perme-
`ability in the basal-to-apical direction significantly high-
`er than that observed in the apical-to-basal direction. We
`also found that equal levels of CsA and TAC interacted
`with their transepithelial flux by different mechanisms.
`Few studies
`investigated the intestinal
`transport
`mechanisms of EVR or SRL [11,12,31–33]. Some of
`these studies, using either rats or cell models, indicated a
`major role of CYP3A in limiting EVR [12] or SRL [31,33]
`oral absorption. On the other hand, the role of P-gp in
`the intestinal absorbtion of ImTORs was not clearly
`
`ª 2011 The Authors Fundamental and Clinical Pharmacology ª 2011 Socie´ te´ Franc¸aise de Pharmacologie et de The´ rapeutique
`Fundamental & Clinical Pharmacology 26 (2012) 463–472
`
`Ex. 1047-0005
`
`

`
`F. Lamoureux et al.
`
`Figure 2 Effects of cyclosporine A,
`tacrolimus and the P-gp inhibitor
`GF120918 on the apically directed efflux
`of everolimus 1 lM (a) and sirolimus
`1 lM (b) (data shown as mean% diff. vs.
`control ± SD; results are representative
`of three independent triplicate experi-
`ments).
`
`+ GF120918 - 2 μM
`
`+ Tacrolimus - 10 μM
`
`+ Cyclosporine A - 10 μM
`
`(b)
`
`200
`
`100
`
`Control
`
`–100
`
`–200
`
`SRL efflux ratio (% dif. vs control)
`
`+ GF120918 - 2 μM
`
`+ Tacrolimus - 10 μM
`
`+ Cyclosporine A - 10 μM
`
`468
`
`(a)
`
`200
`
`100
`
`Control
`
`–100
`
`–200
`
`EVR efflux ratio (% dif. vs control)
`
`demonstrated. Crowe et al. [11] showed that the non-
`specific P-gp inhibitor verapamil decreases the intestinal
`transport of ImTORs across Caco-2 cells and suggested
`using a single-pass rat intestinal perfusion model that the
`permeability of EVR is higher than that of SRL. In
`contrast, a study conducted in both rats and Caco-2 cells
`[33] suggested that SRL does not undergo a P-glycopro-
`tein-mediated transport, while a third study concluded
`that, in contrast to CYP3A4, P-gp is not the major factor
`limiting SRL absorption [31]. In the present study, we
`found that the intestinal absorption of these two drugs is
`subject to an apically directed efflux with a trend toward
`a higher permeability for EVR. In addition, we showed
`using the specific inhibitor GF120918 that P-gp is
`involved in the transepithelial transport of both EVR
`and SRL and that Pgp inhibition resulted in a significant
`decrease in their ER.
`Cyclosporine A and TAC also lead to a significant
`decrease in EVR and SRL ER, which is consistent with
`the inhibition of
`their P-gp-mediated efflux. These
`findings are also consistent with previous clinical obser-
`vations [13–16] of
`increased oral bioavailability of
`ImTORs in the presence of CsA or TAC. Inhibition of
`transport at the intestinal level probably contributes, at
`least partly, to the pharmacokinetic interactions between
`ImTOR and CNI.
`involved in these drug–drug
`The mechanisms
`interactions were further investigated. The inhibitory
`effect of both TAC and CsA on P-gp could be
`demonstrated by incubation with digoxin as a specific
`substrate. CsA and TAC even markedly increased the
`transepithelial absorption flux (Papp,AfiB) of ImTORs,
`whereas no such effect was observed with GF120918
`(which did increase digoxin Papp,AfiB by 2.7-fold) prob-
`ably because of differences in their P-gp affinity. How-
`ever, inhibition of the P-gp-mediated efflux of ImTORs
`does not appear to be the only mechanism associated
`with their increased oral bioavailability in the presence
`
`of CNIs. Indeed, incubation of CsA with either EVR or
`SRL resulted in an unexpected increase of the Papp,BfiA,
`while co-incubation with the P-gp inhibitor GF120918
`decreased this apparent permeability. CsA did not show
`such an effect when incubated with the P-gp substrate
`digoxin. Altogether, this suggests that other mechanisms
`than solely P-gp-mediated efflux might limit SRL and
`EVR intestinal bioavailability. Notably, the TAC inhibi-
`tion profile observed here was very different from that of
`CsA and closer to that of GF120918, both exhibiting a
`decrease in Papp,BfiA. Our data indicate that TAC has a
`different effect on ImTORs intestinal transport, especially
`in the basal-to-apical direction.
`Considering that CsA is a potent inhibitor of CYP3A,
`especially CYP3A4 [12,31,33,34], it can be hypothe-
`sized that its effect on the transepithelial flux of ImTORs
`implies a dual
`inhibition of
`their apical efflux and
`metabolism (Figure 3). Lecointre et al. [34] reported that
`TAC had no effect on any CYP at concentrations below
`1 lM, while at higher concentrations,
`it had only a
`moderate ability to inhibit the two major intestinal
`isoforms CYP3A4 and 3A5.
`A differential inhibition of ImTORs CYP450 metabo-
`lism by CsA and TAC could explain the increased
`apparent permeability of ImTORs in both A to B and B
`to A directions when associated with CsA, owing to less
`drug lost in transcellular transfers. However, it has been
`reported that CYP3A4 is poorly expressed in native
`Caco-2 cells that have not been pretreated by differen-
`tiating agents, such as 1a,25-dihydroxyvitamin D3 and
`vitamin D analogs [35,36]. Therefore,
`the different
`inhibition profiles of CsA and TAC observed here using
`untreated caco-2 cells are unlikely to be as a result of
`their differential inhibition of CYP3A.
`On the other hand, CsA is a known inhibitor of efflux
`transporters other than P-gp such as MRPs and breast
`cancer resistance protein (BCRP) [37,38] also expressed
`in Caco-2 cells [39], as well as of influx transporters such
`
`ª 2011 The Authors Fundamental and Clinical Pharmacology ª 2011 Socie´ te´ Franc¸aise de Pharmacologie et de The´ rapeutique
`Fundamental & Clinical Pharmacology 26 (2012) 463–472
`
`Ex. 1047-0006
`
`

`
`mTOR inhibitors intestinal absorption and interaction with calcineurin inhibitors
`
`469
`
`Apical chamber (A)
`(a)
`
`Everolimus
`
`Caco-2 cells
`monolayer
`
`Porous filter
`
`P-gp
`
`MRPs
`-1/-3/-5/-6
`
`?
`
`Basolateral chamber (B)
`
`Apical chamber (A)
`(b)
`
`Everolimus
`
`Sirolimus
`
`PA(cid:198)B
`
`?
`
`?
`
`MRP2
`
`BCRP
`
`CYP3A
`
`PA(cid:198)B<< PB(cid:198)A
`
`PB(cid:198)A
`
`+ Tacrolimus (TAC)
`
`PA(cid:198)B
`
`Sirolimus
`
`Caco-2 cells
`monolayer
`
`Porous filter
`
`TAC
`
`–
`
`P-gp
`
`MRPs
`-1/-3/-5/-6
`
`?
`
`Basolateral chamber (B)
`
`Apical chamber (A)
`(c)
`
`Everolimus
`
`CsA
`
`–
`
`P-gp
`
`CsA
`–
`?
`
`MRPs
`-1/-3/-5/-6
`
`Caco-2 cells
`monolayer
`
`Porous filter
`
`Basolateral chamber (B)
`
`?
`
`?
`
`MRP2
`
`BCRP
`
`CYP3A
`
`↘
`
`(PB(cid:198)A/ PA(cid:198)B)
`
`PB(cid:198)A
`
`+ Cyclosporine A (CsA)
`
`Sirolimus
`
`PA(cid:198)B
`
`CsA
`
`–
`
`?
`
`–
`
`?
`
`MRP2
`
`BCRP
`
`CsA
`
`-
`
`CYP3A
`
`↘
`
`PB(cid:198)A≈ PA(cid:198)B
`(PB(cid:198)A/ PA(cid:198)B)
`PB(cid:198)A
`
`Figure 3 Schematic overview of the
`study design and the potential mecha-
`nisms implicated in the transport of
`mTOR inhibitors (ImTORs) across Caco-
`2 cells (a), ImTORs in combination with
`tacrolimus (b), and ImTORs in combi-
`nation with cyclosporine A (c).
`
`ª 2011 The Authors Fundamental and Clinical Pharmacology ª 2011 Socie´ te´ Franc¸aise de Pharmacologie et de The´ rapeutique
`Fundamental & Clinical Pharmacology 26 (2012) 463–472
`
`Ex. 1047-0007
`
`

`
`470
`
`F. Lamoureux et al.
`
`as OATPs. We recently showed that ImTORs are not
`transported by OATPs and are not subject to uptake
`mechanisms in Caco-2 cells (which mainly express
`OATP2B1) [9]; therefore, we do not expect OATPs to be
`involved in the ImTORs/CNIs interaction observed here.
`MRP efflux transporters are abundantly expressed in the
`intestine and have been reported to be involved in the
`transport of a number of drugs [40]. MRP-2 and -4, like
`P-gp and BCRP, are localized at the apical side of Caco-2
`cells, whereas MRP-1, -3, -5, and -6 are expressed at the
`basal side of the cells (Figure 3) [40]. Inhibition of apical
`MRPs by CsA would further increase the Papp,AfiB but not
`the Papp,BfiA of ImTORs. In contrast, inhibition of efflux
`transporters located on the basal side of Caco-2 mono-
`layers may explain the increased Papp,BfiA of ImTORS
`observed with CsA. EVR and SRL total concentrations in
`both apical and basal compartments were significantly
`increased by CsA (2.36- and 2.71-fold, respectively;
`P = 0.05) but were not significantly affected by TAC
`(data not shown). These data are compatible with a
`decrease in ImTORs metabolism and/or a decreased basal
`efflux. Altogether, these data suggest that CsA and TAC
`exert a different effect on ImTORs transport across Caco-2
`cells, probably by a different ability to inhibit basal efflux
`transporters in Caco-2 cells. There is currently no
`evidence that these transporters are involved in ImTORs
`intestinal absorption, and further studies should be
`conducted to clarify their possible role.
`Finally, we did not investigate the potential influence
`of ImTORs on CsA or TAC transport across Caco-2 cells.
`However, several clinical studies previously reported that
`simultaneous administration of ImTORs with cyclospor-
`ine or TAC had no significant influence on the overall
`pharmacokinetic and blood levels of
`cyclosporine
`[13,41–43] or TAC [16].
`
`C O N C L U S I O N
`
`In summary, our results are compatible with clinical
`observations of an increased oral bioavailability of
`ImTORs when combined with CsA or TAC. We provide
`evidence that the intestinal permeability of both SRL and
`EVR is influenced by P-gp and that CsA and TAC exert
`different effects on the transepithelial flux of ImTORs:
`CsA apparently inhibits apical and basal efflux trans-
`porters of ImTORs, while TAC appears to inhibit apically
`directed efflux only. Although further studies are needed
`to clarify the involvement of such transporters, these
`results provide a novel
`insight concerning the CNIs/
`ImTORs interaction at the intestinal level.
`
`A C K N O W L E D G E M E N T S
`
`This study was funded by the Limousin Regional
`Council, University of Limoges and Limoges University
`Hospital. We thank Jean-Herve´ Comte and Karine
`Deleaune for their excellent technical assistance.
`
`A B B R E V I A T I O N S L I S T
`
`CNI – calcineurin inhibitor
`CsA – cyclosporine A
`CYP – P450 cytochromes
`ER – efflux ratio
`EVR – everolimus
`ImTOR – mTOR inhibitor
`mTOR – mammalian target of rapamycin
`MRP – multidrug resistance-associated protein
`Papp,AfiB and Papp,BfiA – apparent permeability coeffi-
`cient in the apical-to-basal direction (and respectively in
`the basal-to-apical direction)
`P-gp – P-glycoprotein
`SRL – sirolimus
`TAC – tacrolimus
`TEER – transepithelial electrical resistance.
`
`R E F E R E N C E S
`
`1 Fahr A. Cyclosporin clinical pharmacokinetics. Clin. Pharma-
`cokinet. (1993) 24 472–495.
`2 Venkataramanan R., Swaminathan A., Prasad T. et al. Clinical
`pharmacokinetics of tacrolimus. Clin. Pharmacokinet. (1995)
`29 404–430.
`3 Wacher V.J., Salphati L., Benet L.Z. Active secretion and
`enterocytic drug metabolism barriers to drug absorption. Adv.
`Drug Deliv. Rev. (2001) 46 89–102.
`4 Kamdem L.K., Streit F., Zanger U.M. et al. Contribution of
`CYP3A5 to the in vitro hepatic clearance of tacrolimus. Clin.
`Chem. (2005) 51 1374–1381.
`5 Kronbach T., Fischer V., Meyer U.A. Cyclosporine metabolism
`in human liver: identification of a cytochrome P-450III gene
`family as the major cyclosporine-metabolizing enzyme explains
`interactions of cyclosporine with other drugs. Clin. Pharmacol.
`Ther. (1988) 43 630–635.
`6 Sattler M., Guengerich F.P., Yun C.H., Christians U., Sewing
`K.F. Cytochrome P-450 3A enzymes are responsible for
`biotransformation of FK506 and rapamycin in man and rat.
`Drug Metab. Dispos. (1992) 20 753–761.
`7 Laplante A., Demeule M., Murphy G.F., Beliveau R. Interaction of
`immunosuppressive agents rapamycin and its analogue SDZ-RAD
`with endothelial P-gp. Transplant. Proc. (2002) 34 3393–3395.
`8 Saeki T., Ueda K., Tanigawara Y., H