|Home | About | Journals | Submit | Contact Us | Français|
Fusidic acid (FA), which was approved in the 1960s in many European and Asian countries, has gained renewed interest due to its continued effectiveness against methicillin-resistant Staphylococcus aureus. As rhabdomyolysis has been reported upon coadministration of FA with statins, we aimed to elucidate the underlying molecular mechanisms that contribute to FA-statin drug-drug interactions. Because of the association between rhabdomyolysis and increased exposure to statins, we investigated if cytochrome P450 (CYP) enzymes and transporters involved in the disposition of various statins are inhibited by FA. FA was found to inhibit BCRP and OATP1B1 but not P-gp. In overexpressing cell systems, FA inhibited BCRP-mediated efflux (50% inhibitory concentration [IC50], ~50 to 110 μM) and OATP1B1-mediated uptake (IC50, ~4 to 35 μM) of statins at clinically relevant concentrations achievable in the intestine and liver (based on a 550-mg oral dose of FA, the expected maximum theoretical gastrointestinal concentration is ~4 mM, and the maximum total or unbound concentration in the inlet to the liver was reported to be up to 223 μM or 11 μM, respectively, upon multiple dosing). Similarly, FA inhibited metabolism of statins in human liver microsomes (IC50, ~17 to 195 μM). These data suggest that FA inhibits at least 3 major dispositional pathways (BCRP, OATP1B1, and CYP3A) and thus affects the clearance of several statins. We confirmed that FA is eliminated via phase 1 metabolism (primarily via CYP3A); however, there is also some phase 2 metabolism (mediated primarily by UGT1A1). Taken together, these data provide evidence for molecular mechanisms that may explain the occurrence of rhabdomyolysis when FA is administered with statins.
The evolution of drug-resistant bacteria along with shrinking investments in antimicrobial drug research from pharmaceutical companies has renewed interest in older antibiotics. The Antimicrobial Availability Task Force (AATF) of the Infectious Diseases Society of America (IDSA) has emphasized the urgency to address this public health concern and highlighted methicillin-resistant Staphylococcus aureus (MRSA) as one of the major resistant pathogens without enough treatment options (1). Fusidic acid (FA) has been used to treat MRSA in many countries for several decades due to continued low resistance rates (2).
Cases of drug-drug interactions (DDIs) have been reported upon FA coadministration with drugs of various classes, such as HIV protease inhibitors (ritonavir and saquinavir), immunosuppressants (cyclosporine [CsA]), oral anticoagulants (such as coumarin derivatives), and statins (atorvastatin and simvastatin) (3,–5). While the actual number of cases reported is small, the usage of FA is expected to increase given its continued effectiveness against resistant bacterial strains, with potential for a concomitant rise in DDIs.
Approximately 30 cases of DDIs between FA and statins have been reported in the literature (3, 6,–11). Because the majority of the reports involved atorvastatin and simvastatin, which are known to be cleared primarily by cytochrome P450 3A4 (CYP3A4), and since there is some evidence of involvement of CYP3A4 in FA disposition, it was assumed that FA reduces the clearance of statins via inhibition of CYP3A4 (9,–11). Accordingly, it was suggested that statins which are primarily eliminated by CYP3A4 (atorvastatin and simvastatin) are more likely to result in DDIs when coadministered with FA than statins with minimal elimination by CYP3A4 (rosuvastatin, pravastatin, and fluvastatin) (10). However, there have been two reports with rosuvastatin (3, 7), and alternative mechanisms for the DDIs have been proposed (12,–14). This uncertainty poses a challenge for the coadministration of FA with statins. The Pharmacovigilance Working Party (PhVWP) of the European Medicines Agency (EMA) included a warning against concomitant use of FA with statins in their July 2011 report (http://www.ema.europa.eu/docs/en_GB/document_library/Report/2011/07/WC500109581.pdf).
Given this uncertainty, we set out to identify the molecular mechanism(s) involved in the interactions between FA and statins.
(Part of this work was presented at the Gordon Research Conference, Holderness, NH, July 2010.)
The materials used in this study are described in the supplemental material. In the following experiments, all conditions were optimized and the concentrations chosen with consideration of the clinical relevance and assay conditions, including analytical sensitivity.
MDCK-MDR1 cells (obtained from NKI) and MDCK-breast cancer resistance protein (BCRP) cells were used for p-glycoprotein (P-gp) and BCRP inhibition, respectively. MDCK-BCRP cells were generated by stably transfecting an ABCG2-expressing plasmid into MDCK-II cells, as described previously (15). Transwell experiments were conducted as described previously (16), with minor modifications. Cells (0.25 million cells/well) were grown for 4 to 5 days. Final incubation conditions were set after an initial optimization. Statins, prazosin, or labetalol (2 μM) were dosed in the apical (A → B transport) or basolateral (B → A transport) chamber, and flux across the monolayer was measured. For inhibition experiments, buffer containing CsA (10 μM) or FA (0 to 1,000 μM) was added. Plates were incubated on a shaker at 70 rpm for 120 min at 37°C and 5% CO2. Samples (100 μl) were aliquoted from the apical and basolateral chambers before and after the incubation, precipitated with 250 μl of ice-cold acetonitrile (containing 0.1% formic acid and 250 μg/ml of mevastatin or carbutamide as an internal standard), and retained for liquid chromatography with tandem mass spectrometry detection (LC-MS/MS) (see Table S1 in the supplemental material). The concentration at 120 min was estimated and compared to that at time zero. At the end of the experiment, a lucifer yellow assay was run to determine membrane integrity.
HEK293 cells expressing human OATP1B1 (TransportoCells) and their respective vector control cells were purchased from Corning Life Sciences (Bedford, MA). Cells (0.4 million cells/well) were grown per the manufacturer's recommendations. After preincubation, uptake was initiated by adding 0.3 ml of prewarmed buffer containing 1 μM test compound (statin with or without FA). At the end of the incubation period (5 min), the cells were washed twice with 0.5 ml of prechilled Hank's balanced salt solution (HBSS), the buffer was removed, and 0.4 ml of precipitation solution (100% acetonitrile containing 0.1% formic acid and 250 μg/ml of mevastatin or carbutamide as an internal standard) was added to each well. The plates were incubated at room temperature for approximately 20 min, and cell lysates were harvested and centrifuged. Samples were analyzed by LC-MS/MS (see Table S1 in the supplemental material). Appropriate controls were run based on the manufacturer's recommendations.
Fresh human hepatocyte suspensions were obtained from Invitrogen (Durham, NC). Cell viability was ensured to be >85% for every experiment. [3H]estrone-3-sulfate (E3S) and [3H]rosuvastatin were used as OATP substrates, [3H]propranolol was used as a passive permeability marker, and CsA was used as an inhibitor of active uptake (17, 18).
Hepatocytes were resuspended in modified Williams' E medium at a concentration of 2 × 106 cells/ml. For each reaction, 350 μl of cell suspension was preincubated in the absence or presence of the OATP inhibitor CsA (10 μM) or various concentrations of FA at 37°C for 5 min. The reaction was initiated by adding an equal volume of modified Williams' E medium containing 4 μM substrate to achieve a final concentration of 2 μM. From the reaction mixture, aliquots of 100 μl were transferred to tubes containing a mixture of silicone and mineral oil layered over 1 M potassium hydroxide at various time points (10 to 40 s), and the tubes were immediately centrifuged, snap-frozen in liquid nitrogen, and stored at −80°C overnight. Radioactivity in the pellet was determined using liquid scintillation counting.
Compound stock solutions were prepared in dimethyl sulfoxide, while FA was prepared in prewarmed methanol. The final concentration of organic solvent in the incubation medium was <1% and was matched in controls. Simvastatin (1 μM) was incubated with 0.2 mg/ml of human liver microsomes (HLM), while all other compounds were incubated at 1 μM with 1 mg/ml HLM or 40 pmol recombinant CYP3A4 (rCYP3A4). The final incubation conditions were set after an initial optimization. Incubations were carried out in triplicate in a total volume of 1 ml containing 100 mM potassium phosphate buffer (pH 7.4), Milli-Q water, and 5 mM MgCl2. FA was added at varying concentrations from 0 to 1,000 μM. After a 5-min preincubation, the reaction was initiated by the addition of 1 mM NADPH, and the mixtures were incubated at 37°C. At various time points (0 to 60 min), 0.05-ml aliquots were removed and precipitated with 0.125 ml of ice-cold acetonitrile containing 0.1% formic acid and either mevastatin or carbutamide as an internal standard. Samples were centrifuged for 5 min, and supernatants were diluted with water (1:2 [vol/vol]) and analyzed by LC-MS/MS. Appropriate controls were run for each assay.
The method for determination of FA metabolism with HLM and individual human recombinant CYP or UGT isoforms is described in the supplemental material.
For BCRP- and P-gp-mediated flux in transwell assays, the flux/apparent permeability coefficient (Papp) values were calculated using the following equation: Papp = Cr × Vr/(t × A × C0), where Vr is the volume and Cr the concentration measured in the receiver compartment at the end of incubation, A is the surface area of the membrane (in square centimeters), t is the time of incubation (in seconds), and C0 is the initial concentration. The efflux ratio (ER) was calculated as the ratio of Papp values, as follows: ER = Papp, B → A/Papp, A → B.
For OATP1B1-mediated uptake assays in TransportoCells, uptake measured in the absence of inhibitor was considered maximal, while uptake in the presence of various concentrations of the inhibitor was expressed as a percentage relative to the maximum amount. Half-maximal inhibitory concentrations (IC50s) were calculated from this sigmoid curve.
Additionally, the uptake ratio was calculated as follows: uptake ratio = amount of compound accumulated in OATP1B1-expressing cells/amount of compound accumulated in vector control cells.
For HLM and individually expressed P450 and UGT isozyme assays, intrinsic clearance (CLint) values were calculated by using the following formula: CLint (microliters per minute per milligram of protein) = (V × 0.693)/t1/2.
IC50 values were obtained using GraphPad Prism 5 (GraphPad Software, San Diego, CA). The Hill equation was fit to the relative rate of drug accumulation as a function of inhibitor concentration by using nonlinear regression analysis. The mean IC50 and correlation coefficient (R2) for each substrate-inhibitor pair were determined from at least 2 independent experiments, with 2 or 3 biological replicates in each experiment.
Unless otherwise stated, data are presented as means ± standard deviations (SD). Mean differences were tested with Student's paired t test, with P values of <0.05 considered significant. No adjustments for multiplicity were made.
In the bidirectional transport assay systems, prazosin and labetalol were used as substrates of BCRP and P-gp, respectively. As shown in Fig. 1, the mean ER obtained for prazosin in MDCK-BCRP cells was approximately 7.0, which was reduced to a mean of approximately 3.0 with the addition of 10 μM CsA, a known BCRP inhibitor, confirming the optimal activity of BCRP in the transwell system. Similarly, FA inhibited prazosin transport in a concentration-dependent manner, with a concentration of 100 μM achieving a level of inhibition similar to that with 10 μM CsA. The highest concentration of FA tested (250 μM) resulted in nearly complete inhibition of BCRP. The mean ER obtained for labetalol, a P-gp substrate in the MDCK-MDR1 cells, was approximately 8.0 (Fig. 1), and it remained unaffected in the presence of FA (up to 250 μM). Positive and negative controls (MDCK-WT cells) performed as expected in the assay.
As BCRP is known to be involved in statin disposition (19), inhibition of BCRP-mediated statin transport was tested in vitro. As shown in Fig. 2, mean ERs in the absence of FA were as follows: for pitavastatin, 23.1; for fluvastatin, 17.2; for rosuvastatin, 8.1; and for atorvastatin, 7.4. BCRP-mediated efflux was inhibited by FA in a concentration-dependent manner for all the statins tested. Based on the IC50s, the greatest FA inhibition potency was obtained with atorvastatin (52 μM), and the lowest was obtained with pitavastatin (111.8 μM).
The uptake ratios obtained for various statins in the absence of FA in OATP1B1-overexpressing cells were as follows: for pitavastatin, 45.7; for rosuvastatin, 35.3; for atorvastatin, 18.5; for fluvastatin, 9.2; for cerivastatin, 2.6; and for simvastatin, 1.8. In the wild-type control cells, atorvastatin, fluvastatin, pitavastatin, and rosuvastatin accumulated at <10% of the amounts that accumulated in OATP1B1-overexpressing cells, which shows that there was insignificant passive transport. These data confirmed that the uptake observed in the overexpressing cells was primarily OATP1B1 mediated. However, for cerivastatin, a known substrate of OATP1B1 (20), a relatively large amount of transport was observed in both vector control and OATP1B1-expressing cells, resulting in an uptake ratio of 2.6, which did not decline with the addition of FA (3 to 1,000 μM) (Fig. 3). Simvastatin accumulated to a level only 1.8-fold higher in OATP1B1-overexpressing cells than in vector control cells, but the accumulation declined significantly with the addition of FA, in a concentration-dependent manner (Fig. 3). The lack of significantly more OATP1B1-mediated accumulation of simvastatin is consistent with previously published results (21) by which investigators were unable to demonstrate more OATP1B1-mediated transport of simvastatin in vitro, for reasons that are unclear. For fluvastatin, FA at 100 μM inhibited OATP1B1 uptake by 50%, with a plateau reached at this concentration.
Unlike that of cerivastatin, OATP1B1-mediated uptake of atorvastatin, fluvastatin, pitavastatin, simvastatin, and rosuvastatin (Fig. 3) was inhibited in the presence of FA, in a concentration-dependent manner, with maximal inhibition achieved at approximately 100 μM for most of the statins, except for simvastatin, which showed the most inhibition at 1,000 μM. Based on the calculated IC50s, the greatest FA inhibition potency was obtained with rosuvastatin (~4 μM), and the lowest was seen with simvastatin (~34 μM).
To confirm the inhibition of uptake transporters (at least for OATP1B1) by FA in the native system, we used isolated fresh human hepatocytes. This study was limited to measuring the effects of FA on the uptake of an OATP1B1 substrate, [3H]E3S, and an uptake-dependent statin relatively selective for hepatic OATP transporters, [3H]rosuvastatin, by using the oil spin method. The total uptake amount for each of the substrates varied between the various hepatocyte batches due to interindividual variability. Hence, data from a representative experiment are presented. When hepatocytes were incubated at 37°C, there was an increase in the uptake of both [3H]E3S and [3H]rosuvastatin (Fig. 4) as the exposure time increased (10 to 40 s). In the presence of either CsA (10 μM) or FA (100 μM), up to 4-fold reductions in the uptake of [3H]E3S were observed (Fig. 4). Incubation at 4°C showed approximately 10-fold less uptake of [3H]E3S, which did not increase over 40 s and was unaffected in the presence of CsA and FA (Fig. 4), confirming that the uptake observed at 37°C was dependent on active transport. Furthermore, concentration-dependent decreases in the uptake of [3H]E3S and [3H]rosuvastatin were observed in the presence of FA (Fig. 4). The IC50s of FA for the uptake of [3H]E3S and [3H]rosuvastatin were estimated to be similar, i.e., 24 μM versus 23 μM, respectively (Fig. 4). Uptake of [3H]propranolol, used as a passive marker, was significantly higher at 37°C than at 4°C, but it was unaffected in the presence of the nonspecific active uptake inhibitor CsA, suggesting that the increase in passive uptake of propranolol quite likely was temperature dependent. Furthermore, addition of FA did not affect the uptake of propranolol (Fig. 4).
To understand the specific enzymatic pathways involved in FA elimination, reaction phenotyping was performed with individually expressed human recombinant enzyme systems and HLM by measuring parent disappearance. As shown in Fig. 5, among the cytochrome P450s tested, the highest loss at the end of the incubation (30 min) was observed with rCYP3A4 (~60%), followed by rCYP3A5 (~40%), while rCYP2C9 and rCYP2C19 incubations showed only approximately 15% loss. Negligible metabolism was observed for all the other P450s. These data suggest that the clearance of FA is largely dependent on CYP3A4/CYP3A5 among the P450 enzymes. The percentage of FA remaining after incubation with HLM was approximately 50%, which is also consistent with the values observed with rCYP3A4 and rCYP3A5. For UGTs, the loss of FA appeared to be negligible for all the UGT isoforms included in the experiment except UGT1A1, which showed an average loss of 23% compared to the control (Fig. 5). HLM incubated with alamethicin and UDPGA showed similar losses. While CLint cannot be compared directly between HLM and individually expressed enzymes, given that the unbound fraction (fu inc) has not been calculated for the 2 systems, these data provide an early indication of the involvement of phase 1 metabolism (mediated mainly by CYP3A4 and CYP3A5) in the elimination of FA; however, phase 2 metabolism (mediated by UGT1A1) also plays a minor role in its metabolism.
Cytochrome P450s are involved in the metabolism of some statins (22). Intrinsic clearance of various statins was measured in the absence and presence of increasing concentrations of FA in HLM (Fig. 6). FA was found to decrease the intrinsic clearance of various statins in a dose-dependent manner. Based on the IC50s obtained, the order of decreasing FA potency was as follows: atorvastatin (~18 μM) = cerivastatin (~17 μM) > fluvastatin (~82 μM) > simvastatin (~195 μM).
Given that CYP3A4 appeared to be the major pathway involved in FA metabolism (Fig. 5), we further confirmed if individually expressing recombinant CYP3A4 would demonstrate FA-mediated inhibition of statin metabolism.
Inhibition studies were performed using rCYP3A4 (Fig. 7). We observed inhibition of CYP3A4-mediated intrinsic clearance for both atorvastatin and cerivastatin by addition of increasing concentrations of FA, confirming that CYP3A4 is, at least in part, the likely pathway for interaction. The estimated IC50 was higher with rCYP3A4 (Fig. 7) than with HLM (Fig. 6), with the following order of the statins (for the recombinant system): atorvastatin (75 μM) > cerivastatin (106 μM). Note that with both the rCYP3A4 system and HLM, complete inhibition of cerivastatin intrinsic clearance was not observed. Rosuvastatin is not a substrate for CYP3A4, and as expected, there was no measurable intrinsic clearance for rosuvastatin with either HLM or rCYP3A4, and it was unaffected in the presence of FA (Fig. 7).
The aim of this study was to systematically investigate the mechanism(s) involved in FA-statin interaction. We provide direct evidence suggesting CYP3A4/5-mediated metabolism as the major clearance pathway for FA elimination (Fig. 5). We also show that other enzymes, such as UGT1A1, also expressed in the liver and intestine, may play a role in FA metabolism, but to a lesser extent (Fig. 5). FA is highly protein bound (~98%) and is known to exhibit a complex and nonlinear pharmacokinetic (PK) profile resulting in accumulation upon multiple dosing by autoinhibition of CYP3A4-mediated clearance (23,–25). Consequently, DDIs should be assessed based on exposures following multiple dosing. Added to this is the complication that FA may be administered with rifampin in clinical practice to minimize resistance emergence; rifampin, a potent inducer of CYP enzymes, including CYP3A4, has been shown to reduce the plasma exposure of FA upon concomitant use (26).
Inhibition of intestinal CYP3A4/5 due to a high intestinal concentration of FA would result in increased oral bioavailability of drugs that are metabolized by intestinal CYP3A4/5. The free drug concentration in the gut (shown to be relevant for DDI assessment) (27) is difficult to estimate, but the total intestinal concentration (Igut; calculated as the dose in 250 ml of fluid) of FA was estimated to be up to 4 mM following oral dosing (considerably higher than the IC50 for CYP3A in HLM or the recombinant system). Using HLM, we demonstrated the FA-mediated inhibition of metabolism of both simvastatin and atorvastatin, both known to be metabolized extensively by intestinal CYP3A4, with IC50s much below the estimated Igut value (and unbound hepatic inlet concentration in the case of atorvastatin) that would be expected upon oral dosing of FA in the clinic (Table 1). Hence, an interaction can be expected via both liver and gut metabolism. If we consider the unbound plasma concentration, an interaction may not be expected upon initial dosing, but with each subsequent dose, the unbound plasma concentration of FA will continue to increase due to accumulation, potentially resulting in an interaction upon repeat dosing. Notably, using the recombinant system (Fig. 7), FA IC50s for CYP3A4 were 75 and 106 μM for atorvastatin and cerivastatin, respectively, which are considerably higher than those observed using HLM (17.8 and 17.2 μM, respectively) (Fig. 6). The reasons for this discrepancy are currently unclear.
In addition to metabolic enzymes, it is known that transporters, such as BCRP, play an important role in limiting the absorption of statins, such as atorvastatin, fluvastatin, and rosuvastatin (19, 28). Results from our in vitro study show that FA inhibits BCRP-mediated transport of statins, with IC50s ranging from approximately 50 to 110 μM (Fig. 2), and P450-mediated metabolism, with IC50s ranging from 17 with cerivastatin to 200 μM with simvastatin (Fig. 6 and and7).7). Table 2 provides a summary of the FA IC50s for each of the statins tested in the BCRP, OATP1B1, and HLM systems, based on the data presented in Fig. 2, ,3,3, and and66 and on an assessment of the likely risk of DDIs. Based on the [I2]/IC50 criterion of >10 recommended by the International Transporter Consortium (29), a gut interaction with FA would be expected for the statins that rely on BCRP and CYP3A4 (e.g., atorvastatin, simvastatin, and pitavastatin) (22), resulting in limited oral absorption. However, such an interaction would result in only modest increases (~1.6- to 2.9-fold) in the area under the concentration-time curve (AUC) for these statins (19). Given that hepatic metabolic enzymes and transporters are the predominant dispositional pathways for most statins, one would expect that inhibition of these pathways would contribute to the observed DDIs and that the risk associated with the various statins would depend on the extent of the specific pathway(s) inhibited. OATP1B1-dependent hepatic uptake is another major mechanism involved in the disposition of the majority of the statins. The FA IC50s obtained for OATP1B1-overexpressing cells ranged from 4 for rosuvastatin to 34 μM for simvastatin (Fig. 3), indicating that FA can inhibit OATP1B1-dependent transport of the statins. The FA IC50 (23 μM) against rosuvastatin uptake in fresh human hepatocytes (Fig. 4) was found to be 6-fold higher than that in the overexpressed system. The reasons for this discrepancy are unclear and further investigation into the clinical relevance of these findings is required. It is well known that the IC50 obtained can depend on many variables. While variability is common, the reasons for this are not well understood (30).
Since the majority of the statin-FA DDI case reports involve multiple oral dosing of FA at 500 mg twice a day (b.i.d.) or three times a day (t.i.d.), we performed our DDI analysis based on first-dose and steady-state plasma concentrations (or exposures) achieved using a similar dosing regimen (31), comparing PK after the 1st and 11th FA doses (Table 2). Importantly, we took a conservative approach, utilizing the IC50s obtained with the overexpressed system for OATP1B1-mediated inhibition and those obtained with HLM for P450-mediated inhibition. From these analyses, it is evident that upon single dosing, the plasma or unbound hepatic inlet concentration (~4 μM) may be high enough to inhibit OATP1B1 and P450s, but upon multiple dosing, accumulation of FA will lead to unbound plasma or hepatic inlet levels high enough to inhibit BCRP as well. Consequently, all the statins tested (atorvastatin, simvastatin, rosuvastatin, pitavastatin, fluvastatin, and cerivastatin) are likely susceptible to DDIs, with dependency on the fraction eliminated via BCRP, OATP1B1, or CYP3A4. Given that our study shows that FA is a triple inhibitor (of BCRP, OATP1B1, and CYP3A4) and based on a previously performed analysis (19), the level of interaction may be predicted depending on the fraction of statins eliminated via a specific pathway. Eng et al. (32) suggested, in contrast to our study, that there is no inhibitory effect of FA on BCRP and only a weak inhibition of CYP3A4 activity. The reason for the discrepancy in the reported BCRP IC50 in comparison with our data is unclear. However, the discrepancy in CYP3A4 inhibition may be because the authors tested CYP3A4 inhibition by using midazolam as the probe substrate. We obtained similar results of weak inhibition of CYP3A4 by FA when it was tested against midazolam as well as testosterone (data not shown). In contrast, potent inhibition by FA was observed when it was tested for direct inhibition with various statins. Further, the results from our studies are consistent with literature data on the FA interaction with CYP3A4. The reasons for these differences need to be investigated further. While further modeling efforts may offer more quantitative prediction or a clinical DDI study may ultimately show the extent of DDI, our analyses provide early hints for such interaction.
FA is sold in 23 countries (excluding the United States), with estimated annual prescriptions of approximately 21.3 million, of which 1.3 million are for oral use (33). Statin prescription is also anticipated to increase with patent expirations (34) and changes in guidelines (35). The fact that FA inhibits multiple dispositional pathways highlights the importance of managing the use of FA cautiously due to the potential for DDIs, particularly with statins and HIV protease inhibitors (5). In 2011, the United Kingdom's Medicines and Healthcare Products Regulatory Agency issued a drug safety update warning that systemic FA (Fucidin) should not be given with statins because of a risk of serious and potentially fatal rhabdomyolysis (https://www.gov.uk/drug-safety-update/systemic-fusidic-acid-and-interaction-with-statins). Prescribing information for FA and some statins now acknowledges the potential for DDI. Further studies are required to understand the impact of FA coadministration with other classes of compounds that are substrates of CYP3A4, BCRP, and OATP1B1.
In summary, FA interacts with statins via multiple dispositional pathways, such as BCRP, OATP1B1, and CYP3A4 pathways. This is the first time, to our knowledge, that multiple molecular mechanisms involved in the interaction between various statins and FA have been elucidated.
We dedicate this article to the memory of Scott W. Grimm (1961 to 2014), Executive Director of DMPK, Infection IMED Unit, AstraZeneca. Scott will be remembered for his passion to bring new medicines to the market and to mentor younger scientists to withstand the challenges of drug discovery. In addition to his great scientific contributions, his constant perseverance, support, and guidance enabled the completion of this work.
We thank Ricardo A. Luzietti, Jr., formerly of Drug Metabolism and Pharmacokinetics, Infection Innovative Medicines Unit, AstraZeneca R&D Boston, Waltham, MA, for his contribution to this work.
We acknowledge support from AstraZeneca for the conduct of these studies and Karen McFadden of AstraZeneca, Wilmington, DE, for editorial suggestions during the early stage of manuscript development. Editorial support during the later stages of manuscript development was provided by Valerie Moss, Prime Medica Ltd., Knutsford, United Kingdom, funded by AstraZeneca.
The design and conduct of the study, analysis of the study data, and opinions, conclusions, and interpretation of the data are the responsibility of the authors.
A.G., J.J.H., and S.W.G. participated in research. J.J.H., A.G., J.L., and J.P.B. conducted experiments. A.G., J.J.H., and B.K.B. performed data analysis and interpretation. A.G., J.J.H., B.K.B., S.W.G., J.L., and J.P.B. wrote the manuscript/revised it critically for important intellectual content.
All authors were employed by AstraZeneca at the time of writing of the manuscript.
This article is dedicated to Scott W. Grimm.
Supplemental material for this article may be found at http://dx.doi.org/10.1128/AAC.01335-16.