Abstract
Coleus forskohlii Briq. has been used traditionally for the treatment of several ailments since antiquity in Ayurveda. In the present study, an approach has been made to evaluate the effect of C. forskohlii and its major constituents on cytochrome P450 (CYP3A, CYP2B, and CYP2C) mRNA expression in rat hepatocytes. To gain better understanding of the herb–drug interaction potential of the chemical constituents present in C. forskohlii, the extract was subjected to column chromatography followed by standardization with respect to forskolin, 1-deoxyforskolin, and 1,9-dideoxyforskolin using reversed-phase high-performance liquid chromatography (RP-HPLC). Hepatocytes were treated with extracts, fractions, and phytoconstituents, followed by extraction and purification of total mRNA. Study of mRNA expression was carried out through reverse transcription polymerase chain reaction, followed by agarose gel electrophoresis. Results revealed that the test substances did not show any significant mRNA expression compared to the control against CYP3A, CYP2B, and CYP2C. Positive controls such as dexamethasone and rifampin showed significantly high (p < 0.001) induction potential compared to the control. It can be concluded that C. forskohlii and its major constituents may not be involved in CYP450 induction-based drug interaction.
Keywords: Coleus forskohlii, Cytochrome P450, Drug interaction, mRNA expression, Standardization
Graphical abstract
1. Introduction
Coleus forskohlii Briq. (family: Lamiaceae) has a very long history of use in many traditional herbal medicines, with special reference to Ayurveda. It is considered to possess antianaphylactic, antiobesity, amebicidal, gastroprotective, bronchodilating, antiaging, antioxidant, anti-inflammatory, and anticancer activities. It is being used exclusively for weight management and hypotension.1, 2, 3 Forskolin (1), 1-deoxyforskolin (2), and 1,9-dideoxyforskolin (3) are structurally related bioactive diterpenoids from C. forskohlii (Fig. 1). Compound 1 is exclusively known as a fat burner, commonly used for the treatment of obesity, hypothyroidism, allergies, asthma, eczema, psoriasis, and glaucoma.4 C. forskohlii is one of the commercially important herbal ingredients for weight loss dietary supplements in the global market.5 Its favorable effects on body fat management have been well established.6, 7
Fig. 1.
Structure of forskolin (1), 1-deoxyforskolin (2), and 1,9-dideoxyforskolin (3).
Herb–drug interaction is important for assessing the safety of the use of herbal products. Most interactions are pharmacokinetic, and a major concern would be its effect on drug metabolism through interfering with CYP450.8 More than 50% of clinically used drugs are metabolized by CYP4503A (CYP3A). CYP4502B (CYP2B) is capable of metabolizing 25–30% of substrates metabolized by CYP3A4, whereas CYP4502C (CYP2C) metabolizes 20% of the marketed drugs.9, 10, 11 Induction of CYP450 leads to increases in the rate of metabolite production and hepatic biotransformation of coadministered drugs, and decreases in serum half-life and drug response.9 There is now increasing literature reference to the actions of isolated bioactive constituents, and fractions or standardized whole plant extracts on the regulation of CYP450.12, 13 In the present study, commercial extracts and fractions of C. forskohlii were standardized by reversed-phase high-performance liquid chromatography (RP-HPLC) based on compounds 1, 2, and 3. The quantity of compound 1, 2, and 3 in the extracts and fractions were determined. Further, the effect of extracts, fractions, and phytoconstituents on CYP3A, CYP2B, and CYP2C mRNA expression were evaluated in primary cultures of rat hepatocytes.
2. Materials and methods
2.1. Materials
HPLC-grade anhydrous potassium dihydrogen orthophosphate (KH2PO4), dipotassium hydrogen phosphate (K2HPO4), orthophosphoric acid (H3PO4), water, sodium carbonate (Na2CO3), analytical-grade dimethyl sulfoxide (DMSO) propanol, silica gel (60–120 mesh), and silica gel 60 F254 plates were procured from E. Merck Ltd. (Mumbai, India). Membrane filters (0.45 μm) were obtained from Millipore (Billerica, MA, USA). Standard 1 (assay ≥98%), 2 (assay ≥98%), 3 (assay ≥97%), 3-(4,5–dimethylthiazol-2-yl)-5-diphenyl tetrazolium bromide, fetal bovine serum, phosphate buffered saline, Dulbecco's modified Eagle’s medium—high glucose, Ham's F-10 medium, and trypsin were purchased from Sigma-Aldrich Co. (St Louis, MO, USA). EDTA, glucose, penicillin, and streptomycin were obtained from Hi-Media Laboratories Ltd, (Mumbai, India). Nuclease-free water, dithiothreitol, RiboLock RNAse inhibitor, deoxynucleotide mix, oligo dT, and Revert Aid reverse transcriptase were procured from Thermo Scientific (Waltham, MA, USA). Tri reagent from G Biosciences (St. Louis, MO, USA) was used for the study.
2.2. Plant material
Three different batches of dried C. forskohlii roots were purchased from a regular vendor (Olive Lifesciences Pvt. Ltd., Bangalore, Karnataka, India). Voucher specimens were deposited (VU/H/BD/07/14a, VU/H/BD/07/14b, and VU/H/BD/07/14c) at the Department of Botany and Forestry, Vidyasagar University, Midnapore, West Bengal, India.
2.3. Extraction and fractionation
Roots (2.5 kg) were chopped and extracted with hot ethanol to produce reddish brown oleoresin (batch no. PM13038). The extract yield was 9.14% (W/W). This ethanol extract (CEE) was subjected to column chromatography (60–120 mesh silica) to obtain four different fractions [12.5%, 20%, and 50% ethyl acetate fraction (EAF) and 100% methanol fraction (MEF)]. Column chromatography was started with hexane, followed by ethyl acetate and methanol. Fractions were collected and combined based on the thin layer chromatography (TLC) profile. Another two batches (batch numbers PM13039 and PM13040) of extracts [yield: 8.77% (W/W) and 9.68% (W/W), respectively] were subjected to column chromatography. Ten percent standardized commercial C. forskohlii root extract (CCF) (batch no. OL130769) was prepared at the production unit of Olive Lifesciences Pvt. Ltd., and followed the standard operating procedure (SOP no. MMR-21044).
2.4. RP-HPLC analysis of extracts and fractions
The RP-HPLC system (Shimadzu, Kyoto, Japan) consisted of two LC-20AP controller pumps; an SPD-M20A PDA detector; and a SIL-10AP autosampler with a 20-μL loop and integrated LC solution software. A Phenomenex ODS2 (Phenomenex, Hyderabad, Andhra Pradesh, India) (250 × 4.6 mm2; 5 μm) column (USA) was used for the stationary phase. Separation was achieved using a gradient elution program for pump A (0.136 g of KH2PO4 and 0.5 mL of H3PO4 in 1000 mL water) and pump B (acetonitrile) for 50 minutes, with a flow rate of 1.5 mL/min. Ideal resolution of chromatogram was achieved at 205 nm. Test samples were filtered through a Whatman NYL 0.45 μm syringe filter (Millipore) prior to injection. The amount of the phytoconstituents present in the respective extracts and fractions were determined from the calibration curve.
2.5. Cytotoxicity study
Cytotoxicity of the test samples was assessed based on the method described by Denizot and Lang.14 The monolayer culture was incubated with test solution in Dulbecco's modified Eagle’s medium (high-glucose medium) at 37°C for 3 days in a 5% CO2 atmosphere.
2.6. Isolation of rat hepatocytes
Hepatocytes were isolated from male Sprague–Dawley rats using the modified collagenase digestion method, as described by LeCluyse et al.15
2.7. Treatment of hepatocytes
Hepatocytes were washed and treated with commercial extracts, fractions, and bioactive constituents within nontoxic doses (100 μg/mL) for 24 hours in Ham's F-10 medium. Cultures were processed for total RNA extraction. Dexamethasone was used as a positive inducer of CYP3A and CYP2B.15, 16 Rifampin was used as a positive control against CYP2C.17
2.8. Extraction and purification of total mRNA
Total mRNA was extracted from treated hepatocytes using Tri reagent, according to the protocol described by the manufacturer. The mRNA pellet was dissolved in nuclease-free water. Purity of RNA was estimated using the 260/280 absorbance ratio, and samples were frozen at −80°C for further analysis.
2.9. Reverse transcription polymerase chain reaction
From the total RNA isolated from treated hepatocytes, cDNAs were prepared. The reaction mixture consisted of 1× cDNA synthesis buffer, dithiothreitol (0.5M), RiboLock RNAse inhibitor (20 U), deoxynucleotide mix (1.6mM), oligo dT (100 ng), reverse transcriptase (25 U), and total RNA. Primers for CYP3A and CYP2B were selected as per the method of Park et al.18 Primers for CYP2C and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) were selected following the method of Wauthier et al19 and using the Gene runner software, ver. 3.05 (Hasting Software Inc., Hasting, NY, USA; Eurofins Bangalore, Karnataka, India), respectively.19 The forward primers for CYP3A, CYP2B, CYP2C, and GAPDH were 5′ CCACCAAACCCATGCATACC 3′, 5′ AGCCGGGGTGTCTTCTATCT 3′, 5′ TGCCCCCTTTTTACGAGGCT 3′, and 5′ GTGAAGGTCGGTGTGAACGG 3′, respectively. Reverse primers were 5′ CAGTATCATGCTGCGGCTTC 3′, 5′ CAAAGTGGGCTTTTGCTCAT 3′, 5′GGAACAGATGACTCTGAATTCT 3′, and 5′ CACGCCACAGCTTTCCAGAG 3′ for CYP3A, CYP2B, CYP2C, and GAPDH, respectively. Polymerase chain reaction (MJ mini thermocycler; Bio Rad, Hercules, CA, USA) conditions were initial denaturation at 94°C for 5 minutes, followed by 30 cycles consisting of denaturation at 94°C for 30 seconds, annealing of primers at 62°C for 30 seconds, extension at 72°C for 30 seconds, and final extension at 72°C for 2 minutes.
2.10. Analysis of amplified sequences
The amplified samples were analyzed through 1.8% agarose gel electrophoresis. The gel was scanned with UV illumination using digital imaging (Alpha Digi DOC, Upland, CA, USA). Relative sample expression levels were calculated using Alpha View, version 3.3.1.0. (Cell Biosciences Inc., Santa Clara, CA, USA).
2.11. Statistical analysis
Experimental data were expressed as the mean ± standard deviation or standard error of mean. The results were subjected to one-way analysis of variance, followed by Dunnett's multiple-comparison test by fixing the significance level at p < 0.05, p < 0.01 and p < 0.001. The statistical analyses were performed using GraphPad Prism Version 5.0 (GraphPad Software, Inc., La Jolla, CA, USA).
3. Results and discussion
Interactions of drugs with herbal preparations or foodstuffs present unique challenges due to the involvement of a number of unidentified chemical entities. Therefore, standardization parameters and marker profiling are highly important tools in maintaining batch-to-batch reproducibility of herbal drugs.20, 21, 22 C. forskohlii extracts were subjected to column chromatography. Different fractions were collected, and finally four fractions were selected for a CYP450 induction study based on the TLC profile. Compounds 1, 2, and 3 in extracts and fractions were identified by comparing with the respective retention times of reference standards. Calibration curves showed good linearity in the concentration range of 100–800 μg/mL for compounds 1 and 3. In the case of standard 2, the calibration range was 1–625 μg/mL. The percentages of compounds 1, 2, and 3 in CCF, CEE, and EAF are given in Table 1. The maximum quantity of compounds 1 and 2 was found in 20% EAF. The maximum yield of compound 3 was observed in 12.5% EAF (Table 1). In 50% EAF and MEF, the quantity of compounds 1, 2, and 3 was 0.
Table 1.
RP-HPLC quantification of forskolin (1), 1-deoxyforskolin (2), and 1, 9-dideoxyforskolin (3) in extracts and fractions.
| Sample | % of 1 (W/W) | % of 2 (W/W) | % of 3 (W/W) |
|---|---|---|---|
| CCF | 11.03 ± 1.41 | 0.21 ± 0.04 | 4.60 ± 0.28 |
| CEE | 10.00 ± 0.81 | 1.09 ± 0.05 | 10.33 ± 1.08 |
| 12.5% EAF | 0.00 ± 0.00 | 4.57 ± 0.20 | 16.35 ± 2.17 |
| 20% EAF | 34.68 ± 1.46 | 18.45 ± 1.93 | 0.00 ± 0.00 |
Values are represented as mean ± SEM (n = 3).
CCF = 10% standardized commercial C. forskohlii extract; CEE = C. forskohlii ethanolic extract; EAF = ethyl acetate fraction; RP-HPLC = reversed-phase high-performance liquid chromatography; SEM = standard error of mean.
Relative mRNA expressions of CYP3A, CYP2B, and CYP2C by C. forskohlii extracts, fractions, and phytoconstituents on agarose gel are represented in Figs. 2A, 2B, and 2C, respectively. Values are depicted as arbitrary units normalized to (GAPDH) mRNA. The induction effects of extracts, fractions, and phytoconstituents on CYP3A, CYP2B, and CYP2C are shown in Figs. 3A, 3B, and 3C, respectively. Based on cytotoxicity results (Table 2), the test dose (100 μg/mL) was confirmed for mRNA expression study. Extracts and fractions did not show any significant expression of CYP3A, CYP2B, and CYP2C mRNA compared to the control. Positive controls such as dexamethasone (10μM) and rifampin (50μM) showed significant fold induction (p < 0.001) compared to the control. Results revealed that none of the studied phytoconstituents was involved in CYP3A, CYP2B, and CYP2C mRNA induction. CCF; CEE; 12.5%, 20%, and 50% EAF; and MEF did not show any significant mRNA expression against CYP3A, CYP2B, and CYP2C, compared to positive inducers.
Fig. 2.
Relative (A) CYP3A, (B) CYP2B, and (C) CYP2C mRNA expressions by CC, RIF, 1, 2, 3, CCF, 12.5% EAF, 20% EAF, 50% EAF, and MEF. CC = control; CCF = 10% standardized commercial C. forskohlii extract; CEE = C. forskohlii ethanolic extract; DEX = dexamethasone; EAF = ethyl acetate fraction; MEF = methanol fraction; RIF = rifampin; 1 = forskolin; 2 = 1-deoxyforskolin; 3 = 1,9-dideoxyforskolin.
Fig. 3.
Fold induction of (A) CYP3A, (B) CYP2B, and (C) CYP2C mRNA by RIF, 1, 2, 3, CCF, 12.5% EAF, 20% EAF, 50% EAF, and MEF; mean ± standard deviation; n = 3. One-way ANOVA was carried out, followed by Dunnett's multiple comparison test (p < 0.001 and above). ANOVA = analysis of variance; CC = control; CCF = 10% standardized commercial C. forskohlii extract; CEE = C. forskohlii ethanolic extract; DEX = dexamethasone; EAF = ethyl acetate fraction; MEF = methanol fraction; RIF = rifampin; 1 = forskolin; 2 = 1-deoxyforskolin; 3 = 1,9-dideoxyforskolin.
Table 2.
Cytotoxicity assay of phytoconstituents, extracts, and fractions.
| Sample | CTC50 value (μg/mL) |
|---|---|
| 1 | >1000 ± 0.93 |
| 2 | >1000 ± 0.80 |
| 3 | >1000 ± 0.67 |
| CCF | 346.66 ± 0.52 |
| CEE | 200.66 ± 0.73 |
| 12.5% EAF | 215.49 ± 1.22 |
| 20% EAF | 295.46 ± 1.30 |
| 50% EAF | 360.93 ± 1.91 |
| MEF | 480.10 ± 4.37 |
Values are represented as mean ± SD (n = 3).
CTC50 = 50% cytotoxic concentration; CCF = 10% standardized commercial C. forskohlii extract; CEE = C. forskohlii ethanolic extract; EAF = ethyl acetate fraction; MEF = methanol fraction; SD = standard deviation; 1 = forskolin; 2 = 1-deoxyforskolin; 3 = 1,9-dideoxyforskolin.
Induction of CYP450 is of maximum concern during drug development due to their potential role in the metabolism of drugs. The mRNA expression assay is a widely accepted and reliable tool for determining the induction potential of test substances.23 If a drug produces a change in the catalytic activity or mRNA expression that is >40% of that of the positive control, then it is considered an effective inducer. Moreover, the Food and Drug Administration (FDA) has validated the CYP450 induction study and recommended that induction value of the positive control should be twofold times greater than the blank.24 Yokotani et al25 reported that Coleus extract showed induction potential against CYP2C in vivo, as well as inhibition potential against CYP2C in vitro. In both in vivo and in vitro assays, the effect of main bioactive compound 1 was negligible. In the present in vitro assay, C. forskohlii extracts, fractions, and compounds 1, 2, and 3 did not show any significant CYP3A, CYP2B, and CYP2C induction potential compared to positive controls. Fold induction values of positive controls were two times more compared to the control, which confirmed the assay precision. The detail mechanisms of induction are not within the scope of the present study.
4. Conclusion
Based on the study results, it can be concluded that C. forskohlii and its major bioactive constituents may not be involved in CYP450 mRNA expression-based drug interaction, and safe to consume. Further studies using human hepatic mRNA and clinical trials are need.
Conflicts of interest
The authors have declared that there is no conflict of interest.
Footnotes
Peer review under responsibility of The Center for Food and Biomolecules, National Taiwan University.
References
- 1.Alasbahi R.H., Melzig M.F. Plectranthus barbatus: a review of phytochemistry, ethnobotanical uses and pharmacology—part 2. Planta Med. 2010;76:753–765. doi: 10.1055/s-0029-1240919. [DOI] [PubMed] [Google Scholar]
- 2.Murugesan S., Rajeshkannan C., Sumathi R., Manivachakam P., Suresh Babu D. Bioactivity of root hexane extract of Coleus forskohlii Briq. Labiatae: GC/MS/MS characterization and identification. Eur J Exp Biol. 2012;2:1469–1473. [Google Scholar]
- 3.Shivaprasad H.N., Gopalakrishna S., Mariyanna B., Thekkoot M., Reddy R., Tippeswamy B.S. Effect of Coleus forskohlii extract on cafeteria diet-induced obesity in rats. Pharmacognosy Res. 2014;6:42–45. doi: 10.4103/0974-8490.122916. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Foster S., Johnson R. 1st ed. National Geographic Society; Washington: 2006. Desk Reference to Nature's Medicine. [Google Scholar]
- 5.Virgona N., Yokotani K., Yamazaki Y. Coleus forskohlii extract induces hepatic cytochrome P450 enzymes in mice. Food Chem Toxicol. 2012;50:750–755. doi: 10.1016/j.fct.2011.11.054. [DOI] [PubMed] [Google Scholar]
- 6.Henderson S., Magu B., Rasmussen C. Effects of Coleus forskohlii supplementation on body composition and hematological profiles in mildly overweight women. J Int Soc Sports Nutr. 2005;2:54–62. doi: 10.1186/1550-2783-2-2-54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Godard M.P., Johnson B.A., Richmond S.R. Body composition and hormonal adaptations associated with forskolin consumption in overweight and obese men. Obes Res. 2005;13:1335–1343. doi: 10.1038/oby.2005.162. [DOI] [PubMed] [Google Scholar]
- 8.Foti R.S., Wahlstrom J.L., Wienkers L.C. The in vitro drug interaction potential of dietary supplements containing multiple herbal components. Drug Metab Dispos. 2007;35:185–188. doi: 10.1124/dmd.106.012724. [DOI] [PubMed] [Google Scholar]
- 9.Mukherjee P.K., Ponnusankar S., Pandit S., Hazam P.K., Ahmmed M., Mukherjee K. Botanicals as medicinal food and their effects on drug metabolizing enzymes. Food Chem Toxicol. 2011;49:3142–3153. doi: 10.1016/j.fct.2011.09.015. [DOI] [PubMed] [Google Scholar]
- 10.Lehmann J.M., McKee D.D., Watson M.A., Willson T.M., Moore J.T., Kliewer S.A. The human orphan nuclear receptor PXR is activated by compounds that regulate CYP3A4 gene expression and cause drug interactions. J Clin Invest. 1988;102:1016–1023. doi: 10.1172/JCI3703. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Wang H., Tompkins L.M. CYP2B6: new insights into a historically overlooked cytochrome P450 isozyme. Curr Drug Metab. 2008;9:598–610. doi: 10.2174/138920008785821710. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Foster B.C., Vandenhoek S., Hana J. In vitro inhibition of human cytochrome P450-mediated metabolism of marker substrates by natural products. Phytomedicine. 2003;10:334–342. doi: 10.1078/094471103322004839. [DOI] [PubMed] [Google Scholar]
- 13.Pandit S., Mukherjee P.K., Mukherjee K. Cytochrome P450 inhibitory potential of selected Indian spices—possible food drug interaction. Food Res Int. 2012;45:69–74. [Google Scholar]
- 14.Denizot F., Lang R. Rapid colorimetric assay for cell growth and survival modifications to the tetrazolium dye procedure giving improved sensitivity and reliability. J Immunol Methods. 1986;89:271–277. doi: 10.1016/0022-1759(86)90368-6. [DOI] [PubMed] [Google Scholar]
- 15.LeCluyse E., Bullock P., Madan A., Carroll K., Parkinson A. Influence of extracellular matrix overlay and medium formulation on the induction of cytochrome P-450 2B enzymes in primary cultures of rat hepatocytes. Drug Metab Dispos. 1999;2708:909–915. [PubMed] [Google Scholar]
- 16.Wang K., Zhong L., Satoh Y., Shindon H., Sudoh M. Measurement of cytochrome P450 mRNA expression in primary cultured rat hepatocytes using quantitative real-time reverse transcriptase-polymerase chain reaction. Asian J Drug Metab Pharmacokinet. 2005;5:55–60. [Google Scholar]
- 17.Komoroski B.J., Zhang S., Cai H. Induction and inhibition of cytochromes P450 by the St. John's wort constituent hyperforin in human hepatocyte cultures. Drug Metab Dispos. 2004;32:512–518. doi: 10.1124/dmd.32.5.512. [DOI] [PubMed] [Google Scholar]
- 18.Park D., Jeon J.H., Shin S. Green tea extract increases cyclophosphamide-induced teratogenesis by modulating the expression of cytochrome P-450 mRNA. Reprod Toxicol. 2009;27:79–84. doi: 10.1016/j.reprotox.2008.11.058. [DOI] [PubMed] [Google Scholar]
- 19.Wauthier V., Dubois P., Verbeeck R.K., Calderon P.B. Induction of CYP2C12 expression in senescent male rats is well correlated to an increase of HNF3b expression, while the decline of CYP2C11 expression is unlikely due to a decrease of STAT5 activation. Biochem Pharmacol. 2007;73:923–933. doi: 10.1016/j.bcp.2006.12.003. [DOI] [PubMed] [Google Scholar]
- 20.Zhou S., Gao Y., Jiang W., Huang M., Xu A., Paxton J.W. Interactions of herbs with cytochrome P450. Drug Metab Rev. 2003;35:35–98. doi: 10.1081/dmr-120018248. [DOI] [PubMed] [Google Scholar]
- 21.Pandit S., Mukherjee P.K., Ponnusankar S., Venkatesh M., Srikanth N. Metabolism mediated interaction of α-asarone and Acorus calamus with CYP3A4 and CYP2D6. Fitoterapia. 2011;82:369–374. doi: 10.1016/j.fitote.2010.11.009. [DOI] [PubMed] [Google Scholar]
- 22.Pandit S., Ponnusankar S., Bandyopadhyay A., Ota S., Mukherjee P.K. Exploring the possible metabolism mediated interaction of Glycyrrhiza glabra extract with CYP3A4 and CYP2D6. Phytother Res. 2011;25:1429–1434. doi: 10.1002/ptr.3426. [DOI] [PubMed] [Google Scholar]
- 23.Bjornsson T.D., Callaghan J.T., Einolf H.J., Fischer V., Gan L., Grimm S. The conduct of in vitro and in vivo drug-drug interaction studies: a pharmaceutical research and manufacturers of America (PhRMA) perspective. Drug Metab Dispos. 2003;31:815–832. doi: 10.1124/dmd.31.7.815. [DOI] [PubMed] [Google Scholar]
- 24.Zhang D., Surapaneni S. 1st ed. John Wiley & Sons; New Jersey: 2012. ADME-Enabling Technologies in Drug Design and Development. [Google Scholar]
- 25.Yokotani K., Chiba T., Sato Y. Hepatic cytochrome P450 mediates interaction between warfarin and Coleus forskohlii extract in vivo and in vitro. J Pharm Pharmacol. 2012;64:1793–1801. doi: 10.1111/j.2042-7158.2012.01563.x. [DOI] [PubMed] [Google Scholar]