![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
METABOLISM, TRANSPORT, AND PHARMACOGENOMICS
Division of Molecular Pharmaceutics, University of North Carolina School of Pharmacy, Chapel Hill, North Carolina (F.L., C.C.C., L.M.S., L.H.); Duke University Medical Center, Durham, North Carolina (X.Y.); and University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania (L.M.S.)
Received December 3, 2006; accepted February 20, 2007.
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
). These developments allow for the exploitation of cellular pathways to overcome some of the barriers limiting the success of gene delivery, specifically cell targeting and uptake, and inflammatory response to the delivery vehicles. Viral vectors have been shown to have superior abilities to transfect cells; however, specific immune recognition has always been a concern when considering these systems for clinical trials. Explorations of nonviral systems now focus on both improvement of cell targeting and transfection to achieve levels of transgene activity that meet if not exceed levels observed in viral vectors. Furthermore, formulations of lipid and polymer have been manipulated to minimize nonspecific inflammatory responses induced by the vectors (Niidome and Huang, 2002; Guo and Szoka, 2003). Currently, these approaches have only had limited success. With emergent details of cellular pathways, it is possible to use these advances to improve the efficacy of nonviral vectors for gene delivery.
To increase the efficiency of nonviral vectors, various realms of vector modifications have been investigated. One approach involves specifically altering the structure of the polymers and cationic lipids or addition of other "helper molecules" like polyethylene glycol (Ogris et al., 1999; Wheeler et al., 1999). Alternatively, vectors have been developed to incorporate receptor-specific molecules (Dauty et al., 2002; Maruyama, 2002; Xu et al., 2002; Zuber et al., 2003). In some instances, the targeting molecule is specific for an external receptor, like the incorporation of folate and transferrin molecules into cancer-targeting vectors, because cancer cells tend to have substantially higher numbers of these receptors on their surfaces (Dauty et al., 2002; Maruyama, 2002; Xu et al., 2002; Zuber et al., 2003). These previous studies involving the incorporation of helper molecules have greatly enhanced the efficiency of the vectors by targeting to receptors on the cell surface; however, there are many other possible applications for small-molecule incorporation that have not been explored.
Significant advances in cell signaling have recently revealed a large number of previously unknown nuclear localization sequences (i.e., specific protein sequences that allow passage through the nuclear pore complex) (Nagy and Schwabe, 2004). Nuclear receptors belong to an evolutionary ancient superfamily, which function as ligand-activated transcription factors to assist in regulating processes such as cell growth and differentiation, inflammatory responses, and metabolism. Many studies have shown that small molecules (e.g., retinoids, steroid hormones, fatty acids, cholesterol metabolites) are involved in the regulation of numerous physiological processes, often by binding to small molecules to regulate the activity of proteins (Escriou et al., 2003; Schulman and Heyman, 2004). The ability to vary nuclear receptor-dependent gene expression with small molecules has made nuclear receptors a favored target for gene and drug delivery systems (Hebert, 2003). The additions of these sequences into nonviral vectors are expected to greatly improve gene activity by increasing the internal targeting of the therapeutic gene.
Other small-molecule ligands are also being investigated, specifically for their role in inhibiting inflammatory responses. Although nonviral vectors generally do not elicit a specific humoral immune response, an inflammatory response is frequently observed, increasing levels of TNF-
, IL-6, and IL-12 cytokines. These nonspecific responses increase the lethality of such systems, severely limiting their therapeutic potential as gene delivery vectors. Gaining understanding of inflammatory pathways has allowed for the discovery of many natural and synthetic small molecules that are able to inhibit the pathways leading to cytokine production. Recently, we reported a novel nonviral vector, safeplex, which incorporated various anti-inflammatory drugs, generally inhibiting the NF-
B inflammatory pathway, into the cationic liposomes (Liu et al., 2004). The significantly reduced immune activity has led to further investigation of other small molecules affecting cytokine-inducing pathways.
The current study exploits advances in cell biology by investigating small molecules directly involved in specific cell signaling pathways. First, we incorporated a nuclear localization sequence into the vectors. In the presence of this sequence and a small-molecule ligand, a marked improvement in transgene activity was observed. Furthermore, we investigated two small molecules known to specifically inhibit either the NF-B or the MAPK inflammatory pathway. Both vector systems were found to significantly decrease cytokine levels in the presence of the small molecules. We overcome several of the obstacles affecting nonviral gene delivery through the incorporation of small-molecule ligands.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
strain of Escherichia coli, isolated by alkaline lysis, and purified by ion-exchange column chromatography (QIAGEN Inc., Valencia, CA). Preparation of the Vectors. 1,2-Dioleoyl-3-trimethylammonium-propane (DOTAP) liposomes were prepared from chloroform solution to a concentration of 10 mg/ml. In brief, the lipid solution was placed under a stream of nitrogen to evaporate the solvent until a thin lipid film formed at the bottom of a glass tube. It was further vacuum desiccated for 1 h and then rehydrated in 5% dextrose solution. The lipid suspension was briefly sonicated and then extruded through a polycarbonate membrane with pore size of 0.2 µm. For novel formulations, small molecules (SMs) including chenodeoxycholic acid (CDCA; Sigma, St. Louis, MO), capsaicin (Sigma) and PD98059 (EMD Biosciences, Inc. San Diego, CA) were dissolved in organic solvents at a concentration of 5 mg/ml and mixed with DOTAP before solvent evaporation. The DOTAP liposomes, DOTAP:SM liposomes, and plasmid DNA were diluted with 5% dextrose solution before mixing. Lipoplexes were formed by adding the DNA solution drop-wise into the freshly prepared DOTAP or DOTAP:SM liposome solution in a glass tube while the tube was gently swirled.
Turbidity Studies. Fifty microliters of each sample, CDCA alone, DOTAP liposomes, and DOTAP:CDCA liposomes, was diluted to 500 µl with H2O. Turbidity was measured over 60 min by watching for the appearance of a peak at an absorbance of 600 nm. Absorbances were measured on a DU-800 UV-Vis spectrophotometer (Beckman-Coulter, Fullerton, CA). Stability experiments for DOTAP-capsaicin and DOTAP-PD98059 liposomes were performed under the same conditions.
Encapsulation Efficiency. Two hundred microliters of DOTAP liposome (10 mg/ml) with 5 mg/ml of drug was applied to a Microcon YM-3 spin column (Millipore, Billerica, MA). After a 30-min centrifugation at 14,000g,10 µl of filtrate was combined with 90 µl of 50% alcohol solution. Capsaicin was detected at a wavelength of 281 nm and PD98059 at 238 nm.
In Vitro Gene Transfer. Murine melanoma BL-6 cells (9 x 104 cells/well) were seeded in a 24-well plate and cultured in RPMI 1640 medium with 10% fetal bovine serum for 24 h before transfection. The cells were transfected with 20 µl of the vector containing DOTAP/CDCA and plasmid DNA (2 µg/well) with a charge ratio of 2:1. Luciferase expression was measured 24 h after the transfection using a kit purchased from Promega (Madison, WI) and a luminometer (Autolumat LB953; EG&G, Berthhold, Germany).
In Vivo Gene Transfer. Hydrodynamic injection of CD1 female mice was performed by injecting via the tail vein a large volume (8% of body weight) of solution containing 10 µg of DNA with or without the small-molecule CDCA (0.6 mg i.p.). For inflammatory inhibitor studies by in vivo gene transfer, CD1 female (18–20 g) mice were injected i.v. with the vector containing 25 µg of plasmid DNA with a charge ratio of (±)12:1. Serum was isolated from blood taken by tail vein at 2 h after injection. Mice were sacrificed 6 h after the injection to monitor gene expression. Lungs were collected and placed in 1 ml of ice-cold lysis buffer and homogenized with a Tissue-Tearor (BioSpec Products, Bartlesville, OK) for 20 s at the highest speed. The homogenates were then centrifuged at 14,000g for 5 min at 4°C. Ten microliters of the supernatant was analyzed with the luciferase assay system (Promega).
Analysis of Cytokine Activity. Two hours after the in vivo gene transfer, blood was collected and allowed to clot on ice for at least 4 h. The sample was then centrifuged at 3000g for 20 min at 4°C, and serum was collected for the cytokine assay. The cytokine concentration (TNF-) was determined with mouse cytokine immunoassay kits (R&D Systems, Minneapolis, MN).
Imaging of Liposome Particles. Liposomes were prepared with CDCA as described above. Five microliters of the liposomes was placed on a formvar-coated copper grid (Ted Pella, Redding, CA). The liposomes were visualized using uranyl acetate negative staining, then imaged on a JEOL 100CX (JEOL Ltd., Tokyo, Japan) at the University of Pittsburgh Center for Biological Imaging. Images were taken at 100,000x magnification.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Small-Molecule Ligand Activates a Nuclear Receptor Response Element to Increase Gene Expression in Naked DNA Delivery. FXR is member of the nuclear receptor superfamily, specifically associated with the bile salt circulation pathway and cholesterol homeostasis (Maglich et al., 2003; Redinger, 2003). Pharmaceutical models of FXR regulation are being investigated in hopes of gaining a better understanding of cholesterol/bile acid metabolism for the development of novel drugs to control this pathway. FXR is expressed only in specific tissue types, including liver, gut, adrenal glands, and kidney. In the system described below, FXR is exploited for its ability to target to the nucleus via nuclear localization signal as well as for its relevance in a major metabolic pathway.
FXR must form a functional heterodimer with the retinoid X receptor to be active in bile acid synthesis (Fig. 1). This dimer will bind with the FXRRE in the presence of select ligand molecules to activate transcription factors and initiate synthesis. These ligands are now known to be largely responsible for the regulation of metabolic pathways (Redinger, 2003; Sonoda et al., 2003). The following experiment was designed to specifically employ the aforementioned characteristics of FXR.
|
A plasmid containing the FXRRE upstream from a firefly luciferase gene (referred to as FXRRE-luc) was delivered to the livers of female CD1 mice by hydrodynamic delivery (Liu et al., 1999; Zhang et al., 1999). Luciferase levels were measured to determine the efficiency of gene delivery. When the FXRRE-luc plasmid was delivered, as shown in Fig. 2, luciferase levels were approximately 3 x 108 relative light units/liver due to endogenous FXR. The FXRRE-luc plasmid was then codelivered with a second plasmid containing a strong promoter (CMV) in front of the FXR gene to enhance the production of FXR. No significant increase in luciferase expression was observed. However, when the FXRRE-luc plasmid was injected in the presence of an activator ligand CDCA, specific to FXR, luciferase expression was enhanced by approximately 3-fold. Furthermore, when the FXRRE-luc and CMV-FXR plasmids were codelivered with the ligand, expression was enhanced nearly 6-fold over the levels observed when FXRRE-luc plasmid was delivered alone.
|
|
After verifying the stability of the liposomes, the two formulations were compared for activity in vitro. Lipoplexes, complexes of liposomes and plasmid DNA, were transfected into the mouse melanoma cell line, BL6. DOTAP/FXRRE-luc lipoplexes were not observed to have significant levels of luciferase expression (Fig. 4, n = 6). DOTAP/CDCA/FXRRE-luc lipoplexes also did not show significant levels of luciferase expression. Minimal gene activity was anticipated in both instances because the BL6 cell line was not expected to contain endogenous FXR or other related stimulatory molecules (i.e., CDCA). When the FXRRE-luc plasmid was codelivered with the plasmid containing CMV-FXR in DOTAP liposomes, the luciferase expression increased nearly 3-fold over transfection with the lipoplexes containing only FXRREluc. Moreover, codelivery of the plasmids in the small molecule-containing liposomes enhanced gene expression more than 10-fold over the controls.
|
). Although naked DNA has been found to be nonimmunostimulatory when used alone, complexation with lipids or polymers draws unwanted recognition of the CpG motifs found in plasmid DNA (Krieg et al., 1995; Krieg, 2002). Many approaches have been taken to minimize this effect, including the creation of novel formulations of lipid, pegylation of the molecules (Ogris et al., 1999; Xu et al., 2002; Zuber et al., 2003), and the incorporation of anti-inflammatory molecules conjugated into the vector (Tan et al., 1999; Liu et al., 2004). We have previously described a nonviral vector system that incorporated several general molecules including nonsteroidal anti-inflammatory drugs and steroidal compounds such as dexamethasone and prednisone into the formulation (Liu et al., 2004). In this study, we incorporated a small-molecule ligand to block cytokine formation by specifically inhibiting the NF-B and MAPK pathways (Fig. 5). First, capsaicin, commonly used in over-the-counter anti-inflammatory treatments, was incorporated into the vectors to directly inhibit the NF-B pathway (Singh et al., 1996). A second small molecule, PD98059, a selective noncompetitive inhibitor of the MAPK pathway, was incorporated into the lipid-based vector and also investigated for effectiveness of inhibiting TNF- production (Yoshizumi et al., 2003).
|
The small anti-inflammatory molecules were incorporated into DOTAP liposomes as described under Materials and Methods. The novel formulations were characterized for the stability of the liposomes over time. Data indicates that these liposomes are only stable at ratios below 10:1 (DOTAP/small molecule) based on turbidity studies (data not shown). Optimized complexes were determined to be stable over 1 h, with turbidity curves similar to that of DOTAP:CDCA liposomes (Fig. 3). Encapsulation efficiency of the small-molecule-containing liposomes was examined by centrifugation studies. The complexes were briefly run over a spin column, where any free drug passing into the column eluate was detected by UV spectroscopy (Materials and Methods). Both capsaicin and PD98059 liposomes were found to maintain the drug with greater than 98% efficiency. Incorporation of DNA to form lipoplex was not found to significantly increase leaking of the drug from the complexes (Table 1). It should be noted that ratios of lipid/small molecule are substantially lower than those used for the nuclear receptor-targeting experiments (Fig. 3B) and in our previous studies (Liu et al., 2004). This is largely because the anti-inflammatory molecules were effective at lower concentrations. Concentration studies revealed that the incorporation of higher quantities of drug reduced the stability of the complexes as well as induced toxicity in vivo (data not shown).
|
Transgene activity was examined by measuring luciferase expression in the lungs of mice after tail vein injection of the respective vectors. Luciferase levels for lipoplexes containing capsaicin and PD98059 or DOTAP alone were found to be the same within error for all three formulations (Fig. 6A). Cytokine activity was analyzed by measuring TNF- levels in the serum. DOTAP lipoplexes had TNF- levels at approximately 1000 pg/ml. Lipoplexes containing the anti-inflammatory inhibitors were significantly lower at 225 and 525 pg/ml for capsaicin and PD98059, respectively (Fig. 6B; p < 0.05).
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Increased expression of the transgene was examined by expression of firefly luciferase protein. The aforementioned experiments systematically investigated the role of the FXR nuclear hormone receptor and the associated ligand for activating this promoter region (Figs. 2 and 4). This nuclear receptor (FXR) was chosen to serve as a model for proof of concept. These experiments demonstrate that to some extent all three essential components of the system were still present after cellular uptake because transfection with any two components did not demonstrate optimal transgene activity. In addition, FXR is only expressed in limited tissue types. It was clearly demonstrated that the presence of this nuclear receptor alone does not make significant improvements in transgene activity; however, in the presence of the activating ligand, activity was improved up to 10-fold over the controls. Because several essential elements are required for nuclear receptor activation, this system becomes tunable to specific cell types. Delivery of engineered plasmids containing a hormone receptor response element without both the receptor and activating ligand would be ineffective in tissues that do not already contain both the receptor and the small molecule. Therefore, it is possible to regulate both location and level of transfection in any cell type by incorporating both a nuclear receptor and the ligand into a nonviral formulation. Moreover, this step-wise investigation of the nuclear receptor and ligand stresses the significant role these small molecules play in regulating cellular pathways.
Other small molecules (i.e., capsaicin and PD98059) incorporated into lipid formulations successfully minimized the anti-inflammatory response commonly observed in nonviral vectors. These molecules, selected for their specificity toward inflammatory pathways, targeted the NF-B and MAPK pathways. Incorporation of the small molecules into the liposomes did not alter the efficiency of gene delivery as compared with the DOTAP lipoplexes (Fig. 6A). Because the encapsulation efficiency and turbidity experiments did not indicate any structural defects in the liposomes at these concentrations (Table 1), we presume that no function-altering changes occurred in liposomes containing the small anti-inflammatory molecules. The results suggest that the modified lipoplexes were successfully taken up by cells but also released the free drug after injection, thereby reducing systemic activation of inflammatory cytokines.
The cytokine inhibitors chosen for these experiments are known to inhibit two specific pathways leading toward TNF- production. The results clearly show that TNF- levels are significantly reduced when either the MAPK or the NF-B pathway was inhibited (Fig. 6B). However, it is important to recognize that the level of cytokine reduction was 2-fold lower for the capsaicin-inhibited pathway. This may indicate the inhibition of the NF-B pathway is a more effective method for reducing TNF- levels. Based on these results, it may be possible to inhibit the cytokine levels by affecting one or more activation pathways by the incorporation of small molecules. Thus, manipulation of small molecules to most efficiently down-regulate specific anti-inflammatory pathways may allow vectors to be modified toward limiting specific cytokine production.
Recent advances in cell signaling and trafficking have introduced a much greater understanding into many intra- and intercellular pathways. Gaining perspective on the roles of small molecules to up- or down-regulate these pathways and functions has allowed us to exploit this information to improve gene delivery vectors. The above examples used several small molecules to overcome two main limitations to nonviral gene delivery (i.e., nuclear uptake and inflammation). Incorporation of these small molecules was met with significant success and provides evidence that small-molecule ligands may greatly improve the efficacy of nonviral gene delivery vectors. The nature of the small-molecule ligands and their stability in liposome formulations may also allow for the incorporation of multiple small molecules into liposomes or possibly for use in conjunction with other advances in the field, such as the inclusion of specific cell-targeting molecules (e.g., transferrin or epidermal growth factor). Thus, by combining the breakthroughs in cell biology with advances in gene delivery, it may truly be possible to optimize vectors and to overcome many of the obstacles currently limiting the success of nonviral gene delivery.
|
Footnotes |
|---|
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
ABBREVIATIONS: TNF, tumor necrosis factor; NF, nuclear factor; MAPK, mitogen-activated protein kinase; CMV, cytomegalovirus; FXRRE, farnesoid X receptor response element; FXR, farnesoid X receptor; DOTAP, 1,2-dioleoyl-3-trimethylammonium-propane; SM, small molecule; CDCA, chenodeoxycholic acid; PD98059, 2'-amino-3'-methoxyflavone; IL, interleukin.
Address correspondence to: Dr. Feng Liu, Division of Molecular Pharmaceutics, University of North Carolina School of Pharmacy, 1318 Kerr Hall, CB 7360, Chapel Hill, NC 27599-7360. E-mail: fliu{at}email.unc.edu
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Dauty E, Remy J, Zuber G, and Behr J (2002) Intracellular delivery of nanometric DNA particles via the folate receptor. Bioconjug Chem 13: 831–839.[CrossRef][Medline]
Escriou V, Carriere M, Scherman D, and Wils P (2003) NLS bioconjugates for targeting therapeutic genes to the nucleus. Adv Drug Deliv Rev 55: 295–306.[CrossRef][Medline]
Guo X and Szoka FJ (2003) Chemical approaches to triggerable lipid vesicles for drug and gene delivery. Acc Chem Res 36: 335–341.[CrossRef][Medline]
Hebert E (2003) Improvement of exogenous DNA nuclear importation by nuclear localization signal-bearing vectors: a promising way for non-viral gene therapy? Biol Cell 95: 59–68.[CrossRef][Medline]
Krieg AM (2002) CpG motifs in bacterial DNA and their immune effects. Annu Rev Immunol 20: 709–760.[CrossRef][Medline]
Krieg AM, Yi AK, Matson S, Waldschmidt TJ, Bishop GA, Teasdale R, Koretzky GA, and Klinman DM (1995) CpG motifs in bacterial DNA trigger direct B-cell activation. Nature (Lond) 374: 546–549.[CrossRef][Medline]
Liu F, Shollenberger LM, and Huang L (2004) Non-immunostimulatory nonviral vectors. FASEB J 18: 1779–1781.
Liu F, Song Y, and Liu D (1999) Hydrodynamics-based transfection in animals by systemic administration of plasmid DNA. Gene Ther 6: 1258–1266.[CrossRef][Medline]
Maglich JM, Sluder AE, Willson TM, and Moore JT (2003) Beyond the human genome: examples of nuclear receptor analysis in model organisms and potential for drug discovery. Am J Pharmacogenomics 3: 345–353.[CrossRef][Medline]
Maruyama K (2002) PEG-immunoliposome. Biosci Rep 22: 251–266.[CrossRef][Medline]
Nagy L and Schwabe JW (2004) Mechanism of the nuclear receptor molecular switch. Trends Biochem Sci 29: 317–324.[CrossRef][Medline]
Niidome T and Huang L (2002) Gene therapy progress and prospects: nonviral vectors. Gene Ther 9: 1647–1652.[CrossRef][Medline]
Ogris M, Brunner S, Schuller S, Kircheis R, and Wagner E (1999) PEGylated DNA/transferrin-PEI complexes: reduced interaction with blood components, extended circulation in blood and potential for systemic gene delivery. Gene Ther 6: 595–605.[CrossRef][Medline]
Redinger RN (2003) The coming of age of our understanding of the enterohepatic circulation of bile salts. Am J Surg 185: 168–172.[CrossRef][Medline]
Schulman I and Heyman R (2004) The flip side: Identifying small molecule regulators of nuclear receptors. Chem Biol 11: 639–646.[CrossRef][Medline]
Singh S, Natarajan K, and Aggarwal BB (1996) Capsaicin (8-methyl-N-vanillyl-6-nonenamide) is a potent inhibitor of nuclear transcription factor-kappa B activation by diverse agents. J Immunol 157: 4412–4420.[Abstract]
Sonoda J, Rosenfeld JM, Xu L, Evans RM, and Xie W (2003) A nuclear receptor-mediated xenobiotic response and its implication in drug metabolism and host protection. Curr Drug Metab 4: 59–72.[CrossRef][Medline]
Tan Y, Li S, Pitt BR, and Huang L (1999) The inhibitory role of CpG immunostimulatory motifs in cationic lipid vector-mediated transgene expression in vivo. Hum Gene Ther 10: 2153–2161.[CrossRef][Medline]
Wheeler JJ, Palmer L, Ossanlou M, MacLachlan I, Graham RW, Zhang YP, Hope MJ, Scherrer P, and Cullis PR (1999) Stabilized plasmid-lipid particles: construction and characterization. Gene Ther 6: 271–281.[CrossRef][Medline]
Xu L, Huang CC, Huang W, Tang WH, Rait A, Yin YZ, Cruz I, Xiang LM, Pirollo KF, and Chang EH (2002) Systemic tumor-targeted gene delivery by anti-transferrin receptor scFv-immunoliposomes. Mol Cancer Ther 1: 337–346.
Yoshizumi M, Abe J, Tsuchiya K, Berk BC, and Tamaki T (2003) Stress and vascular responses: atheroprotective effect of laminar fluid shear stress in endothelial cells: possible role of mitogen-activated protein kinases. J Pharmacol Sci 91: 172–176.[CrossRef][Medline]
Zhang G, Budker V, and Wolff JA (1999) High levels of foreign gene expression in hepatocytes after tail vein injections of naked plasmid DNA. Hum Gene Ther 10: 1735–1737.[CrossRef][Medline]
Zhou HS, Liu DP, and Liang CC (2004) Challenges and strategies: the immune responses in gene therapy. Med Res Rev 24: 748–761.[CrossRef][Medline]
Zuber G, Zammut-Italiano L, Dauty E, and Behr J (2003) Targeted gene delivery to cancer cells: directed assembly of nanometric DNA particles coated with folic acid. Angew Chem Int Ed Engl 42: 2666–2699.[CrossRef]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
