Doctoral Thesis
Refine
Document Type
- Doctoral Thesis (4) (remove)
Language
- English (4) (remove)
Has Fulltext
- yes (4)
Is part of the Bibliography
- no (4)
Keywords
- Pharmakokinetik (2)
- ABC-Transporter (1)
- ABC-transporters (1)
- Epilepsie (1)
- Hydrocortison (1)
- Hydroxysteroid-Dehydrogenasen (1)
- Immunoblot (1)
- Klinische Pharmakologie (1)
- Klinisches Experimen (1)
- Lymphozyt (1)
- MRP3 (1)
- OCT1 (1)
- Organischer Kationentransporter (1)
- Osteoporose (1)
- P-Glykoprotein (1)
- Pharmakogenetik (1)
- Real time quantitative PCR (1)
- Rhodaminfarbstoff (1)
- SLC22A1 (1)
- Talinolol (1)
- Transportproteine (1)
- clinical pharmacology (1)
- clinical study (1)
- epilepsy (1)
- genetic variants (1)
- intestinale Absorption (1)
- ligand-transporter interaction (1)
- lymphocyte (1)
- organic cation transporter 1 (1)
- polyspecificity (1)
- species differences (1)
- structure-function relationship (1)
Institute
- Institut für Pharmakologie (4) (remove)
Organic cation transporter OCT1 is strongly expressed in the sinusoidal membrane of hepatocytes. OCT1 mediates the uptake of weakly basic and cationic compounds from the blood into the liver and may thereby facilitate the first step in hepatic metabolism or excretion of many cationic drugs. OCT1 is a polyspecific transporter and has a very broad spectrum of structurally highly diverse ligands (substrates and inhibitors). The exact transport mechanism and the amino acids involved in polyspecific ligand binding of OCT1 are poorly understood.
The aim of this work was to utilize the polyspecificity to better understand the structure-function relationships of OCT1 and to gain first insights into potential mechanisms conferring the polyspecificity. We followed two strategies, analyzing the effects of variability in both ligand and transporter structure on OCT1 function. The effects of ligand structure were analyzed by comparing uptake and inhibitory potencies of structurally similar drugs of the group of opioids. The effects of transporter structure were analyzed by comparing the effects of variability caused by naturally occurring genetic variants or artificial mutations on OCT1 uptake and inhibition of several substrates. Most importantly, the effects of interspecies variability in transporter structure were analyzed by comparing uptake kinetics between human and mouse OCT1 orthologs. To this end, we used stably or transiently transfected HEK293 cells overexpressing OCT1 and different chimeric and mutant variants thereof.
Focusing on OCT1 ligands, we compared the uptake and inhibitory potencies of structurally similar opioids. Only minor changes of the ligand structure strongly affected the interaction with OCT1. The presence of the ether linkage between C4 and C5 of the morphinan ring was associated with reduced OCT1 inhibitory potencies, while passive membrane permeability was the major negative determinant of OCT1-mediated uptake among structurally highly similar morphinan opioids. Only minor structural changes strongly increased the inhibitory potency by 28-fold from the lowest IC50 of 2004 µM for oxycodone to 72 µM for morphine. Additional removal of the ether linkage between C4-C5 increased the inhibitory potency by a total of 313-fold to the lowest IC50 of 6 µM for dextrorphan. Consequently, our data demonstrates that despite its polyspecificity, OCT1-mediated uptake and inhibition of this uptake is still somewhat very specific.
Focusing on OCT1 protein structure, we first analyzed the effects of variability caused by naturally occurring genetic variants on OCT1 uptake and inhibition. OCT1 transport was strongly affected by OCT1 genetic variants and these effects were often substrate-specific. Correlation of these effects revealed several substrates that were similarly affected by the variants and may therefore be suggested to share similar or overlapping binding sites in OCT1. In addition, the effects of the genetic variants OCT1*2 and OCT1*3 on different substrates correlated well which may suggest that the structural variability caused by these two variants similarly affects substrate uptake. OCT1 genetic variants also affected the inhibition of OCT1, with both substrate and genotype-specific differences. Ranitidine inhibited the uptake of several substrates, among them the clinically relevant drugs metformin and morphine. Moreover, the inhibition was more potent (about 2-fold) on the uptake mediated by the common genetic variant OCT1*2 than on the uptake mediated by the reference OCT1*1.
Second, we analyzed the effects of artificial mutations of key amino acids. Tyr222 and Asp475 in rat OCT1 had strongly substrate-specific and also species-specific effects on both OCT1-mediated uptake and inhibition. Mutation of these amino acids strongly decreased OCT1-mediated uptake, which further underscored an important role especially of Asp475. Interestingly, despite a proposed essential role of this amino acid, we observed Asp475-independent transport. This transport was observed in mouse, but not in human OCT1 and was substrate-specific. TMH10 was identified to be involved in determining the Asp475-independent uptake of mouse OCT1.
Finally and most importantly, we analyzed the effects of sequence differences between human and mouse OCT1 on the transport kinetics of several OCT1 substrates. The transport kinetics differed strongly between human and mouse OCT1 orthologs. These differences were substrate-specific and affected both the affinity (KM) and capacity (vmax) of transport. Human OCT1 had an 8-fold higher capacity of trospium transport, while mouse OCT1 had an 8-fold higher capacity of fenoterol transport. Furthermore, mouse OCT1 had a 5-fold higher affinity for metformin transport compared to human OCT1. The difference between Phe32 in human and Leu32 in mouse OCT1 in TMH1 was identified to confer a higher capacity of transport by human compared to mouse OCT1, while the difference between Cys36 in human and Tyr36 in mouse OCT1 in TMH1 was identified to confer a higher capacity of transport by mouse compared to human OCT1. Furthermore, Leu155 in human OCT1, corresponding to Val156 in mouse OCT1 in TMH2, in concert with TMH3 were identified to confer the differences in affinity for metformin transport between the species.
It may be speculated that ligand binding in OCT1 involves a core binding region that includes Asp474/475 and that polyspecific ligand binding is enabled by providing further binding partners (different amino acids) in more peripheral regions that different ligands can selectively interact with. This mechanism may also be a first step in explaining the substrate-specific effects of genetic variants with clinical relevance. Based on our findings, these “polyspecificity regions” may include TMH1, TMH2, and TMH3. Further analyses are warranted to characterize and narrow down these regions to unravel the structure-function relationships and with that the polyspecificity of OCT1.
To summarize, variability in both ligand and transporter structure strongly affected OCT1 function and we were able to identify ligand structures that affect inhibitory potency and protein structures that confer species-specific differences in OCT1 transport. This work emphasizes again the complexity of OCT1 transport and structure-function relationships. We also showed that, in spite of the difficulties for experimental analysis and data interpretation that arise from the polyspecific nature of OCT1, polyspecificity can also be used as a tool to better understand the structure-function relationships of this transporter.
About 30 % of epileptic patients are non-responsive to multidrug antiepileptic therapy. One of non-responsiveness in epilepsy hypothesis claims that non-responsiveness occurs because of reduced access of antiepileptic drugs to their targets, as a result of increased efflux of antiepileptic drugs away from these targets. Transporters believed to be involved in non-responsiveness in epilepsy are mainly but not exclusively the members of the ABC superfamily including P-gp (MDR1, ABCB1), MRP1 (ABCC1), MRP2 (ABCC2) and others. These proteins are normally found in the blood-brain barrier and the blood-cerebrospinal fluid barrier where they function as protectors. There is emerging evidence that P-gp, MRP1 and MRP2 are up-regulated in epileptogenic brain tissue. The risk of non-responsiveness could be related also to the MDR1 or MRP2 gene polymorphisms. We hypothesised that changes in expression and function of multidrug transporters involved in non-responsiveness of epilepsy might be detectable not only in the brain but also in other tissues such as lymphocytes. Therefore we evaluated the expression of MDR1, MRP1 and MRP2 and function of P-gp in lymphocytes in patients with epilepsy and healthy subjects. Three groups of epileptic patients and 15 healthy subjects as a control group were included in the study. The patients’ group was defined as follows: Monotherapy – patients treated with carbamazepine monotherapy, without seizures - corresponded to group responders. Combined therapy – patients after monotherapy (two different medicines have been tried) and combined therapy (two trials of combined therapy), not free of seizures. Monotherapy and combined therapy groups each embraced 15 patients. Neurosurgery – patients who had undergone neurosurgery, afterwards were or were not additionally treated with carbamazepine, with or without seizures. This group comprised 24 patients. Combined therapy and neurosurgery groups composed the group of non-responders. The mRNA expression of MRP1, MRP2 and MDR1 by means of quantitative real-time PCR as well as MRP2 and P-gp protein content by Western blot in lymphocytes was measured. For P-gp functional analysis rhodamine efflux from lymphocytes and natural killer (NK) cells was performed. The influence of the polymorphisms C3435T, G2677T/A in the MDR1 gene and C24T, G1249A, C3972T in the MRP2 gene for the transporters expression, function and their association with non-responsive epilepsy phenotype was investigated. Our results showed that MRP1 expression in lymphocytes was significantly lower in epileptics than in healthy subjects. Non-responders had lower MRP1 mRNA content in lymphocytes than responders. We did not find any difference in MRP2 expression between epileptics and healthy volunteers. MRP2 mRNA levels in lymphocytes were higher in non-responders than in responders. However, at protein level epileptic patients had significantly lower MRP2 content in lymphocytes than controls. MRP2 protein content did not differ in responders and non-responders. There was no reliable correlation between MRP2 mRNA expression and MRP2 protein content in lymphocytes. Epileptics had significantly lower MDR1 expression in lymphocytes than healthy individuals. MDR1 expression was decreasing according to the consumption of antiepileptic drugs and seizures frequency: patients after neurosurgery had significantly lower MDR1 expression than patients after combined therapy and monotherapy. MDR1 expression was significantly lower in non-responders than in responders. At protein level epileptics had lower P-gp content than controls. Detected P-gp amount in lymphocytes did not differ between responders and non-responders. Rhodamine efflux from lymphocytes and NK cells did not differ significantly between epileptics and healthy subjects, but it was higher in patients after neurosurgery than in patients after monotherapy. Rhodamine efflux from NK cells, which are known to express the highest levels of P-gp, was significantly higher in non-responders than in responders. In this study, we showed that MRP1 mRNA expression in lymphocytes was significantly correlated to its expression in the brain. We detected also a significant co-correlation between MRP1 expression in the hippocampus and MDR1 expression in lymphocytes. We found no evidence regarding the impact of the MDR1 polymorphisms on mRNA expression, P-gp content and rhodamine efflux from lymphocytes. Our data showed lack of evidence regarding the impact of the MRP2 polymorphisms on mRNA expression and protein content. We did not detect any association between MDR1 or MRP2 polymorphisms and non-responsiveness in epilepsy or epilepsy in the main. In conclusion, our results suggest that lymphocytes are an appropriate surrogate for studies on changes of multidrug transporters expression in epilepsy. Lymphocytes as an easily accessible tissue might serve as a marker for responsiveness to antiepileptic drug therapy in epilepsy studies.
In Germany, around 5.7 million people suffer from osteoporosis. Osteoporosis is characterised by a reduced bone mineral density that leads to an increased risk of fractures. The 11β-hydroxysteroid dehydrogenase 1 (11β-HSD1) is an important regulator of local cortisol metabolism. It converts biologically inactive cortisone to biologically active cortisol, but can also catalyse the reverse reaction. 11β-HSD1 is strongly expressed in liver, but 11β-HSD1 expression and activity were also reported in bone. Moreover, polymorphisms in intron 5 of HSD11B1 (the gene encoding for 11β-HSD1) are associated with bone mineral density (BMD) and risk of fractures.
This work aimed to confirm and refine the associations between polymorphisms in intron 5 of HSD11B1 and BMD, and to identify the underlying molecular and cellular mechanisms. To this end, analyses were performed on three different levels:
i) studies in humans, to confirm and refine the association of polymorphisms in intron 5 of HSD11B1 with BMD, suppressed cortisol levels (PDC) and stiffness index,
ii) cellular analyses, to identify the role of 11β-HSD1 in differentiation of the immortalised human mesenchymal stem cell line SCP-1,
iii) molecular genetic analyses, to reveal the effect of intron 5 polymorphisms on transcriptional regulation.
Fine-mapping analyses of already existing clinical data from 452 osteoporosis patients (HSD study) did not point to another intron 5 SNP as being causative for the observed clinical association. A second prospective clinical study (OsteoGene) was performed to confirm the association of rs11811440 and rs932335 with PDC levels and BMD. A trend to decreased PDC levels and increased BMD was observed in homozygous carriers of the minor A-allele of rs11811440 in patients above the age of 65 years. Pooled analyses of the HSD and the OsteoGene studies revealed a significant association of the minor A-allele with increased Z-scores of the left femoral neck. No associations of rs11811440 and rs932335 with stiffness index, BMI and fat depots were detected the general population using data from the SHIP study.
To analyse the effect of 11β-HSD1 on differentiation of mesenchymal stem cells, HSD11B1 overexpressing and HSD11B1 knockout SCP-1 cells were generated. HSD11B1 was stably overexpressed in SCP-1 cells using targeted chromosomal integration. The successful overexpression was shown by 243-fold increased HSD11B1 mRNA expression levels and a 9 fold increased 11β-HSD1 activity, compared to the wildtype cells. Knockout cells were generated by CRISPR-Cas9 mediated gene editing targeting exon 2 and exon 5 of HSD11B1. Using next generation sequencing, the clones 1C4 and 2D10 were confirmed to carry two inactive HSD11B1 alleles and were chosen for further analyses. mRNA expression was unchanged in both knockout clones. However, a clear enzyme activity was detected in the 2D10 clone, whereas no cortisol production was detected in the 1C4 clone. SNaPshot analyses revealed the presence of wildtype cells in the 2D10 clone that became predominant with increased passages. Therefore, further analyses were focused on the 1C4 clone only. The protein expression in the 1C4 clone decreased to 30% of the expression of the wildtype cells.
HSD11B1 expression and cortisol production were compared between wildtype, knockout and overexpressing SCP-1 cells under three differentiation conditions: adipogenic, osteogenic with 1α,25-dihydroxyvitamin D3 and osteogenic with dexamethasone. HSD11B1 expression increased upon adipogenic differentiation and in the presence of cortisone in the wildtype and the overexpressing, but not in the knockout cells. Also, the cortisol production from cortisone increased over time in the overexpressing and the wildtype cells, but not in the knockout cells. The increase was dependent on the differentiation used between 3-fold and 9-fold higher in the overexpressing than in the wildtype cells.
The generated and validated overexpressing and knockout cell lines were used to analyse the influence of 11β-HSD1 on adipogenic and osteogenic differentiation. Upon adipogenic differentiation, the overexpressing cells accumulated significantly more lipid droplets than the wildtype cells. The accumulation of lipid droplets was not abolished in the knockout. However, when dexamethasone was substituted by cortisone, the knockout cells accumulated less lipid droplets than in the presence of dexamethasone, supporting the involvement of 11β-HSD1 in adipogenic differentiation. Expression of the adipogenic markers FABP4 and LPL increased upon adipogenic differentiation, but a distinct influence of the presence or absence of HSD11B1 on the FABP4 and LPL expression was not detected. Upon osteogenic differentiation with 1α,25-dihydroxyvitamin D3, ALP activity increased only in the knockout cells (more than 5-fold). Accordingly, the strongest increase in ALPL expression was detected also in the knockout cells. Both, ALP activity and gene expression were independent of cortisone. Addtionally, BGLAP expression was increased upon osteogenic differentiation. Unexpectedly, in the presence of cortisone, BGLAP expression increased in the overexpressing cells. Expression of the Wnt inhibitor DKK1 also increased in the overexpressing cells in the presence of cortisone indicating a decreased osteogenic differentiation. Moreover, expression of the adipogenic markers FABP4 and LPL increased in the overexpressing cells in the presence of cortisone indicating a switch from osteogenic to adipogenic differentiation. Upon osteogenic differentiation with dexamethasone, ALP activity and matrix mineralisation was lowest in the overexpressing cells.
Finally, the effects of the SNPs rs11811440, rs11119328, rs1000283 and rs932335 in intron 5 of HSD11B1 on transcriptional regulation were analysed by reporter gene assays and electrophoretic mobility shift assays. All four SNPs are genetically linked and are localized within evolutionary conserved regions. The minor C-allele of rs932335 significantly increased luciferase activity. In contrast, the major G-allele of rs932335 showed strong protein binding. However, no transcription factor binding sites were identified at the SNP sites. Additionally, bioinformatics analyses of publicly available RNA-Seq data of adipose tissue and liver confirmed the absence of alternative splicing. Alignment of HSD11B1 intron 5 to the Rfam database predicted the presence of non-coding RNAs (ncRNAs) in intron 5. However, none of the ncRNAs overlapped with the SNP sites.
In conclusion, 11β-HSD1 was shown to be involved in adipogenic differentiation and peripheral cortisol production by 11β-HSD1 promotes a switch from osteogenic to adipogenic differentiation. Moreover, among osteoporosis patients, homozygous carriers of the minor A-allele of rs11811440 have increased Z-scores of the femoral neck. Furthermore, HSD11B1 knockout and overexpressing cell lines were successfully generated and validated. These cell lines could be a useful tool in future analyses of the role of peripheral cortisol activation by 11β-HSD1 in differentiation of mesenchymal stem cells.
The role of uptake and efflux transporters in the pharmacokinetics of ß1-receptor blocker talinolol
(2016)
Introduction: The β1-adrenergic receptor antagonist talinolol is a probe drug for P-glycoprotein (P-gp). It is absorbed erratically and incompletely from the gastrointestinal tract. However, its pharmacokinetics might also be influenced by further uptake and efflux transporters as concluded from interaction studies with naringin and verapamil in human. Additionally, the transcellular transport through the different tissues, including enterocytes, hepatocytes and kidney tubular cells, is not completely understood so far. Therefore, we aimed to measure the affinity of talinolol to drug transporting proteins (OCT1-3, PEPT1, OCTN2, ASBT, NTCP, MRP 1-3 and P-gp as well as OATP 1B1, 1B3, 2B1 and 1A2) and some of their genetic variants known to be of pharmacokinetic relevance (OATP1A2 *2 and*3 as well as OATP2B1 V201M, R312Q and S486F). In a further step, we retrospectively evaluated the impact of clinically relevant genetic polymorphisms of transporters on the pharmacokinetics of talinolol in healthy subjects. Materials and Methods: Time and concentration-dependent uptake assays with [3H]-talinolol were performed either in stable transfected HEK293 or MDCKII cells expressing OATP1A2 *1, *2 and *3, OATP1B1, OATP1B3, OATP2B1 (and its genetic variants p.V201M, p.R312Q and p.S486F), NTCP, ASBT, PEPT1, OCTN2, OCT 1-3 and the respective vector control or in inside-out lipovesicles expressing the efflux transporters MRP1-3 and P-gp. Talinolol was quantified by liquid scintillation counting. The transport rates were then corrected by the transporter proteomics measured in the cellular membrane. Regarding the pharmacogenomic evaluation, it was carried out retrospectively in 39 healthy subjects who had participated in former pharmacokinetic studies with talinolol. This evaluation included a variety of transporter related genetic variants, known to be of a clinical meaning for their substrates. Results: Among the uptake transporters, talinolol was shown to be a substrate of OATP1B3 (Km= 153 ± 137 μmol/l; Vmax= 168 ± 30.3 μmol/mgxmin), OATP1B1 (Km= 301 ± 133 μmol/l; Vmax= 1135 ± 348 μmol/mgxmin), OATP2B1 (Km= 459 ± 260 μmol/l; Vmax= 4.32 ± 1.33 μmol/mgxmin), OATP1A2 (Km= 477 ± 158 μmol/l; Vmax= 0.61 ± 0.1 μmol/mgxmin) and NTCP (Km= 2560 ± 781 μmol/l; Vmax= 15944 ± 3741 μmol/mgxmin) but not a substrate of OCT1-3, OCTN2, PEPT1 or ASBT. When it comes to the efflux transporters, talinolol was transported by both P-gp (Km = 175 ± 206 mol/l; Vmax = 14 ± 10.8 nmol/mgxmin) and MRP3 (Km= 86.8 ± 62.8 μmol/l; Vmax= 133 ± 51.5 μmol/mgxmin) but not by MRP2. The pharmacogenomic analysis supported the in-vitro results, as it showed a significant decrease in talinolol absorption (AUC and Cmax) in subjects with the loss of function variant MRP3 211C>T and in those with a decreased P-gp function due to having less than 5 T-allels in the haplotype P-gp 1236-2677-3435-TTT. No significant changes were found associated with other transporters’ genetic variants. Conclusion: Our in-vitro results suggested the vectorial transport of talinolol through the enterocytes to consist mainly of apical OATP2B1 and P-gp and basolateral MRP3. Additionally in the hepatocytes, apical OATP1B1, OATP1B3 and NTCP seem to be involved as well. This vectorial transport was demonstrated in-vivo for the first time by our pharmacogenomic analysis, where talinolol absorption was significantly influenced by both P-gp and MRP3 genetic variants.