Doctoral Thesis
Refine
Year of publication
- 2009 (116) (remove)
Document Type
- Doctoral Thesis (116) (remove)
Language
- English (116) (remove)
Has Fulltext
- yes (116)
Is part of the Bibliography
- no (116)
Keywords
- Wirkstoff-Rezeptor-Bindung (4)
- Licht-Sammel-Komplex (3)
- Ligand <Biochemie> (3)
- Membranproteine (3)
- Theoretische Physik (3)
- Algebraische Geometrie (2)
- Biochemie (2)
- Crystallography (2)
- FT-IR-Spektroskopie (2)
- Gestaltwahrnehmung (2)
Institute
- Biochemie und Chemie (33)
- Physik (24)
- Biowissenschaften (19)
- Geowissenschaften (6)
- Medizin (6)
- Pharmazie (5)
- Mathematik (4)
- Psychologie (4)
- Informatik (3)
- Frankfurt Institute for Advanced Studies (FIAS) (2)
‘The whole is more than the sum of its parts.’ This idea has been brought forward by psychologists such as Max Wertheimer who formulated Gestalt laws that describe our perception. One law is that of collinearity: elements that correspond in their local orientation to their global axis of alignment form a collinear line, compared to a noncollinear line where local and global orientations are orthogonal. Psychophysical studies revealed a perceptual advantage for collinear over non-collinear stimulus context. It was suggested that this behavioral finding could be related to underlying neuronal mechanisms already in the primary visual cortex (V1). Studies have shown that neurons in V1 are linked according to a common fate: cells responding to collinearly aligned contours are predominantly interconnected by anisotropic long-range lateral connections. In the cat, the same holds true for visual interhemispheric connections. In the present study we aimed to test how the perceptual advantage of a collinear line is reflected in the anatomical properties within or between the two primary visual cortices. We applied two neurophysiological methods, electrode and optical recording, and reversibly deactivated the topographically corresponding contralateral region by cooling in eight anesthetized cats. In electrophysiology experiments our results revealed that influences by stimulus context significantly depend on a unit’s orientation preference. Vertical preferring units had on average a higher spike rate for collinear over non-collinear context. Horizontal preferring units showed the opposite result. Optical imaging experiments confirmed these findings for cortical areas assigned to vertical orientation preference. Further, when deactivating the contralateral region the spike rate for horizontal preferring units in the intact hemisphere significantly decreased in response to a collinear stimulus context. Most of the optical imaging experiments revealed a decrease in cortical activity in response to either stimulus context crossing the vertical midline. In conclusion, our results support the notion that modulating influences from stimulus context can be quite variable. We suggest that the kind of influence may depend on a cell’s orientation preference. The perceptual advantage of a collinear line as one of the Gestalt laws proposes is not uniformly represented in the activity of individual cells in V1. However, it is likely that the combined activity of many V1 neurons serves to activate neurons further up the processing stream which eventually leads to the perceptual phenomenon.
Manual development of deep linguistic resources is time-consuming and costly and therefore often described as a bottleneck for traditional rule-based NLP. In my PhD thesis I present a treebank-based method for the automatic acquisition of LFG resources for German. The method automatically creates deep and rich linguistic presentations from labelled data (treebanks) and can be applied to large data sets. My research is based on and substantially extends previous work on automatically acquiring wide-coverage, deep, constraint-based grammatical resources from the English Penn-II treebank (Cahill et al.,2002; Burke et al., 2004; Cahill, 2004). Best results for English show a dependency f-score of 82.73% (Cahill et al., 2008) against the PARC 700 dependency bank, outperforming the best hand-crafted grammar of Kaplan et al. (2004). Preliminary work has been carried out to test the approach on languages other than English, providing proof of concept for the applicability of the method (Cahill et al., 2003; Cahill, 2004; Cahill et al., 2005). While first results have been promising, a number of important research questions have been raised. The original approach presented first in Cahill et al. (2002) is strongly tailored to English and the datastructures provided by the Penn-II treebank (Marcus et al., 1993). English is configurational and rather poor in inflectional forms. German, by contrast, features semi-free word order and a much richer morphology. Furthermore, treebanks for German differ considerably from the Penn-II treebank as regards data structures and encoding schemes underlying the grammar acquisition task. In my thesis I examine the impact of language-specific properties of German as well as linguistically motivated treebank design decisions on PCFG parsing and LFG grammar acquisition. I present experiments investigating the influence of treebank design on PCFG parsing and show which type of representations are useful for the PCFG and LFG grammar acquisition tasks. Furthermore, I present a novel approach to cross-treebank comparison, measuring the effect of controlled error insertion on treebank trees and parser output from different treebanks. I complement the cross-treebank comparison by providing a human evaluation using TePaCoC, a new testsuite for testing parser performance on complex grammatical constructions. Manual evaluation on TePaCoC data provides new insights on the impact of flat vs. hierarchical annotation schemes on data-driven parsing. I present treebank-based LFG acquisition methodologies for two German treebanks. An extensive evaluation along different dimensions complements the investigation and provides valuable insights for the future development of treebanks.
This study analyses five British translations of Bertolt Brecht's 'Mutter Courage und ihre Kinder'. Two of these translations were written by speakers of German, and three by well-known British playwrights with no knowledge of the source text language. Four have been produced in mainstream British theatres in the past twenty-five years. The study applies translation studies methodology to a textual analysis which focuses on the translation of techniques of linguistic "Verfremdung", as well as linguistic expression of the comedy and of the political dimension in the work. It thus closes the gap in current Brecht research in examining the importance of his idiosyncratic use of language to the translation and reception of his work in the UK. The study assesses the ways in which the translator and director are influenced by Brecht's legacy in the UK and in turn, what image of Brecht they mediate through the production on stage. To this end, the study throws light on the formation of Brecht's problematic reputation in the UK, and it also highlights the social and political circumstances in early twentieth century Germany which prompted Brecht to develop his theory of an epic theatre. The focus on a linguistic examination allows the translator's contribution to the production process to be isolated. Together with an investigation of the reception of each performance text, this in turn facilitates a more accurate assessment of the translator and director's respective influence in the process of transforming a foreign-language text onto a local stage. The analysis also sheds light on the different approaches taken by speakers of German, and playwrights creating an English version from a literal translation. It pinpoints losses in translation and adaptation, and suggests how future versions may avoid these.
In mitochondrial respiration, the soluble protein cytochrome c accepts an electron from the membrane bound cytochrome bc1. The interaction between cytochrome bc1 and cytochrome c is highly transient in nature, enabling turnover numbers greater than 160 s-1. Yeast cytochrome bc1 has been successfully crystallised with bound cytochrome c with the help of an antibody fragment (Lange and Hunte 2002; Solmaz and Hunte 2008). In all crystal structures of the complex, the homodimeric cytochrome bc1 binds only one cytochrome c, with the binding site located on subunit cytochrome c1. Univalent cytochrome c binding is correlated with conformational changes of the Rieske protein head domain and subunit QCR6p. The interface of the complex is small. The haem moieties are centrally located in a mainly non-polar contact site that includes a cation–! interaction and is surrounded by complementary charged residues. The crystal structure is in agreement with the general architecture of the interfaces of transient redox complexes and also reveals several interesting features unique to the cytochrome bc1. On the basis of the crystal structures, an extensive thermodynamic and kinetic characterisation of the interaction was carried out in this work to challenge the static snapshot of the bound proteins in the crystal structure as the relevant physiological electron transfer. The thermodynamic parameters of the interaction between the redox partners were determined using isothermal titration calorimetry (ITC). The association constant for cytochrome bc1 and cytochrome c in oxidised state under physiological ionic strength of 120 mM at 25 °C, was determined to be 5 " 103 M-1 by direct ITC titration. So, the partners interact with an affinity of 200 #M. In spite of the low affinity the complex has a life time ($ = 1/koff) of 5 #second, sufficiently long to enable the theoretically calculated electron transfer rates of 1.0 " 106 to 2.6 " 107 s%1 with a lifetime ($ = 1/rate) of 1-0.04 μseconds and experimentally determined rate of 7.7 " 104 s%1 with a lifetime of 13 μseconds. The low affinity makes it difficult to ascertain the stoichiometry of binding. The enthalpy of the interaction is endothermic, which is consistent with the nature of an interface where hydrophobic interactions are dominant. The enthalpy and entropy is 3.6 kJmol-1 and 83 kJmol-1K-1, respectively. The importance of key interface residues was also investigated. The role of the interface residue G89 of cytochrome c which might have a role in the dissociation of the complex has been probed by site-directed mutagenesis. The interface contains a cation-! interaction between F230 of cytochrome bc1 and R19 of cytochrome c, which is thought to provide the specificity to the interaction between the otherwise promiscuous partners. To analyse the role of this interaction pair in electron transfer, F230L and F230W mutants were used to measure direct electron transfer rates by flash photolysis and steady state kinetics. The findings indicate that another ! system can work as functional substitution of F230, while deleting the ! system has a deleterious effect on the complex formation. The inability of F230L to achieve the transient and steady state turnover rates as wild type protein indicates a scenario where the variant achieves an altered bound state with inefficient electron transfer pathways and higher edge-to-edge distance. The role of supernumerary subunit QCR6p in complex formation was investigated by steady state kinetics measurements. Subunit QCR6p does not interact directly with cytochrome c but is positioned in such a way that it could electrostatically steer cytochrome c in a reactive ensemble. The highly acidic and disordered N-terminus of QCR6p could interact with a patch of conserved lysine residues on cytochrome c. The role of subunit QCR6p has been assessed using QCR6p deleted cytochrome bc1 and a lysine variant of cytochrome c. The results show that QCR6p not only affects the kinetics of the interaction but is also important for the stability of cytochrome bc1. The kinetic and thermodynamic data obtained during this study provide evidence for the functional importance of non-catalytic cytochrome bc1 subunit QCR6p, show that the entropy driven interaction is indeed of low affinity and highly transient in nature and indicate that the interface is well suited to ensure the high turnover of the electron transfer chain where cytochrome c interacts with multiple partners using overlapping interfaces. The suggested role of the cation-! interaction as a highly specific interaction has been validated.
Understanding the dynamics of recurrent neural networks is crucial for explaining how the brain processes information. In the neocortex, a range of different plasticity mechanisms are shaping recurrent networks into effective information processing circuits that learn appropriate representations for time-varying sensory stimuli. However, it has been difficult to mimic these abilities in artificial neural models. In the present thesis, we introduce several recurrent network models of threshold units that combine spike timing dependent plasticity with homeostatic plasticity mechanisms like intrinsic plasticity or synaptic normalization. We investigate how these different forms of plasticity shape the dynamics and computational properties of recurrent networks. The networks receive input sequences composed of different symbols and learn the structure embedded in these sequences in an unsupervised manner. Information is encoded in the form of trajectories through a high-dimensional state space reminiscent of recent biological findings on cortical coding. We find that these self-organizing plastic networks are able to represent and "understand" the spatio-temporal patterns in their inputs while maintaining their dynamics in a healthy regime suitable for learning. The emergent properties are not easily predictable on the basis of the individual plasticity mechanisms at work. Our results underscore the importance of studying the interaction of different forms of plasticity on network behavior.
In this study, I investigate the crustal and upper mantle velocity structure beneath the Rwenzori Mountains in western Uganda. This mountain range is situated within the western branch of the East African Rift and reaches altitudes of more than 5000 m. I use four different approaches that belong to the travel-time tomography method. The first approach is based on the isotropic tomographic inversion of local data, which contain information about 2053 earthquakes recorded by a network of up to 35 stations covering an area of 140×90 km2. The LOTOS-09 algorithm described here is used to realize this approach. The second approach is based on the anisotropic tomographic inversion of the same local dataset. This method employs the tomographic code ANITA, developed with my participation, which provides 3D anisotropic P and isotropic S velocity distributions based on P and S travel-times from local seismicity. For the P anisotropic model, four parameters for each parameterization cell are determined. This represents an orthorhombic anisotropy with one vertically-oriented predefined direction. Three of the parameters describe slowness variations along three horizontal orientations with azimuths of 0°, 60° and 120°, and one is a perturbation along the vertical axis. The third approach is based on tomographic inversion of the teleseismic data, which contain information about the traveltimes of P-waves coming from 284 teleseismic events recorded by the seismic network stations. The TELELOTOS code, which is my own modification of the LOTOS-09 algorithm, is used in this approach. The TELELOTOS code is designed to iteratively invert the local and/or teleseismic datasets. Finally, I present the results of the new tomographic approach, which is based on the simultaneous inversion of the joint local and teleseismic data. The simultaneous use of these datasets for the tomographic inversion has several advantages. In this case, the velocity structure in the study area can be resolved as deep as in the teleseismic approach. At the same time, in the upper part of the study volume, the resolution of the obtained models is as good as in the local tomography. The TELELOTOS algorithm is used to perform the joint tomographic inversion. Special attention is paid in this work to synthetic testing. A number of different synthetic and real data tests are performed to estimate the resolution ability and robustness of the obtained models. In particular, synthetic tests have shown that the results of the anisotropic tomographic inversion of the local data have to be considered as unsatisfactory. For all approaches used in this study, I present synthetic models that reproduce the same pattern of anomalies as that obtained by inverting the real data. These models are used to interpret the results and estimate the real amplitudes of the obtained anomalies. The obtained models exhibit a relatively strong negative P anomaly (up to -10%) beneath the Rwenzori Mountains. Low velocities are found in the northeastern part of the array at shallower depths and are most likely related to sedimentary deposits, while higher velocities are found beneath the eastern rift shoulder and are thought to be related to old cratonic crust. The presence of low velocities in the northwestern part of the array may be caused by a magmatic intrusion beneath the Buranga hot springs. Relatively low velocities were observed within the lower crust and upper mantle in the western and southern parts of the study area (beneath the rift valley and the entire length of the Rwenzori range). The higher amplitude of the low-velocity anomaly in the south can be related to the thinner lithosphere in the southern part of the Albertine rift. In the center of the study area, a small negative anomaly is observed, with the intensity increasing with depth. This anomaly is presumably related to a fluids rising up from a plume branch in the deeper part of the mantle. According to the interpretation of the local earthquake distribution, the Rwenzori Mountains are located between two rift valleys with flanks marked by normal faults. The Rwenzori block is bounded by thrust faults that are probably due to compression.
In Philadelphia Chromosome (Ph) positive ALL and CML the fusion between BCR and ABL leads to the BCR/ABL fusion proteins, which induces the leukemic phenotype because of the constitutive activation of multiple signaling pathways down-stream to the aberrant BCR/ABL fusion tyrosine kinase. Targeted inhibition of BCR/ABL by ABL-kinase inhibitors induces apoptosis in BCR/ABL transformed cells and leads to complete remission in Ph positive leukemia patients. However, a large portion of patients with advanced Ph+ leukemia relapse and acquire resistance. Kinase domain (KD) mutations interfering with inhibitor binding represent the major mechanism of acquired resistance in patients with Ph+ leukemia. Tetramerization of BCR/ABL through the N-terminal coiled-coil region (CC) of BCR is essential for the ABL-kinase activation. Targeting the CC-domain forces BCR/ABL into a monomeric conformation, reduces its kinase activity and increases the sensitivity for Imatinib. Here we show that i.) targeting the tetramerization by a peptide representing the Helix-2 of the CC efficiently reduced the autophosphorylation of both WT BCR/ABL and its mutants; ii.) Helix-2 inhibited the transformation potential of BCR/ABL independently of the presence of mutations; iii.) Helix-2 efficiently cooperated with Imatinib as revealed by their effects on the transformation potential and the factor-independence related to BCR/ABL with the exception of mutant T315I. These findings suggest that BCR/ABL harboring the T315I mutation have a transformation potential which is at least partially independent from its kinase activity. Targeted inhibition of BCR/ABL by small molecule inhibitors reverses the transformation potential of BCR/ABL. We definitively proved that targeting the tetramerization of BCR/ABL mediated by the N-terminal coiled-coil domain (CC) using competitive peptides, representing the Helix-2 of the CC, represents a valid therapeutic approach for treating Ph+ leukemia. To further develop competitive peptides for targeting BCR/ABL, we created a membrane permeable Helix-2 peptide (MPH-2) by fusing the Helix-2 peptide with a peptide transduction tag. In this study, we report that the MPH-2: (i) interacted with BCR/ABL in vivo; (ii) efficiently inhibited the autophosphorylation of BCR/ABL; (iii) suppressed the growth and viability of Ph+ leukemic cells; and (iv) was efficiently transduced into mononuclear cells (MNC) in an in vivo mouse model. The T315I mutation confers resistance against all actually approved ABL-kinase inhibitors and competitive peptides. It seems not only to decrease affinity for kinase inhibitors but to confer additional features to the leukemogenic potential of BCR/ABL. To determine the role of T315I in resistance to the inhibition of oligomerization and in the leukemogenic potential of BCR/ABL, we investigated its influence on loss-of-function mutants with regard to the capacity to mediate factor-independence. Thus we studied the effects of T315I on BCR/ABL mutants lacking functional domains in the BCR portion indispensable for the oncogenic activity of BCR/ABL such as the N-terminal coiled coil (CC), the tyrosine phosphorylation site Y177 and the serine/threonine kinase domain (ST), as well as on the ABL portion of BCR/ABL (#ABL-T315I) with or without the inhibitory SH3 (delta SH3-ABL) domain. Here we report that i.) T315I restored the capacity to mediate factor independence of oligomerization_deficient p185BCR/ABL; ii.) resistance of p185-T315I against inhibition of the oligomerization depends on the phosphorylation at Y177; iii.) autophosphorylation at Y177 is not affected by the oligomerization inhibition, but phosphorylation at Y177 of endogenous BCR parallels the effects of T315I; iv.) the effects of T315I are associated with an intact ABL_kinase activity; v.) the presence of T315I is associated with an increased ABL_kinase activity also in mutants unable to induce Y177 phosphorylation of endogenous BCR; vi.) there is no direct relationship between the ABL-kinase activity and the capacity to mediate factor_independence induced by T315I as revealed by the #ABL-T315I mutant, which was unable to induce Y177 phosphorylation of BCR only in the presence of the SH3 domain. In contrast to its physiological counterpart c-ABL, the BCR/ABL kinase is constitutively activated, inducing the leukemic phenotype. The N-terminus of c-ABL (Cap region) contributes to the regulation of its kinase function. It is myristoylated, and the myristate residue binds to a hydrophobic pocket in the kinase domain known as the myristoyl binding pocket in a process called “capping”, which results in an auto-inhibited conformation. Because the cap region is replaced by the N-terminus of BCR, BCR/ABL “escapes” this auto-inhibition. Allosteric inhibition by myristate “mimics”, such as GNF-2, is able to inhibit unmutated BCR/ABL, but not the BCR/ABL that harbors the “gatekeeper” mutation T315I. Here we investigated the possibility of increasing the efficacy of allosteric inhibition by blocking BCR/ABL oligomerization. We demonstrate that inhibition of oligomerization was able not only to increase the efficacy of GNF-2 on unmutated BCR/ABL, but also to overcome the resistance of BCR/ABL-T315I to allosteric inhibition. These results strongly suggest that the response to allosteric inhibition by GNF-2 is inversely related to the degree of oligomerization of BCR/ABL. Taken together these data suggest that the inhibition of tetramerization inhibits BCR/ABL-mediated transformation and can contribute to overcome Imatinib-resistance. The study provides the first evidence that an efficient peptide transduction system facilitates the employ-ment of competitive peptides to target the oligomerization interface of BCR/ABL in vivo. Further the data show that T315I confers additional leukemogenic activity to BCR/ABL, which might explain the clinical behavior of patients with BCR/ABL -T315I-positive blasts. In summary, our observations establish a new approach for the molecular targeting of BCR/ABL and its resistant mutants represented by the combination of oligomerization and allosteric inhibitors.
Signal-dependent regulation of actin dynamics is essential for many cellular processes, including directional cell migration. In particular, cell migration is initiated by lamellipodia, actin-based protrusions of the plasma membrane. The formation of these protruding structures require incessant assembly and disassembly of actin filaments. The Arp2/3 complex and WAVE proteins are essential for both lamellipodium formation and its dynamics. WAVEs mediate the activation of the Arp2/3 complex downstream of the small GTPase Rac, thus being critical for Rac- and RTK-induced actin polymerization and cell migration. The WAVE-family proteins are always found associated with multiprotein complexes. The most abundant WAVE-based complex is referred to as the WANP (WAVE2-Abi-1-Nap1-PIR121) complex. IQGAP1 is a huge scaffolding protein with multiple protein-interacting domains. IQGAP1 participates in many fundamental activities, including regulation of the actin cytoskeleton, mitogenic, adhesive and migratory responses, as well as in cell polarity and cellular trafficking. IQGAP1 binds to N-WASP, thus raising the possibility that it might control actin nucleation by the Arp2/3 complex. In this study, IQGAP1 was found co-immunoprecipitated not only with WAVE, but also with the endogenous WANP-complex subunits. Correspondingly, IQGAP1 associated to both anti-WAVE and anti-Abi-1 immuno-complexes. Pull-down experiments proved that IQGAP1 binds directly to the WANP-complex subunits. Physical interaction between IQGAP1 and the reconstituted WANP complex could also be demonstrated. Together, these data indicate that IQGAP1 is an accessory component of the WANP complex. Interestingly, the IQGAP-WANP complex disassembled after either EGF stimulation or transfection with constitutively active Cdc42 and Rac1. HeLa cells devoid of IQGAP1 showed diminished and less persistent ruffling upon EGF, but not HGF, stimulation in comparison with the control. This phenotype was accompanied by a strong reduction in chemotaxis towards both growth factors, which was as dramatic as in WANP-complex knockdown (KD) cells. Moreover, GM130 and Giantin showed a polarized and flat ribbon-like pattern in control cells, as it is expected for cis- and cis/medial-Golgi markers. Conversely, small and dispersed vesicular structures were found in both IQGAP1 KD and WANP-complex KD cells. Importantly, Arp2/3-complex silencing resulted in the same phenotypes. Consistently, Brefeldin A-induced disassembly of the Golgi strongly inhibited the IQGAP1-WANP-complex interaction and chemotaxis towards EGF in wild-type cells. The re-expression of an RNAi-resistant wild-type IQGAP1 in IQGAP1 KD cells fully rescued both the ruffling abilities and Golgi structure. A constitutively active mutant, unable to bind to neither Rac1 /Cdc42 nor the WANP complex, could reconstitute only the former defect. Hence, this study shows that actin dynamics regulated by the IQGAP1-WANP complex controls Golgi-apparatus architecture and its contribution to cell chemotaxis. The working model here proposes that at the Golgi apparatus, recruitment of the WANP complex by IQGAP1 leads to the assembly of actin filaments required to maintain the appropriated Golgi morphology. The dissociation of the complex may be required to allow the remodeling of the Golgi membranes in order to respond following a chemoattractant gradient.
A framework for the analysis and visualization of multielectrode spike trains / von Ovidiu F. Jurjut
(2009)
The brain is a highly distributed system of constantly interacting neurons. Understanding how it gives rise to our subjective experiences and perceptions depends largely on understanding the neuronal mechanisms of information processing. These mechanisms are still poorly understood and a matter of ongoing debate remains the timescale on which the coding process evolves. Recently, multielectrode recordings of neuronal activity have begun to contribute substantially to elucidating how information coding is implemented in brain circuits. Unfortunately, analysis and interpretation of multielectrode data is often difficult because of their complexity and large volume. Here we propose a framework that enables the efficient analysis and visualization of multielectrode spiking data. First, using self-organizing maps, we identified reoccurring multi-neuronal spike patterns that evolve on various timescales. Second, we developed a color-based visualization technique for these patterns. They were mapped onto a three-dimensional color space based on their reciprocal similarities, i.e., similar patterns were assigned similar colors. This innovative representation enables a quick and comprehensive inspection of spiking data and provides a qualitative description of pattern distribution across entire datasets. Third, we quantified the observed pattern expression motifs and we investigated their contribution to the encoding of stimulus-related information. An emphasis was on the timescale on which patterns evolve, covering the temporal scales from synchrony up to mean firing rate. Using our multi-neuronal analysis framework, we investigated data recorded from the primary visual cortex of anesthetized cats. We found that cortical responses to dynamic stimuli are best described as successions of multi-neuronal activation patterns, i.e., trajectories in a multidimensional pattern space. Patterns that encode stimulus-specific information are not confined to a single timescale but can span a broad range of timescales, which are tightly related to the temporal dynamics of the stimuli. Therefore, the strict separation between synchrony and mean firing rate is somewhat artificial as these two represent only extreme cases of a continuum of timescales that are expressed in cortical dynamics. Results also indicate that timescales consistent with the time constants of neuronal membranes and fast synaptic transmission (~10-20 ms) appear to play a particularly salient role in coding, as patterns evolving on these timescales seem to be involved in the representation of stimuli with both slow and fast temporal dynamics.
The mTOR kinase inhibitor rapamycin (sirolimus) is a drug with potent immunosuppressive and antiproliferative properties. We found that rapamycin induces the TGF/Smad signaling cascade in rat mesangial cells (MC) as depicted by the nuclear translocation of phospho-Smads 2, -3 and Smad-4, respectively. Concomitantly rapamycin increases the nuclear DNA binding of receptor (R)- and co-Smad proteins to a cognate Smad-binding element (SBE) which in turn causes an increase in profibrotic gene expression as exemplified by the connective tissue growth factor (CTGF) and plasminogen activator inhibitor 1 (PAI-1). Using small interfering (si)RNA we demonstrate that Smad 2/3 activation by rapamycin depends on its endogenous receptor FK-binding protein 12 (FKBP12). Mechanistically, Smad induction by rapamycin is initiated by an increase in active TGF1 as shown by ELISA and by the inhibitory effects of a neutralizing TGF antibody. Using an activin receptor-like kinase (ALK)-5 inhibitor and by siRNA against the TGF type II receptor TGF-RII) we furthermore demonstrate a functional involvement of both types of TGF receptors. However, rapamycin did not compete with TGFfor TGF-receptor binding as found in radioligand-binding assay. Besides SB203580, a specific inhibitor of the p38 MAPK, the reactive oxygen species (ROS) scavenger N-acetyl-cysteine (NAC) and a cell-permeable superoxide dismutase (SOD) mimetic strongly abrogated the stimulatory effects of rapamycin on Smad 2 and 3 phosphorylation. Furthermore, the rapid increase in Dichlorofluorescein (DCF) formation implies that rapamycin mainly acts through ROS. In conclusion, activation of the profibrotic TGFSmad signaling cascade accompanies the immunosuppressive and antiproliferative actions of rapamycin. Keywords: FK506 binding protein; p38 MAP kinase; rapamycin; renal fibrosis; Smads; TGFβ