Plasmocytoid dentritic cell
Myeloid dentritic cell
- CpG motif:
Unmethylated CG dinucleotide
Synthetic oligodeoxynucleotide with CpG motif
Human plasmacytoid dendritic cells (DC) (PDC, CD123+) and myeloid DC (MDC, CD11c+) may be able to discriminate between distinct classes of microbial molecules based on a different pattern of Toll-like receptor (TLR) expression. TLR1–TLR9 were examined in purified PDC and MDC. TLR9, which is critically involved in the recognition of CpG motifs in mice, was present in PDCbut not in MDC. TLR4, which is required for the response to LPS, was selectively expressed on MDC. Consistent with TLR expression, PDC were susceptible to stimulation by CpG oligodeoxynucleotide (ODN) but not by LPS, while MDC responded to LPS but not to CpG ODN. In PDC, CpG ODN supported survival, activation (CD80, CD86, CD40, MHC class II), chemokine production (IL-8, IP-10) and maturation (CD83). CD40 ligand (CD40L) and CpG ODN synergized to activate PDC and to stimulate the production of IFN-α and IL-12 including bioactive IL-12 p70. Previous incubation of PDC with IL-3 decreased the amount of CpG-induced IFN-α and shifted the cytokine response in favor of IL-12. CpG ODN-activated PDC showed an increased ability to stimulate proliferation of naive allogeneic CD4 T cells, butTh1 polarization of developing T cells required simultaneous activation of PDC by CD40 ligation and CpG ODN. CpG ODN-stimulated PDC expressed CCR7, which mediates homing to lymph nodes. In conclusion, our studies reveal that IL-12 p70 production by PDC is under strict control of two signals, an adequate exogenous microbial stimulus such as CpG ODN, and CD40L provided endogenously by activatedT cells. Thus, CpG ODN acts as an enhancer of T cell help, while T cell-controlled restriction to foreign antigens is maintained.
Bacterial DNA represents a pathogen-associated molecular pattern, which is recognized by the vertebrate immune system leading to a coordinated set of immune responses that includes innate and acquired immunity 1. Recognition of bacterial DNA is based on the presence of unmethylated CG dinucleotides in particular sequence contexts (CpG motifs). The presence of bacterialDNA can be mimicked by synthetic oligodeoxynucleotides that contain such CpG motifs (CpG ODN) 2.
The effects of CpG DNA on the murine immune system are well characterized. As a vaccine adjuvant CpG DNA is at least as effective as the gold standard complete Freund's adjuvant (CFA), but hashigher Th1 activity and lower toxicity 3–5. Animal models suggest that therapeutic applications of CpG DNA include the induction of protective innate immunity ininfectious diseases, the redirection of a Th2 response in asthma or allergic diseases and different applications in immunotherapy of cancer 6.
Recent progress has been made regarding the understanding of the CpG DNA-mediated effects in the human immune system. The identification of a CpG motif which is recognized by human immune cells allowed the design of a highly active CpG ODN with a stable phosphorothioate backbone (CpG ODN 2006) which can be applied in vivo7, 8. Excellent adjuvant activity of CpG ODN 2006 in primates was confirmed by immunizing chimpanzees and orangutans against hepatitis B 8, 9. In human monocytes CpG ODN does not directly stimulate the production of the proinflammatory cytokines TNF or IL-6 which are known to mediate toxicity induced by LPS in vivo10, 11.
Dendritic cells (nature's adjuvant) link innate and adaptive immunity by their ability to induce appropriate immune responses upon recognition of invading pathogens 12. We previously demonstrated that CD4-positive DC in human peripheral blood but not monocyte-derived DC respond to CpG DNA 13. CD4-positive DC in peripheral blood comprise CD123+ plasmacytoid dendritic cells (PDC) and CD11c+ myeloid dendritic cells (MDC) 14, 15. Upon viral infection immature PDC produce large amounts of type I interferon 15, 16. PDC have been named DC2 based on studies demonstrating that CD40L-activated PDC promote Th2 responses 17. Others have questioned the view of PDC as DC2 by showing that PDC stimulated by a virus can induce a Th1 response or Th0 response 18, 19.
Toll-like receptors (TLR) are essential for the recognition of pathogen-associated molecular patterns by the innate immune system. Ten members of the mammalian TLR family have been reported so far (TLR1–10) 20–23. Cooperation between different TLR allows this restricted family of receptors to detect a wide spectrum of microbial molecules. TLR4, which is expressed on monocytes, specifically recognizes LPS derived from gram-negative bacteria 24. Human monocytes express TLR1, 2, 4, 5 and 6, and newly acquire TLR3 when they develop to monocyte-derived DC 25. TLR9 is critically involved in the recognition of CpG motifs by murine B cells, macrophages and bone marrow-derived DC 26. The expression and functional role of TLR9 in the human system has not been demonstrated. * * During the editorial process a paper was published showing that TLR9 is also involved in CpG motif recognition in humans: Bauer, S. et al., Human TLR9 confers responsiveness to bacterial DNA via species-specific CpG motif recognition. PNAS. 2001. 98: 9237 – 9242.
In the present study we found that TLR9 expression is associated with selective recognition of CpG ODN by PDC but not by MDC. The availability of CpG ODN as a unique microbial stimulus for PDC enabled us to demonstrate that upon adequate stimulation, such as the combination of CpG ODN and CD40 ligand (CD40L), PDC are capable to produce large amounts of Th1 cytokines and to induce an IL-12-dependent Th1 response. Our results suggest that PDC are critically involved in the favorable properties of CpG ODN as a powerful Th1 adjuvant.
2.1 CpG ODN activate purified PDC but not purified MDC
Isolated PDC and MDC were incubated with different stimuli and analyzed for the expression of costimulatory molecules at different time points (Fig. 1A). PDC but not MDC responded to CpG ODN by a rapid increase of CD80, CD86 and CD40 expression (Fig. 1A, upper panel). In contrast, LPS activated MDC but not PDC (Fig. 1A, lower panel). As compared to PDC without stimulus, CpG ODN-stimulated PDC at 48 h (Fig. 1B, upper panel) showed a 4.3-fold increase in CD80, an 8.9-fold increase in CD86, a 2-fold increase in HLA-DR and a 2.6-fold increase in CD40 (n=7, * p<0.001). CpG ODN was a more potent stimulus than IL-3 and CD40L to up-regulate costimulatory molecules in PDC. MDC exclusively responded to LPS and GMCSF (Fig. 1B, lower panel).
We examined the activation of MDC and PDC by CpG ODN in a mixed population of CD4+ blood DC. As shown in Fig. 2A CpG but not LPS strongly up-regulated CD80 expression on both CD11c– PDC and CD11c+ MDC. Furthermore, CD80 and CD86 were significantly up-regulated by CpG ODN in both the PDC and MDC subset within PBMC enriched for DC (Fig. 2B). These results suggest an indirect PDC-mediated effect of CpG ODN on MDC.
2.2 PDC and MDC display different patterns of TLR expression
We hypothesized that the different response of PDC and MDC to CpG ODN and LPS is due to a characteristic pattern of TLR in each cell type. Expression of TLR1–9 transcripts in freshly isolated PDC and MDC was detected by reverse transcription (RT)-PCR (Fig. 3). For comparison TLR expression in monocytes is shown. PDC and MDC clearly differ in their TLR expression pattern: TLR1 and TLR7 mRNA were equally expressed in both cell types. Transcripts of TLR3 and TLR4 were only detected in MDC, whereas TLR9 was exclusively expressed in PDC. Expression of TLR9, but lack of TLR4 in PDC correlates with the response of PDC to CpG ODN but not LPS. Detection of TLR4, but not TLR9 in MDC may explain their response to LPS but not to CpG ODN.
2.3 CpG ODN promote survival, maturation and lymph node homing of PDC
CpG ODN was more potent than IL-3 to maintain survival of PDC (3-fold and 1.4-fold as compared to medium control), whereas CD40L and LPS did not improve survival (Fig. 4A). PDC acquired a mature phenotype after 48 h of incubation with CpG ODN as demonstrated by increased CD83 expression which was not seen with IL-3 alone (Fig. 4B). Constitutive expression of CXCR3 protein on PDC was not affected by CpG, whereas CCR7 mRNA and protein was rapidly up-regulated by CpG ODN (Fig. 4C, E). CpG ODN also induced the expression of IP-10 mRNA (Fig. 4D) and the production of IL-8 in response to CpG ODN (23 ng/ml with CpG ODN as compared to 0.5 ng/ml without stimulus, n=3, not in figure).
2.4 Synergy of CpG ODN and CD40L is required for the production of bioactive IL-12 in PDC
As demonstrated in Fig. 5 (upper panel) PDC produced small amounts of IFN-α and total IL-12 in the presence of CpG ODN, IL-3, LPS or CD40L alone. CpG-ODN and CD40L together showed strong synergy to induce IFN-α and total IL-12 production by PDC (IFN-α: 26,091±8,065 pg/ml, p=0.03, n=5; IL-12: 6,518±952 pg/ ml, p=0.0004, n=6). Bioactive IL-12p70 was exclusively detected in the supernatant of PDC that were incubated with CpG and CD40L (IL-12 p70: 137.3±71.4 pg/ml, n=5) but not with CpG or CD40L alone (IL-12 p70: <3.1 pg/ml, lower detection limit, n=5). MDC produced IL-12 in response to LPS or CD40L, but did not respond to CpG ODN (Fig. 5, right lower panel). Bioactive IL-12 p70, however, was not detectable in these samples (<3.1 pg/ml, n=2). The combination of CpG ODN and CD40L did not enhance IL-12 production in MDC (Fig. 5, left lower panel).
2.5 The ratio of IFN-α and IL-12 produced by PDC in response to CpG and CD40L depends on the stage of differentiation
Freshly isolated PDC stimulated with CpG ODN alone showed a transient production of both IFN-α and IL-12 mainly during the first 12 h of stimulation (Fig. 6). After removal of the supernatant and reculturing of the cells without stimulus some residual IL-12 production could be detected in the following 12 h, but not thereafter (right panel).
The capacity of PDC to produce IFN-α and IL-12 upon stimulation was influenced by prior culture of PDC with IL-3 (Fig. 7). IFN-α production induced by CD40L or CD40L and CpG ODN markedly decreased when PDC were preincubated with IL-3 for longer periods of time (Fig. 7A). In contrast, the ability of PDC to produce IL-12 upon stimulation was markedly enhanced by prior treatment with IL-3 (Fig. 7B, 21 ng/ml total IL-12 after 67 h of preincubation in response to CD40L and CpG ODN). Again bioactive IL-12 p70 was only produced by PDC stimulated with the combination of CpG ODN and CD40L (291 pg/ml IL-12 p70).
2.6 CpG ODN increase the capacity of PDC to stimulate proliferation of naive CD4 T cells
Purified PDC, preincubated with CpG ODN, CD40L or without stimulus for 6 h, were co-cultured with CFSE-stained allogeneic naive CD4 T cells at the indicated ratios for 6 days. Proliferation was measured as the percentage of CFSE-low T cells (Fig. 8). CpG-activated PDC were more potent stimulators of T cell proliferation than CD40L-activated PDC or unstimulated PDC. Similar results were obtained when PDC were stimulated for 24 h, harvested, washed and counted before co-culture with allogeneic T cells, demonstrating that the increase in T cell proliferation is not solely due to enhanced survival of CpG ODN-stimulated PDC.
2.7 PDC activated by CpG ODN and CD40L induce an IL-12-dependent Th1 response
As shown in Fig. 9 (upper panels), PDC activated by CpG ODN or CD40L alone were not able to clearly polarize the T cell response. The combination of both stimuli, however, decreased the percentage of IL-4-producing T cells and increased the percentage of IFN–γ-producing T cells. The observed Th1 polarization was dependent on IL-12 but not type I IFN as shown by blocking Ab. An increase in T cells which produce both IFN-γ and IL-10 or T cells which produce only IL-10 (regulatory T cells) was not detected (Fig. 9, lower panels).
The family of TLR has the capacity to establish a combinatorial repertoire to discriminate among the large number of pathogen-associated molecules found in nature 20. Dendritic cells (DC) use a broad array of TLR 25, 27. Teleologically, it would make sense that different types of DC with distinct sets of TLR induce immune responses appropriate to defeat the corresponding type of pathogen. We demonstrate that PDC and MDC express a characteristic pattern of TLR. Freshly isolated PDC and MDC showed a mutually exclusive expression of TLR9 and TLR4, which was associated with the biological response to CpG ODN and LPS. PDC but not MDC expressed TLR9 and responded to CpG ODN. This is in accordance with a recent study in mice showing that TLR9 is essential for the recognition of CpG motifs 26. On the other hand, TLR4 was expressed in MDC which responded to LPS. TLR4 is known to be required for response to LPS derived from gram-negative bacteria 24. TLR3, which is thought to be DC specific 25, 27, was expressed in MDC, but not in PDC.
According to our results the PDC is the first human cell type described to express TLR9. One might speculate that TLR9 is specialized to detect intracellular pathogens such as viruses and intracellular bacteria or parasites based on CpG motifs within their DNA. PDC are the major producers of IFN-α, which plays an important role in the clearance of viruses 28 and other intracellular pathogens 29. Indeed, CpG ODN has been shown to protect mice against lethal infections with intracellular bacteria and parasites 6.
The present study reveals that purified PDC but not MDC represent the primary target for CpG ODN. CpG ODN increased survival, activation and maturation of PDC. Of note, activation of MDC by CpG ODN was seen only in the presence of PDC, suggesting that MDC are not directly susceptible to CpG ODN, but rather are activated by PDC-derived cytokines. It is well established that PDC are capable of producing IFN-α 15, 16. In contrast, the ability of PDC to synthesize IL-12 is controversial 18, 19. Our data show that in PDC CpG ODN and CD40L synergize at inducing large amounts of IL-12 including its bioactive form IL-12 p70.
The CD40 pathway has been thought to play a greater role than microbial stimulation in eliciting IL-12 production by DC, implying that the secretion of this cytokine by DC is primarily regulated by feedback signals from activated T cells 30, 31. However, there is recent evidence that the production of bioactive IL-12 p70 heterodimer by DC in vivo requires both microbial and T cell-derived stimuli 32. In vitro, mechanical stress during isolation of cells or cell culture conditions may substitute for a second signal 32, 33. In agreement with this, in our study MDC showed spontaneous activation and maturation in cell culture and released significant amounts of IL-12 upon stimulation with either LPS or CD40L alone. The situation was different in PDC, which in the absence of IL-3 showed no activation but rather spontaneous apoptosis in cell culture. CpG ODN rescued PDC from apoptosis and markedly increased the expression of CD40, but high levels of IL-12 including bioactive IL-12 p70 were only produced after PDC received a second signal through CD40 ligation. Thus, IL-12 production of PDC is strictly controlled by two independent pathways, the presence of an exogenous microbial stimulus, and CD40 ligation which is provided endogenously by activated T cells.
IFN–α and IL-12 are two alternative cytokines used by the human immune system to promote Th1 responses 34. In our studies simultaneous production of both Th1-promoting cytokines IFN–α and IL-12 was a characteristic feature of PDC. Similarly to IL-12, the induction of high amounts of IFN–α by the CpG ODN used (ODN 2006) depended on the presence of both CpG ODN and CD40 ligation. In a recent study we identified distinct CpG ODN sequences (prototype ODN 2216), which in contrast to ODN 2006 do not require CD40 cross-linking for the induction of high amounts of IFN–α in PDC 35. Unlike ODN 2006, ODN 2216 similar to viruses seems to activate a separate signaling pathway which is sufficient to fully activate IFN–α production by PDC.
In agreement with our earlier results and two other studies 36, 37, ODN 2006 in the absence of CD40 cross-linking induced relatively low amounts of IFN–αin PDC (2.5 ng/ml, Fig. 5) when compared to maximal stimulation with ODN 2216 (428 ng/ml, 35). Consequently, induction of IFN–α by ODN 2006 in whole PBMC is hardly detectable (0.021 ng/ml, 0.2% PDC in PBMC, 35). Of note, PDC prepared from tonsils or isolated by protocols using anti-CD4 Ab show a diminished capacity to produce IFNα (unpublished observation).
Interestingly, the ratio of IFN-γ and IL-12 induced by CpG ODN and CD40L shifts toward IL-12 as PDC differentiate over time in culture with IL-3. The later PDC encounter CpG ODN the more IL-12 and the less IFN–α is produced. The biological consequence of this is that the predominant production of IFN–α at the beginning of an immune response stimulates the proliferation of CD8+ memory T cells in vivo38–40 and promotes innate effector mechanisms including activation of NK cells 28 which represent a first barrier to the pathogen. Later predominant IL-12 production supports priming of antigen-specific T cell responses. Besides other sources, IL-3 is produced by monocytes/macrophages and by TCR-triggered T cells 41, 42. Activated T cells may thus promote the development of IL-12-producing Th1-inducing PDC in two ways, via production of IL-3 and via expression of CD40L.
Expression of chemokines and chemokine receptors plays an important role in the interaction of DC with T cells. We found that PDC rapidly up-regulate CCR7 and IP-10 upon recognition of CpG ODN, whereas constitutive expression of CXCR3 was not affected. The inflammatory chemokine IP-10 is known to recruit CXCR3-expressing cells such as activated T cells and unstimulated PDC 15, 43. Expression of CCR7 is known to mediate homing of immune cells to lymph nodes by binding of CCR7 to SLC or ELC expressed on high endothelial venules and lymphatic endothelium 44. Due to CCR7 expression, PDC may colocalize with naive T cells and central memory T cells 45 in the T cell areas of secondary lymphatic tissues.
In conclusion, the availability of CpG ODN as a specific microbial stimulus for PDC has changed our picture of the functional role of PDC. Our data support the view that PDC in the presence ofthe appropriate microbial stimulus and CD40 ligation induce an IL-12-dependent Th1 response, and may function as DC2 or tolerogenic DC only in the absence of adequate stimulation. Selective amplification of T cell-derived signals might explain why CpG ODN acts as a potent Th1 adjuvant in vivo without causing major toxicity or autoimmunity.
4 Materials and methods
4.1 Media and reagents
IMDM (PAA laboratories, Linz, Austria) supplemented with 8% human serum (BioWhittaker, Walkersville, MD, USA), 1.5 mM L-glutamine, 100 U/ml penicillin and 100 μg/ml streptomycin (Sigma, Munich, Germany) was used. CpG ODN 2006 (24-mer, 5′-TCG TCG TTT TGT CGT TTT GTCGTT-3prime;, provided by Coley Pharmaceutical Group, Wellesley, MA) wascompletely phosphorothioate modified and was free of endotoxin (below 0.03 EU/ml, LAL-assay, BioWhittaker). Based on previous studies ODN 2006 was used at a final concentration of 6 μg/ml. IL-3and GMCSF were purchased from R&D Systems (Wiesbaden, Germany). LPS (S. typhimurium) was from Sigma.
4.2 Isolation of cells
Peripheral blood mononuclear cells (PBMC) were obtained from buffy coats of healthy blood donors by Ficoll-Hypaque density gradient centrifugation (Biochrom) as described 46. PDC and MDC were isolated by magnetically activated cell sorting using the BDCA-4 and BDCA-1 DC isolation kits from Miltenyi Biotec (Bergisch-Gladbach, Germany). The purity of isolated PDC and MDCwas >95%. CD4+ blood DC were isolated from PBMC by depletion of T cells, NK cells and monocytes followed by selection of CD4+ cells (Miltenyi Biotec). For preparation of highly purified untouched monocytes PBMC were first depleted of PDC by direct magnetic labeling with anti-BDCA-4-coupled magnetic beads. The remaining cells were depleted of T cells, NK cells, B cells and basophils resulting in >96% pure monocytes with less than 0.05% PDC (reagents from Miltenyi Biotec). Untouched CD4+ T cells were prepared by magnetic depletion of other cells withinPBMC using hapten-coupled antibodies to CD8, CD11b, CD16, CD19, CD36, CD56 and anti-hapten magnetic beads. In a second step naive CD4+ T cells were isolated from CD4+ T cells by depletion of CD45RO+ T cells (>90% CD4+ CD45RA+ T cells) (Miltenyi Biotec).
4.3 Culture of dendritic cells
Isolated PDC and MDC (105 cells/200 μl/well, in 96-well round-bottom plates) were incubated with IL-3 (10 ng/ml), GM-CSF (800 U/ml), LPS (1 μg/ml) or CD40L-transfected BHKcells (ATTC 79814, mycoplasma-negative, 10,000 cells/well, irradiated with 30 Gy) as indicated.
4.4 Flow cytometry
Surface antigen staining was performed as previously described 47. Fluorescence-labeled mAb against CD3, CD123, CD11c, CD40, CD80, CD83, CD86, HLA-DR and CXCR3, were purchased from Becton Dickinson (Heidelberg, Germany). Anti-BDCA-2 was purchased from Miltenyi Biotec. CCR7 expression was detected by incubation with anti-CCR7 mAb (rat IgG2a, clone 3D12; kindly provided by R. Förster), followed by biotinylated anti-rat IgG2a mAb (clone RG7/1.30) and streptavidin-APC (both from PharMingen). Dead cells were stained with TO-PRO-3 iodide (2 nM, Molecular Probes, Leiden, The Netherlands).
4.5 Detection of cytokines by ELISA
The human IFN-α multi-species ELISA, range 20–5,000 pg/ml (PBL Biomedical Laboratories, New Brunswick, NJ, USA), the human IL-12 (p40/p70) ELISA, range 62.5–2,000 pg/ml (Bender Med Systems, Vienna, Austria) and the human IL-12 p70 ELISA, range 3.1–200 pg/ml (Bender Med Systems, Vienna, Austria) were used.
4.6 T cell proliferation assay
Purified PDC were incubated with CpG ODN or CD40L in 96-well round-bottom plates for 6 h. Allogeneic naive CD4+ T cells, stained with CFSE [5-(and-6-)-carboxyfluorescein diacetate succinimidyl ester, Molecular Probes] as described 13 were added (8×104 T cells/well; PDC: T cell ratio as indicated). After 6 days of co-culture the percentage of proliferating CFSE-low CD4+ T cells was measured by flow cytometry.
4.7 Detection of T helper cell responses by intracellular cytokine staining
PDC were preincubated with IL-3 for 60 h (105 cells/well), washed, stimulated at 104 cells/well with CpG, CD40L or CpG plus CD40L for 6 h and co-cultured with allogeneic naive CD4+ T cells (105 cells/well, ratio 1:10) for 5 days. T cells were expanded with IL-2 (50 U/ml) for additional 5 days. Anti-IL-12 p40/p70 mAb (clone C8.6, PharMingen, 3 μg/ml) or a combination of polyclonal rabbit anti-IFN-α (5,000 neutralizing U/ml), anti-IFN-β (2,000 neutralizing U/ml) and mouse anti-IFN-α/-β receptor mAb (20 μg/ml, PBL, New Brunswick, NJ) were used as indicated. Cells were restimulated with PMA (50 ng/ml) and ionomycin (500 ng/ml) for 6 h and 1 μg/ml brefeldin A was added during the last 3 h (all from Sigma). Cells were harvested, stained with APC-labeled anti-CD3, fixed, permeabilized (Caltag Laboratories, Burlingame, CA, USA) and stained with anti-IFN-γ-FITC, anti-IL-4-PE or anti-IL-10-PE antibodies (PharMingen).
Isolated PDC (3×105–5×105/sample) were lysed and RNA was extracted (High-Pure lysis solution, total RNA isolation kit, Roche, Mannheim, Germany) and reverse transcribed (First Strand cDNA synthesis kit, Roche). For quantitation of CCR7 and IP-10 mRNA by real-time RT-PCR, target sequences were amplified using LightCycler Primer Sets (Search-LC, Heidelberg, Germany) with the LightCycler FastStart DNA Sybr Green I Kit (Roche Diagnostics) according to the manufacturers protocol. Input was normalized by the average expression of four housekeeping genes: β-actin, HPRT, G6PDH and cyclophilin B. The copy number is expressed as the number of transcripts/μl cDNA. For detection of TLR mRNA, primers were designed to produce transcript specific bands for TLR1–9 (Table 1). As positive control, primers for GAPDH were used. PCR conditions were applied as described earlier 48 using a RoboCycler Gradient 40 (Stratagene, Heidelberg, Germany). Products were separated by agarose gel electrophoresis, stained with ethidium bromide and viewed on a UV transilluminator.
4.9 Statistical analysis
Data are expressed as means ± SEM. Statistical significance of differences was determined by the paired two-tailed Student's t-test (* indicates p<0.05). Statistical analyses were performed using StatView 4.51 software (Abacus Concepts Inc., Calabasas, CA).
G. Hartmann is supported by a grant from the BMBF and Coley Pharmaceutical GmbH, Langenfeld (03-12235-6). Additional support was provided by DFG Ha 2780/1–1, a grant from the University of Munich FöFoLe No. 44, Dr. Mildred Scheel-Stiftung 10–1309-En2, and the German-Israeli Foundation No. I-021–203.05/96. This work is part of the thesis of A. Towarowski atthe Ludwig-Maximilians-University, Munich, Germany.