Bioorganic synthesis
Group Prof. Dr. Oliver Seitz
Research
Fields of Research
DNA-directed chemistrya) DNA-directed chemistryabasic site, DNA catalysis, DNA-directed synthesis, DNA-directed transfer reactions, ligation assay, FRET, multiplex assay, native chemical ligation, PCR-less DNA detection, peptide nucleic acids, product inhibition, signal amplification, single base mutation, single base mutations, SNP, templates, template catalysis, turnover The Watson-Crick base pairing between complementary nucleotides is the universal binding mode that governs biological key processes such as replication, transcription and translation. In the terminology of chemistry, both DNA and RNA act as templates which organize substrates such that subsequent ligation reactions are facilitated. Ligation reactions that proceed under the control of a DNA-like template illustrate how chemical reactivities can be controlled by Watson-Crick base-pairing. The study of DNA-directed ligation reactions is of concern not only because of a general interest in template-directed synthesis. It is the potential regarding gene-targeted chemistry that intrigues us. For example, we examine whether it would be possible to design a chemical ligation reaction such that product formation would prove indicative for a certain single base mutation in DNA or RNA. We use the DNA-analogue peptide nucleic acid (PNA, see Fig. 6) which binds complementary DNA with higher affinity and sequence specificity than DNA itself. We have shown that PNA-amino acid conjugates can be ligated in a DNA-controlled fashion. The ligation strategy is that of a peptide coupling. We favor the use of native chemical ligation, a very powerful reaction known in protein chemistry. One PNA-probe is equipped with a thioester, the other features an N-terminal cysteine or isocysteine residue. In presence of a complementary nucleic acid template a ternary complex forms and the peptide ligation becomes accelerated (Fig 1).  Figure 1: DNA-controlled ligation The native chemical PNA-ligation developed by us proceeds as rapidly and more selective than ligase mediated oligonucleotide ligations. The DNA template can induce a more than 40.000-fold rate acceleration and the sequence specificity can be higher than 4.000-fold in discriminating matched from single mismatched DNA. This high selectivity is the result of the particular ligation architecture which involves an unpaired DNA-base opposite to the ligation site. The enormous sequence specificity can facilitate the detection of early cancer onset. As an example we have shown that as little as 0.2% of single-base mutated DNA in presence of 99.8% wild-type DNA can be detected by mass-spectrometric analysis of the PNA-native chemical ligation. We have also demonstrated the application of the PNA-ligation chemistry to double stranded DNA-templates produced by PCR.  Figure 2: DNA-directed ligation is higly selective as seen by the HPLC traces We are currently taking advantage of the extreme sequence specificity of native chemical PNA ligation in single base mutation analysis of "real world" DNA. Towards this end, we are developing ligation reactions that proceed at the "extreme" conditions applied in the analysis of genomic DNA by the real-time polymerase chain reaction (real-time PCR, qPCR). This requires the design of probes that signal product formation by enhancements of fluorescence. Figure 3 shows an example of a ligation detected by fluorescence resonance energy transfer (FRET).  Figure 3: Ligation detected by fluorescence resonance energy transfer (FRET)
"Mass-spectrometrical Monitoring of a PNA-based Ligation Reaction for the Multiplex Detection of DNA Single Base Mutations" b) Reactions, in which DNA acts as catalyst (chemical amplification)One drawback of using chemical methods for ligation of oligonucleotides and
analogues is product inhibition. Usually the products of ligation bind to the template with higher
affinity than the probes before ligation. Thus, conventional methods of chemical oligonucleotide
ligation result in only one product molecule per mole of target template. However, amplification of
product signals is desired for example when the DNA or RNA to be detected is present at low
concentration. We design reactions that are amenable to catalysis by non-structured DNA. It is our
long-term goal to obtain catalytic efficiencies that may allow intracellular applications and
"PCR-less" DNA-detection.  Figure 4: Two-step ligation-rearrangement sequence At current, we are developing strategies that aim for destabilization of
product-template complexes by more global changes of product formation.  Figure 5: DNA-catalyzed transfer reaction to detect small amounts of DNA In addition to potential applications in DNA sequence analysis we are now exploring DNA-catalyzed peptidyl transfer reactions.
"Reducing Product Inhibition in DNA-Template Controlled Ligation Reactions" Molecular Diagnosticsa) New hybridization probes for DNA- and RNA-detectioncyanine dyes, DNA arrays, energy transfer, FIT probes, FRET, intercalation, light-up probes, NIR fluorophore, PNA Beacons, RNA imaging, single base mutation, SNP, solid-phase synthesis, real-time PCR Many diseases with different forms of appearance such as cystic fibrosis, Tay Sachs
disease, Huntington disease, familial hypercholesterolemia and cancer have genetic determinants. It is one
of the chief aims of molecular diagnostics to detect a developing disease before any symptoms appear.
DNA-targeted analyses play a very important role and are used in various clinical settings. The
binding of a probe molecule to the com-plementary nucleic acid target is the molecular basis for most
of the current methods in DNA-based diagnostics. In principal there are two different approaches,
heterogeneous and homogeneous formats. Heterogeneous assays rely on an immobilization of either the
analyt or the probe molecule to a solid- or gel-phase, which facilitates the removal of unbound
binding partners. Areas in which binding had occurred are detectable by means of a reporter-group
that is usually appended to the soluble binder. In contrast, homogeneous assays are comprised of only
a solution phase and separation of un-bound from bound molecules is not possible. We design new
homogenous assays, which are conceptually more demanding since the hybridization event has to be
coupled with the alteration of a detectable variable. One of the advantages of homogeneous
DNA-detection is that nucleic acid hybridization can be monitored in real-time even within a living
cell. Furthermore, single closed-tube assays are feasible which reduces the contamination risk and
speeds up analysis.  Figure 6: Peptide nucleic acid (PNA) - a DNA-analogue with interesting properties Forced intercalation of cyanine dyes as base surrogates in PNA and DNA: We extended the functional properties of nucleic acids by replacing nucleobases with fluorescent dyes. The "dye base" can respond to alterations of the local duplex structure such as those imposed by DNA damage or single base mutations. We introduced the cyanine dye thiazole orange as base surrogate in PNA (Fig 7). This is in contrast to other thiazole orange containing probes, which carry the fluorophore appended via flexible tethers. We coined the term "forced intercalation probes = FIT probes" to emphasize one of the essential features. The cyanine dye can be stacked to specific base pairs, even to mismatched ones, which under normal circumstances provide a low affinity site for binding of intercalators. Binding of the FIT-PNA probe with a perfectly complementary sequence results in strong enhancements of thiazole orange emission because intercalation between the base-pairs of the newly formed target-probe duplex restricts the torsional flexibility around the central methine bridge which closes an important non-radiative decay channel. A mismatched base pair greatly reduces the efficiency of probe hybridzation which leads to lower melting temperature of mismatched probe-target duplexes. In addition to hybridization specificity, FIT-probes offer an additional level of sequence discrimination that has its origin in the high flexibility of mismatched base-pairs. This low viscosity environment allows torsions around the methine bridge between the two ring systems of thiazole orange. Thus, thiazole orange fluorescence remains low even when a mismatched probe-target duplex had formed. As result the mismatch fluorescence is always lower than match fluorescence even at temperatures below the Tm of the mismatched probe-target duplexes.  Figure 7: The intercalator dye thiazole orange as base surrogate in PNA Currently we are investigating applications of FIT-probes in Real-Time PCR technology. We also surmise that FIT-probes should provide advantages in the detection of viral RNA within live cells. One active field of research involves the search for other dyes with thiazole orange like properties in order to increase the repertoire of colours and to allow multiplexed assays. We would like to transfer the concept of forced intercalation probes also to DNA-based probes.
"A Convergent Strategy for the Modification of Peptide Nucleic Acids (PNA). Novel Mismatch-Specific PNA-Hybridisation Probes." b) Peptide Beaconscellular imaging, fluorescence, FRET, hairpins, peptides, solid-phase synthesis In drug screening and in diagnostics, signaling probe molecules are required that report on specific protein-peptide or protein-protein interactions. A generic method for constructing simple assays - as available in DNA-targeted analyses - does not exist for peptide and proteins. We develop a new type of biosensor, "Peptide Beacons". Peptide beacons are designed to signal the presence of a specific protein by enhancements of fluorescence. The new signaling probes will enable real-time detection of protein function in vitro and in living cells. Switching of protein activitya) Reversible switching of peptide conformation and bioactivityconformational switching, hybridization, kinase, phosphopeptides, PNA, SH2 do-mains, signal transduction Conformational switching is a key-biological process that modulates the function
of proteins in proteinprotein interaction networks. We are interested in interfering in signal
transduction pathways at controlled time points. We develop peptide conjugates, switchable by an
external stimulus, as ligands of cellular proteins. We envision DNA or RNA hybridization as a means
to control the conformation and bioactivity of a peptide. Ultimately we would like to use cell
endogenous RNA as a stimulus of activation/deactivation processes. This would allow a) the
asignment of new function to RNA and b) the coupling of normally uncoupled intracellular
processes.  Figure 10: Exerting control over conformations of peptides  Figure 11: Switching of inhibitory activity
"DNA-Controlled Reversible Switching of Peptide Conformation and Bioactivity" Peptide and protein chemistryIt is one of the foremost challenges in the "post genome era" to unravel the function of newly discovered protein molecules. We believe that a true understanding of protein function can only be obtained at the molecular level. Gaining knowledge at the molecular level of chemistry requires the ability to pinpoint designed modifications to defined positions of the protein molecules. However, the variability of a protein that is generated by biosynthesis is limited and with a few exceptions restricted to the exchange of the 21 proteinogenic amino acids. In contrast, chemical synthesis allows for the implementation of almost any kind of modification at both the protein backbone and the protein side chains. The chemical total synthesis of proteins, even of small proteins, is by no means trivial. Solid phase peptide synthesis is the most powerful method for the synthesis of small- to medium-sized peptides (5-50 amino acids). The iterative coupling steps accumulate by-products, the separation of which is difficult with longer sequences. Convergent methods avoid the "cumulative disaster" of linear synthesis. Access to large peptides can be provided by employing medium-sized peptide segments that are easily available by solid phase synthesis. The most successful approach involves the segment coupling of unprotected peptide segments. At present, it appears that the "native chemical ligation (NCL)" as developed by Dawson and Kent is the most powerful method for the coupling of two unprotected peptide segments (Fig 12).  Figure 12: Native Chemical Ligation Native chemical ligations draw upon the reaction of a peptidyl thioester with a cysteinyl peptide. Our work is focussed on the development of efficient methods for thioester synthesis and on the extension of the native chemical ligation chemistry to non-cysteine ligation sites. We are focussing on methods that may enable the fabrication of functional protein chips by chemical protein synthesis. a) New methods for the Fmoc-based solid-phase synthesis of peptide thioestersacyl transfer, auto-purification, latent thioester, macrocyclization, native chemical ligation, protein synthesis, safety-catch linker, self-activation, solid-phase peptide synthesis Self-purifying solid-phase synthesis of peptide thioesters: Peptide
thioesters are invaluable building blocks for segment ligation techniques such as the powerful native
chemical ligation chemistry. However, the methods available for the solid-phase thioester synthesis
are not as efficient in terms of yield and purity as techniques for the synthesis of peptide acids and
peptide amides. Our aim is to develop a synthesis method that enables a fully automated preparation of
peptide thioesters without the need to employ the often cumbersome HPLC purification. We anticipate
that such a method should eliminate a significant bottleneck in divergent protein synthesis, that is
high throughput synthesis of peptide thioesters.  Figure 13: Self-purification protocol in the solid-phase synthesis of peptide thioesters  Figure 14: comparison of conventional linear synthesis with our self-purification approach We are currently applying the self-purifying method to the synthesis of difficult peptides. We also search for milder methods of sulfonamide activation. For facilitating fully auto-mated synthesis we explore the application of other than allyl-based protecting groups. We will also explore the self-purifying method in a parallel format to enable the synthesis of protein arrays via divergent segment ligation. Intramolecular acyl shifts for activation of stable peptide (thio)esters: In this project we are aiming for the development of a strategy that enables the solid-phase synthesis of peptide thioesters by applying only "standard techiques". Towards this end, we are exploring stable (thio)esters that can be activated at the conditions applied in native chemical ligation chemistry.
"Self-Purifying Solid-Phase Synthesis of Peptide Thioesters" b) New methods for peptide segment couplings in waterchemical protein synthesis, convergent synthesis, ligation auxiliary, native chemical ligation New auxiliaries for the "extended native chemical ligation": The restriction to the formation of X-Cys peptide bonds is a major limitation of native chemical ligation. Directing thiol groups have been attached to amino acids to enable native chemical ligation with amino acids other than cysteine. For this purpose, acid-labile auxiliaries of the benzyl type have been used. These auxiliaries allowed the coupling to non-demanding amico acids such as glycine, alanine or serine. We are developing thiol-containing directing groups with the aim to minimize the sterical demand of the auxiliary. DNA-Protein-Interactionsa) Exploring the base flipping mechanism and selective inhibition of DNA-methyltransferasesbase flipping, base surrogates, C-glycosylation, DNA methylation, enzyme inhibitors, epigenomics, molecular enzymology DNA adopts a double-helical structure with the nucleobases buried in the interior of the double helix. For gaining steric access to innerhelical target structures DNA-modifying enzymes have evolved binding modes that lead to a local disruption of hydrogen-bonding and base-stacking interactions. For example, DNA-methyltransferases (DNA-MTases) commence the methyl group transfer by rotating the target base completely out of the helix. As a result of the enzymatic base flipping process an unpaired nucleobase remains in the duplex (Fig 8). In a minimum model enzyme binding has to compensate for 1) the dissociation of a Watson-Crick base pair and 2) the formation of an apparent abasic site that interrupts contiguous base-stacking. We envision that aromatic base surrogates restore the contiguous base stack, which is interrupted upon enzymatic base flipping, and thus tighten the MTase-substrate complex. In addition, a polycyclic base surrogate destabilizes the innerhelical conformation of the opposing base. The DNA MTase could thus bind to double-stranded DNA containing a preorganized unstacked target base without paying the energetic penalty for disrupting a Watson-Crick base pair and base-stacking interactions. Indeed, the incorporation of pyrene into DNA resulted in 400-fold enhanced binding to M.TaqI MTase.  Figure 8: Targeting base flipping enzymes with base modified oligonucleotides New methods for the synthesis of 1-aryl-2-deoxy-C-ribosides: The incorporation of polycyclic base surrogates into duplex oligonucleotides is achieved by means of C-glycosidically modified nucleotide building blocks. The development of a powerful C-glycosylation chemistry is hence one of the key requirements. Figure 9 shows successful examples. The use of organocuprates in the reaction with chloro-sugars provides C-ribosides in high yields and in a very reliable manner. This cuprate-mediated C-glycosylation is the work horse of our C-riboside program. Usually, a subsequent acid-promoted epimerization is required to convert the a-anomers (formed in ex-cess) to the more desired a-anomers. In case we require the C-riboside of an electron rich arene, Friedel-Crafts type alkylation provides most rapid access to predominant formation of b-ribosides. Sometimes it is difficult to separate the b-anomers from the a-anomeric by-product. For these cases, we have developed aluminate-promoted C-glycosylation reactions. The 1,2-anhydroarabinoside can be opened in syn-selective fashion by aryldimethylaluminium reagents. This reaction exclusively delivers the b-anomers.  Figure 9: The development of a powerful C-glycosylation chemistry
"Polycyclic aromatic DNA-Base surrogates: High-affinity binding to an adenine-specific base-flipping DNA methyltransferase." |
