Gotowa bibliografia na temat „Beta Hybrid Peptides”

Three randomly derived sequences that can substitute for the signal peptide of Saccharomyces cerevisiae invertase were tested for the efficiency with which they can translocate invertase or beta-galactosidase into the endoplasmic reticulum. The rate of translocation, as measured by glycosylation, was estimated in pulse-chase experiments to be less than 6 min. When fused to beta-galactosidase, these peptides, like the normal invertase signal sequence, direct the hybrid protein to a perinuclear region, consistent with localization to the endoplasmic reticulum. The diversity of function of random peptides was studied further by immunofluorescence localization of proteins fused to 28 random sequences: 4 directed the hybrid to the endoplasmic reticulum, 3 directed it to the mitochondria, and 1 directed it to the nucleus.

Three randomly derived sequences that can substitute for the signal peptide of Saccharomyces cerevisiae invertase were tested for the efficiency with which they can translocate invertase or beta-galactosidase into the endoplasmic reticulum. The rate of translocation, as measured by glycosylation, was estimated in pulse-chase experiments to be less than 6 min. When fused to beta-galactosidase, these peptides, like the normal invertase signal sequence, direct the hybrid protein to a perinuclear region, consistent with localization to the endoplasmic reticulum. The diversity of function of random peptides was studied further by immunofluorescence localization of proteins fused to 28 random sequences: 4 directed the hybrid to the endoplasmic reticulum, 3 directed it to the mitochondria, and 1 directed it to the nucleus.

Abstract Autoimmune diabetes results from the destruction of insulin-producing beta cells, a process initiated by CD4 T cells recognizing MHC-II bound beta cell-derived peptides. We identified MHC-II bound peptides in the islets and throughout the peripheral lymphatic system in order to better understand diabetes initiation and progression. Toward this end, we performed unbiased nanoflow liquid chromatography-tandem mass spectrometry (nLC-MS/MS) analysis to obtain sequence information for MHC-II peptides isolated from islets, pancreatic lymph nodes and spleens of nonobese diabetic (NOD) mice. Peptides derived from beta cell proteins were further examined for their relative I-Ag7 binding strength and CD4 T cell autoreactivity. For a discussion of the major immunogenic peptides derived from the B chain of insulin and from C peptide, see the abstract of Wan et al. Herein we discuss the finding of a deamidated C peptide fragment and our search for fused peptides. We identified the deamidated insulin-1 C peptide fragment Ins1C:51–61, LQTLALEVARE. Evidence in the literature supports this C-terminal deamidation as being spontaneous rather than enzyme-mediated. Deamidation improved I-Ag7 binding compared to the wild type peptide and increased its immunogenicity, demonstrating the biological importance of this modification. The homologous peptide from insulin-2 was not identified, indicating a possible sequence bias for spontaneous deamidation. We identified only one fused peptide bound to I-Ag7, an insulin C peptide-islet amyloid polypeptide hybrid insulin peptide (HIP) that matches the previously reported HIP6.9. We also identified several free fused peptides in crinosomes and posit this as the probable site of their formation.

Abstract The transcripts encoding two strongly alloantigenic class I mutant molecules, Kdm4 and Kdm5, were characterized and found to encode products that differ from the parental Kd glycoprotein by single amino acid substitutions. The Kdm4 molecule has an amino acid change at position 114, an integral component of a beta-sheet associated with pockets D and E of the peptide binding site. The basis for strong alloantigenicity of the variant molecule can be attributed to differences in peptide binding that were visualized by HPLC analysis of eluted peptides. In contrast, the Kdm5 molecule differs from the parent at position 158, a component of the alpha-helix that is not associated with any of the pockets of the peptide binding site. No differences in peptide binding by Kdm5 in comparison with the parent Kd molecule were seen by HPLC, suggesting that the variant and parent molecules bind the same set of peptides. The ability of (dm4 x dm5) F1 hybrid mice to recognize and lyse BALB/c stimulator cells indicates that the alloantigenic properties determined by the 158 substitution result from the interactions of the alpha-helix regions (changed in dm5) with the pockets of the binding site (changed in dm4). We conclude that self peptides shared by the F1 hybrid and the BALB/c stimulator cells are recognized in the context of structural features of the helices of the Ag-presenting molecule as alloantigenic determinants.

Abstract Anti-I-A mAb were used as probes of functional epitopes for both the presentation of hen egg lysozyme (HEL) peptides to I-Ak-restricted T cell hybridomas and the direct binding of the HEL (46-61) peptide. When mAb directed to polymorphic regions of I-Ak were used as inhibitors of Ag presentation, several different patterns of inhibition were observed among T cells specific for the same HEL peptide as well as among T cells specific for different fragments of HEL. Although there appears to be a conserved usage of some TCR V beta gene segments among the T cell hybrids specific for the same HEL peptide, no correlation is evident between a single V gene usage and susceptibility to blocking of Ag presentation by a particular anti-I-Ak mAb. Several of the mAb demonstrated T cell "clonotypic blocking" of Ag presentation, whereas others blocked presentation to every T cell hybrid tested, regardless of the peptide specificity. When mAb directed to nonpolymorphic regions of the I-A molecule were tested for their ability to block Ag presentation, little or no inhibition was observed. In addition, Fab' fragments of inhibitory mAb functioned identically to their intact homologous counterparts in their ability to block Ag presentation indicating that "nonspecific" steric hindrance was not playing a major role in the inhibitions observed. When the polymorphic region-directed anti-I-A mAb were tested for their ability to block the direct binding of the lysozyme peptide HEL(46-61) to I-Ak, those mAb that block HEL presentation to all T cell hybrids were found to block the binding of this peptide. However, anti-I-A mAb that demonstrate selective inhibition of T cell hybrid stimulation during Ag presentation, i.e., those directed to polymorphic serologic specificities Ia.15 and Ia.19, do not block the binding of HEL(46-61) to I-Ak. These data indicate that functionally independent epitopes exist on the I-Ak molecule for the binding of antigenic peptides and for interaction with the TCR.

Alzheimer’s disease (AD) is the most common neurodegenerative pathology among progressive dementias, and it is characterized by the accumulation in the brain of extracellular aggregates of beta-amyloid proteins and neurofibrillary intracellular tangles consisting of τ-hyperphosphorylated proteins. Under normal conditions, beta-amyloid peptides exert important trophic and antioxidant roles, while their massive presence leads to a cascade of events culminating in the onset of AD. The fibrils of beta-amyloid proteins are formed by the process of fibrillogenesis that, starting from individual monomers of beta-amyloid, can generate polymers of this protein, constituting the hypothesis of the “amyloid cascade”. To date, due to the lack of pharmacological treatment for AD without toxic side effects, chemical research is directed towards the realization of hybrid compounds that can act as an adjuvant in the treatment of this neurodegenerative pathology. The hybrid compounds used in this work include moieties of a hydroxytyrosol, a nitrohydroxytyrosol, a tyrosol, and a homovanillyl alcohol bound to the N-benzylpiperidine moiety of donepezil, the main drug used in AD. Previous experiments have shown different properties of these hybrids, including low toxicity and antioxidant and chelating activities. The purpose of this work was to test the effects of hybrid compounds mixed with Aβ 1–40 to induce fibrillogenesis and mimic AD pathogenesis. This condition has been studied both in test tubes and by an in vitro model of neuronal differentiated human SH-SY5Y neuroblastoma cells. The results obtained from test tube experiments showed that some hybrids inhibit the activity of the enzymes AChE, BuChE, and BACE-1. Cell experiments suggested that hybrids could inhibit fibrillogenesis, negatively modulating caspase-3. They were also shown to exert antioxidant effects, and the acetylated hybrids were found to be more functional and efficient than nonacetylated forms.

Abstract Structural polymorphism of murine factor H protein was demonstrated by using three different methods. 1) By prolonged agarose electrophoresis and immunofixation, factor H protein was visualized in the beta region as a single, distinct protein band in freshly bled EDTA-plasmas from many laboratory and wild mice. Two variants were detected among a large number of tested strains; one, referred to as H.1, moved faster to the anodal region (type strain, BALB/c), and the other, referred to as H.2, moved more slowly to the anodal region (type strain, STR). The F1 hybrid between BALB/c and STR exhibited a combining type of factor H protein, which was observed in each parent. 2) Two-dimensional peptide mapping analysis was carried out with tryptic peptides of these two factor H allotypes. Almost all of the spots in the maps of tryptic peptides were common to both allotypes. However, three distinct spots among the 57 spots detected in the map of tryptic peptides of the H.1 allotypes were not detected in that of H.2 allotype, whereas two spots among the 56 spots in the map of H.2 allotype were unique for this allotype. The F1 hybrid between BALB/c and STR showed a combining type of the map of parent. 3) Alloantisera against each of H allotypes were successfully produced in BALB/c or BALB/c-H.2 (a congenic strain with H.2 allotype) by repeated injection of each purified factor H protein either from the BALB/c or the STR strain. These findings indicated that the observed variants of factor H represent antigenically and structurally distinguishable allotypes. The allotypes of murine factor H protein are controlled by a single codominant locus located between the Hc locus and the beta 2M locus on the second chromosome of the mouse. This was shown by phenotyping the Hc locus and H locus with backcross progenies between A/J (one of strain with H.1) and MoA (one of strain with H.2). The recombination frequency between these two loci was 0.17 +/- 0.046.

Newly assembled heavy chain-beta 2m heterodimers of class I histocompatibility molecules associate with the endoplasmic reticulum (ER) peptide transporter, TAP, and subsequently dissociate from TAP in parallel with their transport from the ER to the Golgi apparatus. It appears that TAP-associated class I molecules are waiting to bind appropriate peptides before they dissociate from TAP and leave the ER since binding of high affinity peptides to class I molecules in vitro leads to dissociation of TAP-class I complexes. In further support of this notion, we report that limiting peptide supply through inhibition of proteasome activities prolongs the association of mouse class I molecules with TAP and concomitantly slows their transport to the Golgi apparatus. By using a series of deletion mutants and hybrid class I molecules we demonstrate that the extracellular domains of class I molecules are sufficient for their peptide-regulated interaction with TAP. Furthermore, based on the inability of an alpha 3 domain-specific mAb to recognize TAP-class I complexes and the fact that a point mutant of the Dd molecule at residue 222 is unable to bind to TAP, it is likely that a major site of interaction with TAP resides in the membrane-proximal region of the heavy chain alpha 3 domain. Finally, we examined the relationship between the interaction of mouse heavy chain-beta 2m heterodimers with TAP and with the resident ER chaperone, calnexin. Most heterodimers that bound to TAP were found to associate simultaneously with calnexin. Upon delivery of peptide to class I molecules in permeabilized cells, dissociation from TAP was observed but the interaction with calnexin was largely maintained. Therefore, both TAP and calnexin may participate in the ER retention of peptide-deficient class I molecules. However, since release from calnexin occurs after dissociation from TAP, it appears that calnexin ultimately determines if a class I molecule is to be exported from the ER.

Abstract MHC proteins are polymorphic cell surface glycoproteins involved in the binding of peptide Ag and their presentation to T lymphocytes. The polymorphic amino acids of MHC proteins are primarily located in the N-terminal domains and are thought to influence T cell recognition both by influencing the binding of peptide Ag and by direct contact with the T cell receptor. In order to determine the relative importance of individual polymorphic amino acids in Ag presentation, a number of groups have taken the approach of interchanging polymorphic amino acids between different alleles of MHC protein in an attempt to define which of the polymorphisms influence peptide binding and which influence T cell recognition by direct contact with the TCR. The peptide OVA323-339 has been previously shown to bind to the MHC class II protein Ad and to have a much lower affinity for Ak, whereas the peptide hen egg lysozyme 46-61 binds well to Ak and poorly to Ad. In the present report, we have analyzed the ability of purified wild-type MHC class II proteins as well as the ability of three different hybrid molecules between Ad and Ak to bind and present these peptides. We find that the alpha-chain of the MHC class II protein plays a critical role in the binding of HEL46-61 and confers the specificity for binding OVA323-339, regardless of which beta-chain is present. We also find that the beta-chain region 65-67 does not control the specificity of peptide binding to the MHC protein, but is important in T cell responses to preformed MHC-peptide complexes, suggesting a role for this region in contacting the TCR.

Abstract A segmental analysis of the key regions of HLA-DR1 that control T cell allorecognition was performed by using a series of transfected cell lines expressing the products of recombinant DRB/H-2Eb genes, paired with either DR alpha or H-2E alpha. Four of eight human T cell clones tolerated substitution of the H-2E alpha chain, but only one clone showed any response to the DR alpha/H-2E beta k dimer. Both the membrane-proximal and the membrane-distal domains of the beta-chain played an important part in stimulating these clones. The response of four of eight clones was markedly inhibited by substitution of the H-2E beta 2 for the DR beta 2 domain. This inhibition showed a complete correlation with the sensitivity of the clones to inhibition by anti-CD4 mAb. Taken together, these results suggest that the interaction site for CD4 may include residues on the beta 2-domain. Introduction of H-2Ek sequence into either half of the beta 1-domain led to a complete loss of response by all but two of the clones. This is consistent with these clones having dual specificity for exposed DR1-specific polymorphisms and for DR1-bound peptides. The pattern of response of one of the clones suggested that indirect conformational effects on the alpha 1-domain may also contribute to the influence of the amino-terminal half of the beta 1-domain on T cell recognition. In the presence of H-2E alpha, this clone responded more strongly when the amino-terminal half of the beta 1-domain was of H-2Ek rather than DR1 sequence. This implies that species matching of the floor of the beta 1-domain with the alpha-chain is more important than the presence of the alpha-chain of the parental species.

Structural studies of peptides are of great importance in developing novel and effective biomaterials ranging from drugs and vaccines to nano materials with industrial applications. In addition, they provide model systems to study and mimic the protein conformations. The ability to generate folded intramolecularly hydrogen bonded structures in short peptides is essential for peptide design strategies, which rely on the use of folding nuclei in the construction of secondary structure modules like helices and β-hairpins. In these approaches, conformational choices at selected positions are biased, using local stereochemical constraints, that limit the range of accessible backbone torsion angles. X-ray crystallographic studies of designed peptides provide definitive proof of the success of a design strategy, and provide essential structural information that can be utilized in the future design of biologically and structurally important polypeptides. Recent trends in peptide research focus on the incorporation of β-, γ- and higher homologs of the α-amino acid residues in designed peptides as they confer more proteolytic stability to the polypeptides. X-ray crystallographic studies of such modified peptides containing non-protein residues are essential, since information on the geometric and stereochemical properties of modified amino acids can only be gathered from the systematic structural studies of synthetic peptides incorporating them. This thesis reports a systematic study of the structures and conformations of amino acid derivatives and designed peptides containing stereochemically constrained α-, β- and γ-amino acid residues and the structural studies of polymorphic peptide helices. The structures described in thesis contain the Cα,α-dialkyalted α-residues α-aminoisobutyric acid (Aib) and 1-aminocyclohexane-1-carboxylic acid (Ac6c), the β-amino acid residue 1-aminocyclohexane acetic acid (β3,3Ac6c) and the γ-amino acid residue 1-aminomethylcyclohexaneacetic acid (gabapentin, Gpn). The crystal structure determination of peptides incorporating conformationally constrained α-, β- and γ- amino acid residues permitted the characterization of new types of hydrogen bonded turns and polymorphs. The studies enabled the precise determination of conformational and geometric parameters of two ω-amino acid residues, gabapentin and β 3,3Ac6c and provided detailed information about the conformational excursions possible for peptide molecules. This thesis is divided into 10 chapters. Chapter 1 gives a general introduction to the stereochemistry of the polypeptide chain, description of backbone torsion angles of α- and ω- amino acid residues and the major secondary structures of α-peptides, β-peptides, γ-peptides and hybrid peptides. A brief introduction to polymorphism and weak interactions, in particular aromatic interactions, is also provided, followed by a discussion on X-ray diffraction and solution to the phase problem. Chapter 2 describes the crystal structures of gabapentin zwitterion and its eight derivatives (Ananda, Aravinda, Vasudev et al., 2003). The crystal structure of the gabapentin zwitterions determined in this study is identical to that previously reported (Ibers, J. A. Acta Crystallogr. 2001, C57, 641-643). Eight of the nine achiral compounds crystallized in centrosymmetric space groups P21/c, C2/c or Pbca, while one derivative (Tos-Gpn-OH) crystallized in non-centrosymmetric space group Pna21 with four independent molecules in the asymmetric unit.The structural studies presented in this chapter reveal that the geminal substituents on the Cβ atom limits the values of dihedral angles θ1 and θ2 to ±60°, resulting in folded backbone conformations in all the examples. Intramolecular hydrogen bonds with 7-atoms in the hydrogen bond turn (C7) are observed in three derivatives, gabapentin hydrochloride (GPNCL), Boc-Gpn-OH (BGPNH) and Piv-Gpn-OH (PIVGPN), while a 9-atom hydrogen bonded turn (C9) is observed in Ac-Gpn-OH (ACGPH). Unique structural features, such as an unusual anti conformation of the COOH group (in ACGPH) and positional disorder of the cyclohexane ring (in BGPNN), indicating the co-existence of both the interconvertible chair conformations, are revealed by the crystal structure analyses. Chapter 3 describes the structural characterization of novel hydrogen bonded conformations of homo oligomers of Gpn. The crystal structures of three peptides, Boc-Gpn-Gpn-NHMe (GPN2), Boc-Gpn-Gpn-Leu-OMe (GPN2L) and Boc-Gpn-Gpn-Gpn-Gpn-NHMe (GPN4) provide the first crystallographic characterization of two new families of polypeptide structures, the C9 helices and C9 ribbons (Vasudev et al., 2005, 2007), in which the molecular conformations are stabilized by contiguous C9 turns formed by the hydrogen bonding between the CO group of residue (i) and the NH group of residue (i+2). The C9 hydrogen bond is characterized by a specific combination of the four torsion angles for the Gpn backbone, with the torsion angles θ1 and θ2 adopting g+/g+ or g /g- conformations. The structural analysis also permits precise determination of hydrogen bond geometry for the C9 structures, which is highly linear in contrast to the analogous γ-turn hydrogen bonds in α-peptides. A comparison of the backbone conformations in the three peptides reveals two classes of C9 hydrogen bonded secondary structures, namely C9 helices and C9 ribbons. The packing arrangement in these γ-peptides follows the same patterns as the helix packing in crystals of α-peptides. Chapter 4 describes ten crystal structures of short hybrid peptides containing the Gpn residue (Vasudev et al., 2007). In addition to the C7 and C9 hydrogen bonded turns which are defined by the backbone conformations at the Gpn residue, hybrid turns defined by a combination of backbone conformations at the α and γ-residues or at the β and γ-residues have been determined. Peptides Boc-Ac6c-Gpn-OH (ACGPH), Piv-Pro-Gpn-Val-OMe (PPGPV) and Boc-Val-Pro-Gpn-OH (VPGPH) reveal molecular conformation stabilized by intramolecular C9 hydrogen bonds, while Boc-Ac6c-Gpn-OMe (ACGPO) and Boc-Gpn-Aib-OH (GPUH) are stabilized by a C7 hydrogen bonded turn at the Gpn residue. An αγ hybrid turn with 12 atoms in the intramolecular hydrogen bonded rings (C12 turns) has been observed in the tripeptide Boc-Ac6c-Gpn-Ac6c-OMe (ACGP3), while βγ hybrid turns with 13 atoms in the hydrogen bonded ring (C13 turns) have been characterized in the tripeptides Boc-βLeu-Gpn-Val-OMe (BLGPV) and Boc- βPhe-Gpn-Phe-OMe (BFGPF). The two βγ C13 turns belong to two different categories and are characterized by different sets of backbone torsion angles for the β and γ residues. A γα C10 hydrogen bond, which is formed in the N→C direction (NHi ••• COi+2), as opposed to the regular hydrogen bonded helices of α-peptides, has also been observed in BFGPF. The Chapter provides a comparison of the backbone torsion angles of the Gpn residue in various hydrogen bonded turns and a brief comparison of the observed hydrogen bonded turns with those of the α-peptides. Chapter 5 describes the crystal structures of three αγ hybrid peptides which show C12/C10 mixed hydrogen bond patterns (Vasudev et al., 2007, 2008a; Chatterjee, Vasudev et al.,2008a). The insertion of gabapentin in the predominantly α-amino acid sequences in Boc-Ala-Aib-Gpn-Aib-Ala-OMe (AUGP5) and Boc-Leu-Gpn-Aib-Leu-Gpn-Aib-OMe results in the observation of helices stabilized by αα C10 (310-turn) and αγ C12 turns. The tetrapeptide Boc-Leu-Gpn-Leu-Aib-OMe reveals a novel conformation, stabilized by C12 (αγ) and C10 (γα) hydrogen bonds of opposite hydrogen bond directionalities. The conformations observed in crystals have been extended to generate C12 helix and C12/C10 helix with alternating hydrogen bond polarities in ( αγ)n sequences. The structure determination of three crystals, providing five molecular conformations, presented in this chapter provides the first crystallographic characterization of two types of helices predicted for the regular αγ hybrid peptides from theoretical calculations. The crystal structure of Boc-Ala-Aib-Gpn-Aib-Ala-OMe also provides an example for the co-existence of left-handed and right-handed helix in the asymmetric unit. Chapter 6 describes the structural studies of αγ hybrid peptides containing Aib and Gpn residues, and is divided into two parts. The first part presents the crystal structure analysis of peptides of sequence length 2 to 4, with alternating Aib and Gpn residues, and illustrates the conformational variability in αγ hybrid sequences as evidenced by the observation of conformational polymorphs (Chatterjee, Vasudev et al., 2008b; Vasudev et al., 2007; Ananda, Vasudev et al., 2005). The peptide Boc-Gpn-Aib-NHMe (GUN), Boc-Aib-Gpn-Aib-OMe (UGU), Boc-Gpn-Aib-Gpn-Aib-OMe (GU4O), Boc-Aib-Gpn-Aib-Gpn-OMe (UG4O) and Boc-Aib-Gpn-Aib-Gpn-NHMe (UG4N), all of which are potential candidates for exhibiting αγ C12 hydrogen bonds, reveal molecular conformations stabilized by diverse hydrogen bonded turns such as C7, C9, C12 and C17 in crystals. The conformational heterogeneity in this class of hybrid peptides is further evidenced by the observation of three polymorphs in the monoclinic space group P21/c for the tetrapeptide Boc-Aib-Gpn-Aib-Gpn-NHMe (UG4N), providing four independent peptide molecules adopting two distinct backbone conformations. In one polymorph, C12 helices terminated with an unusual three residue ( γαγ) C17 turn is observed, while the unfolding of helical conformation by solvent insertion into the backbone is observed in the other two polymorphs. The studies indicate the possible utility of Gpn residue in stabilizing locally folded conformations in the folding pathway, thus permitting their crystallographic characterization in multiple crystal forms. A discussion of the structural and conformational features of Gpn residues determined from all the crystal structures is presented in the Chapter, along with a φ-ψ plot for the Gpn residue. Part 2 of Chapter 6 describes the crystal structures of two octapeptides, Boc-Gpn-Aib-Gpn-Aib-Gpn-Aib-Gpn-Aib-OMe (GU8) and Boc-Leu-Phe-Val-Aib-Gpn-Leu-Phe-Val-OMe (LFVUG8), featuring C12 turns at the Aib-Gpn segments (Chatterjee, Vasudev et al., 2009). GU8 folds into a C12 helix flanked by C9 hydrogen bonds at both the termini, while LFVUG8 adopts β-hairpin conformation with a chain-reversing C12 turn at the central Aib-Gpn segment. A remarkable feature of the Aib-Gpn turn in the β-hairpin structure is the anti conformation about the Cβ-Cα (θ2) bond, which is the only example of a Gpn residue not adopting gauche conformation for both θ1 and θ2. The crystal structures of the two peptides, mimicking the two major secondary structural elements of α-peptides in hybrid polypeptides, permits a comparative study of the mode of molecular packing in crystals of α-peptides and hybrid peptides. The chapter also discusses theoretical calculations on αγ hybrid sequences, which reveal new types of C12 hydrogen bonded turns. Chapter 7 describes the crystal structures of conformationally biased tert-butyl derivatives of Gpn. The crystallographic characterization of the E (trans) and Z (cis) isomers of the residue,three protected derivatives and a tripeptide provides examples of C7 and C9 hydrogen bonded conformations, suggesting that the C7 and C9 hydrogen bonds can be formed by Gpn residues with both the chair conformations of the cyclohexane ring. Chapter 8 describes the systematic structural studies of the derivatives and peptides of the stereochemically constrained β- amino acid residue, β3,3Ac6c (Vasudev et al., 2008c). The backbone torsion angles φ and θ adopt gauche conformation in majority of the examples, owing to the presence of a cyclohexane ring on the Cβ atom. In contrast to Gpn, β3,3Ac6c does not show strong preference for adopting intramolecularly hydrogen bonded conformations. Of the 16 crystal structures determined, intramolecular hydrogen bonds involving the β-residue are observed only in 4 cases. The amino acid zwitterion (BAC6C), the hydrochloride (BACHCL) and the dipeptide Boc-β3,3Ac6c-β3,3Ac6c-NHMe (BAC62N) form N-H•••O hydrogen bonds with 6-atoms in the hydrogen bond ring (C6 turns). An αβ hybrid C11 hydrogen bonded turn is characterized in the dipeptide Piv-Pro-β3,3Ac6c-NHMe, which is distinctly different from the C11 hydrogen bonds observed in αβ hybrid peptide helices. Several unique structural features such as a dynamic disorder of the hydrogen atom of the carboxylic acid group (in BBAC) and cis geometry of the urethane bond (in BBAC, BAC62N and BPBAC) have been observed in this study. A comparison of the backbone conformations of β3,3Ac6c with other β- amino acid residues is also provided. Chapter 9 describes the crystallographic characterization of a new polymorph of gabapentin monohydrate and crystal structures of the zwitterions of E and Z isomers of tert-butylgabapentin and its hydrochloride and hydrobromide (Vasudev et al., 2009). A comparison of the crystal structures of the monoclinic form (Ibers, J. A. Acta Crystallogr. 2001, C57, 641-643) of gabapentin monohydrate and the newly characterized orthorhombic form reveals identical molecular conformations and intermolecular hydrogen bond patterns in both the polymorphs. The two polymorphs show differences in the orientation of molecules constituting a layer of hydrophobic interactions between the cyclohexyl side chains. A comparison of the packing arrangements of the zwitterionic amino acid molecules in the crystal structures of gabapentin monohydrate, the tert-butyl derivatives and other co-crystals of gabapentin that had been characterized so far, is provided which would facilitate prediction of new polymorphs of the widely used drug molecule, Gpn. Chapter 10 describes the crystallization of α-peptide helices in multiple crystal forms (Vasudev et al., 2008b). Crystal structures of two peptides, Boc-Leu-Aib-Phe-Phe-Leu-Aib-Ala-Ala-Leu-Aib-OMe (LFF), Boc-Leu-Aib-Phe-Ala-Leu-Ala-Leu-Aib-OMe (D1) in two crystal forms and the crystal structure of a related sequence, Boc-Leu-Aib-Phe-Ala-Phe-Aib-Leu-Ala-Leu-Aib-OMe (D10) permit an analysis of the molecular conformation and packing patterns of peptide helices in crystals. The two polymorphs of LFF, crystallized in the space groups P21 and P22121, reveal very similar molecular conformation (α/310-helix) in both the polymorphic crystals; the two forms differ significantly in the pattern of solvation. The crystal structure determination of a monoclinic (P21) and an orthorhombic polymorph (P21212) of D1 provides five different peptide conformations, four of which are α-helical and one is a mixed 310/α-helix. The crystal structure determination of the three peptides provide an opportunity to compare the nature and role of aromatic interactions in stabilizing molecular conformation and packing and its significance in the observation of polymorphism. An analysis of the Cambridge Structural Database and a model for nucleation of crystals in hydrophobic peptide helices are also discussed.

This thesis set out to explore the conformational properties of short designed peptide sequences, in which transitions between structural states may be anticipated. The use of conformationally constrained residues like α-aminoisobutyric acid (Aib) and D-proline (DPro) permits the design of model sequences for structural studies. The principle of imposing conformational constraints by multiple substitutions at backbone atoms in aminoacid residues may also be extended to the higher homologs of α-amino acids, namely β and residues. The experimental results presented in this thesis also examine the potential of using cross-strand interactions between aromatic residues as a probe of structure in designed peptide β-hairpins. Chapter 1 provides a very brief introduction to the necessary background on which the experimental studies in this thesis are based. Chapter 2 describes studies aimed at establishing chain length effects on helix-hairpin conformational distributions in short synthetic sequences, containing centrally positioned Aib-DAla and Aib-Aib segments.The Aib-DAla dipeptide segment has a tendency to form both type-I'/III' and type-I/III β-turns. The occurrence of prime turns facilitates the formation of β-hairpin conformations, while type-I/III turns can nucleate helix formation. The octapeptide Boc-Leu-Phe-Val-Aib-DAla-Leu-Phe-Val-OMe (1) has been previously shown to form a β-hairpin in the crystalline state and in solution. The effects of sequence truncation have been examined using the model peptides Boc-Phe-Val-Aib-Xxx-Leu-Phe-NHMe (2, 6), Boc-Val-Aib-Xxx-Leu-NHMe (3, 7) and Boc-Aib-Xxx-NHMe (4, 8), where Xxx = DAla, Aib. For peptides with central Aib-Aib segments, Boc-Phe-Val-Aib-Aib-Leu-Phe-NHMe (6), Boc-Val-Aib-Aib-Leu-NHMe (7) and Boc-Aib-Aib-NHMe (8) local helical conformations have been established by NMR studies in both hydrogen bonding (CD3OH) and non-hydrogen bonding (CDCl3) solvents. In contrast, the corresponding hexapeptide Boc-Phe-Val-Aib-DAla-Leu-Phe-Val-NHMe (2) favors helical conformations in CDCl3 and β-hairpin conformations in CD3OH. β-Turn conformations (type-I /III) stabilized by intramolecular 4 1 hydrogen bonds are observed for the peptide Boc-Aib-DAla-NHMe (4) and Boc-Aib-Aib-NHMe (8) in crystals. The tetrapeptide Boc-Val-Aib-Aib-Leu-NHMe (7) adopts an incipient 310-helical conformation stabilized by three 4 1 hydrogen bonds. The peptide Boc-Val-Aib-DAla- Leu-NHMe (3) adopts a novel -turn conformation, stabilized by three intramolecular hydrogen bonds (two 4 1 and one 5 1). The Aib-DAla segment adopts a type-I' β-turn conformation. The observation of the NOE Val(1) NH HNCH3 (5), in CD3OH, suggests that the solid state conformation of peptide 3 is maintained in methanol solutions. Peptide hairpins provide an ideal scaffold for exploring cross-strand interactions between residues on facing antiparallel strands. Chapter 3 reports studies directed towards probing, aromatic interactions between facing Phe residues, positioned at the non-hydrogen bonding positions in designed octapeptide β-hairpins. The studies described in this Chapter employ ring current shifted aromatic proton resonances as a means of probing aromatic ring orientations. Crystal structures of eight peptide -hairpins with the sequence Boc-Leu-Phe-Val-Xxx-Yyy-Leu-Phe-Val-OMe revealed that the Phe(2) and Phe(7) aromatic rings are in close spatial proximity, with a centroid-centroid distance (Rcen) of 4.4Å to 5.4Å between the two phenyl rings. Proton NMR spectra in chloroform and methanol solutions reveal a significant upfield shift of the Phe(7) C , ′ H2 protons (6.65 ppm to 7.04 ppm). Specific assignments of the aromatic protons have been carried out in the peptide Boc-Leu-Phe-Val-DPro-LPro-Leu-Phe-Val-OMe (6). The anticipated ring current shifts have been estimated from the aromatic ring geometries observed in crystals for all eight peptides. Only one of the C , ′ H proton lies in the shielding zone, with rapid ring flipping, resulting in averaging between the two extreme chemical shifts. An approximate estimate of the population of conformations which resemble crystal state orientations may be obtained. Key nuclear Overhauser effects (NOEs) between facing Phe sidechains provide support for close similarity between the solid state and solution conformations. Temperature dependence of aromatic ring proton chemical shifts and line widths for peptide 6 (Boc-Leu-Phe-Val-DPro-LPro-Leu-Phe-Val-OMe) and the control peptide Boc-Leu-Val-Val-DPro-Gly-Leu-Phe-Val-OMe establish an enhanced barrier to ring flipping, when the two Phe rings are in proximity. Modeling studies suggest that small, conformational adjustments about the C -C ( 1), and C -C ( 2) bonds of the Phe residues may be required in order to permit unhindered, uncorrelated flipping of both the Phe rings. The maintenance of specific aromatic ring orientations in organic solvents provides evidence for significant stabilizing interactions. Earlier studies from this laboratory established that a centrally positioned DPro-LPro-DAla segment could induce hairpin formation in nonapeptide sequences, facilitated by a three residue loop segment. The DAla residue at position 6 in the nonapeptide Boc-Leu-Phe-Val-DPro-LPro-DAla-Leu-Phe-Val-OMe has been shown to adopt a left handed helical (αL) conformation. The studies described in Chapter 4, examine the effects of aminoacid replacements at positions 5 and 6. NMR studies on eight nonapeptides, with the general sequence Boc-Leu-Phe-Val-DPro-Xxx-Yyy-Leu-Phe-Val-OMe are described. In the case of peptides with a central DPro-LPro-Yyy sequence, two kinds of hairpin conformations are formed in solution. These are; i) β-hairpin structures with a central three residue loop, resulting in registered antiparallel tripeptide strands, and ii) a slipped hairpin structure, nucleated by a central DPro-LPro type-II β-turn, with residue 6 being incorporated into the C-terminal strand. The three residue loop β-hairpins are favored for DAla(6) and Aib(6), while the LAla(6) peptide favors a “slipped” hairpin structure. Replacement of the Pro(5) residue by LAla results in a reduced population of three residue hairpins in the nonapeptide with the DPro-LAla-DAla segment. Replacement of Pro(5) by Aib, abolished hairpin formation. Aromatic proton chemical shifts provide a convenient diagnostic for the presence of three residue loop hairpin conformations in these nonapeptides. A great deal of current interest has focused on the conformations of peptides incorporating β and γ aminoacid residues. Earlier studies from this laboratory have focused on the conformational properties of the β,β -disubstituted γ residue gabapentin (1-aminomethylcyclohexane acetic acid). Subsequent work with the related β aminoacid β3,3Ac6c (1-aminocyclohexaneacetic acid) revealed that intramolecularly hydrogen bonded conformations are infrequently observed in short peptides. The studies described in Chapter 5, examine the conformational properties for model peptides containing the isomeric β-aminoacid, β2,2Ac6c (1-aminomethylcyclohexane-1-carboxylic acid). The effect of gem dialkyl substituents on the backbone conformations of amino acid residues in peptides, has been investigated using four model peptides, Boc-Xxx-2,2Ac6c-NHMe [Xxx = Leu (1), Phe(2)] and Boc-Xxx- 3,3Ac6c-NHMe [Xxx = Leu (3), Phe(4)]. Tetrasubstituted carbon atoms restrict the ranges of stereochemically allowed a C11 helical turn, which is a backbone expanded analog of the type III -turn in sequences. The crystal structure of the peptide Boc-Phe- 3,3Ac6c-NHMe (4) establishes a the asymmetric unit adopt backbone torsion angles of opposite signs. In one of the molecules, the Phe residue adopts an unfavourable backbone conformation, with the energetic penalty being offset by favourable aromatic interactions between proximal molecules in the crystal. NMR studies provide evidence for the maintenance of folded structures in solution, in these hybrid sequences. The result presented in this thesis suggests that it should be possible to construct designed synthetic peptides, which can undergo transitions between two distinct and energetically favourable conformational states. The ability to design peptide sequences that can undergo switching between helical and β-hairpin states, or between hairpin structures with variations in connecting loop length may prove valuable in providing further insights into the factors influencing conformational dynamics.

This thesis describes NMR studies which probe weak interactions between amino acid side chains in folded peptide structures. Aromatic/aromatic interactions between facing phenylalanine residues have been probed in antiparallel β-sheets, while aromatic/proline interactions have been examined using cyclic peptide disulfides that occur in the venom of marine cone snails. Novel intramolecular hydrogen bonded structures in hybrid peptides containing backbone homologated residues, specifically γ-amino acids, are also described. Chapter 1 provides a brief background to the principles involved in the design of antiparallel β-sheet structures and an introduction to previous studies on aromatic/aromatic and aromatic/proline interactions in influencing peptide conformations. A summary of the NMR methods used is also presented. Chapter 2 discusses the structural characterisation of a designed 14 residue, three stranded β-sheet peptide, Boc-LFVDP-PLFVADP-PLFV-OMe (LFV14). The results described in this Chapter support the presence of multiple conformational states about the χ1 (Cα-Cβ) torsional degree of freedom for the interacting aromatic pairs in solution. Chapter 3 presents the structural characterisation of a designed 19 residue three stranded hybrid β-sheet peptide, Boc-LVβFVDPGLβFVVLDPGLVLβFVV-OMe (BBH19). β-amino acid residues (β-phenylalanine, βPhe) were incorporated at facing positions on antiparallel β-sheets. The BBH19 structure provides an example of interaction between the N and C-terminal strands in a three stranded structure with an α/β hybrid backbone. Chapter 4 focuses on studies of the conformations of the contryphan In936 (GCVDLYPWC*) from Conus inscriptus and the related peptide Lo959 (GCPDWDPWC*) from Conus loroissi. Both peptides possess a macrocyclic 23 membered ring, with multiple accessible conformational states. Chapter 5 describes conformational analysis of a novel 20 membered cyclic peptide disulfide, CIWPWC (Vi804), from Conus virgo. NMR structures were calculated for Vi804 and an analog peptide, CIDWPWC, DW3-Vi804. Chapter 6 explores the solution conformation of hybrid sequences containing α and γ residues. Oligopeptides of the type (αγ)n and (αγγ)n have been studied in solution by NMR methods. Chapter 7 provides a summary of the results described in this thesis and highlights the major conclusions.

Chemical conjugates of fibrin-specificantibodies and plasminogen activators. Urokinase or tPA were linked covalently toamonoclonal antibody specific for the amino terminus of the beta chain of human fibrin (59D8) by means of the unidirectionalcross-linking reagent SPDP. The fibrinolytic potency of the conjugates at equal amidolytic activities was compared to the native plasminogen activators in an assay measuring lysis of 1251-fibrin monomer covalently linked to Sepharose CL-4B. Urokinase was least potent, tPA exhibited a 10fold increase in fibrinolysis whereas both the urokinase and tPA antibody conjugates and a urokinase-Fab conjugate were 250fold more potent than urokinase and 25 fold more potent than tPA. Enhanced fibrinolysis was fully inhibited by b peptide indicating its dependence on antigen binding. In a plasma assay conjugates of tPA orUK to antibody produced a 3.2- to 4.5-fold enhancement in clot lysis in human plasma over that of the respective unconjugated plasminogen activator. However, the UK-59D8 conjugate was only as potent as tPAalone. Antibody-conjugated tPA or UK consumed less fibrinogen, alpha 2-antiplasminand plasminogen than did the unconjugatedactivators, at equipotent thrombolytic concentrations. In a quantitative rabbit thrombolysis model, the activity of the purified conjugate was compared with that oftPA alone and that of a conjugate betweentPA and a digoxin-specific monoclonal antibody. After correction for spontaneous lysis, tPA-59D8 was shown to be 2.8 to,9.6times more potent than tPA alone. Unconjugated tPA and tPA-digoxin were equipotent.At equivalent thrombolytic concentrations, tPA-59D8 degraded less fibrinogen and consumed less alpha 2-antiplasmin than did tPA alone. These results suggest that tPA can be efficiently directed to the site of a thrombus by conjugation to an antifibrin monoclonal antibody, resulting in both more potent and more selective thrombolysis.A recombinant fusion protein comprising a fibrin-specific antibody site and theB chain of tPA. The rearranged 59D8 heavychain gene was cloned and combined in theexpression vector pSV2gpt withsequence coding for a portion of the Gamma 2b constant region and the catalytic beta chain of t-PA. This construct was transfected into heavy chain loss variant cells derived from the 59D8 hybridoma. Recombinant protein was purified by affinitychromatography and analyzed with Western blots. These revealed a 65-kD heavy chain-t-PA fusion protein that is secreted in association with the 59D8 light chain in the form of a 170-kD disulfide linked dimer. A chromogenic substrate assay showed the fusion protein to have 70 percent of the peptidolytic activity of native t-PA and to activate plasminogen as efficiently as t-PA. In a competitive binding assay, reconstituted antibody was shown to have a binding profile similar to that of native 59D8. Thus by recombinant techniques we have produced a novel hybrid protein capable of high affinity fibrin binding andplasminogen activation.Chemical conjugates between a fibrin-specific and a tPA-specific antibody. A heteroantibody duplex (duplex) with specificities for both tPA and fibrin was synthesized by conjugating iminothiolane-modified anti-tPA monoclonal antibody (TCL8) toantifibrin antibody 59D8. Addition of both duplex and tPA to a plasma clot assay gave more lysis (200 units produced 23.1 lysis; 400 units, 29.5 lysis) than did tPAalone (200 units, 1.8% lysis; 400 units,19% lysis). Despite increased potency associated with duplex addition, fibrinogen and alpha-2-antiplasmin levels at equal tPA concentrations did not differ. Thus, itis possible to concentrate tPA (added separately) to the site of a thrombus in plasma using a heteroantibody duplex with specificities for both tPA and fibrin.Biosynthetically produced heteroduplexantibodies that are both fibin and tPA-specific. The bispecific antibodies were prepared in two ways. First, polyethylene glycol-mediated fusions were performed with two different hybridoma cell lines: anti-fribrin b chain producer, 59D8 and anti-tPA producer, TCL8. TCL8 cells were selected for HPRT-minus variants and then fused with TK-deficient 59D8 cells. One cell line, F36.23, possessed both anti-human fibrin and anti-human t-PA immunoreactivities. A second method yielded another bispecific antibody, F32.1. This cell line was selected after fusing TCL8 (HPRT-minus) cells with spleen cells from a mouseimmunized with a fibrin-like peptide corresponding to the amino terminus of fibrinalpha-chains. Affinity-purified F32.1 andF36.23 retained anti-fibrin and anti-t-PAactivity and enhanced fibrinolytic potency of tPA by a factor of 10.