Susceptibility of glycans to β-elimination in Fmoc-based O-glycopeptide synthesis

order to investigate the possible extent of β-elimination occuring in Fmoc-based continuous-flow solid-phase glycopeptide synthesis, the influence of the pK_b of the base used for N^α-deprotection has been studied. A glycosylated pentapeptide as synthesized using 50%morpholine, 10%, piperidine or 2% DBU, respectively, in DMF for deprotection. The dehydropentapeptide N^α-.Ac-Thr-Thr-δAba-Val-Thr-NH₂, which would be formed in the case of β-elimination, was prepared independently and used as a control in HPLC analysis; however, this product was not formed under any of the deprotection conditions applied. Furthermore, a 23 amino acid long glycopeptide from human intestinal mucin was prepared using 2% DBU as a base for Fmoc cleavage, and similarly no β-elimination was observed. The glycopeptide products were subjected to a prolonged treatment with sodium hydroxide in methanol/water without significant formation of byproducts, and the pure glycopeptides were isolated and characterized by ¹H-NMR spectroscopy.     

New Mutants of Epsilon Toxin from Clostridium perfringens with an Altered Receptor-Binding Site and Cell-Type Specificity

Epsilon toxin (Etx) from Clostridium perfringens is the third most potent toxin after the botulinum and tetanus toxins. Etx is the main agent of enterotoxemia in ruminants and is produced by Clostridium perfringens toxinotypes B and D, causing great economic losses. Etx selectively binds to target cells, oligomerizes and inserts into the plasma membrane, and forms pores. A series of mutants have been previously generated to understand the cellular and molecular mechanisms of the toxin and to obtain valid molecular tools for effective vaccination protocols. Here, two new non-toxic Etx mutants were generated by selective deletions in the binding (Etx-ΔS188-F196) or insertion (Etx-ΔV108-F135) domains of the toxin. As expected, our results showed that Etx-ΔS188-F196 did not exhibit the usual Etx binding pattern but surprisingly recognized specifically an O-glycoprotein present in the proximal tubules of the kidneys in a wide range of animals, including ruminants. Although diminished, Etx-ΔV108-F135 maintained the capacity for binding and even oligomerization, indicating that the mutation particularly affected the pore-forming ability of the toxin.     

9-Fluorenylmethyl esters

N ^α-Protected amino acid 9-fluorenylmethyl esters (Fm esters) were prepared by imidazole-catalyzed transesterification of active esters with 9-fluorenylmethanol (9-hydroxymethylfluorene). The new carboxyl protection is unaffected by acids, but is efficiently removed by β-elimination under the influence of secondary and tertiary amines. Primary amines and ammonia can cause slight amide formation. Deblocking was achieved also by catalytic hydrogenation.     

Interlocking activities of DNA polymerase β in the base excision repair pathway

Base excision repair (BER) is a major cellular pathway for DNA damage repair. During BER, DNA polymerase β (Polβ) is hypothesized to first perform gap-filling DNA synthesis by its polymerase activity and then cleave a 5′-deoxyribose-5-phosphate (dRP) moiety via its dRP lyase activity. Through gel electrophoresis and kinetic analysis of partial BER reconstitution, we demonstrated that gap-filling DNA synthesis by the polymerase activity likely occurred after Schiff base formation but before β-elimination, the two chemical reactions catalyzed by the dRP lyase activity. The Schiff base formation and β-elimination intermediates were trapped by sodium borohydride reduction and identified by mass spectrometry and X-ray crystallography. Presteady-state kinetic analysis revealed that cross-linked Polβ (i.e., reduced Schiff base) exhibited a 17-fold higher polymerase efficiency than uncross-linked Polβ. Conventional and time-resolved X-ray crystallography of cross-linked Polβ visualized important intermediates for its dRP lyase and polymerase activities, leading to a modified chemical mechanism for the dRP lyase activity. The observed interlocking enzymatic activities of Polβ allow us to propose an altered mechanism for the BER pathway, at least under the conditions employed. Plausibly, the temporally coordinated activities at the two Polβ active sites may well be the reason why Polβ has both active sites embedded in a single polypeptide chain. This proposed pathway suggests a corrected facet of BER and DNA repair, and may enable alternative chemical strategies for therapeutic intervention, as Polβ dysfunction is a key element common to several disorders.

One of the major cellular pathways for repair of DNA damage is base excision repair (BER) (1–5). In this pathway (Scheme 1A), DNA lesions are removed by glycosylases (e.g., uracil by uracil–DNA glycosylase [UDG]) before the damaged DNA strand is incised by apurinic/apyrimidinic (AP) endonuclease, resulting in a single-nucleotide gap flanked by a 3′-OH and a 5′-deoxyribose-5-phosphate (dRP) moiety. Subsequently, DNA polymerase β (Polβ) is presumed to first catalyze gap-filling DNA synthesis through its DNA polymerase activity and then perform dRP cleavage via its dRP lyase activity, leaving a nicked DNA substrate for ligation by either Ligase III/XRCC1 or Ligase I (6–12).Open in a separate windowScheme 1.The BER pathway. (A) The BER pathway in the literature as cited in the Introduction. (B) Our proposed BER pathway.The dRP lyase active site resides within the 8-kDa N-terminal domain of Polβ, whereas the polymerase active site sits at the palm subdomain (SI Appendix, Fig. S1A) (7). Previously, Polβ has been shown to remove a dRP moiety through a Schiff base–mediated β-elimination reaction (13) rather than through hydrolysis (7–10, 14, 15). A Schiff base is generated following nucleophilic attack by the side chain of an active site lysine residue on the sugar C1′ atom of the dRP moiety (step 1 in Scheme 2). Whereas biochemical data suggest that K72 in human polymerase-β (hPolβ) acts as the active site nucleophile, conflicting evidence as well as a lack of supporting structural data have complicated understanding of the dRP cleavage mechanism (10, 14, 15). For instance, mutation of K72 to alanine does not fully abrogate the dRP lyase activity, suggesting that a different residue may support the nucleophilic attack on the C1′ (11). Furthermore, only limited conclusions can be drawn from the existing binary crystal structures of hPolβ bound to either a single-nucleotide gapped DNA substrate (hPolβ•DNA^P) (SI Appendix, Fig. S2B) containing only a 5′-phosphate, rather than a full dRP moiety, on the downstream primer (SI Appendix, Fig. S2 A, i) (16) or a nicked DNA substrate (hPolβ•DNA^THF) (SI Appendix, Fig. S2C) containing a 5′-dRP mimic (SI Appendix, Fig. S2 A, ii) (11). Due to the lack of the deoxyribose moiety in the structure of hPolβ•DNA^P, information about the dRP cleavage mechanism is lacking. On the other hand, in the structure of hPolβ•DNA^THF, the nonnatural dRP mimic was bound in a nonproductive docking site stabilized through the interaction between its 5′-phosphate and K68 (SI Appendix, Fig. S2C). This nonproductive site is distinct from the putative dRP lyase active site as the Nε atom of K72 is more than 10 Å from the dRP sugar C1′ (11). In fact, from this position, the dRP must rotate ∼120° around the 3′-phosphate to be in close-enough proximity to the Nε atom of K72 for nucleophilic attack to occur (SI Appendix, Fig. S2C). Furthermore, the active site residues responsible for stabilizing the reactive ring-opened aldehyde state of the dRP moiety and abstracting a proton from the ribose C2′ atom to facilitate β-elimination (Scheme 2) remain unidentified.Open in a separate windowScheme 2.Proposed chemical mechanism for the dRP lyase activity of hPolβ. Specific water molecules are denoted as X, Y, and Z.Biochemical studies of the processing of dRP moieties in yeast cell-free extract (17), steady-state kinetic studies of fully reconstituted human BER (4), and investigation of the numbers of endogenous AP sites in genomic DNA of rats and human tissue (5) all suggest that dRP cleavage is the rate-limiting step of the entire BER pathway. However, there is no experimental evidence to indicate that all potential steps associated with dRP cleavage by the lyase activity of Polβ (Scheme 2) occur after gap-filling DNA synthesis catalyzed by the polymerase activity. For example, if facile Schiff base formation occurs before and faster than nucleotide incorporation, the covalently linked Polβ–DNA intermediate, rather than the noncovalent binary complex Polβ•DNA, may catalyze gap-filling DNA synthesis. This possibility has never been investigated, and all previously published in vitro studies have used DNA substrates like either DNA^P (SI Appendix, Fig. S2 A, i) (18–24) or a gapped DNA substrate containing a dRP mimic (SI Appendix, Fig. S2 A, ii) (11, 25).Here, we generated a natural dRP moiety by using either UDG to process a nicked DNA substrate containing a 2′-deoxyuridine or UDG and apurinic/apyrimidinic endonuclease 1 (APE1) to initiate BER on a double-stranded DNA substrate containing a 2′-deoxyuridine. Addition of hPolβ, correct deoxynucleoside triphosphate (dNTP), and then, sodium borohydride (NaBH₄) to the dRP-containing DNA products allowed for the capture of a reduced Schiff base and a β-elimination intermediate produced via hPolβ-catalyzed dRP cleavage (Scheme 2). Through X-ray crystallographic, kinetic, and mass spectrometric (MS) analysis of these cross-linked hPolβ complexes, we envisioned a detailed chemical mechanism for the dRP lyase activity of hPolβ. In addition, we utilized presteady-state kinetic methods to evaluate the impact of the reduced Schiff base intermediate on the efficiency and fidelity of gap-filling DNA synthesis by the polymerase activity of hPolβ. Finally, we employed time-resolved X-ray crystallography to structurally characterize intermediates of gap-filling DNA synthesis by cross-linked hPolβ. Based on several lines of experimental evidence, we proposed a modified BER pathway (Scheme 1B), which posits an interlocking mechanism in which gap-filling DNA synthesis by the polymerase activity occurs between Schiff base formation and β-elimination, the two steps catalyzed by the lyase activity.     

Reduced CO2-elimination during combined high-frequency ventilation compared to conventional pressure-controlled ventilation in surfactant-deficient piglets

Background: Combined high-frequency ventilation (CHFV) combines a conventional low-frequency component with superimposed high-frequency jet pulses. The intention is to overcome the limited CO₂-elimination of high-frequency ventilation, and to decrease airway pressures and enhance hemodynamic performance by reducing the conventional component. The present study was performed to compare the effects of conventional continuous positive-pressure ventilation (CPPV) on gas exchange, airway pressures and cardiac output to those of CHFV at matched minute volume (MV) and mean airway pressure (MPAW). Methods: Sixteen anaesthetised piglets with lavage-induced surfactant deficiency were ventilated with CPPV, with positive end-expiratory pressure (PEEP) set to obliterate the lower inflection point of the inspiratory pressure-volume loop. This setting was compared to CHFV during which 50% of the total MV was applied as superimposed jet pulses of 20 Hz at otherwise unchanged settings, and to CPPV at a PEEP level which was reduced (CPPV_red) until MPAW matched MPAW during CHFV. Gas exchange, airway pressures and hemodynamics were measured after the ventilatory setting had been applied for 20 min. Results: MPAW decreased from (median) 2.7 kPa with CPPV to 2.4 kPa with CHFV (P≤0.05). Peak inspiratory pressure was 3.6 kPa with CPPV, 3.2 kPa with CHFV, and 3.2 kPa with CPPV_red (P≤0.05 for differences to CPPV), respectively. PaCO₂ was comparable during CPPV (5.9 kPa), CPPV_red and CHFV_CO2, while it increased during CHFV (6.8 kPa, (P≤0.05)). Cardiac output did not differ significantly between the settings. Conclusions: In the porcine lavage model, CO₂-elimination is reduced during CHFV compared to CPPV at matched minute volume. At matched mean airway pressure, CHFV fails to reduce peak inspiratory airway pressure and to improve hemodynamic performance compared to CPPV.     

Semisynthesis of the native sequence of horse heart cytochrome c-(66–104)-nonatriacontapeptide

The syntheses of some derivatives of horse cytochrome c-(66–79)-tetradeca-peptide are presented. The syntheses are so designed that analogues of this phylogenetically well preserved sequence can be obtained also. The compounds were intended as synthons for the semisynthesis of 65-homoserine-cytochrome c, which we described earlier. A requisite for this project was the C-terminal tetracosapeptide fragment of the protein, accessible through degradation of cytochrome c with cyanogen bromide. The five e amino groups in this compound are reversibly protected with the 2-(methylsulfonyl)ethyloxycarbonyl function, which is resistant to acid and causes little impairment of solubility. The condensation of the fragments leading to the native sequence of horse cytochrome c-(66–104)-nonatriacontapeptide is presented also. The syntheses were performed using the solution strategy. Some unexpected ring closing reactions involving tyrosine and tert.-butyl prolylasparaginylcarbazate, are described.     

O-glycosidic linkage in glycoprotein isolates from human ocular mucus

Alkaline beta-elimination and sodium borohydride reduction were used to study the O-glycosidic linkage of N-acetylgalactosamine to seryl and threonyl residues in the high molecular weight crude glycoprotein fractions isolated from human ocular mucus. Pure mucins, BSM and OSM, were used as models. Data are presented for the existence of such O-glycosidic linkages. It was estimated from sodium borohydride reaction that at least 22% of the total peptide bonded hydroxyamino acid residues are linked O-glycosidically. These findings, as a continuation of our previous work on the isolation and chemical characterization of human ocular mucus, provided additional evidence of the mucin nature of the glycoprotein isolates.     

Moisture-Induced Aggregation of Lyophilized Insulin

A critical problem in the storage and delivery of pharmaceutical proteins is aggregation in the solid state induced by elevated temperature and moisture. These conditions are particularly relevant for studies of protein stability during accelerated storage or for proteins loaded in polymeric delivery devices in vivo. In the present investigation, we have found that, when exposed to an environment simulating these conditions, lyophilized insulin undergoes both covalent and noncovalent aggregation. The covalent process has been elucidated to be intermolecular thiol-catalyzed disulfide interchange following -elimination of an intact disulfide bridge in the insulin molecule. This process is accelerated by increasing the temperature and water content of the insulin powder or by performing lyophilization and/or dissolution of insulin in alkaline media. The aggregation can be ameliorated by the presence of Cu ²⁺ , which presumably catalyzes the oxidization of free thiols. The water sorption isotherm for insulin reveals that the extent of aggregation directly correlates with the water uptake by the lyophilized insulin powder, thus pointing to the critical role of protein conformational mobility in the aggregation process.     

Degradation Pathways for Recombinant Human Macrophage Colony-Stimulating Factor in Aqueous Solution

Recombinant human macrophage colony-stimulating factor (rhM-CSF) promotes macrophage proliferation and activity. rhM-CSF clinical trials are currently in progress and require a stable, pharmaceutically acceptable dosage form. This report documents pH effects on rhM-CSF degradation profiles in aqueous solution, with an emphasis on identifying degradation products. Thus, highly purified rhM-CSF was maintained at 30 to 50°C in solutions adjusted to pH 2 to 10. Stressed samples were analyzed by SDS-PAGE, reverse-phase HPLC, size exclusion HPLC, scanning microcalorimetry, and murine bone marrow activity. The results show maximal protein stability in the region pH 7 to 8. Degradation product chromatographic and electrophoretic analyses show distinctly different degradation product profiles in acidic versus alkaline solution. For samples stressed in acidic solution, degradation products were isolated chromatographically and electrophoretically. These degradation products were characterized by N-terminal amino acid sequencing, fast-atom bombardment mass spectrometry, and peptide mapping. The results show that the major degradation pathway in acidic solution involves peptide cleavage at two sites: aspartate₁₆₉-proline₁₇₀ and aspartate₂₁₃-proline₂₁₄. A third potential cleavage site (aspartate₄₅-proline₄₆) remains intact under conditions that cleave Asp₁₆₉-Pro₁₇₀ and Asp₂₁₃-Pro₂₁₄. In alkaline solution, degradation proceeds via parallel cleavage and intramolecular cross-linking reactions. A -elimination mechanism is proposed to account for the degradation in alkaline solution. Consistent with literature observations, the rhM-CSF N-terminal cleavage products retain biological activity.