FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2163727669> ?p ?o ?g. }

• W2163727669 endingPage "6189" @default.
• W2163727669 startingPage "6183" @default.
• W2163727669 abstract "Recent studies from many laboratories revealed remarkable structural, distributional, and functional diversities of oligo/polysialic acids (OSA/PSA) that exist in organisms ranging from bacteria to man. These diversities are further complicated by the fact that OSA/PSA spontaneously form lactones under even mildly acidic conditions. By using high performance anion-exchange chromatography (HPAEC) with nitrate eluents, we found that lactonization of α2,8-linked OSA/PSA (oligo/poly-Neu5Ac, oligo/poly-Neu5Gc and oligo/poly-KDN) proceeds readily, and the lactonization process displays three discrete stages. The initial stage is characterized by limited lactonization occurring between two internal sialic acid residues, reflected by a regular pattern of lactone peaks interdigitated with non-lactonized peaks on HPAEC. In the middle stage, multiple lactonized species are formed from a molecule with a given degree of polymerization (DP), in which the maximum number of lactone rings formed equals DP minus 2. At the final stage, completely lactonized species become the major components, resulting in drastic changes in the physicochemical properties of the sample. Interestingly, the smallest lactonizable OSA are tetramer, trimer, and dimer at the initial, middle, and final stages, respectively. At any of the stages, OSA/PSA of higher DP lactonize more rapidly, but all the lactone rings rapidly open up when exposed to mild alkali. Lactonized OSA/PSA are resistant to both enzyme- and acid-catalyzed glycosidic bond cleavage. The latter fact was utilized to obtain more high DP oligo/poly(α2,8-Neu5Gc) chains from a polysialoglycoprotein. Our results should be useful in preparation, storage, and analysis of OSA/PSA. Possible biological significance and bioengineering potentials of lactonization are discussed. Recent studies from many laboratories revealed remarkable structural, distributional, and functional diversities of oligo/polysialic acids (OSA/PSA) that exist in organisms ranging from bacteria to man. These diversities are further complicated by the fact that OSA/PSA spontaneously form lactones under even mildly acidic conditions. By using high performance anion-exchange chromatography (HPAEC) with nitrate eluents, we found that lactonization of α2,8-linked OSA/PSA (oligo/poly-Neu5Ac, oligo/poly-Neu5Gc and oligo/poly-KDN) proceeds readily, and the lactonization process displays three discrete stages. The initial stage is characterized by limited lactonization occurring between two internal sialic acid residues, reflected by a regular pattern of lactone peaks interdigitated with non-lactonized peaks on HPAEC. In the middle stage, multiple lactonized species are formed from a molecule with a given degree of polymerization (DP), in which the maximum number of lactone rings formed equals DP minus 2. At the final stage, completely lactonized species become the major components, resulting in drastic changes in the physicochemical properties of the sample. Interestingly, the smallest lactonizable OSA are tetramer, trimer, and dimer at the initial, middle, and final stages, respectively. At any of the stages, OSA/PSA of higher DP lactonize more rapidly, but all the lactone rings rapidly open up when exposed to mild alkali. Lactonized OSA/PSA are resistant to both enzyme- and acid-catalyzed glycosidic bond cleavage. The latter fact was utilized to obtain more high DP oligo/poly(α2,8-Neu5Gc) chains from a polysialoglycoprotein. Our results should be useful in preparation, storage, and analysis of OSA/PSA. Possible biological significance and bioengineering potentials of lactonization are discussed. Recent studies on oligo/polysialic acids 1In this paper, we refer to those with DP equal to or higher than 10 as polymers since they can be differentiated from the homologues of DP <10 by antibodies, as reported elsewhere, and can be easily prepared by precipitating polylactone from water as shown in this paper.(OSA/PSA) 2The abbreviations used are: OSA, oligosialic acids; PSA, polysialic acids; HPAEC, high performance anion-exchange chromatography; PAD, pulsed amperometric detection/detector; DP, degree of polymerization; Neu5Ac, 5-N-acetylneuraminic acid; Neu5Gc, 5-N-glycolylneuraminic acid; KDN, 2-keto-3-deoxy-d-glycero- d-galactonononic acid; PSGP, polysialoglycoprotein; 5L1, 5L2, 5L3, etc., lactones of pentameric Neu5Ac with one, two, and three lactone rings per molecule, respectively; GD3, II3(NeuAc)2-LacCer; GD1b, II3(NeuAc)2-GgOse4Cer. have revealed remarkable structural (1Sato C. Kitajima K. Tazawa I. Inoue Y. Inoue S. Troy II, F.A. J. Biol. Chem. 1993; 268: 23675-23684Abstract Full Text PDF PubMed Google Scholar, 2Qu B. Ziak M. Zuber C. Roth J. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 8995-8998Crossref PubMed Scopus (24) Google Scholar, 3Kitazume S. Kitajima K. Inoue S. Haslam S.M. Morris H.R. Dell A. Lennarz W.J. Inoue Y. J. Biol. Chem. 1996; 271: 6694-6701Abstract Full Text PDF PubMed Scopus (59) Google Scholar), tissue/cell distributional (4Angata K. Nakayama J. Fredette B. Chong K. Ranscht B. Fukuda M. J. Biol. Chem. 1997; 272: 7182-7190Abstract Full Text Full Text PDF PubMed Scopus (129) Google Scholar), functional (5Rutishauser U. Landmesser L. Trends Neurosci. 1996; 19: 422-427Abstract Full Text Full Text PDF PubMed Scopus (424) Google Scholar, 6Shen H. Watanabe M. Tomasiewicz H. Rutishauser U. Magnuson T. Glass J.D. J. Neurosci. 1997; 17: 5221-5522Crossref PubMed Google Scholar, 7O'Connell A.W. Fox G.B. Barry T. Murphy K.J. Fichera G. Foley A.G. Kelly J. Regan C.M. J. Neurochem. 1997; 68: 2538-2546Crossref PubMed Scopus (75) Google Scholar), and evolutionary diversities ranging from bacteria to man (8Troy II, F.A. Glycobiology. 1992; 2: 5-23Crossref PubMed Scopus (322) Google Scholar, 9Roth J. Kempf A. Reuter G. Schauer R. Gehring W.J. Science. 1992; 256: 673-675Crossref PubMed Scopus (145) Google Scholar). Structural diversities of OSA/PSA are further complicated by their ease of lactonization. The existence of lactones in OSA/PSA was suggested first many years ago (10McGuire E.J. Binkley S.B. Biochemistry. 1964; 3: 247-251Crossref PubMed Scopus (118) Google Scholar). It was suspected that lactonization could significantly alter the physicochemical and biological properties of OSA/PSA such as charge density, conformation, and antigenicity (11Lifely M.R. Gilbert A.S. Moreno C. Carbohydr. Res. 1981; 94: 193-203Crossref PubMed Scopus (69) Google Scholar, 12Lifely M.R. Gilbert A.S. Moreno C. Carbohydr. Res. 1984; 134: 229-243Crossref PubMed Scopus (40) Google Scholar). Both α2,8- and α2,9-linked oligo/poly-Neu5Ac lactonized rapidly under acidic conditions (11Lifely M.R. Gilbert A.S. Moreno C. Carbohydr. Res. 1981; 94: 193-203Crossref PubMed Scopus (69) Google Scholar, 12Lifely M.R. Gilbert A.S. Moreno C. Carbohydr. Res. 1984; 134: 229-243Crossref PubMed Scopus (40) Google Scholar). Similar phenomenon was observed in gangliosides containing α2,8-linked Neu5Ac dimer (13Bassi R. Riboni L. Sonnino S. Tettamanti G. Carbohydr. Res. 1989; 193: 141-146Crossref PubMed Scopus (33) Google Scholar, 14Tsuda M. Terabayashi T. Kawanishi Y. Chem. Phys. Lipids. 1994; 70: 95-99Crossref PubMed Scopus (4) Google Scholar). We have also demonstrated that passing a solution of sodium salt of colominic acid (a mixture of oligo/poly(α2,8-Neu5Ac) homologues) through a Dowex 50 (H+-form) column or simply dialyzing it against water can induce lactonization (15Zhang Y. Lee Y.C. Glycobiology. 1997; 7 (abstr.): 1027Crossref Scopus (45) Google Scholar). Failure to completely de-lactonize OSA/PSA mixtures after partial hydrolysis under acidic conditions severely limited their separation by capillary electrophoresis (16Cheng M.-C. Lin S.-L. Wu S.-H. Inoue S. Inoue Y. Anal. Biochem. 1998; 260: 154-159Crossref PubMed Scopus (36) Google Scholar). Similarly, during the separation of OSA from hydrolysate of colominic acid on an anion-exchange column, we observed that each OSA peak of an expected degree of polymerization (DP) contained a small amount of OSA with a higher than expected DP (e.g. the OSA of DP5 in the expected DP4 peak). Co-existence of OSA of two different DP was also detected in samples obtained elsewhere. Apparently, loss of negative charges upon lactonization of OSA caused them to elute at the positions of lower DP. The natural occurrence of sialic acid lactones in glycolipids also underscores the significance of lactonization in vivo(17Gross S.K. Williams M.A. McCluer R.H. J. Neurochem. 1980; 34: 1351-1361Crossref PubMed Scopus (62) Google Scholar, 18Riboni L. Sonnino S. Acquotti D. Malesci A. Ghidoni R. Egge H. Mingrino S. Tettamanti G. J. Biol. Chem. 1986; 261: 8514-8519Abstract Full Text PDF PubMed Google Scholar, 19Kawashima I. Kotani M. Ozawa H. Suzuki M. Tai T. Int. J. Cancer. 1994; 58: 263-268Crossref PubMed Scopus (24) Google Scholar, 20Kielczynski W. Bartholomeusz R.K. Harrison L.C. Glycobiology. 1994; 4: 791-796Crossref PubMed Scopus (14) Google Scholar). However, it is not clear whether lactonization in vivo occurs spontaneously by merely acid-catalyzed chemical reactions or is actively controlled by enzymatic processes, although lactonization could be indirectly regulated byO-substitutions and de-O-substitutions on sialic acids, which are suggested to be tightly controlled in living systems (21Varki A. Glycobiology. 1992; 2: 25-40Crossref PubMed Scopus (487) Google Scholar, 22Shi W.-X. Chammas R. Varki A. J. Biol. Chem. 1996; 271: 31517-31525Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar, 23Shi Wei X. Chammas R. Varki A. Glycobiology. 1998; 8: 199-205Crossref PubMed Scopus (44) Google Scholar). Therefore, systematic investigation on acid-catalyzed lactonization of OSA/PSA will not only provide useful information for proper sample handling in vitro but also shed light on our understanding of related processes in vivo. It has been suggested by NMR and substitution studies (11Lifely M.R. Gilbert A.S. Moreno C. Carbohydr. Res. 1981; 94: 193-203Crossref PubMed Scopus (69) Google Scholar) that lactonization of oligo/poly(α2,8-Neu5Ac) occurs between two adjacent sialic acid residues, the carboxyl group of one residue esterifying the 9-hydroxyl group of the residue at the reducing side to form a 6-membered ring (Fig. 1). The same lactone ring was also observed between the two α2,8-linked sialic acid residues in GD3 and GD1b ganglioside lactones (24Ando S. Yu R.K. Scarsdale J.N. Kusunoki S. Prestegard J.H. J. Biol. Chem. 1989; 264: 3478-3483Abstract Full Text PDF PubMed Google Scholar, 25Acquotti D. Fronza G. Ragg E. Sonnino S. Chem. Phys. Lipids. 1991; 59: 107-126Crossref PubMed Scopus (51) Google Scholar). If this is the only type of lactone ring, there will be three possible lactone species from a trimer, i.e.(Neu5Acα2,8)3-(1′:9)-lactone (A), (Neu5Acα2,8)3-(1“:9′)-lactone (B), and (Neu5Acα2, 8)3-(1′:9, 1”:9′)-di-lactone (AB) (see Fig. 1). It is obvious that the possible patterns of lactonization will become progressively complicated as DP increases. The lack of detailed understanding of the lactonization most likely is due to unavailability of effective methodology. Spectrometric methods such as IR (11Lifely M.R. Gilbert A.S. Moreno C. Carbohydr. Res. 1981; 94: 193-203Crossref PubMed Scopus (69) Google Scholar, 12Lifely M.R. Gilbert A.S. Moreno C. Carbohydr. Res. 1984; 134: 229-243Crossref PubMed Scopus (40) Google Scholar), NMR (11Lifely M.R. Gilbert A.S. Moreno C. Carbohydr. Res. 1981; 94: 193-203Crossref PubMed Scopus (69) Google Scholar, 12Lifely M.R. Gilbert A.S. Moreno C. Carbohydr. Res. 1984; 134: 229-243Crossref PubMed Scopus (40) Google Scholar), and CD (26Terabayashi T. Ogawa T. Kawanishi Y. Carbohydr. Polym. 1996; 29: 35-39Crossref Scopus (11) Google Scholar) usually reveal only the averaged properties such as the ratio of lactonized and non-lactonized species, whereas more detailed and specific information such as the positions of the lactone rings and the distribution of different lactone species is difficult to obtain. We have developed a highly sensitive and efficient method for analysis of OSA/PSA using high performance anion-exchange chromatography (HPAEC) and pulsed amperometric detection (PAD) by utilizing either neutral or alkaline conditions for separation (27Zhang Y. Inoue Y. Inoue S. Lee Y.C. Anal. Biochem. 1997; 250: 245-251Crossref PubMed Scopus (39) Google Scholar). By using this method, we systematically studied acid-catalyzed lactonization of OSA/PSA as reported in this paper. Quaternary methylamine anion exchanger is from Millipore Waters Chromatography Division (Milford, MA). Sephacryl S-300 and Sephadex G-25 resins are from Sigma. Bio-Gel P-2 resin is from Bio-Rad. Oligo(α2,8-Neu5Ac) of DP2–6, colominic acid (Na+-form) as well as Arthrobacter ureafaciensneuraminidase are gifts from Drs. Y. Tsukada and Y. Ohta, Kyoto Research Laboratories, Marukin Shoyu Co., Uji, Japan. The enzyme activity was defined and determined as reported (28Myers R.W. Lee R.T. Lee Y.C. Thomas G.H. Reynolds L.W. Uchida Y. Anal. Biochem. 1980; 101: 166-174Crossref PubMed Scopus (148) Google Scholar). Oligo(α2,8-Neu5Gc), Oligo(α2,8-KDN), and polysialoglycoprotein (PSGP) from salmon eggs are gifts from Drs. S. and Y. Inoue, Academia Sinica (Taipei, Taiwan). Oligo-Neu5Acs were also prepared as described (29Kitazume S. Kitajima K. Inoue S. Inoue Y. Anal. Biochem. 1992; 202: 25-34Crossref PubMed Scopus (62) Google Scholar) with some modifications. Briefly, a quaternary methylamine anion-exchange column (1 × 29 cm) was used in 10 mmphosphate buffer (pH 7.5) with a linear gradient eluent from 0 to 0.4m NaCl. Samples were adjusted to pH 12 for a few minutes prior to the chromatography (see “Discussion”). To remove the contaminating free oligo/poly-Neu5Gc chains (resulted from auto-hydrolysis) as well as peptides free of oligo/poly-Neu5Gc from PSGP, 10 mg of the PSGP was fractionated on a Sephacryl S-300 column (1 × 118 cm) equilibrated and eluted with 10 mm phosphate buffer (pH 7.6). The effluent was monitored byA 215 nm, and the absorption peaks were checked with HPAEC by injecting small aliquots of the fractions before and after partial hydrolysis with 0.1 m HCl at 80 °C for 15 min. Fractions containing intact PSGP showed the diagnostic peak patterns of oligo/poly-Neu5Gc by HPAEC (27Zhang Y. Inoue Y. Inoue S. Lee Y.C. Anal. Biochem. 1997; 250: 245-251Crossref PubMed Scopus (39) Google Scholar) only after partial hydrolysis. Such fractions were therefore pooled and desalted on a Sephadex G-25 column (0.7 × 19 cm) equilibrated and eluted with water, freeze-dried, and stored at −20 °C. Dionex (Sunnyvale, CA) Bio-LC was used with the CarboPac PA-1 and PA-100 columns in combination with a pulsed amperometric detector (PAD-2) using the SC-PAD-2 detector cell with a gold working electrode and a silver/AgCl reference electrode. The detector sensitivity was set at 1 μA. Potential and time settings of the detector were as follows: E 1 = +0.05 V (t 1 = 0.42 s), E 2 = +0.65 V (t 2 = 0.18 s),E 3 = −0.10 V (t 3 = 0.36 s). These settings are optimal for nitrate eluent and also give excellent results when eluting with NaOH only or NaOH with sodium acetate. A Spectra SYSTEM AS3000 autosampler (Thermo Separation Products, San Jose, CA) was used for sample injection and maintenance of sample temperature at 4 °C prior to injection. Data were collected via Dionex ACI (advanced computer interface) and Dionex AI-450 software. For the separation of the lactones, neutral nitrate gradients were produced by mixing 0.5 m NaNO3 and water. Freshly prepared 0.5 m NaNO3, typically of pH 6.0–6.5, was used directly for HPAEC. However, the pH of the solution stored in the reservoir of the eluent degas module (Dionex) must be tested to confirm that a proper pH is maintained. 3The pH may rise occasionally because of contamination from the NaOH-containing reservoirs. To ensure detector response and to maintain a stable base line, a postcolumn pneumatic controller (Dionex) driven by pressured nitrogen gas was used to add a solution of 0.5 m NaNO3 in 1m NaOH at 0.5 ml/min to the postcolumn eluent before it enters the detector. When it was desirable to have lactone rings opened, alkaline elutions were used that contained a constant flow of 0.1 m NaOH in addition to a gradient generated with 0.5m NaNO3. All eluents were sparged and pressured under helium using the Dionex eluent degas module. Lactonization of α2,8-linked OSA/PSA were induced with such mild acidic solutions as 40 mm sodium acetate (pH 4.8), 10 mm sodium phosphate (pH 3.2), and 20 mm HCl. Lactonization at 4 °C, room temperature (23 °C), 37, 55, and 80 °C were tested. Samples were also lactonized by passing through a Dowex 50 (H+-form) column at room temperature. Lactonization in strong acid was performed by treating the samples with 1 m HCl at ambient temperature for 2 h or at 4 °C overnight. The samples were diluted 100 times with water before injection. When 250 mg of colominic acid (Na+-form) was dissolved in 1 ml of 1 m HCl and kept at ambient temperature for 2 h, a white, milky suspension of extensively lactonized PSA was formed, which will be referred to as “polylactone.” Polylactone was separated by centrifugation and washed with 1 ml of distilled de-ionized water three times by repeated suspension and centrifugation. After freeze-drying, the polylactone samples were stored at −20 °C. Colominic acid (100 mg, Na+-form) was dissolved in 1 ml of water and dialyzed against 1 liter of distilled de-ionized water at 4 °C for 3 days in a dialysis tubing (molecular mass cut-off = 1 kDa) with a daily change of water. After dialysis, the sample was lyophilized and stored at −20 °C. Freeze-drying and subsequent storage of the dried sample at −20 °C did not change the lactonization pattern (data not shown). Dialysis of pentamer and hexamer of oligo-Neu5Ac was performed on a micro scale. 4Details of this procedure will be published elsewhere. Briefly, 50 μl each of the samples (Na+-form) at 1 mg/ml was transferred into a 250-μl polypropylene screw-top microvial (Sun Brokers, Wilmington, NC). A cut piece of dialysis membrane (molecular mass cut-off = 1 kDa) was placed on top of the vial and screwed tight with an open-top cap. The vial was immersed upside-down in 500 ml of distilled de-ionized water and dialyzed with stirring at 4 °C for 1 day. After dialysis, the samples were diluted with water before analysis by HPAEC. OSA of DP2–6 were purified on a quaternary methylamine column as described earlier and desalted by passing through a Bio-Gel P-2 column equilibrated and eluted with 10 mm phosphate buffer (pH 7.5). The sample solutions were frozen and stored at −20 °C for 6 months to 1 year and analyzed by HPAEC. Neuraminidase digestion was performed at 4 °C to minimize possible opening or rearrangement of the lactone rings as well as auto-hydrolysis of glycosidic linkages. Dialyzed OSA and colominic acid were diluted with 80 mm sodium acetate buffer (pH 4.8) to a final concentration of 40 mm sodium acetate and digested with a neuraminidase (20–80 milliunits of enzyme per μg of substrate) at 4 °C for different periods and directly injected to the CarboPac PA-1 column. Following the analysis, the column was cleaned by eluting with 0.2 m NaOH at 0.2 ml/min overnight to remove any neuraminidase that might have accumulated under neutral eluting condition. The stability of colominic acid polylactones in acid was tested by hydrolyzing lactonized (experimental) and freshly NaOH-treated (control) samples. For experimental samples, to a mixture of 57 μl of 1 mHCl and 10 μl of 1 m NaOH in a 0.5-ml screw-capped polypropylene microcentrifuge tube (USA/Scientific Plastics, Ocala, FL) was added 400 μl of 3 mg/ml polylactone. For control samples, 10 μl of 1 m NaOH and 400 μl of 3 mg/ml polylactone were mixed and kept at ambient temperature for 5 min to open all lactone rings. To the mixture was then added 57 μl of 1 m HCl. The experimental and control, both in capped tubes with a final concentration of 0.1 m HCl, were heated at 80 °C in a glycerol bath in a heating block. After different times, the tubes were cooled on ice, and a calculated amount of NaOH was added to neutralize the solutions. Aliquots of the neutralized samples were analyzed with HPAEC using an alkaline sodium nitrate as eluent. Acid hydrolysis of purified PSGP was carried out in a similar manner. For the experimental, 20 μl of 10 mg/ml PSGP and 10 μl of 0.1m NaOH were mixed and kept at ambient temperature for 10 min to remove any possible O-acetylation that may prevent lactonization. To the mixture was added 10 μl of 4 m HCl and left at ambient temperature for 2 h (to induce lactonization) followed by dilution with water to a final concentration of 0.1m HCl. For the control, equivalent amounts of solutions were used except that the NaOH and HCl were pre-mixed and diluted first before adding to the PSGP. The samples were then heated at 80 °C for 15 min before they were neutralized and analyzed by HPAEC. We have shown that colominic acid can be separated into a series of peaks by HPAEC using a neutral nitrate eluent (27Zhang Y. Inoue Y. Inoue S. Lee Y.C. Anal. Biochem. 1997; 250: 245-251Crossref PubMed Scopus (39) Google Scholar). Under neutral elution conditions, retention of OSA/PSA on the CarboPac PA-1 column is mainly determined by the number of ionized carboxyl groups. Lactonization reduces the number of negative charges and should cause the elution time to decrease. Indeed, we observed that the peak distribution pattern of OSA/PSA changed upon lactonization. However, fragmentation of OSA/PSA due to glycosidic bond cleavage can also change the peak distribution pattern. To distinguish between the two possible causes, we treated the sample with NaOH (final concentration of 20 mm over and above neutralization of acids) immediately before injection or carried out the elution using an alkaline nitrate eluent. Brief exposure to alkali such as these opened lactone rings but retained glycosidic linkages. Representative chromatograms from a series of periodic injections of a colominic acid sample kept at 4 °C in 20 mm HCl solution are shown in Fig. 2. The original peak distribution pattern of the sample (Fig. 2 A) changed within 40 min in 20 mm HCl (Fig. 2 B). The peak height of high DP homologues decreased dramatically, and new peaks (which are unlabeled) appeared between the two adjacent low DP homologue peaks (which are labeled with DP). Continued incubation resulted in further decrease of the high DP peaks, and the pattern of new peaks between the two adjacent low DP homologues became more complex (Fig.2 C). However, in the subsequent analysis (Fig.2 D) in which the same sample after 10 h incubation was treated with NaOH prior to injection, all the original high DP peaks were restored concomitant with disappearance of the new peaks found between the two adjacent low DP homologues. The peak distribution patterns of Fig. 2, A and D, are nearly the same except that in Fig. 2 D monomer and dimer peaks increased somewhat, apparently due to limited glycosidic cleavage. These results suggest that there was no significant degradation of the high DP homologues under the acidic condition used here. Therefore, the apparent diminishing of high DP homologues in Fig. 2, B andC, was mostly caused by lactonization and not by hydrolysis. The lactonized species not only appeared as the alkali-sensitive new peaks between two adjacent low-DP homologues (Fig. 2, B andC) but also fused together, forming a broad raised base line (Fig. 2 C) similar to the pattern shown by capillary electrophoresis (16Cheng M.-C. Lin S.-L. Wu S.-H. Inoue S. Inoue Y. Anal. Biochem. 1998; 260: 154-159Crossref PubMed Scopus (36) Google Scholar). By comparing samples with and without NaOH treatment, significant lactonization of OSA/PSA was also observed under conditions that were frequently used for acid hydrolysis. For example, at pH 4.8, a lactonization pattern similar to that of Fig. 2 B was observed at both 37 and 55 °C (not shown). At 80 °C in 0.1m HCl, which is a typical hydrolytic condition for de-sialylation of glycoconjugates, instead of fragmentation of OSA/PSA chains revealed by the NaOH-treated hydrolysate, extensive lactonization occurred as indicated by broad peaks around monomer, dimer and trimer, respectively (not shown). Since lactonization requires protonation of the carboxyl group, we tested whether colominic acid in free acid form can spontaneously form lactones. A colominic acid (Na+-form) solution was passed through a Dowex 50 (H+-form) column and examined by HPAEC. A pattern similar to that of Fig. 2 C was observed (not shown), indicating that lactonization proceeded quickly and extensively. This result was consistent with that obtained by IR spectroscopy (11Lifely M.R. Gilbert A.S. Moreno C. Carbohydr. Res. 1981; 94: 193-203Crossref PubMed Scopus (69) Google Scholar, 12Lifely M.R. Gilbert A.S. Moreno C. Carbohydr. Res. 1984; 134: 229-243Crossref PubMed Scopus (40) Google Scholar). Interestingly, mere dialysis of the colominic acid sample (Na+-form) against water also induced limited and selective lactonization, which gave a pattern (not shown) similar to that of Fig. 2 B. To ascertain if the shifting of individual peaks is due to lactonization and also to assign proper DP values to these peaks, we studied lactonization of purified individual oligo(α2,8-Neu5Ac) of DP2–6. All incubations were at 4 °C to minimize glycosidic cleavage. A set of chromatograms after 10 h incubation at pH 3.2 is shown in Fig.3. Each chromatogram (Fig. 3,B–E) except for that of dimer (Fig. 3 A) contains three kinds of peaks as follows: 1) the latest eluting original OSA peak; 2) the small peaks of monomer, dimer, and trimer, etc. (resulted from hydrolysis); and 3) the broad and skewed peaks labeled withL. When these samples were treated with NaOH, all the L peaks disappeared, whereas the original OSA peaks were restored nearly to their original height (not shown). Therefore, the L peaks are lactonized species. As a specific example, Neu5Ac pentamer (Fig. 3 D) formed three groups of lactonized species, designated as 5L3, 5L2, and 5L1 (see Footnote 2). We interpret that 5L3, 5L2, and 5L1 are pentamers with different numbers of lactone rings since they eluted in close proximity to non-lactonized dimer, trimer, and tetramer (indicating that they have 2, 3, and 4 residual carboxyl groups, see “Discussion”), respectively. Within each L peak there are lactone ring position isomers, which is reflected by the broad and skewed nature of the L peaks. Such assignments were supported by data obtained under higher chromatographic resolution, which was accomplished by using a more moderate gradient or isocratic elution. For example, the two lactone ring position isomers (see Fig. 1; (Neu5Acα2, 8)3-(1′:9)-lactone (A) and (Neu5Acα2, 8)3-(1“:9′)-lactone (B)) in the peak 3L1 (Fig.3 B) were isolated using isocratic elution. They both reverted to the original trimer upon the NaOH treatment (data not shown). Interestingly, Neu5Ac dimer did not form any lactone even when kept at 4 °C for 3 days (data not shown). We also quantified each species at timed intervals, on the basis that the molar response factors of lactonized species are equal to or close to those of their parent non-lactonized species (see “Discussion”). Quantitative analysis of the average lactonization rate at potential lactonizable sites on each OSA revealed that OSA with higher DP lactonized more rapidly and had higher percentage of potential sites being lactonized (not shown). Fig.4 shows the lactonization patterns of oligo-Neu5Acs of DP2–6 after incubation in 1 m HCl at 4 °C overnight. Treating these samples with NaOH reversed all the lactone (L-labeled) peaks to their non-lactonized precursors (see dash-lined chromatograms in Fig. 4). Compared with the patterns shown in Fig. 3, lactonization was more extensive, and more than half of dimers eluted as an alkali-sensitive peak that had the same elution time as monomer, which indicated that the lactonized dimers contained one lactone ring per molecule (labeled as 2L1). Lactone peaks (i.e. 3L2, 4L3, 5L4, and 6L5) having approximately the same elution time with monomer also appeared as the major species in each of other OSA, suggesting that a significant fraction of these molecules underwent complete lactonization, which involved all carboxyl groups except that on the reducing-terminal residue (see the (Neu5Acα2,8)3-(1′:9, 1“:9′)-di-lactone (AB) in Fig. 1). Similar to the results observed at pH 3.2, OSA with higher DP lactonized more rapidly and extensively. For instance, whereas all the pentamer and hexamer molecules lactonized, the original non-lactonized molecules still existed in OSA with DP <5. Surprisingly, degradation by acid hydrolysis was hardly detected in oligo-Neu5Acs kept at 4 °C in 1 m HCl, suggesting that formation of lactone rings protects against glycosidic cleavage (also see other results shown later). The above results show that lactonization of OSA/PSA can occur in discrete stages (see “Discussion”). The most interesting stage is the “initial stage,” which is manifested by a limited and selective lactonization pattern (Fig. 2 B). This stage can be reached also by simply dialyzing the sample against water or by prolonged storage in 10 mm phosphate buffer (pH 7.5) at −20 °C. The lactonization patterns in Fig. 5,A1 and B1, show that dialyzed pentamer and hexamer each formed one major lactone peak that eluted near the non-lactonized tetramer and pentamer peaks, respectively, indicating only mono-lactone was formed from each oligomer. Similar patterns were observed in oligo-Neu5Acs of DP4–6 after long term storage at −20 °C but that dimer and trimer did not form any lactone. NaBH4-reduced colominic acid can still be lactonized in a similar fashion as non-reduced colominic acid (not shown), supporting th" @default.
• W2163727669 created "2016-06-24" @default.
• W2163727669 creator A5037398755 @default.
• W2163727669 creator A5082117229 @default.
• W2163727669 date "1999-03-01" @default.
• W2163727669 modified "2023-10-18" @default.
• W2163727669 title "Acid-catalyzed Lactonization of α2,8-Linked Oligo/Polysialic Acids Studied by High Performance Anion-exchange Chromatography" @default.
• W2163727669 cites W1500103283 @default.
• W2163727669 cites W1506739991 @default.
• W2163727669 cites W1539757478 @default.
• W2163727669 cites W1569183316 @default.
• W2163727669 cites W1592948505 @default.
• W2163727669 cites W1672932894 @default.
• W2163727669 cites W1971643460 @default.
• W2163727669 cites W1972670966 @default.
• W2163727669 cites W1979969301 @default.
• W2163727669 cites W2001637853 @default.
• W2163727669 cites W2005121006 @default.
• W2163727669 cites W2005503279 @default.
• W2163727669 cites W2012930902 @default.
• W2163727669 cites W2016596101 @default.
• W2163727669 cites W2022501524 @default.
• W2163727669 cites W2024187848 @default.
• W2163727669 cites W2029587634 @default.
• W2163727669 cites W2030966251 @default.
• W2163727669 cites W2032007065 @default.
• W2163727669 cites W2032402267 @default.
• W2163727669 cites W2034664516 @default.
• W2163727669 cites W2046821291 @default.
• W2163727669 cites W2052222940 @default.
• W2163727669 cites W2055708695 @default.
• W2163727669 cites W2057241909 @default.
• W2163727669 cites W2069767991 @default.
• W2163727669 cites W2072733441 @default.
• W2163727669 cites W2079531828 @default.
• W2163727669 cites W2080658483 @default.
• W2163727669 cites W2083743613 @default.
• W2163727669 cites W2091483994 @default.
• W2163727669 cites W2093463554 @default.
• W2163727669 cites W2098120364 @default.
• W2163727669 cites W2120012498 @default.
• W2163727669 cites W2120591238 @default.
• W2163727669 cites W2120899418 @default.
• W2163727669 cites W2146813978 @default.
• W2163727669 cites W2150380845 @default.
• W2163727669 cites W2157265748 @default.
• W2163727669 cites W2172243326 @default.
• W2163727669 cites W4252141089 @default.
• W2163727669 doi "https://doi.org/10.1074/jbc.274.10.6183" @default.
• W2163727669 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/10037703" @default.
• W2163727669 hasPublicationYear "1999" @default.
• W2163727669 type Work @default.
• W2163727669 sameAs 2163727669 @default.
• W2163727669 citedByCount "44" @default.
• W2163727669 countsByYear W21637276692012 @default.
• W2163727669 countsByYear W21637276692013 @default.
• W2163727669 countsByYear W21637276692014 @default.
• W2163727669 countsByYear W21637276692016 @default.
• W2163727669 countsByYear W21637276692017 @default.
• W2163727669 countsByYear W21637276692018 @default.
• W2163727669 countsByYear W21637276692019 @default.
• W2163727669 countsByYear W21637276692020 @default.
• W2163727669 countsByYear W21637276692021 @default.
• W2163727669 crossrefType "journal-article" @default.
• W2163727669 hasAuthorship W2163727669A5037398755 @default.
• W2163727669 hasAuthorship W2163727669A5082117229 @default.
• W2163727669 hasBestOaLocation W21637276691 @default.
• W2163727669 hasConcept C117760992 @default.
• W2163727669 hasConcept C122871604 @default.
• W2163727669 hasConcept C145148216 @default.
• W2163727669 hasConcept C1491633281 @default.
• W2163727669 hasConcept C161790260 @default.
• W2163727669 hasConcept C178790620 @default.
• W2163727669 hasConcept C185592680 @default.
• W2163727669 hasConcept C206967267 @default.
• W2163727669 hasConcept C21951064 @default.
• W2163727669 hasConcept C2778993516 @default.
• W2163727669 hasConcept C43617362 @default.
• W2163727669 hasConcept C55493867 @default.
• W2163727669 hasConcept C85789140 @default.
• W2163727669 hasConceptScore W2163727669C117760992 @default.
• W2163727669 hasConceptScore W2163727669C122871604 @default.
• W2163727669 hasConceptScore W2163727669C145148216 @default.
• W2163727669 hasConceptScore W2163727669C1491633281 @default.
• W2163727669 hasConceptScore W2163727669C161790260 @default.
• W2163727669 hasConceptScore W2163727669C178790620 @default.
• W2163727669 hasConceptScore W2163727669C185592680 @default.
• W2163727669 hasConceptScore W2163727669C206967267 @default.
• W2163727669 hasConceptScore W2163727669C21951064 @default.
• W2163727669 hasConceptScore W2163727669C2778993516 @default.
• W2163727669 hasConceptScore W2163727669C43617362 @default.
• W2163727669 hasConceptScore W2163727669C55493867 @default.
• W2163727669 hasConceptScore W2163727669C85789140 @default.
• W2163727669 hasIssue "10" @default.
• W2163727669 hasLocation W21637276691 @default.
• W2163727669 hasOpenAccess W2163727669 @default.
• W2163727669 hasPrimaryLocation W21637276691 @default.
• W2163727669 hasRelatedWork W1544804321 @default.

• next