Isolation and Characterization of Gram-Positive Cyanophycin

Nov 12, 2003 - This study is the first report on the extracellular degradation of cyanophycin (CGP) by Gram-positive bacteria. Three different Gram-po...
0 downloads 0 Views 165KB Size
Biomacromolecules 2004, 5, 153-161

153

Isolation and Characterization of Gram-Positive Cyanophycin-Degrading BacteriasKinetic Studies on Cyanophycin Depolymerase Activity in Aerobic Bacteria Martin Obst,‡,# Ahmed Sallam,‡,# Heinrich Luftmann,§ and Alexander Steinbu¨chel*,‡ Institut fu¨r Molekulare Mikrobiologie und Biotechnologie, Westfa¨lische Wilhelms-Universita¨t Mu¨nster, Corrensstrasse 3, 48149 Mu¨nster, Germany, and Institut fu¨r Organische Chemie, Westfa¨lische Wilhelms-Universita¨t Mu¨nster, Corrensstrasse 40, 48149 Mu¨nster, Germany Received August 5, 2003; Revised Manuscript Received October 9, 2003

This study is the first report on the extracellular degradation of cyanophycin (CGP) by Gram-positive bacteria. Three different Gram-positive bacteria were isolated from forest soil that were able to utilize CGP as the sole carbon source for growth. The isolates were assigned to species of the genera Bacillus and Micromonospora. From one of the isolates, which was taxonomically affiliated as Bacillus megaterium strain BAC19, the extracellular CGP depolymerase (extracellular CGPase; CphEBm) was purified to electrophoretic homogeneity by fast protein liquid chromatography and affinity binding to an arginineagarose column. The purified enzyme was specific for hydrolytic cleavage of CGP, and inhibitor studies indicated that CphEBm is a serine-type peptidase. As CGP degradation products, (β-Asp-Arg)2 tetrapeptides in addition to β-Asp-Arg dipeptides occurred, which were identified by electrospray ionization mass spectrometry analysis. Furthermore, a novel quantitative enzyme assay was developed for kinetic studies on CGP depolymerases. For CphEBm, as well as for the extracellular CGPase of Pseudomonas anguilliseptica strain BI (CphEPa), KM values of 2.2 and 1.0 µM, respectively, for CGP were determined. Introduction The poly(amino acid) cyanophycin (cyanophycin granule polypeptide, CGP) is a wide spread biopolymer that is synthesized by most cyanobacteria1,2 and also by several nonphotosynthetic bacteria.3,4 In these bacteria, CGP serves as a nitrogen, carbon, and energy storage compound.1-3 CGP is synthesized and accumulated in the cytoplasm as insoluble membraneless inclusions5 when growth is limited, and it is mobilized and degraded in the cells when growth is resumed. The CGP molecule is a comblike polymer with R-aminoR-carboxy-linked L-aspartic acid residues representing the poly(aspartic acid) backbone and L-arginine residues bound to the β-carboxylic groups of aspartic acids.6,7 Most previous studies on CGP metabolism focused on the biosynthesis of this biopolymer with CGP synthetases (CphA) as the enzymes analyzed in most detail (reviewed in ref 8). In contrast, only a few reports are available on the intracellular CGP degradation catalyzed by CGPases of cyanobacteria, which are referred to as CphB.7,9 Dipeptides consisting of arginine and aspartic acid and free arginine were identified as products of CGP degradation by CphB of Synechocystis sp. PCC6803 in addition to small amounts of aspartic acid.9 The first and so far only study on the extracellular degradation of CGP employing Pseudomonas * To whom correspondence should be addressed. Tel: +49 (251) 8339821. Fax: +49 (251) 8338388. E-mail: [email protected]. ‡ Institut fu ¨ r Molekulare Mikrobiologie und Biotechnologie Mu¨nster. § Institut fu ¨ r Organische Chemie Mu¨nster. # These authors have equally contributed to the work.

anguilliseptica strain BI and several other Gram-negative bacteria was recently published.10 This study also described the occurrence of β-Asp-Arg dipeptides as final products of CGP degradation mediated by the extracellular CGPase (CphE) of P. anguilliseptica strain BI. Until this study, nothing was known about the extracellular degradation of CGP by bacteria or other microorganisms. Because CGP occurs in many more bacteria than previously known, it is potentially an abundant compound in several natural habitats. The capability for extracellular degradation of this biopolymer by different microorganisms should be studied in more detail because of the potential occurrence of industrially relevant degradation products such as poly(aspartic acid) that may result from cleavage of CGP by a β-amide bond hydrolyzing CGPase, which, unlike all other known CGPases, releases free arginine from the poly(aspartic acid) backbone. In this study, we demonstrate for the first time the isolation of Gram-positive aerobic bacteria that are able to utilize CGP as the sole carbon source for growth. We report on the taxonomic determination of a Bacillus megaterium strain and on the substrate utilization capabilities of this bacterium. Furthermore, we describe purification and biochemical characterization of the extracellular CGPase CphEBm of this strain and the CGP cleavage products formed by this enzyme. The development of a new quantitative enzyme assay allowed analysis of CGP turnover, and the KM values of the extracellular CGPases of B. megaterium and P. anguilliseptica for CGP were determined.

10.1021/bm034281p CCC: $27.50 © 2004 American Chemical Society Published on Web 11/12/2003

154

Biomacromolecules, Vol. 5, No. 1, 2004

Obst et al.

Table 1. Strains and Plasmids Used in This Study strains and plasmids

Escherichia coli TOP10

Escherichia coli DH1 Synechocystis sp. strain PCC6308a SK19 Bacillus megaterium strain BAC19 Bacillus sp. strain BAC16 Streptomyces sp. strain PASI Micromonospora sp. strain SK22

relevant characteristics Bacterial Strains F- araD139 ∆(ara leu)7697 ∆lacX74 galU galk rpsL deoR Φ80dlacZ∆M15 endA1 nupG recA1 mcrA ∆(mrr hsdRMS mcrBC) F-, supE44, hsdR17, recA1, gyrA96, relA1, endA1, thi-1, λ CGP overproducing mutant of strain PCC6308 CGP degrading strain CGP degrading strain CGP degrading strain CGP degrading strain Plasmids pMa/c5-914 carrying a 2.6-kb PCR product from Synechocystis sp. strain PCC6803 genomic DNA harboring cphA Apr, lacPOZ′, transcription initiation site of T7 and SP6

pMa/c5-914::cphA

pGEM-T Easy

reference or source Invitrogen (San Diego, CA) Hanahan 198314 personal gift from T. Haib this study this study Obst et al. 200210 this study Frey et al. 200213

Promega (Mannheim, Germany)

a Pasteur Culture Collection of Cyanobacteria at the Institut Pasteur, Paris, France. b Institut fu ¨ r Molekulare Mikrobiologie und Biotechnologie, Mu¨nster, Germany.

Experimental Section Isolation of CGP Degrading Gram-Positive Bacteria, Bacterial Strains, Plasmids, and Culture Condition. CGPdegrading Gram-positive bacteria were isolated from soil samples suspended in 50 mM sodium phosphate buffer (pH 8.3) after either pasteurization (90 °C, 30 min) or treatment with 6% (w/v) KOH. Samples were collected from soil of a forest at Mu¨nster-Nienberge (Germany). After pH adjustment, samples were spread on solid basic inorganic medium B11 supplemented with trace element solution SL 712 and overlaid with 0.5% (w/v) agar containing 0.2% (w/v) of CGP, which was prepared as described by Obst et al.10 Bacteria isolated in this study are listed in Table 1. CGP-degrading strains were grown either on standard I complex medium (MERCK, Darmstadt, Germany) or on basic inorganic medium B (see above) for CGP degradation and substrate utilization experiments. Concentrations of CGP and other carbon sources applied in this study are indicated in the text. For all experiments, CGP produced either by Synechocystis sp. strain PCC6308 or by Escherichia coli DH1 harboring plasmid pMa/c5-914::cphA (cphA from Synechocystis sp. strain 6803) was used. The apparent molecular weight range of CGP produced by E. coli DH1 was between 25 and 35 kDa as determined by SDS-PAGE; the molecular weight of CGP from Synechocystis sp. strain PCC6308 ranged from 43 to 100 kDa. Determination of the Gram behavior of the isolated bacteria (Gram staining, LAAP test employing test strips from Merck, Darmstadt, Germany, and KOH test) was performed according to standard protocols. All isolates were grown at 30 °C. Plasmid pGEM-T Easy (Promega, Mannheim, Germany) was used for cloning of PCR-amplified 16S rRNA genes. Competent cells of E. coli TOP10, prepared according to the method of Hanahan,14 were used as recipients of plasmid DNA. E. coli was grown at 37 °C in Luria-Bertani (LB) medium containing 75 µg/mL ampicillin,15 and 0.004% (w/v) X-Gal was added to the medium for identification of

E. coli clones harboring DNA fragment containing plasmids. Culture supernatants and cell pellets were obtained by centrifugation (15 min, 2800g, 4 °C). Isolation, Manipulation, Transfer and Analysis of DNA. Plasmid DNA was isolated from E. coli cells by the alkaline lysis method.16 Total genomic DNA of new Gram-positive isolates was obtained employing the method of Rao et al.17 The 16S rRNA genes of new isolates were amplified from total DNA (see above) using oligonucleotide primers as described before.28 After purification of PCR products, employing a NucleoTrapCR kit (Macherey-Nagel, Du¨ren, Germany), 5′-IRD 800-labeled synthetic oligonucleotide primers (MWG-Biotech, Ebersberg, Germany) and a Sequi Therm EXCEL TM II long-read cycle sequencing kit (Epicenter Technologies, WI) were used for DNA-sequence analysis according to the “primer-hopping strategy”.29 Analysis was done in 6% (w/v) acrylamide gels using Sequagel XR (acrylamide/urea) solution, Complete (buffer reagent) solution (National Diagnostics, Sommerville, NJ) and buffer containing 89 mM TRIS, 89 mM boric acid, and 2 mM EDTA in a LI-COR 4000L automatic sequencing apparatus (MWG-Biotech, Ebersberg, Germany). Nucleic acid sequence data were analyzed with the sequence analysis software CAP (Contig Assembly Program, www.infobiogen.fr/ services/analyseq/cgi-bin/cap_in.pl) and ClustalX 1.8.20 Phylogenetic trees were constructed using the program TREE 1.6.5. The 16S rDNA sequences were aligned with published sequences from representative bacteria from the National Centre for Biotechnology Information (NCBI) database. The 16S rRNA gene sequence data of Bacillus megaterium BAC19 were deposited in the NCBI database under accession number AY180964. Analytical Methods. Reversed-phase HPLC was used to determine the products of enzymatic CGP degradation after derivatization of amino group containing products with OPA (o-phthaldialdehyde) reagent as described before.21 Electrospray ionization mass spectrometry (ESI/MS/MS) was applied for the identification of degradation products of CGP by mass determination and structural analysis.22 All

Gram-Positive Cyanophycin-Degrading Bacteria

measurements were performed employing a Quattro LCZ system (Micromass, Manchester, U.K.) with a nanospray inlet. Quantitative CGP determination was performed employing the method of Bradford23 originally developed for the determination of protein concentrations. Samples of degradation experiments, which contained a fine suspension of CGP particles, were applied to 0.1 N HCl to inactivate CGPases and to dissolve CGP. The employed CGP suspension was prepared from CGP that was dissolved in 0.1 N HCl and subsequently precipitated under vigorous stirring by the addition of 0.1 N NaOH in a small cuvette using a miniature magnetic stirrer for homogenization. A calibration curve covering a range of CGP concentrations from 0 to 64 µg/ mL was used to calculate concentrations of remaining CGP during degradation experiments. Two microliters of stock solutions of purified CphE from P. anguilliseptica strain BI10 (0.8 µg protein/µl) or CphEBm (estimated concentration from silver-stained gels of 0.05-0.08 µg protein/µl) were added in each degradation experiment. Purification of the Extracellular CGPase from Bacillus megaterium Strain BAC19, Electrophoresis. One liter of cell-free supernatant from a Bacillus megaterium BAC19 culture grown for 12 h on CGP was obtained by sedimentation and subsequent filtration of the supernatant through a 0.2 µm nitrocellulose membrane. Enzyme purification steps were carried out at 4 °C in the presence of 50 mM sodium phosphate buffer (pH 8.3). After concentration of the buffered enzyme solution in an ultrafiltration chamber (Amicon, Beverly, MA) employing a YM10 membrane, the enzyme was applied onto a MonoQ HR5/5 anion-exchange column (Amersham Pharmacia LKB Biotechnology, Uppsala, Sweden). Subsequently, the column was washed with two bed volumes [BV] of buffer. A linear NaCl gradient (0-1 M) with an increase of NaCl concentration of 17 mM/ml was employed for elution of CGPase at a total flow rate of 1 mL/min. CphEBm was eluted at NaCl concentrations of 100400 mM. Active fractions of 1 mL were detected by the CGP-overlay agar plate assay, that is, the occurrence of halos on CGP-overlay agar plates within 10-60 min after application of 10 µL aliquots of the respective samples and incubation at 30 °C was monitored. Fractions exhibiting high enzyme activity were combined and desalted by ultrafiltration (see above). Combined fractions were applied onto an arginine-agarose column (5 mL BV; Sigma Chemical, St. Louis, MO). For selective elution of the enzyme, an arginine gradient (0-1 M) was applied. To avoid nonspecific protein binding, the buffer contained in addition 100 mM NaCl. Enzyme activity occurred in fractions containing 80-130 mM arginine. The resulting eluate was concentrated by ultrafiltration (see above) and used as enzyme stock solution in all experiments conducted to characterize CphEBm biochemically (estimated concentration from silver-stained gels of 0.05-0.08 µg protein/µl). SDS-PAGE of active enzyme fractions or CGP samples was performed in 11.5% (w/v) polyacrylamide gels according to standard protocols.23 CGP was visualized by the Coomassie-staining method.25 For visualization of proteins, the silver-staining method was employed.26 To determine the

Biomacromolecules, Vol. 5, No. 1, 2004 155

N-terminal amino acid sequence, CphEBm was blotted from an SDS-polyacrylamide gel onto a poly(vinylidene difluoride) (PVDF) membrane (Millipore, Bedford, MA) according to the method of Towbin et al.27 employing a semidry-fastblot 32/33 apparatus (Biometra, Go¨ttingen, Germany) and analyzed by automated Edman degradation. Characterization of the Purified CGPase. To reveal the substrate specificity of purified CphEBm, the enzyme was incubated at 30 °C in 1 mL of 50 mM sodium phosphate buffer (pH 8.3) with various polypeptide substrates. Each reaction contained 1 mg of the respective substrate and 10 µl of enzyme stock solution. The reaction was stopped after 2 h by inactivation of CphEBm at 70 °C for 10 min. After centrifugation, 100 µl aliquots of supernatant were incubated at 95 °C for 5 min in the presence of 1.25% (w/v) ninhydrin (Merck, Darmstadt, Germany) in 1 mL of total reaction volume. Subsequently, the samples were assayed photometrically at 570 nm for the presence of released hydrolysis products. Bovine casein (Hammersten-grade) was from Merck (Darmstadt, Germany), bovine serum albumin (BSA) from Roth (Karlsruhe, Germany), and poly(R,β-D/Laspartic acid) (Mr ) 11 000) was obtained from Bayer (Leverkusen, Germany). Results and Discussion Isolation and Taxonomic Classification of GramPositive CGP-Degrading Bacteria. Several Gram-positive CGP-degrading bacteria, utilizing CGP as the sole carbon source for growth, were isolated from forest soil samples in this study after pasteurization for enrichment of endosporeforming bacteria or by treatment with 6% (w/v) KOH, a newly developed method for selection of bacteria possessing a multilayered and therefore KOH-resistant murein sacculus. Three isolates (BAC16, BAC19, and SK22), which were identified as CGP-degrading organisms because of their ability to form degradation halos on CGP-overlay agar plates, were analyzed in more detail. Whereas BAC16 grew neither on arginine nor on aspartic acid, BAC19 and SK22 exhibited moderate growth on arginine and SK22 also weak growth on aspartic acid. Because of the inability of isolate BAC19 to grow on aspartic acid but ability to grow on arginine, this strain was a potential producer of a β-amide bond hydrolyzing enzyme, which does not hydrolyze the poly(aspartic acid) backbone of CGP. Therefore, this isolate was chosen for detailed biochemical characterization. For the same reason, another Gram-positive bacterium from the culture collection of our institute, Streptomyces sp. PASI,10 was included in this study. Colony growth and light micrographic investigations indicated that BAC16 and BAC19 belong to the group of Bacillaceae. Cells of both isolates were straight and rodshaped and formed heat-resistant endospores, whereas isolates SK22 and Streptomyces sp. PASI showed mycelial growth. These observations were in good accordance with data obtained by 16S rRNA gene sequence determination, which identified BAC16 and BAC19 as species of the genus Bacillus and SK22 as a species of the genus Micromonospora, and also the assignment of strain PASI to the genus

156

Biomacromolecules, Vol. 5, No. 1, 2004

Figure 1. Taxonomic classification of the isolated Bacillus megaterium strain BAC19 and its affiliation with other Bacillus strains based on 16S rRNA gene sequence homologies to closely related Bacillus strains and other bacteria. The bar represents the ratio of nucleotide substitutions for the given branches.

Streptomyces by Obst et al.10 was confirmed. Isolate BAC19 was classified on the species level as a strain of Bacillus megaterium as revealed by the phylogenetic tree shown in Figure 1. An NCBI database search revealed highest sequence homology of strain BAC19 to Bacillus megaterium type strain DSM 32T (98% identical nucleotides). Purification of the Extracellular CGPase (CphEBm) from B. megaterium Strain BAC19 and N-Terminal Sequence Determination. The extracellular CGPase of B. megaterium strain BAC19, which was referred to as CphEBm, was purified to electrophoretic homogeneity from CGPgrown cultures by the application of only two steps, that is, anion-exchange chromatography on MonoQ HR5/5 and affinity chromatography on an L-arginine-agarose matrix (Figure 2). The low protein content of the culture supernatant in combination with an unexpectedly high purity of CphEBm in CGP mineral medium cultures prevented the detection of extracellular proteins in SDS-polyacrylamide gels stained with the Coomassie dye and also prevented determination of protein concentrations by the application of Bradford reagent during enzyme purification. Therefore, the silverstaining method with about 10-fold higher sensitivity was used for determination of the apparent molecular mass of CphEBm subunits (Figure 2) and for the estimation of the protein content of the CphEBm stock solution. This staining method revealed that in culture supernatant and after the first purification step protein content was very low and CphEBm was hardly detected even after a 100-fold concentration of the applied samples. Purified CphEBm exhibited an apparent

Obst et al.

Figure 2. Analysis of purification of extracellular CGPase from B. megaterium strain BAC19 (CphEBm) by SDS-PAGE: (lane 1) molecular mass standard proteins; (lane 2) culture supernatant after sterile filtration; (lane 3) combined active enzyme fractions after anionexchange chromatography on MonoQ; (lane 4) combined active fractions after affinity chromatography on an arginine-agarose matrix. The silver-staining method was used for visualization of proteins.

molecular weight of 37 kDa (Figure 2), which is 6 kDa less than the molecular weight of 43 kDa determined for CphE from P. anguilliseptica strain BI10 and therefore closer to the size of 27 kDa of cyanobacterial intracellular CGPases.9 In contrast, the gene product of the putative intracellular CGPase of Acinetobacter sp. (CphI) exhibits a calculated molecular mass of 80 kDa.3 Automated Edman-degradation of purified CphEBm identified the amino acid sequence at the N-terminus of the enzyme as ASAPYTLYQVGSTAD. The comparison of this amino acid sequence with sequences of the NCBI database gave no significant homologies to any known protein. The extracellular location of CphEBm, as well as the absence of a methionine residue at the N-terminus of the mature form of the enzyme, suggests the presence of a leader peptide in the intracellular precursor molecule of CphEBm, which is cleaved off during export of the protein, comparable to the leader peptide described for CphE of P. anguilliseptica strain BI.10 CphEBm revealed a very high affinity to immobilized arginine residues of the arginine-agarose matrix, which required the addition of 100 mM NaCl during protein elution. The affinity was comparable with the affinity of the corresponding extracellular enzyme of P. anguilliseptica strain BI.10 Furthermore, during cultivation of B. megaterium strain BAC19 in CGP liquid medium, strong binding of CphEBm to the substrate CGP occurred, which led to complete or partial absence of detectable enzyme activity in particlefree supernatant. Only after all CGP particles in the medium had visibly disappeared, the enzyme became soluble in the culture supernatant, and activity of CphEBm could be detected. No CGPase activity was detected if Bacillus megaterium strain BAC19 was cultivated on other carbon sources than CGP such as glucose or standard 1 complex medium.

Gram-Positive Cyanophycin-Degrading Bacteria

Thermal Stability and Substrate Specificity of CphEBm. For determination of the heat stability of CphEBm, a wide range of temperatures (30-75 °C) was applied to sodium phosphate buffered (50 mM, pH 8.3) solutions of the purified enzyme, and the enzyme activity remaining after 20 min of incubation was detected on CGP-overlay agar plates. Constant CGPase activity was observed at temperatures of 3042 °C. Temperatures of more than 42 °C led to partial inactivation of the enzyme as indicated by reduced diameters of the degradation halos. At temperatures of 68 °C or higher, CphEBm was completely inactivated. The substrate specificity of CphEBm was investigated by detection of the released degradation products from different polypeptide substrates employing the ninhydrin reagent. After 2 h incubation of CGP, BSA, bovine casein (Hammerstengrade), and poly(R/β-D/L-aspartic acid) in the presence of purified CphEBm, only CGP samples showed a release of ninhydrin-positive degradation products, whereas the others revealed no release of detectable products (data not shown) indicating the high specificity of purified CphEBm for CGP. This is consistent with the intracellular CGPase (CphB) of Synechocystis sp. PCC 68039 and the extracellular CphE of P. anguilliseptica BI.10 CphEBm is another example supporting the general assumption that CGP is exclusively hydrolyzed by specialized CGPases and not by proteases. This conclusion is also confirmed by total resistance of CGP to a variety of commercially available proteases6,28 and by the failure of typical protease producing bacilli like B. megaterium (DSM 319) or B. subtilis 168+ (DSM 402) to hydrolyze CGP.10 Analysis of Degradation Products after Incubation of CGP with CphEBm. For analysis of the time-dependent formation of high molecular weight CGP degradation products, CGP isolated from Synechocystis PCC6308 was incubated in the presence of purified CphEBm, and the degradation products were separated by SDS-PAGE (Figure 3). HPLC analysis (data not shown) and electrospray ionization mass spectrometry (ESI/MS/MS) (Figures 4 and 5) were also employed to identify low molecular weight CGP degradation products. Degradation of the high molecular weight, polydisperse cyanobacterial CGP (about 43-100 kDa; lane 2 in Figure 3) to products of lower molecular weight (lanes 3-10) was demonstrated. In contrast to SDS-PAGE data obtained for CGP degradation by CphEPa, which showed the occurrence of a molecule population in the MW range of about 30 kDa and a low degree of polydispersity,10 the CphEBm reaction resulted in the formation of CGP molecules that showed no distinct bands but a more dispersed molecule distribution. Nevertheless, the degradation of high molecular weight CGP of about 100 kDa was almost complete after 40 min (lane 5 in Figure 3), whereas degradation products within a MW range of about 50-60 kDa remained and even accumulated to some extent as indicated by a slight transient increase of the color intensity after 20-40 min (lanes 6-7 in Figure 3). The lack of detectable high molecular weight material after 24 h of incubation (lane 11) was clearly shown (Figure 3). Analysis of similar degradation samples, employing purified CphEBm and CGP produced in recombinant E. coli DH1 cells by

Biomacromolecules, Vol. 5, No. 1, 2004 157

Figure 3. Decrease of the molecular mass of CGP during incubation of the polymer in the presence of purified CphEBm: (lanes 1 and 12) molecular mass standard proteins; (lane 2) 25 µg of cyanobacterial CGP (isolated from Synechocystis PCC6308) in 50 mM sodium phosphate buffer (pH 8.3); (lanes 3-10) CGP after 2, 5, 10, 20, 40, 70, 110, 150, and 1440 min of incubation in the presence of CphEBm. CGPase was inactivated by 5 min incubation at 70 °C before applying the samples to SDS-PAGE. CGP was visualized by Coomassie staining.

Figure 4. ESI/MS positive ion spectrum of an overnight degradation sample of CGP (1 mg/mL) produced by recombinant E. coli cells incubated in the presence of CphEBm. The peaks I (m/z 290), II (m/z 281), III (m/z 562), and IV (m/z 175) represent the predominant ions in the spectrum and correspond to the β-Asp-Arg dipeptide (I), the (β-Asp-Arg)2 tetrapeptide (II, III), or arginine (IV). Fragmentation of the tetrapeptide [M2 + 2H]2+ at m/z 281 and of the β-Asp-Arg dipeptide [M1 + H]+ at m/z 290 confirmed their identity (compare Figure 5 for details).

HPLC confirmed the correlation between the decrease of the molecular mass of CGP, that is, the initial formation of high molecular weight degradation products, and subsequent increase of the amounts of low molecular weight degradation products. The main degradation products detected by HPLC analysis were β-Asp-Arg dipeptides (data not shown), which were identified because of their characteristic retention time

158

Biomacromolecules, Vol. 5, No. 1, 2004

Figure 5. Structural analysis of ions represented by the peaks at m/z 281 and m/z 290 (see positive ion spectrum in Figure 4) by ESI/ MS/MS. The fragmentation pattern of both ions confirmed the proposed structures for the (β-Asp-Arg)2 tetrapeptide [M2 + 2H]2+ [1] and the β-Asp-Arg dipeptide [M1+H]+ [2]. One of the daughter ions (Ia) detected after fragmentation of the tetrapeptide (IIa in panel 1) matches exactly the m/z value of the dipeptide (Ib) shown in panel 2, indicating that the dipeptide is a structural element of the ion detected at m/z 281. In addition, in both spectra, panels 1 and 2, the peak at m/z 175 ([arginine + H]+; IIIa and IIIb) was found. Peak IIb represents the β-Asp-Arg dipeptide after the loss of NH3, peaks IVa and IVb at m/z 158 correspond to the arginine ion after the loss of NH3 [arginine + H - 17]+, whereas peaks Va and Vb represent the residual part of the aspartic acid molecule after fragmentation of the β-Asp-Arg dipeptide at the β-amide bond.

of 2.2 min.10 The HPLC chromatograms consistently showed additional peaks, which were probably representing oligomers of the β-Asp-Arg dipeptide (see below), indicating incomplete degradation of CGP after 24 h of incubation. The occurrence of dipeptides was also confirmed by electrospray ionization mass spectrometry (ESI/MS/MS) analysis (Figures 4 and 5). This method was employed for identification of the final degradation products of CGP by mass determination and structural analysis. Figure 4 shows the spectrum of the positively charged predominant molecules with m/z values