Antibacterial activity of glycosylated and phosphorylated chromogranin A-derived peptide 173-194 from bovine adrenal medullary chromaffin granules.

Recently, we have isolated from bovine chromaffin granules and identified two natural peptides possessing antibacterial activity: secretolytin (chromogranin B 614-626) and enkelytin (proenkephalin-A 209-237). Here, we characterize a large natural fragment, corresponding to chromogranin A 79-431, that inhibits growth of both Gram-positive and Gram-negative bacteria. The aim of the present work was to determine the shortest active peptide located in the 79-431 chromogranin A region. Three peptides, which shared the same 173-194 chromogranin A sequence (YPGPQAKEDSEGPSQGPASREK) but differed in post-translational modifications, including O-glycosylation and tyrosine phosphorylation, were isolated. A detailed study using microsequencing and mass spectrometry allowed us to correlate their antibacterial activity with these post-translational modifications. The chromogranin A precursor fragment (79-431) and the active glycosylated and phosphorylated peptides were, respectively, named prochromacin and chromacin (P, G, and PG for phosphorylated, glycosylated, and phosphorylated-glycosylated form).

Secretory granules from bovine adrenal medullary chromaffin cells contain a complex mixture of secretory products that include low molecular mass constituents such as catecholamines, ascorbate, nucleotides, calcium, enkephalins, and several water-soluble proteins. Among the latter, dopamine-␤hydroxylase, proenkephalin-A, and the chromogranins/secretogranins have been extensively studied. Chromogranins, which are widely distributed in endocrine and neuroendocrine cells and in neurons, are a family of acidic soluble proteins (1)(2)(3)(4). At the subcellular level, chromogranins are exclusively found in the soluble core of hormone and neurotransmitter storage vesicles and are released during exocytosis. In bovine adrenal medulla, chromogranin A (CGA) 1 (40% of total soluble granule proteins) is a single polypeptide chain of 431 residues. The difference between the apparent (70 kDa) and theoretical molecular mass (48 kDa) probably results from post-translational modifications (i.e. glycosylation, phosphorylation, and sulfation) and the abundance of acidic residues (25%), which cause a slower migration during electrophoresis in the presence of sodium dodecyl sulfate (see Ref. 5, for review). The function of CGA is still elusive, although the different members of the chromogranin family (CGA, chromogranin B/secretogranin I (CGB/SgI), chromogranin C/secretogranin II (CGC/SgII), secretogranin III (SgIII), secretogranin IV (SgIV), or HISL-19 antigen and secretogranin V (SgV) or 7B2) are now considered as precursor proteins that are actively processed into peptides within the secretory granules.
In 1993, we reported a detailed study of the intracellular and extracellular processing of CGA in bovine chromaffin granules (6). The proteolytic processing of CGA is a topic of growing interest as biological activities have been attributed to specific domains of CGA. Several of these CGA-derived peptides act predominantly as inhibitors of hormone and neurotransmitter release, either as autocrine or paracrine controls. For example, CGA is the precursor of pancreastatin (248 -293), a peptide that negatively modulates insulin secretion from endocrine pancreatic islets (7,8), amylase release from exocrine pancreas (9), and acid secretion from parietal cells (10). Another CGAderived peptide, parastatin (347-419), inhibits parathyroid cell secretion (11). As early as 1988, it was established that CGA is the precursor of a peptide that inhibits the secretory activity of chromaffin cells (12). More recently, a peptide corresponding to the sequence 1-113 has been shown to inhibit hormone secretion in the bovine parathyroid gland (13); a homologous peptide named betagranin corresponding to the sequence 1-115 has been isolated from rat pancreas, but its function has not yet been defined (14). Peptides containing the N-terminal domain (1-76), vasostatins, have been characterized in bovine adrenal medulla (15) and have been found to exhibit vasoinhibitory activity of isolated human blood vessels (16,17).
Recently, we have shown that peptides with antibacterial activity are present as water-soluble components of bovine chromaffin granules and are released during secretion. In material released from stimulated cultured chromaffin cells, we have identified secretolytin (18,19), a peptide corresponding to the C-terminal sequence (614 -626) of bovine chromogranin B. A second antibacterial peptide, enkelytin (20), derived from proenkephalin (209 -237), was then identified among the nu-merous peptides present in the intragranular matrix. As a continuation of these studies, the present paper reports the antibacterial activity of a large CGA fragment (residues 79 -431) generated by natural cleavage at the previously described site 78 -79 (6). This CGA fragment was detectable in the intragranular matrix, was released during exocytosis and furthermore inhibited the growth of both Gram-positive (Micrococcus luteus) and Gram-negative (Escherichia coli) bacteria.
In view of the large size of this antibacterial CGA fragment, the aim of the present paper was to characterize the shortest active peptide derived from this fragment. After proteolysis of the whole CGA molecule, active peptides located in the 79 -431 region were analyzed using a combination of microsequencing and mass spectrometry. We isolated several peptides that shared the same 173-194 sequence but differed by glycosylation and phosphorylation modifications. Their sequence has no homology with the previously described sequences of antibacterial peptides. Structural features and more particularly posttranslational modifications are discussed in relation with their antibacterial activity.

EXPERIMENTAL PROCEDURES
Purification of CGA-derived Peptides 173-194 -Secretory granules were isolated from bovine adrenal medulla (21), and soluble proteins were separated from membranes after lysis and centrifugation (22). CGA was purified by HPLC on a Macherey Nagel Nucleosil 300 -5C18 column (4 ϫ 250 mm; particle size: 5 m and pore size: 100 nm) with the Applied Biosystems HPLC system 140 B as described previously (18). Then, CGA (10 nmol) was digested for 2 h at 37°C with endoproteinase Lys-C at a protein-to-proteinase weight ratio of 1000:1 in 100 mM Tris-HCl, pH 8.3. Generated peptides were then separated on a Macherey Nagel 300 -5C18 column. Absorbance was monitored at 214 nm, and the solvent system consisted of 0.1% trifluoroacetic acid in water (solvent A) and 0.1% trifluoroacetic acid, 30% water, 69.9% acetonitrile (solvent B). Material was eluted at a flow rate of 0.7 ml/min using successively a gradient of 0 -25% B in A over 10 min followed by a gradient of 25-75% over 50 min. Each peak fraction was manually collected and concentrated by evaporation, but not to dryness.
Sequence Analysis-The sequence of purified CGA-derived peptides was determined in our laboratory by automatic Edman degradation on an Applied Biosystems 473 A microsequencer. Samples (100 pmol) were loaded onto polybrene-treated and precycled glass-fiber filters (6). In order to identify phosphorylated residues, samples were modified with ethanethiol according to the method previously described (23). Before sequencing, reagents were removed, using the ProSpin sample preparation cartridge (Applied Biosystems, a division of Perkin-Elmer), and the modified peptide was loaded on the ProBlott polyvinylidene difluoride membrane (Applied Biosystems).
Mass Spectrometry Analysis-Determination of mass was carried out on a Brucker BIFLEX™ matrix-assisted laser desorption time-of-flight mass spectrometer (MALDI-TOF) equipped with the SCOUT™ high resolution optics with X-Y multisample probe, a gridless reflector and the HIMAS™ linear detector. This instrument has a maximum accelerating potential of 30 kV and may be operated either in the linear or reflector mode. Ionization was accomplished with a 337-nm beam from a nitrogen laser with a repetition rate of 3 Hz. The output signal from the detector was digitized at a sampling rate of 250 MHz in linear mode and 500 MHz in reflector mode using a 1-GHz digital oscilloscope (Lecroy model). The instrument control and data processing were accomplished with software supplied by Bruker using a Sun Sparc workstation. These studies were realized using as the matrix ␣-cyano-4hydroxycinnamic acid obtained from Sigma and prepared as a saturated solution in acetone. Aliquots (1-2 l) of the sample matrix solution were deposited onto probe tips and dried by ambient air. After fast spreading and fast evaporation of the solvent, we obtained a thin layer of matrix crystals (24,25). A micromolar analyte solution was applied to the matrix and allowed to dry under moderate vacuum. This preparation was washed by applying 1 l of a 0.5% trifluoroacetic acid in water solution and then flushed after a few seconds. This cleaning procedure often allows an increase in sensitivity and mass accuracy by removing the remaining alkali cations.
Antibacterial Assays-Bacteria were grown aerobically at 37°C in yeast extract-free Luria-Bertani medium (1% Bactotryptone and 0.5% NaCl (m/v), pH 7.5). Antimicrobial activity was determined by measuring growth inhibition of M. luteus (strain A270), Bacillus megaterium (strain M A) and E. coli (strain D22) in Luria-Bertani medium (30). Aliquots of peptide extract (200 pmol; 10 l) were incubated in microtiter plates with 100 l of a midlogarithmic phase culture of bacteria with starting absorbance at 620 nm of 0.001. Microbial growth was assessed by the increase of A 620 nm after 16 h incubation at 37°C (M. luteus and E. coli) or 20°C (B. megaterium). The A 620 nm value of control cultures grown in the absence of peptide (10 l of water in place of peptide solution) was taken as 100%.
Lysis of Erythrocytes-Bovine erythrocytes were isolated and the buffy coat removed by centrifugation of freshly collected blood at 1000 ϫ g for 5 min and washing three times with 10 mM sodium phosphate, pH 7.5, containing 0.9% NaCl (NaCl/P i ). Erythrocytes (45 l) were incubated at 37°C over 40 min in NaCl/P i with 4 and 8 M of a mixture of glycosylated/phosphorylated CGA-derived peptides (5 l). Then, incubation media were centrifuged at 1000 ϫ g for 5 min and aliquots (30 l) of supernatant were diluted in water (1 ml). The absorbance of the diluted solution was measured at 420 nm. The absorbance obtained after treating erythrocytes with 2% SDS was taken as 100%.
Peptide Synthesis-Peptides were synthesized in our laboratory on an Applied Biosystems 432 A peptide synthesizer, SYNERGY, using the stepwise solid-phase synthetic approach and Fmoc (9-fluorenylmethoxycarbonyl) chemistry (31). All residues were incorporated with double coupling. Peptides were further purified by reverse phase HPLC on a Brownlee Aquapore OD-300 (7 ϫ 250 mm) and lyophilized.
Sequence Comparisons-Sequence alignment of bovine CGA (173-194) with corresponding fragments of CGA and CGB for different species was performed using the Clustal V multiple sequence alignment program (32) using default parameters. Chromogranin sequences were retrieved from the Swiss-Prot data base.

RESULTS
Bovine chromaffin granules contain a large number of chromogranin-and proenkephalin-A-derived peptides. When the soluble intragranular material or the secreted fragments from K ϩ -depolarized chromaffin cells were separated by HPLC and their effects on the growth of M. luteus (Gram-positive bacteria) and E. coli (Gram-negative bacteria) tested, antibacterial activity was detected in several fractions. With soluble granule matrix proteins as starting material (Fig. 1A), active fractions inhibiting the growth of both these bacteria were recovered in a group of peaks (a1 in Fig. 1A1), where proenkephalin-A and CGA-derived peptides were identified. After a second purification step by HPLC, an active fraction (a2) corresponding to a fragment with N-terminal end starting on position 79 in CGA sequence was identified. After monodimensional gel electrophoresis, the apparent molecular mass of this fragment was estimated to 63 kDa. This value corresponds to the large CGA fragment 79 -431 reported previously (6). We also looked for antibacterial activity in material released from K ϩ -stimulated cultured chromaffin cells (Fig. 1B). The same active CGA 79 -431 fragment was isolated (fraction b2). Microsequencing revealed a unique N-terminal end, and electrophoresis separation showed a single component with an apparent molecular mass of 63 kDa. This fragment displays antibacterial activity, inducing a 100% inhibition of bacterial growth at a concentration of 1.5 M.
In order to characterize the shortest active sequence included in this large fragment 79 -431, CGA was submitted to proteolysis with endoproteinase Lys-C and after separation by HPLC on a reverse-phase C18 column, the generated fragments were tested for their antibacterial activity, and peptide structure was determined by combination of microsequencing and mass spectrometry analysis. As shown in Fig. 2, peptides inhibiting M. luteus growth were found in three successive peaks numbered 1 to 3. Complete sequencing of material present in peaks 1-3 indicated that it was the same peptide located in position 173-194 (Fig. 2B). This peptide was also recovered in the neighboring peak 4, but no bacterial activity was found in this fraction. The sequence (173-194) of this peptide was included in the large fragment described previously (Fig. 1), and its elution as four different subspecies suggested posttranslational modifications.
Isolation of Antibacterial CGA-derived Peptides-The sequencing of peptides present in peaks 1 and 4 showed unambiguously PTH-(Tyr 173 , Ser 182 , and Ser 191 ), whereas in peaks 1 and 2, the PTH-Ser 186 was not detectable, thus indicating a possible O-glycosylation on this residue. According to structure analysis reported by Wilson (33), the statistically significant patterns PS/T and S/TXXP predict O-glycosylation sites, and the sequence around residue Ser 186 (PSQGP) fits with these two characteristic patterns. In addition, Ser 182 (KEDSEG) is located in a consensus phosphorylation site by protein kinase C (K/RXXS/T).
To further determine the structure of the 173-194 peptide subspecies, mass spectrometry analyses were initiated using the MALDI-TOF technique.
Mass  2. Purification and structural characterization of the four different forms of CGA-derived fragment (173-194) found in fractions 1-4. A, CGA-derived peptides after digestion with endoproteinase Lys-C were separated on a Macherey Nagel reverse-phase C18 column (4 ϫ 250 mm). The elution profile is shown; absorbance was monitored at 214 nm, and elution was performed with a linear gradient as indicated on the right-hand scale. Numbers 1-4 indicate fractions containing the amino acid sequence reported in B. B, sequence determination of the 194 -The material present in peak 3 ( Fig. 2A) was first treated with alkaline phosphatase treatment to verify whether this peptide was phosphorylated (additional mass: 79 Da) or sulfated (additional mass: 80 Da). After mass spectrometry analysis, we observed the removal of a mass of 79 Da, indicating that this peptide was phosphorylated. To identify the amino acid bearing the post-translational phosphorylation, the 173-194 peptide from peak 3 was submitted to the Meyer modification in presence of ethanethiol (23) prior to sequencing, a method which identified the phosphoserine residue as PTH-Sethylcysteine. We found that residues Ser 182 , Ser 186 , and Ser 191 were not phosphorylated whereas the N-terminal residue Tyr 173 was modified. However, PTH-Leu was unexpectedly recovered instead of PTH-Tyr. Sequencing of peptide present in peak 4 clearly showed that Tyr 173 was unmodified. Similarly, PTH-Leu was obtained as the N-terminal residue when peptide present in peak 1 was submitted to the Meyer modification before sequencing.
It is noteworthy that Ser 182 residue, which is a putative substrate for protein kinase C, was not phosphorylated in any fraction.
Structural Characterization of the Glycosylation of the 173-194 Peptide-The molar carbohydrate composition was performed on the 173-194 peptide isolated from peak 2 using a gas chromatography. The carbohydrates NeuAc, Gal, and GalNAc were detected in a molar ratio of 1:1:1. Using similar experimental conditions, a similar molar ratio was also obtained for the whole purified CGA protein, suggesting a unique structure for the different glycan chains distributed along the protein. To obtain more structural information, the strategy was to electrotransfer purified CGA onto nitrocellulose sheet and to use a panel of lectins, the specificity of which is detailed in Fig. 3A. As shown in Fig. 3B, CGA was not recognized by SNA-dig lectin (lane 1), whereas a 70-kDa band was revealed with MAA-dig lectin (lane 2); the binding of this lectin to CGA is abolished with sialidase treatment (lane 6). This result indicated that sialic acid residues are in a ␣2-3 linkage and that no ␣2-6-linked sialic acid residues are present in CGA glycans.
In addition, ACA-dig lectin decorated a similar band prior to or after sialidase treatment (lanes 4 and 8, respectively), whereas PNA-dig lectin only bound to the glycoprotein after desialylation (lane 7). The two lectins recognized the T-antigen (Gal␤1-3GalNAc␣1-O-Ser) with the difference that ACA lectin is also able to bind to the cryptic T-antigen. Desialylation clearly indicated that all the Gal␤1-3GalNAc␣1-O-Ser sequences in CGA are sialylated. Taken together these data suggest that the different glycosylation sites in CGA are on serine residues and are composed of the trisaccharide NeuAc␣2-3Gal␤1-3GalNAc␣1 O-linked to a serine residue. In the glycopeptide 173-194 this glycan moiety is linked to Ser 186 . The calculated molecular mass of this glycan is 655 and corresponds to the experimental value determined from mass spectrometry (Fig. 2). In 1991, Wilson reported the structural requirements for the addition of O-linked N-acetylgalactosamine to proteins (33): the most prominent feature in the vicinity of O-glycosylated sites is a significantly increased frequency of proline residues, especially at positions Ϫ1 and ϩ3 relative to the glycosylated residue. In the glycopeptide 173-194, a characteristic sequence pattern PSQGP surrounds the Ser 186 Oglycosylation site. Table I  (v) in all preparations of CGA digested with endoproteinase Lys-C, the ratio of peaks 2 and 3 appears constant, whereas the ratio of peaks 1 and 4 was more variable between preparations.

Structural Characterization of the Different Post-translational Modified Species of 173-194 Peptide-
Antibacterial Activity of CGA 173-194 Subspecies-As shown in Fig. 4, CGA peptides present in fractions 1-3 inhibited the growth of M. luteus (strain A 270), but were inactive toward E. coli (strain D22). B. megaterium (strain M A) was affected only by the peptide present in fraction 1. Complete inhibition of bacteria growth was reached for a peptide concentration around 1.5 M. In contrast, unmodified CGA 173-194 from fraction 4 was totally inactive. These results suggest the importance of these post-translational modifications for antibacterial activity. As a control, in order to demonstrate that the antibacterial activity was not due to nonpeptidic material present in the fractions, we digested extract aliquots (  PSQGPASR, and EK as determined by microsequencing) had no effects on M. luteus and B. megaterium growth (Fig. 4), showing that the N-terminal moiety 173-179 (phosphorylated or nonphosphorylated) and the two C-terminal residues (E, K) are essential for antibacterial activity.
Antibacterial peptides corresponding to the phosphorylated/ glycosylated forms of CGA 173-194 did not cause hemolysis of bovine erythrocytes. We propose to name these antibacterial peptides chromacin: chromacin-P, chromacin-G, and chromacin-PG, respectively, for phosphorylated, glycosylated, and phosphorylated-glycosylated form. Since this sequence is included in a larger CGA fragment (79 -431), this precursor, which has inhibitory effects on M. luteus and E. coli, could be named prochromacin.

DISCUSSION
Bovine CGA from adrenal medulla is a glycoprotein containing 5.4% carbohydrate (1). It has been reported previously that sugars are mainly present as O-glycosidically-linked tri-and tetrasaccharides composed of N-acetylgalactosamine, galactose, and sialic acid (38). In addition, it has also been established that adrenal CGA is a phosphoprotein containing five phosphoserine residues per molecule (39) and that it also incorporates [ 35 S]sulfate during protein biosynthesis (40), most of it appearing bound to carbohydrates.
The present paper reports the location of two post-translational modifications on bovine CGA isolated from adrenal medulla, since we clearly demonstrate that tyrosine 173 and serine 186 are, respectively, phosphorylation and O-glycosylation sites. The glycan moiety bound to serine 186 was shown to be NeuAc␣2-3Gal␤1-3GalNAc␣1. We also demonstrated that these post-translational modifications are required to confer antibacterial activity to the CGA 173-194 fragment. Its antibacterial activity is directed against Gram-positive bacteria.
Analysis of the CGA 173-194 fragment indicates that it is a negatively charged peptide with a net charge of Ϫ1. The presence of the phosphate group on tyrosine 173 and the NeuAc␣2-3Gal␤1-3GalNAc␣1 moiety on serine 186 increases the acidic nature of the corresponding antibacterial chromacin-P, -G, and -PG species. This property contrasts with the idea that antibacterial peptides need to be positively charged in order to bind to bacterial surfaces; however, we reported previously that enkelytin, which is also a highly negatively charged peptide derived from proenkephalin-A, has potent antibacterial activity (20). In CGA 173-194 sequence, the total proportion of charged residues is 32%, and its hydropathy profile corresponds to a very hydrophilic peptide in contrast with the amphipathic structure generally described for antibacterial peptides.
In the chromacin peptide, the 4 proline residues located in positions 174, 176, 185, 189, and the three glycine-proline bonds (175-176, 184 -185, 188 -189) will induce conformational constraints on the peptide chain: the side chain is probably bent back onto the backbone amide position. Inside an ␣-helix, proline residues can no longer establish hydrogen bonds with the preceding turn, thereby introducing a kink. Furthermore, the specific properties of proline render its presence in ␣-helix energetically unfavorable. It is interesting to note, however, that in the cecropin model (41), the two ␣-helices are joined by a hinge region containing a glycine-proline doublet. In the CGA 173-194 fragment, the close proximity of the three glycineproline doublets prevents the formation of ␣-helices. Nevertheless, proline has been found with relatively high frequency in the putatively ␣-helical transmembrane segments of many integral membrane proteins; transmembrane proline residues have a tendency to be conserved among homologous proteins that function as receptor subunits or transporters (42). In addition, functioning as a built-in signal, proline residues in polypeptide chain may function as a contributing to a diversity of processes (43) such as immunomodulation (44 -46), coagulation (47), homeostasis (43), inflammation, and microbial/viral infections (48,49). Many neuro-and vasoactive peptides have proline residues in their sequences. An examination of the amino acid sequences of proteins registered in international data banks reveals that numerous cytokines and growth factors share an X-proline sequence at their amino terminus (i.e. interleukins IL-1␤, IL-2). Alignment of bovine CGA peptide 173-194 with homologous fragments from several species (Fig.  5) shows the conservation of this X-proline sequence at the N-terminal end of bovine, porcine, and human CGA as well as human CGB. Proline may not only determine secondary structural properties necessary for their biological activity, but may also hinder nonspecific proteolytic modifications such as Cterminal amidation, acetylation, or N-terminal cyclization to pyroglutamic acid. The striking degree of conservation seems to reflect an evolutionary pressure toward this X-proline motif. Alternatively, proline-containing motifs may serve as recognition sites for specific peptidases (46). Some other inducible antibacterial peptides can also be described as proline-rich peptides (2-3 kDa) and glycine-rich peptides . It is important to point out that the chromacin sequence is different from these antibacterial molecules, including apidaecin (50), abaecin (51), drosocin (30), pyrrhocoricin (52), diptericin (53), and lebocin (54).
Drosocin, pyrrhocoricin, diptericin, and lebocin carry O-gly-  cosylated substitutions located on threonine residues. Interestingly, the removal of these glycan moieties suppresses their antibacterial activity, indicating that this post-translational modification is essential for their biological activity (53,54). The target of lebocin and its related peptides has been suggested to be the bacterial membrane as they strongly affect the permeability of liposome preparations (54). One apparently universal consequence of O-glycosylation is the relative resistance to proteases of the region neighboring the O-glycosylated sites. The most likely explanation for protease resistance is that the attached carbohydrate located in a ␤-turn secondary structure blocks access to the peptide core. A second consequence of O-glycosylation is the induction of a specific conformation. According to the method of Chou and Fasman (55, 56), we can predict that the O-glycosylated serine residue 186 is located in a ␤-turn (results not shown), which permits the formation of a stem in the polypeptide structure. This conformation is probably stabilized by diverse electrostatic interactions between lysine 179, glutamic acid residues 180 and 183 within the first arm and arginine 192, glutamic acid 193, and lysine 194 within the second arm. Interestingly, photoaffinity cross-linking studies have previously established that the proline-rich glycoprotein from human parotid saliva interacts with carbohydrate units (NeuAc␣2-3Gal␤1-3GalNAc) of the glycoproteins and lectin(s) located on bacteria cell surfaces (57). These results indicate that glycoprotein binding is mediated by a bacterial adhesin that recognizes complex oligosaccharide structures consisting of sialic acid, galactose, and N-acetylgalactosamine. In enkelytin, we have shown (20) that phosphorylation of two serine residues provokes the opening of a boomerang-like structure, which allows the C-terminal amphipatic helix to interact with the bacterial membrane. Curiously, in phosphorylated chromacin, the phosphorylation site is N-terminal tyrosine residue 173. Phosphorylation of tyrosine residues is not very common, representing only 0.03% of the phosphorylated amino acids in normal cells (58). Phosphorylation of tyrosine 183 is one of the essential modifications necessary for chromacin-P antibacterial activity. However, the nonphosphorylated, glycosylated chromacin subspecies (chromacin-G) was also fully active. The structural modification of chromacin induced by phosphorylation of N-terminal tyrosine residue is not understood yet. At this stage we can only speculate that active chromacin peptides interact with a structural bacterial "receptor," which accommodates only the glycosylated and/or phosphorylated forms. An interesting consequence of O-glycosylation and phosphorylation is that the (chromacin-P, -G, and -PG) become relatively resistant to proteases and therefore are stable antibacterial agents. In addition, the larger fragment named prochromacin (CGA 79 -431) has antibacterial activity directed against both Gram-positive and Gram-negative bacteria. This double activity is lost in chromacin-P, -G, and -PG, suggesting that other domains in prochromacin are responsible for the antibacterial activity directed against Gram-negative bacteria.
In Fig. 5, the alignment of chromacins (59) with corresponding peptides of chromogranins raises several points. First, comparison of these different CGA peptides shows a relatively high degree of identity between porcine CGA (60) (66.7%), human CGA (61) (70.8%), and also mouse (62) and rat CGA (63) (45.8%). Interestingly, some similitude with CGB sequence and more particularly with human and mouse CGB (64, 66) (33.3 and 25%, respectively) was also observed, but only a weak identity (12, 5%) was obtained with bovine CGB (65). Second, the simultaneous presence of phosphorylated tyrosine and Oglycosylated serine occurs only in the bovine CGA sequence. A potential O-glycosylation site is present in the porcine CGA sequence, while the N-terminal sequence including the first tyrosine residue is identical in human CGA. Third, the Nterminal domains (residues 173-183 in bovine CGA) seem to be more conserved than the C-terminal region 184 -194. A ratio of 90% is obtained for comparison with human CGA. In addition, alignment of the N-terminal sequence 173-183 with sequences contained in data banks show 100% of homology with a short peptide (YPAPQGRE) of metavinculin (VINC_XENLA), a protein known to be localized in the subplasmalemmal region in many cells and to be involved in the binding of cytoskeletal filaments to membranes.
Finally, our studies indicate that the different antibacterial peptides isolated from bovine chromaffin granules, enkelytin, secretolytin, and prochromacin are co-released with catecholamines and other neuropeptides during exocytosis following K ϩ stimulation of cultured chromaffin cells. In view of the widespread distribution of the chromogranins and proenkephalin-A, these peptides may also be present and secreted from other endocrine and neuroendocrine chromogranin-containing cells. The identification of different classes of antibacterial peptides in diverse range of organisms, including prokaryotes, insects, frogs, and mammals, suggest they play a potentially important role in the host defense against microbial infections. Experiments currently in progress in our laboratory aim to establish the physiological relevance of this bovine chromaffin granule antibacterial activity.