Identification of Posttranslational Modifications and cDNA Sequencing Errors in the Rat S100 Proteins MRP8 and 14 Using Electrospray Ionization Mass Spectrometry

ANALYTICAL BIOCHEMISTRY ARTICLE NO.

258, 285–292 (1998)

AB972601

Identification of Posttranslational Modifications and cDNA Sequencing Errors in the Rat S100 Proteins MRP8 and 14 Using Electrospray Ionization Mass Spectrometry Mark J. Raftery1 and Carolyn L. Geczy Cytokine Research Unit, School of Pathology, University of New South Wales, Kensington, New South Wales, 2052, Australia

Received September 18, 1997

MRP8 and 14 are S100 proteins expressed by myeloid cells and are predicted to have important functions in inflammation. The proteins were isolated from spleens from three rat strains. Electrospray ionization mass spectrometry indicated masses of 10149 6 2 Da for MRP8 and 13,069 6 2 Da for MRP14 compared to masses calculated from proteins derived from their cDNA sequences of 10,211 and 13214 Da, respectively, indicating posttranslational modifications and/or errors in the derived protein sequences. Several endoprotease digest peptides did not correspond to any theoretical digest products after comparison of ESI masses with those derived from the theoretical digest. Both proteins were Nterminally acetylated after deletion of the initiator Met, reducing the theoretical masses by 89 Da. A peptide with mass 28 Da greater than the theoretical was isolated from the Asp N digestion of MRP8. Nterminal sequencing indicated translated Val instead of the predicted Ala at position 72 of MRP8. A peptide 56 Da less than the theoretical was isolated from the chymotryptic digestion of MRP14, and the carboxyamidomethylated form was N-terminally sequenced and found to have translated Ser instead of the predicted Arg at position 105. In addition, His106 was methylated. The corrected theoretical masses, incorporating the posttranslational modifications and sequencing errors, are 10,149.4 and 13,069.9 Da for MRP8 and 14, respectively, in good agreement with the experimental masses. © 1998 Academic Press

1 To whom correspondence should be addressed. Fax: 61-2-93851389. E-mail: [email protected].

0003-2697/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.

MRP82 and 14, also known as p8 and p14, L1 light and heavy chain, calgranulins A and B, and S100A8 and A9 (1) are small acidic proteins containing two Ca21 binding EF hands. They belong to the highly conserved S100 protein family (2– 4) which may influence diverse cellular processes including cell cycle progression, cell differentiation, regulation of protein phosphorylation, and cytoskeletal–membrane interactions (5, 6). The precise functions of human (h) MRP8 and 14 are unclear, although they have been associated with inflammation and may contribute to the pathologies of diseases such as rheumatoid arthritis and cystic fibrosis (5). MRP14 inhibits casein kinase I and II activity (7) and has extracellular antimicrobial activity, possibly due to its ability to sequester Zn21 from the surrounding environment (8, 9). Murine CP10 [known as murine MRP8 (10)] was the first of three S100 proteins described as chemotactic (11–13) and the rat MRP8/14 complex (calprotectin), at high concentrations, causes apoptosis (14). These proteins comprise a high proportion of the cytoplasm of neutrophils and the complex is translocated to the membrane following stimulation, suggesting essential roles in neutrophil activation (15, 16). Moreover, the extended C-terminal domain of hMRP14 is identical to the neutrophil immobilizing factor (4) and has high sequence homology to the contact domain of high-molecular-weight kininogen, suggesting that MRP14 may promote the localiza2 Abbreviations used: MRP14, migration inhibitory factor (MIF)related protein, 14 kDa; MRP8, migration inhibitory factor (MIF)related protein, 8 kDa; cDNA, complementary DNA; ESI/MS, electrospray ionization mass spectrometry; MALDI/MS, matrix-assisted laser desorption ionization mass spectrometry; CP10, chemotactic protein, 10 kDa; RP-HPLC, reverse-phase high-performance liquid chromatography; TBS, Tris-buffered saline; TFA, trifluoroacetic acid; DTT, dithiothreitol; PTH, phenylthiohydantoin; TFA, trifluoroacetic acid.

285

286

RAFTERY AND GECZY

tion of neutrophils and monocytes at sites of inflammation (17). Human MRP8 and 14 were isolated from supernatants of activated peripheral blood mononuclear cells as part of a complex (18) and amino acid sequences confirmed from the derived cDNA sequence (18). Two forms of hMRP14 were identified by MALDI/MS (19) and after incorporating posttranslational modifications, the calculated masses of hMRP8 and 14 derived from the cDNA sequences were identical to experimental masses determined by ESI/MS (20). CP10 (21) is a potent chemoattractant for myeloid cells and is implicated in leukocyte recruitment in inflammatory lesions such as Gram-negative bacterial infection and atherosclerosis (11). CP10 and murine MRP14 were isolated from concanavalin-A-stimulated spleen cell supernatants and amino acid sequences determined biochemically and from the derived cDNA sequence (10, 21–23). The calculated and experimental masses of CP10 (determined by ESI/MS) were identical. No N-terminally truncated murine MRP14 has been identified, although a combination of ESI/MS and Edman sequencing indicates posttranslational modifications of N-terminal acetylation and specific 1-methylation of His106 (23). The precise function for the latter unusual modification is not known. Expression of rat MRP8 and 14 mRNA has been associated with the development of chronic arthritis in susceptible strains and amino acid sequences were derived from cDNA sequences obtained from a subtractive cDNA library, indicating unique expression of the genes in macrophages from susceptible animals (24). Supporting evidence for a role for these proteins in host surveillance came from the isolation, from lysates of rat inflammatory peritoneal exudates, of a complex with cytotostatic and cytolytic properties for leukemic cells and characterized as MRP8 and 14 (14, 25). Moreover, MRP14 is associated with a neutrophil phenotype associated with the metastasis-enhancing capacity of rat mammary adenocarcinoma (26) indicating important novel functions for these proteins. Although partial sequencing of internal cleavage peptides indicated sequences identical with those derived from the cDNA (14, 26), there are no reports of the isolation and characterization of the full-length rat proteins. We have isolated rat MRP8 and 14 from spleen cell lysates. Comparison of theoretical masses derived from the published cDNA sequences with those obtained after ESI/MS indicated differences of 62 Da for MRP8 and 145 Da for MRP14. Electrospray ionization mass spectrometry is a rapid and precise method for determining the mass of proteins and peptides and can validate protein sequences (27). We have identified and characterized posttranslational modifications and cDNA sequencing errors in both proteins; the corrected

theoretical and experimental masses were identical after incorporating these modifications. EXPERIMENTAL

General. Reagents and chemicals were analytical grade (Sigma, St. Louis, MO; Bio-Rad, Hercules, CA) and solvents HPLC grade (Mallinckrodt, Clayton South, Vic., Australia). Rats (Lewis, Fischer, and Sprague–Dawley) were from the UNSW animal facility. SDS/PAGE was performed using a Mini Protean II apparatus (Bio-Rad) with 15% gels and a Tris/Tricine buffer system (28). Proteins were transferred to polyvinylidene difluoride (Immobilon-P, Millipore, Bedford, MA) for Western blotting analysis using polyclonal antibodies to mouse CP10 and MRP14, horseradish peroxidase-conjugated second antibodies, and enhanced chemiluminescence (Amersham, Buckinghamshire, UK) as previously described (23, 29). Liquid chromatographic separations were performed using a nonmetallic LC6-26 HPLC system (Waters, Bedford, MA) and UV absorbance was monitored at 214 and 280 nm with a Waters 996 photodiode array detector. Protein extraction from spleens. Spleens from rats (Lewis, Fischer, or Sprague–Dawley; 200 –250 g) were stored at 280°C until used. After homogenizing three spleens in TBS (50 ml; 25 mM Tris, 250 mM NaCl, pH 7.5) with protease inhibitors (complete tablets, Boehringer-Mannheim, Castle Hill, NSW), the mixture was freeze/thawed (33) and centrifuged (5000g, 5 min) and the supernatant removed. The pellet was resuspended in TBS (25 ml), freeze/thawed, and centrifuged (5000g, 5 min) and combined supernatants were diluted with TBS (25 ml) and dialyzed against NaCl (25 mM), Tris (25 mM), pH 8 (Buffer A), at 4°C for 18 h. Carboxyamidomethylated MRP8 and MRP14 were isolated after reduction of lysates with DTT (50 mM) for 60 min at 23°C and alkylation with iodoacetamide (200 mM) for 90 min at 23°C in the dark, before dialysis. Column chromatography. The dialysate was loaded onto an AP1 column (1 3 20 cm, Waters) containing DEAE-FF Sepharose (Pharmacia, Uppsala, Sweden) preequilibrated with Buffer A at 1 ml/min for 25 min and the flowthrough pumped directly onto a heparin Econo-Pak cartridge (Bio-Rad). The heparin column was connected to an LC-626 and washed with Buffer A for 20 min. A gradient of Buffer A to 1 M NaCl, Tris (25 mM), pH 8, over 20 min was applied at 1 ml/min and fractions were collected at 2-min intervals. Fractions from the heparin column which reacted positively by Western blotting with polyclonal antibodies to murine CP10 and MRP14 were analyzed by C4 RP-HPLC (5 mm, 300 Å, 4.6 3 250 mm, Separation Group, Hesperia, CA) using a gradient of 25 to 70% acetonitrile (0.1% TFA) or 25 to 50% acetonitrile (0.1% TFA) over 30 min. Fractions with major A214 nm were collected manually

CHARACTERIZATION OF RAT MYELOID S100 PROTEINS

and analyzed by SDS/PAGE/Western blotting and mass spectrometry. Endoprotease digests. Proteins (;50 mg) isolated from C4 RP-HPLC were lyophilized to approximately 200 ml (Speedvac, Savant, Farmingdale, NY) and digested in ammonium bicarbonate [250 ml, 100 mM, pH 8.0 (150 mM DTT with endoprotease Arg C)] using endoprotease Asp N, Arg C, Glu C, or chymotrypsin (sequencing grade, Boehringer) at an enzyme to substrate ratio of approximately 1:100 (w/w) at 37°C for 2–20 h. The pH of each digest was lowered to approximately 2 with 1% TFA and the mixture applied directly to a C18 RP-column (5 mm, 300 Å, 4.6 3 250 mm, Separations Group). Peptides with major A214 nm were eluted in a gradient of 5 to 75% acetonitrile (0.1% TFA) at 1 ml/min over 30 min and collected manually, then lyophilized and dissolved in acetonitrile/water/0.1% TFA (;100 ml) for mass determination by ESI/MS. Mass spectrometry. Electrospray ionization mass spectra were acquired using a single-quadrupole mass spectrometer equipped with an electrospray ionization source (Platform, VG-Fisons Instruments, Manchester, UK). Samples (;50 pmol, 10 ml) were injected into a moving solvent (10 ml/min; 50:50 water:acetonitrile, 0.1% TFA) coupled directly to the ionization source via polyetheretherketone tubing (127 mm 3 40 cm). The source temperature was 50°C and nitrogen was used as the nebulizer gas. Sample droplets were ionized at a positive potential of approximately 3 kV and transferred to the mass analyzer with a cone voltage of 50 V. The peak width at half height was 1 Da. Spectra of proteins were acquired in multichannel acquisition mode over the mass range of 700 to 1800 Da in 5 s and then calibrated with horse heart myoglobin (Sigma). Spectra of peptides were also acquired in multichannel acquisition mode over the mass range 250 to 2200 Da in 10 s. Automated N-terminal sequencing. Peptides (typically 50 –200 pmol) were N-terminally sequenced using an Applied Biosystems Model 470A automated protein sequencer (Applied Biosystems, Burwood, Vic.) at the School of Biochemistry, La Trobe University (Bundoora, Vic.). RESULTS AND DISCUSSION

The amino acid sequences of rat MRP8 and 14 were originally derived from two cDNA clones obtained from a subtractive cDNA library produced using peritoneal exudate cells from Lew/N (susceptible) and F344/N (resistant) rats challenged with streptococcal cell wall preparations. The clones encoded proteins of 89 and 113 amino acids, with calculated masses of 10,211 and 13,214 Da, designated as rat MRP8 and 14, respectively (24). The amino acid sequences were partially confirmed by N-terminal sequencing of cyanogen bro-

287

FIG. 1. C4 RP-HPLC trace of the heparin fraction which eluted between 250 and 300 mM NaCl. Rat MRP8 and 14 are indicated. The inset shows silver-stained SDS/PAGE of major peaks which were collected manually. Major bands in lanes 3 (;14,000) and 6 (;8000) were identified as MRP14 and MRP8, respectively (see text for details).

mide cleavage products of MRP8 and 14 isolated from inflammatory peritoneal cell lysates from Wistar rats. Both proteins were N-terminally blocked (25). Rat (Fischer) MRP14 associated with neutrophils infiltrating a highly metastatic adenocarcinoma was identified by 2D-PAGE and its identity confirmed after partial N-terminal sequencing of internal tryptic peptides (26). Because definition of the precise structure of these proteins is pivotal to structure/function analysis, we adapted an efficient protocol for purification of the murine proteins (23) to isolate rat MRP8 and 14. The proteins were then biochemically characterized using ESI/MS and Edman sequencing. Isolation of MRP8 and MRP14. MRP8 and 14 were isolated from the soluble fraction of spleen homogenates from Lewis, Sprague–Dawley, and Fischer rats in order to eliminate possible differences in amino acid sequences due to allelic variations between strains. The proteins were not retained by DEAE-Sepharose and eluted as a single peak with 250 –300 mM NaCl from a heparin affinity column for which they had moderate affinity. The proteins were readily separated by C4 RP-HPLC (Fig. 1) in yields, based on relative UV absorbance, of approximately 50 mg MRP8/14 per spleen, confirming their constitutive expression in spleen. SDS/PAGE of the fractionated proteins indicated major bands at approximately 14,000 for MRP14 and 8000 for MRP8 (lanes 3 and 6, Fig. 1). The 14,000 component (lane 3) reacted with a polyclonal antimouse MRP14 antibody and that at 8000 (lane 6) with a polyclonal anti-mouse CP10 antibody by Western blotting, confirming their identity (not shown). Spleen lysates from the three rat strains gave essentially the same elution profiles after heparin affinity and C4

288

RAFTERY AND GECZY TABLE 1

Comparison of the Electrospray Ionization Masses of Rat MRP8 and 14 Isolated from Spleens from Lewis and Fischer Rats Protein

Predicted mass

ESI mass

Calc massa

RatMRP8Lewis RatMRP14Lewis RatMRP8Fischer RatMRP14Fischer

10,210.3 13,214.1

10,149 13,070 10,150 13,069

10,149.4 13,069.9 10,149.4 13,069.9

b b

Note. The theoretical masses determined from the cDNA sequences of the proteins from Lewis rats are also shown together with corrected theoretical masses. a Mass calculated after incorporation of the posttranslational modifications and sequencing errors (see text for details). b The protein sequences derived from the cDNA sequences of Fischer and Sprague–Dawley rats have not been determined.

RP-HPLC with approximately equal quantities of MRP8 and 14 in each. Elution characteristics were similar to those of CP10 and murine MRP14 isolated from activated spleen cell supernatants (23) and elution times of the rat proteins from C4 RP-HPLC were within 30 s of the murine homologues. The masses determined from the predicted protein sequence of Lewis rat cDNA and those obtained from electrospray ionization mass spectra of the proteins from the other two rat strains are compared in Table 1 and Fig. 2. The ESI masses of each were the same, confirming translation from identical mRNA sequences and indicating that allelic variations would not contribute to the mass differences. The theoretical mass of MRP8 is 10,210.5 Da, whereas the ESI mass of each sample was 10,149 6 2 Da (62 Da difference) and the theoretical mass of MRP14 is 13,214.1 Da, 145 Da greater than the experimental mass (13,069 6 2 Da). Masses failed to correspond to predicted masses even after allowing for common posttranslational modifications. If, as previously suggested (25), the proteins were N-terminally acetylated (together with removal of the initiator Met), their corrected masses would be 10,121.5 and 13,125.0 Da, (i.e., a difference of 128 Da for MRP8 and 256 Da for MRP14) suggesting additional posttranslational modifications or errors in the derived amino acid sequences. Endoprotease digests were performed to locate and identify the discrepancies. Modifications of MRP8. Endoprotease Asp N and Arg C digests were required to obtain peptides spanning the entire sequence of MRP8. Masses of digest peptides isolated by C18 RP-HPLC were determined by ESI/MS and each was assigned a probable sequence by comparison with the calculated masses of peptides obtained by theoretical digestion of MRP8 (Tables 2A and 2B). Sequences for two peptides in the Asp N digest could not be assigned. The peptide which eluted at

21.65 min (mass 3469 Da) was identified as the Nterminal cleavage product as the ESI mass was identical to that of the theoretical peptide after modification by removal of the initiator Met and N-terminal acetylation (Table 2A). The same peptide (R2) was isolated after Arg C digestion (Table 2B) confirming that it corresponded to the N-terminal, posttranslationally modified fragment. In contrast, human MRP8 is not acetylated and retains the translation-initiating N-terminal Met, whereas CP10 is not acetylated and the N-terminal Met is removed (20). The reason for these differences is unclear, but may reflect the relative susceptibilities of the N-terminal amino acids in these proteins (Fig. 3A) to undergo N-terminal modifications (30). Additional modifications were indicated because the theoretical mass of MRP8 was still some 28 Da less than the calculated mass. The peptide eluting at 23.1 min (Table 2A) was 28 Da greater than the Asp N product MRP8(62– 83) and the amino acid sequence, determined by Edman sequencing, was DNAINFEEFLVLVIRVGVAAHK. Peptide R4 (MRP849 –76) from the Arg C digest was also 28 Da more than expected (Table 2B), indicating that the anomaly occurred between residues 62 and 76. A single difference between the cDNAderived protein sequence and that obtained by Edman degradation occurred at position 72 (residue 11 in peptide D3) and indicated translation of Val instead of a predicted Ala. This substitution would account for the additional 28 Da observed in D3, R4, and MRP8. Analysis of the cDNA indicated that the codon for Ala was GCG. The closest Val codon would be GTG, suggesting a sequencing error incorporating C instead of T. The calculated mass of MRP8, after incorporating all modifications, is 10,149.4 Da, in close agreement with the experimental mass (10149 Da). Alignment of the amino acid sequences of CP10 (mouse MRP8), rat, and human MRP8 (Fig. 3A) shows that CP10 and rat MRP8 share 80% identity and 91% similarity,3 whereas hMRP8 is 63% identical and 82% similar to rat MRP8. Greatest homology is apparent in the highly conserved Ca21 binding domains. The ‘‘hinge’’ region of the murine and rat proteins is also highly conserved. Functional comparisons of the chemotactic capacity of the murine and human hinge peptides (amino acids 42–55) indicate that only murine CP10 has this property (22); functional characterization of the rat protein and hinge region awaits further analysis. Modifications of MRP14. Chymotryptic and endoprotease Asp N digests of MRP14 were performed to obtain peptides covering the entire sequence, except for 3

Sequences were compared using the Gap program contained within the Genetics Computer Group Package, Version 8.1 (Madison, WI).

CHARACTERIZATION OF RAT MYELOID S100 PROTEINS

289

FIG. 2. Electrospray ionization mass spectra of MRP8 (A) and MRP14 (B) from spleen from Sprague–Dawley rats isolated by C4 RP-HPLC (deconvoluted spectra are shown as insets).

MRP14(65–70). This region was analyzed by Glu C digestion and a peptide corresponding to MRP14(57–77) had an identical mass (2523 Da) to the theoretical mass (not shown). Digest peptides isolated by C18 RPHPLC were assigned probable sequences after comparison with theoretical masses (Tables 3A and 3B). Sequences for three chymotrypsin digest products (peptides C3, C6, and C7) could not be assigned. Pep-

tides eluting at 21.55 min (mass 2076 Da) and 22.19 min (mass 2505 Da) were identified as posttranslationally modified N-terminal cleavage products; theoretical masses of both were identical to the ESI mass after removal of the initiator Met and N-terminal acetylation (Table 3A). The two peptides were formed due to incomplete chymotryptic digestion; the relative proportion of C3 increased with prolonged incubation (not

290

RAFTERY AND GECZY TABLE 2

Electrospray Ionization Mass Spectrometric Analysis of Peptides Formed after AspN (A) and ArgC (B) Digestion of Rat MRP8 HPLC peak No.

Retention time (min)

Peptide structure

Predicted residue Nos.

Calc mass (Da)

ESI mass (Da)

D1 D2 D3

18.2 21.65 23.1

DDFRKMVTTECPQFVQNKNTESLFKEL (N-Ac)ATELEKALSNVIEVYHNYSGIKGNHHALYR DNAINFEEFLVLVIRVGVAAHKb

31–57 1–30 62–83

3248.7 3469.9a 2454.7a

3247 3469 2453

R1 R2 R3 R4

9.68 21.6 22.65 23.6

VGVAAHKDSHKE (N-Ac)ATELEKALSNVIEVYHNYSGIKGNHHALYR NTESLFK NTESLFKELDVNSDNAINFEEFLVLVIR

77–88 1–30 49–55 49–76

1277.4 3469.9a 837.9 3269.7a

1277 3470 839 3267

A

B

Note. C18 HPLC retention times, theoretical masses, and probable sequences are also shown. a Masses were calculated after incorporating posttranslational modifications and sequencing errors. b This peptide was N-terminally sequenced; Val was identified at position 11 (see text for details).

shown). The experimental mass of Asp N digest peptide D4 (Table 3B) confirmed the modifications. The recalculated theoretical mass is thus reduced to 13,125.0 Da, 56 Da more than the experimentally determined value, indicating additional modifications. A peptide eluting at 11.7 min was 56 Da less than the theoretical chymotryptic product 89 –112. Because this

FIG. 3. Alignment of the amino acid sequences of (A) CP10 (mouse MRP8), rat and human MRP8; and (B) mouse, rat, and human MRP14. The C-terminal domains of the rat and mouse proteins are highly conserved. For optimum alignment gaps were introduced into mouse and rat MRP14 at residue 96 and residue 4 in human MRP14.

contained two Cys residues not normally detected by Edman sequencing unless derivatized, MRP14 was reduced and alkylated and this increased its mass by 115 Da (13,184 Da), indicating alkylation of two of the possible three Cys residues. After chymotryptic digestion all peptides eluted from C18 RP-HPLC at the same retention times as those of the underivatized protein digest, except for C3 which decreased to 11.02 min. The mass of C3 was 2775 Da (113 Da greater than the underivatized peptide) indicating alkylation of both C-terminal Cys residues. The carboxyamidomethylated peptide was N-terminally sequenced and a confidant prediction of amino acids made through to cycle 23. The sequence was identical to that predicted by the cDNA except for amino acids 17 and 18 (residues 105 and 106 in MRP14). The cDNA predicted Arg at position 17, whereas Edman sequencing indicated a translated Ser. The mRNA codon predicting Arg was AGG, whereas the closest codons translating Ser are AGT or AGC, indicating that a sequencing error may have occurred by replacement of T or C with G. A Ser at this position would reduce the mass of MRP14 by 69 Da, still 14 Da less than the experimental mass (Table 1) suggesting an additional modification. The cDNA sequence predicted His as the residue after Ser (His106), whereas Edman sequencing indicated Ala. Incorporation of Ala was excluded because the predicted mass of MRP14 would be 12,989.8 Da. Our previous report demonstrated an unusual His modification at position 106 of mouse MRP14 and the PTH derivative (PTH-1 methyl His) elutes near PTH-Ala after Edman degradation (23). The mass of rat MRP14 incorporating methyl His at position 106 would include an additional 14 Da (i.e., a total mass of 13,070 Da), suggesting that rat MRP14 contains a similarly modified His106. The

291

CHARACTERIZATION OF RAT MYELOID S100 PROTEINS TABLE 3

Electrospray Ionization Mass Spectrometric Analysis of Peptides Formed after Chymotryptic (A) and Asp N (B) Digestion of Rat MRP14 HPLC peak No.

Retention time (min)

Peptide structure

Predicted residue Nos.

Calc mass (Da)

ESI mass (Da)

A C3 C4 C5 C6 C7

11.7 13.77 17.05 21.55 22.19

ACHEKLHENNPRGHDHS(HMe)GKGCGKb GHPDTLNKAEF EECMMLMGKLIF (N-Ac)AAKTGSQLERSISTIINVF (N-Ac)AAKTGSQLERSISTIINVFHQY

89–112 27–37 77–88 1–19 1–22

2662.9a 1228.3 1444.9 2077.4a 2505.8a

2662b 1228 1446 2076 2505

D1 D2 D3 D4 D4

11.77 13.77 16.25 19.02 21.85

DIME DTLNKAEFKEMVNK DLPNFLKREKRNENLLR DNQLSFEECMMLMGKLIFACHEKLHENNPRGH (N-Ac)AAKTGSQLERSISTIINVFHQYSRKYGHP

61–64 30–43 44–60 71–102 1–29

506.6 1666.9 2155.5 3773.4 3331.8a

507 1668 2155 3771 3332

B

Note. C18 HPLC retention times, theoretical masses, and probable sequences are also shown. a Masses were calculated after incorporating identified posttranslational modifications and sequencing errors. b After reduction with DTT and alkylation with iodoacetamide, the mass of peak C3 increased to 2775 Da (masscalc 2777.0 Da). N-terminal amino acid sequencing indicated Ser at position 17 and methylated His at position 18 (see text for details).

calculated mass of rat MRP14, incorporating all modifications is 13,069.9 Da, in close agreement with the experimental mass (13,069 Da). Alignment of the amino acid sequences of rat and mouse MRP14 indicates 78% identity and 89% similarity (Fig. 3B) with striking differences apparent in the N- and C-terminal domains The N-terminal sequences (residues 1 to 60) are 64% identical (80% similar), whereas the C-terminal sequences (residues 61 to 112) are almost identical (96%) with conserved substitutions of Ala (mouse) for Gly (rat) at position 74 and Gly (mouse) for Asp (rat) at position 103 (Fig. 3B). Alignment of the sequences of rat and human MRP14 shows 63% identity and 76% similarity, with no particular predilection for the N-terminal or C-terminal domains (Fig. 3B). Thus, the C-terminal domains of mouse and rat MRP14 may have diverged more slowly than the N-terminal domain, suggesting that these residues may be functionally more significant than the N-terminal region.

CONCLUSION

MRP8 and 14 from the cytosol of normal spleen cells from three rat strains were isolated in high yield. Although the results confirm constitutive expression, the cell type expressing these proteins was not identified but is most likely to be the neutrophil. MRP8 and 14 mRNA expression may increase in macrophages in response to an appropriate proinflammatory stimulus (24), although the mediators involved in their regula-

tion, such as those described by us for the induction of CP10 in macrophages (31), and associations between mRNA expression and protein translation require more detailed analysis. There is evidence that the proteins need not necessarily be coordinately expressed (26, 31) and, unlike S100b (32), little is known concerning effects of homo- or heterodimer formation on function. The proteins were posttranslationally modified and the cDNA-derived protein sequences of both contained errors. Modifications were readily identified by ESI/MS because of its ability to rapidly determine the precise mass of proteins and peptides. The calculated and theoretical masses of rat MRP8 and 14 were identical after incorporating the modifications. The function of the preserved N-methylation of His106 in rat MRP14 is unclear, but it may play a role in MRP14’s ability to bind Zn21 with high affinity (23). This may explain the observed differences in the antimicrobial capacities of murine and human MRP14 (8, 33). The latter is active, is zinc-dependent, and contains no methylated His residues. Knowledge of the precise structure of this group of S100 proteins will provide a solid basis for analysis of domains critical for both intra- and extracellular functions. ACKNOWLEDGMENTS This work was supported in part by grants from the National Health and Medical Research Council of Australia. Members of the Cytokine Research Unit are acknowledged for helpful discussions.

292

RAFTERY AND GECZY

REFERENCES 1. Scha¨fer, B. W., Wicki, R., Engelkamp, D., Mattei, M. G., and Heizmann, C. W. (1995) Genomics 25, 638 – 643. 2. Scha¨fer, B. W., and Heizmann, C. W. (1996) TIBS 21, 134 –140. 3. Zimmer, D. B., Cornwall, E. H., Landar, A., and Song, W. (1995) Brain Res. Bull. 37, 417– 429. 4. Hessian, P. A., Edgeworth, J., and Hogg, N. (1993) J. Leukocyte Biol. 53, 197–204. 5. Kligman, D., and Hilt, D. C. (1988) TIBS 13, 437– 443. 6. Fano, G., Biocca, S., Fulle, S., Mariggio, M. A., and Belia, S. (1995) Prog. Neurobiol. 46, 71– 82. 7. Murao, S., Collart, F. R., and Huberman, E. (1989) J. Biol. Chem. 264, 8356 – 8360. 8. Murthy, A. R. K., Lehrer, R. I., Harwig, S. S. L., and Miyasaki, K. T. (1993) J. Immunol. 151, 6291– 6301. 9. Clohessy, P. A., and Golden, B. E. (1996) J. Leukocyte Biol. 60, 674. 10. Lagasse, E., and Weissman, I. L. (1992) Blood 79, 1907–1915. 11. Geczy, C. L. (1996) Biochim. Biophys. Acta 1313, 246 –253. 12. Komada, T., Araki, R., Nakatani, K., Yada, I., Naka, M., and Tanaka, T. (1996) Biochem. Biophys. Res. Commun. 220, 871– 874. 13. Tan, J. Q., Vorum, H., Larsen, C. G., Madsen, P., Rasmussen, H. H., Gesser, B., Etzerodt, M., Honore, B., Celis, J. E., and Thestruppedersen, K. (1996) J. Invest. Dermatol. 107, 5–10. 14. Yui, S., Mikami, M., and Yamazaki, M. (1995) J. Leukocyte Biol. 58, 650 – 658. 15. Lemarchand, P., Vaglio, M., Mauel, J., and Markert, M. (1992) J. Biol. Chem. 267, 1937–19382. 16. van den Bos, C., Roth, J., Koch, H., Hartmann, M., and Sorg, C. (1996) J. Immunol. 156, 1247–1254. 17. Hessian, P. A., Wilkinson, L., and Hogg, N. (1995) FEBS Lett. 371, 271–275. 18. Odink, K., Cerletti, N., Bru¨ggen, J., Clerc, R. G., Tarcsay, L., Zwadlo, G., Gerhards, G., Schlegel, R., and Sorg, C. (1987) Nature 330, 80 – 82.

19. Teigelkamp, S., Bhardwaj, R. S., Roth, J., Meinardus Hager, G., Karas, M., and Sorg, C. (1991) J. Biol. Chem. 266, 13462– 13467. 20. Raftery, M. J., and Geczy, C. L. (1996) in Proceedings of the 44th ASMS Conference on Mass Spectrometry and Allied Topics, pp. 1421, Portland, OR. 21. Lackmann, M., Cornish, C. J., Simpson, R. J., Moritz, R. L., and Geczy, C. L. (1992) J. Biol. Chem. 267, 7499 –7504. 22. Lackmann, M., Rajasekariah, P., Iismaa, S. E., Cornish, C. J., Simpson, R. J., Reid, G., Moritz, R. L., and Geczy, C. L. (1992) J. Immunol. 150, 2981–2991. 23. Raftery, M. J., Harrison, C. A., Alewood, P., Jones, A., and Geczy, C. L. (1996) Biochem. J. 316, 285–293. 24. Imamichi, T., Uchida, I., Wahl, S. M., and McCartney Francis, N. (1993) Biochem. Biophys. Res. Commun. 194, 819 – 825. 25. Yui, S., Mikami, M., and Yamazaki, M. (1995) J. Leukocyte Biol. 58, 307–316. 26. McGary, C. T., Yu-Ching, E. P., Hanspeter, M., Guntrum, W. D., Neri, A., and Weltch, D. R. (1997) Anticancer Res. 17, 1– 6. 27. Ashton, D. S., Beddell, C. R., Green, B. N., and Oliver, R. W. A. (1994) FEBS Lett. 342, 1– 6. 28. Scha¨gger, H., and von Jagow, G. (1987) Anal. Biochem. 166, 368 –379. 29. Iismaa, S. E., Hu, S., Kocher, M., Lackmann, M., Harrison, C. A., Thliveris, S., and Geczy, C. L. (1994) DNA Cell Biol. 13, 183–192. 30. Bachmair, A., Finley, D., and Varshavsky, A. (1986) Science 234, 179 –186. 31. Hu, S. P., Harrison, C., Xu, K., Cornish, C. J., and Geczy, C. L. (1996) Blood 87, 3919 –3928. 32. Barger, S. W., Wolchok, S. R., and Van Eldik, L. J. (1992) Biochim. Biophys. Acta. 1160, 105–112. 33. Kocher, M., Kenny, P. A., Farram, E., Majid, K., Finlay-Jones, J. J., and Geczy, C. L. (1996) Infect. Immunol. 64, 1342–1350.