J Pepride R ~ J . 49. 1997. S82 58.5 Printed in UK ~ all rrghtc reserved
Copyright 0 Munksgaard 1997 JOURNAL OF PEPTIDE RESEARCH
ISSN 1397-002X
Complex formation between a formyl peptide and 24p3 protein with
a blocked N-terminus of pyroglutamate
SIN-TAK CHU
',
HAN-JIA LIN and YEE-HSIUNG CHEN' J
'
Institute of Biological Chemistry, Academia Sinica and'
Institute of Biochemical Science, College of Science, National Taiwan Universirj; Taipei, Taiwan, Republic of CliinaReceived 17 October, revised 7 December 1996, accepted for publication 9 February 1997
We have purified 24p3 protein from mouse uterine fluid (Biocheni. J. 316, 545-550, 1996). It is a 25.8-kDa glycoprotein with a N-blocked terminus. This work demonstrated the N-blocked residue to be pyrogluta- mate, supporting the post-translational cleavage site at Ala-Gln in the precursor protein to generate a putative protein of 180 amino acid residues. Consequently, the two cysteines, Cys" and Cys'", and the two tryptophans, Trp3' and Trp", are assigned along the polypeptide chain. No free thiol group was detected in the protein. The presence of formyl-Met-Leu-Phe in the protein solution causes a considerable decrease in the protein fluorescence due to Trp3' and Trp" . Analysis of the fluorescence data supports the idea that the protein can be complexed with the formyl peptide. The association constant for the complex formation is (4.8k0.29) x lo5 M - ' at pH 7.4. 0 Munksgaard 1997.
Key words; fluorescence; formyl peptide; 24p3 protein: pyroglutamate; pyroglutamate amino peptidase
Rodents have been used as important experimental animals for studying the reproductive biology of mammals. The accumulation of uterine luminal fluid ( U L F ) in the proestrus phase of rodent reproductive circle is well known. The phenomenon also occurs in the immature female rodent stimulated by estrogen or its analogues. Thus, establishing the structure and function of protein components in ULF becomes important in order to understand their roles in the growth and development of the reproductive tract (s) concerned. In this regard, we have demonstrated recently a glycoprotein derived from 24p3 mRNA in mouse ULF (1). 24p3 cDNA was originally cloned from SV40-infected mouse kidney primary culture cells (2). The protein derived from this gene (hereafter referred to as 24p3 protein) was found also in lipopolysaccharide-stimulated mouse PU5.1.8 macrophage cells ( 3 ) and bFGF- stimulated 3T3 cells (4). The results of Liu and Nilson-Hamilton reveal 24p3 protein to be an acute phase protein (5). Its structure and function have not yet been reported. Here we report the complex formation of formyl-Met-Leu-Phe and 24p3 protein that was demonstrated to have a blocked N-terminus of pyroglutamate.
EXPERIMENTAL PROCEDURES
Materials. Outbred ICR mice were purchased from Charles River Laboratory (Wilmington, MA) and
582
were bred in the animal center at the College of Medicine, National Taiwan University. Animals were treated following institutional guidelines for the care and use of experimental animals. Diethylstilbesterol (DES), 5,5'-dithiobis( 2-nitrobenzoic acid) (Ellmen's regent) and formyl-Met-Leu-Phe (formyl peptide) were from Sigma (St. Louis, MO, USA). Calf-liver pyroglutamate aminopeptidase (E.C.3.4.19.3) was from Boehringer Mannheim G.m.b.H. (Germany).
Preparation of 24p3 protein. Mice were housed under
controlled lighting (14 h light, 10 h dark) at 21-22 "C and provided with water and National Institutes of Health 31 laboratory mouse chow ad libitum. DES dissolved in corn oil was administrated subcutane- ously to female mice (3 wk old) with a daily dose of 100 ng/g body weight for three consecutive days. Subsequently, the animals were killed and ULF was collected. 24p3 Protein was purified from mouse ULF according to our previous procedure ( 1 ).
Protein analysis. 24p3 Protein was digested with
pyroglutamate aminopeptidase by a modified method of Henze et al. (6). After the protein (20pg) in 100 pL of sodium phosphate (pH 7.0) containing 0.1 mM EDTA was digested with the enzyme (2 pg) at 4 T for 18 h, the solution was incubated further at 25 'C for
4
h. The solution was subjected to HPLCComplex between formy peptide and 24p3 protein
NGAL QDSTSDLIPAPPLSKVPLQQNFQDN
I I I I I I I I I I I I I I I 2 4 ~ 3 MALSVMCLGLALLGVLQSQAQDSTQNLIPAPSLLTVPLQPDFRSD
*DSTQNLIPAPSLLTVPLQ
(from the Wroglut-te-armno-peptldase
-digested protein sanple)
-10 +1 10 2 0 NGAL Q F Q G K W Y W G L A G N A I L R E D K D P Q K M Y A T I Y D L K E D K S Y N V 24133 QFRGRWYWGLAGNAVQKKTEGSF?MYSTIYELQEN”VSIL 30 40 50 60 7 0 NGAL FRKK--KCDYWIRTWPGCQPEFTLGNIKSYPGLTSYLVRT 2 4 ~ 3 VRDQDQGCRYWIRTWPSSRQFTLGNMHSYPQVQSY~WQVATT I I I I I I I I I I I I I I1 I l l I I I I I I I I I I I I I I I I I I I l l l l l l I l l I l l 1 I 80 90 1 0 0 110 NGAL NYNQHAMVFFKKVSQNREYFKITLYGRTKELTSELKENFIRFSKS 24p3 DYNQFAMVFFFXTSENKQYFKITLYGRTKELSPELKERFTS I l l I I I I I I I I I I I I I I I I I I I I I I l l 1 I I1 I1 120 13 0 140 150 160 NGAL LGLPFNiIVFPVPIEQCID3 24p3 LGLKDDNIIFSVPPDQCIDN I l l I I I I I I I I I 170 180 FIGURE 1
Alignment of amino acid sequence of 24p3 protein and human NGAL. 24p3 protein was digested with pyroglutamate amino peptidase. The partial sequence derived directly from the enzyme- digested protein sample by Edman degradation was compared with 24p3 cDNA-deduced sequence of precursor protein. The cleavage point for the generation of mature protein is indicated by a star, *. The amino acid sequences of 24p3 protein were aligned with that of human NGAL and the identical residues are indicated by vertical bars.
on a c18column and the fraction corresponding to
24133 protein was collected. Amino acid sequences were determined by automated Edman degradation with a gas-phase microsequencer [ 477A protein sequencer with on-line 120A analyser (Applied Biosystems, Foster City, CA)].
Measurement offltlorescence spectra. The fluorescence intensity, expressed in arbitrary units, was measured at room temperature with a Hitachi F-4010 fluores- cence spectrophotometer. Both the excitation and the emission slit width were 10 nm. Raman emission due to the scattering of solvent was minimized by adjusting the intensity scale. The protein solution was freshly prepared. It took no more than 5 min to scan a spectrum to prevent the protein precipitation.
Fi:
represented the fluorescence intensity at wavelength 2, (nm) when the fluorophore was excited at wave- length A1 (nm).Analysis offluorescence data. The modified Scatchard plot (7) was constructed to analyze the fluorescence data of a complex formed by a formyl peptide and 24p3 protein:
IAFI/[Llf~,,=(K,F,)-KL
I
AFI (1) where A F is the change in protein fluorescence on adding formyl peptide, L, and Fm the protein fluores- cence in the absence of ligand. KL is the association constant of the complex concerned. Throughout the titration,1
AFI/
[LItotal was plotted againstI
AFI, since [Llfr,, was close to [Lltotal.RESULTS AND DISCUSSION
The blocked N-terminus of 24p3 protein
There are three cysteines and two tryptophans in the precursor protein according to the 24p3 cDNA- deduced 200 amino acid sequence (Fig. 1 ). The post- translational cleavage site cannot be assigned directly from Edman degradation of 24p3 protein that has a blocked N-terminus (1). We digested the homogen- eous protein as characterized by SDS/PAGE (1) with pyroglutamate amino peptidase. Automated Edman degradation of the digested protein sample for 18 cycles gave reliable amino acid sequence D-S-T-Q-N- L-I-P-A-P-S-L-L-T-V-P-L-Q, which is identical with the corresponding peptide sequence deduced from 24p3 cDNA in all positions (Fig. 1). Apparently, the blocked N-terminus is pyroglutamate and the post- translational cleavage should take place at Ala-Gln in the precursor protein. We analyzed the precursor protein sequence by the methods of McGeoch (8) and von Heijne (9) and concluded the same post- translational cleavage site. The results enable us to assign the positions of the two cysteines, Cys” and C Y S ’ ~ ~ , and the two tryptophans, Trp3’ and Trpsl, along the polypeptide chain (Fig. 1). Removal of the
signal peptide, which contains one cysteine, from the precursor protein gives a putative protein comprising 180 amino acid residues that sum up to have a molecular mass of 20839 Da. Therefore, the core protein together with the glycoconjugate (1 ) should give M , ~ 2 3 . 5 kDa for the glycoprotein. Previously, a 23-kDa band of the deglycosylated 24p3 protein on SDS/PAGE (1) seemed to overestimate the molecu- lar size.
We found that 24p3 protein was not reactive to Ellman’s reagent, indicative of the absence of a free thiol group on the protein surface. The protein also did not react with Ellman’s reagent when the experi- ment was performed in the presence of 6.0 M urea. Apparently, there are no free cysteines that are par- tially or fully buried in the protein molecule, sup- porting the formation of a disulfide bond between 24p3 protein shows a high degree of similarity to human neutrophil gelatinase-associated lipocalin (NGAL) (10): the two proteins have 70% identity (Fig. 1 ). Human NGAL is crosslinked with gelatinase by a disulfide bridge ( 11). This protein has three cysteines, namely C Y S ~ ~ , CysS7 and C Y S ” ~ (Fig. 1). The sequence aligment of human NGAL and 24p3 protein suggests a disulfide bond formed by Cys76 and CYS’~’ in human NGAL. Thus, CysS7 has a free
583
S-T. Chu et al.
thiol group that is reactive to associate covalently with gelatinase. However, the situation would not happen to 24p3 protein, since it is devoid of free thiol group.
The interaction of 24p3 protein and formy1 peptide
Figure 2 displays the emission spectra of 24p3 protein under several conditions. Excitation was at 295 nm to ensure the fluorescence due to Trp31 and Trp*l. The native protein in phosphate-buffered saline at pH 7.4 (Fig. 2, curve I ) exhibits a peak at 332 nm. The peak shifts to 350 nm, and the intensity increases greatly when the protein is in 6.0 M guanidinium
hydrochloride (Fig. 2, curve 11). Futher reduction of the unfolded protein with 1 ,Cdithiothreitol shifts the peak to 352 nrn (Fig. 2, curve 111). Apparently, the two tryptophan residues are restricted into a config- uration that differs from that of free tryptophan in aqueous solution.
Human NGAL is a glycoprotein with the binding capacity for formyl-Met-Leu-Phe (12). It raises the question of whether 24p3 protein can be complexed with the formyl peptide. Figure 3 gives the fluores- cence emission spectrum of 24p3 protein at a concentration of 0.8 PM in the presence of 80 mM
formyl peptide. Relative to the protein fluorescence, the formyl peptide itself shows very weak fluores- cence intensity. The formyl peptide in the protein solution does not shift the emission peak of protein but effects a considerable decrease in the protein intrinsic fluorescence. Fitting the data of f';;:
obtained from adding formyl peptide to the protein solution with eqn. (1) shows a linear curve in the
900 I I I I I
2
600J
C $ 300- v) OJ 0 3 LI
I I I I I II
0 300 320 3LO 360 380 400 Wavelength,nm FIGURE 2Fluorescence emission spectra of 24p3 protein. The emission spectra were scanned with excitation wavelength at 295 nm. The protein was at 1.0 PM in phosphate-buffered saline at pH 7.4 (curve I ) or 6.0 M guanidinium hydrochloride in the absence (curve 11) and in the presence of 1 0 m ~ 1.4-dithiothreitol (curve 111). 584
v
1
,
,L-,,-. .-.-
\ LL 0 300 350 400Wavelength
, n m FIGURE 3Effect of formyl-Met-Leu-Phe on the fluorescence of 24133 protein in phosphate-buffered saline at pH 7.4. The emission spectra were scanned with excitation wavelength at 295 nm. The protein was at 0.8 PM alone ( - - - ) and in the presence of 80 mM formyl peptide
(...) . The fluorescence of 80mM formyl peptide alone was very
weak (-.- ). The modified Scatchard plot for the binding of formyl peptide to 24p3 protein is given in the inset. The data obtained from adding formyl peptide to the protein solution were analyzed by using eqn. ( 1 ) with linear-regression fitting. The correlation coefficient was calculated to be more than 0.96.
modified Scatchard plot (Fig. 3, inset), manifesting that the protein has one singular type of formyl peptide-binding site. The association constant for the affinity site was estimated to be (4.8f0.29) x lo5 M - ' , which is comparable with
that of the complex formed by formyl peptide and human NGAL ( 12).
ACKNOWLEDGEMENT
This work was supported in part from National Science Council, Taipei, Taiwan, Republic of China (grants NSC85-2311-B001-038 and NSC85-23 1 1-B001-064).
REFERENCES
1 . Chu, S.T., Huang, H.L., Chen, J.M. & Chen. Y.H. (1996)
Biochrm. J. 316, 545-550
2. Hraba-Renevey, S., Turler, H., Kress, M., Salomon, C. & Weil, R. (1989) Oncogene 4, 601-608
3 . Meheus, L.A., Fransen, L.M., Raymackers, J.G., Blockx, H.A., van Beeumen, J.J., van Bun, S.M. & van de Voorde, A. (1993) J. Immunol. 151, 1535-1547
4. Davis, T.R., Tabatabai, L., Bruns, K., Hamilton, R.T. & Nilsen-Hamilton, M. ( 1991 ) Biochem. Biophys. Actu 1095,
5. Liu, Q. & Nilsen-Hamilton, M. (1995) J. Bid Chem. 270,
6. Henze, W.J., Rodriguez, H., Singer, A.G., Stults, J.T., Macrides, F., Agosta, W.C. & Niall, H. (1988) J. Biol. Chem.
145-152
22565-22570
Complex between formy peptide and 24p3 protein
7. Epstein, M., Levitzki, A. & Renben, J. (1974) Biochemistry Address:
Dr, Yee-Hsiung Chen
Institute of Biological Chemistry Academia Sinica Taipei Taiwan Republic of China Fax: (02)-363-5038 13, 1777-1782
8. McGeoch, D.J. (1985) Virus Res. 3, 271-275 9. von Heijne, G. (1986) Nucl. Acid Res. 14, 4683-4690
10. Kjeldsen, L., Johnsen, A.H., Sengelerv, H. & Borregaard, N. 11. Triebel, S., Blaser, J., Reinke, H. & Tschesche, H. (1992) 12. Allen, R.A., Erickson, R.W. & Jesaitis, A.J. (1989) Biochem.
23-106 (1993) J. B i d . Chem. 268, 10425-10432
FEES Lett. 314, 386-388
Biophys. Acta 991, 123-133
