A synthetic nonapeptide corresponding to the NH2-terminal sequence ...

2 downloads 58 Views 491KB Size Report
a chain of human iC3b by limited proteolysis with plasma kallikrein was shown to exhibit several biolog- ical functions. This C3-derived cleavage product, C3d-.
THEJOURNAL os BIOLOGICAL CHEMISTRY Vol. 260. No. 5. Issue of Mareh 10, pp. 2597-2600,1985 0 1985 by The American Society of Biologxd Chemists Inc.

Communication

Printed in

UTSA.

Fragments from various complement components are thought to exert immunoregulatory effects and those derived from the C3 molecule are perhaps the best characterized in chemical terms (1-4).Recently, a novel polypeptide generated from iC3b by limited enzymatic cleavage with human plasma kallikrein has been isolated and characterized (5).This fragment, (Received for publication, September 27,1984) C3d-K, was shown to suppress mitogen- and antigen-induced Paul D. Hoeprich, Jr.@, Clemens A. DahindenS, proliferation of cultured human T-lymphocytes and to cause Peter J. LachmannW, Alvin E. Davis, 11111, and leukocytosis when injected intravenously into rabbits. LeuTony E. HugliS kocytosis activity, ie. an increase in the number of leukocytes From the $Department of Immunology, Scripps Clinic and in peripheral blood, has been previously associated with two Research Foundation, La JoUq California 92037, the C3-derivedfragments: C3e (1)and leukocyte mobilizing factor (IUniversity of Cambridge, Cambridge,England, and (6). Interestingly, these two factors share several character11 Harvard Medical School, Boston, Massachusetts 02115 istics. They are both anionic fragments of similar size, M, Numerous biologically active fragments have been 10,000,and thought to arise in the later stages of C3b breakdescribed that are derived from the C3 molecule. Re- down. It is speculated that C3d-K, a M, 41,000fragment, may cently, a polypeptide ( X 41,000) generated from the contain all or part of these two smaller fragments of C3. a chainofhumaniC3bbylimitedproteolysis with Davis et al. (7) recently reported the characterization of a plasma kallikrein was shown to exhibit several biolog- M,40,000fragment called C3d,g. This polypeptide, generated ical functions. This C3-derived cleavage product, C3dK, suppresses mitogen- and antigen-induced prolifer- by Factor I inactivation of fluid phase Q,is structurallyquite ation of human T-lymphocytes and induces leukocyto-similar to C3d-K. The NH2-terminal sequence of fragment C3d,g overlaps with C3d-K beginning at residue 10.Therefore sis in rabbits. We have identified and synthesized a portion of C3d-K that is associated with the leukocy- C3d-K contains an additional nine amino acids (Fig. 1).On a tosis phenomenon. A nonapeptide corresponding to the functional level, the smaller C3d,g is capable of suppressing amino-terminal nine residues of C3d-K was synthe- mitogen-induced T-cell proliferation’ but unable to induce sized using conventional Merrifield solid-phase peptide leukocytosis in rabbits (5). It was therefore concluded that chemistry; the structure of this peptide is Thr-Leu- the site responsible for inducing leukocytosis resides in the Asp-Pro-Glu-Arg-Leu-Gly-Arg (TLDPERLGR).At a NH2-terminal nonapeptide portion of the C3d-K molecule final concentration of4 X 10“ M, both the nonapeptide and likewise represents the active center of C3 fragments and the des-Arg octapeptide (TLDPERLG) were ca- C3d-K, and probably C3e and leukocyte mobilizing factor. pable of inducing leukocytosis in rabbits. Additionally, both peptides enhance vascular permeability when in- This communication describes a synthesis of jected in guinea pig skin. These activities are similar TLDPERLGR, a nonapeptide that represents the NHz-terto those previously attributedto a C3 fragment iden- minal sequence of C3d-K, and provides evidence of its ability to induce leukocytosis in rabbits. Additionally, the nonapeptified as C3ebyGhebrehiwetandMuller-Eberhard (Ghebrehiwet, B., and Muller-Eberhard,H. J. (1979) tide was tested for its ability to enhance vascular permeability J. Immunol. 123,616-621). We concludethatthe because factor C3e reportedly expresses vasoactivity. nonapeptide TLDPERLGR represents the active center of the C3-derived leukocytosis factors C3e and C3dEXPERIMENTAL PROCEDURES K. This active synthetic analogueofC3d-Kshould prove valuable in elucidating the mechanism of actionSolid-phose Peptide Synthesis-Reagents used in the synthesis for complement-dependent leukocyte mobilization i n were purchased from the indicated vendors: Boc-amino acids (Vega

A Synthetic Nonapeptide Corresponding to the NH2terminal Sequence of C3d-K Causes Leukocytosis in Rabbits*

vivo.

Several biologic activities have been associated with physiologic fragments derived from complement (C’) proteins.

* This work was supported in part by Public Health Service Training Grant HL07195-08 (toP. D. H.)and Research Grants HL16411, AI 17354, and HL 25658 (to T. E. H.). This is publication number 3590 IMM from the Department of Immunology, Scripps Clinic and Research Foundation, La Jolla, CA. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solelyto indicate this fact. 5 To whom correspondence should be addressed. The abbreviations used are: C, complement; PMN, polymorphonuclear leukocytes; Boc, t-butoxycarbonyl. Accepted symbols for complement components used are recommended by the World Health Organization ((1981) Eur. J. ImmunoL 11.668-669). Those not listed are: C3d-K, M. 41,000 fragment derived from limited kallikrein digestion of iC3b of a chain origin; C3c-K, M,127,000fragment from iC3b missing C3d-K, C3d,g, M, 40,000 fragment derived from proteolysis (Factor I) of iC3b, same as a-2D.

and Bachem), diisopropylcarbodiimide (Aldrich) and [“Clglycine (ICN). The tert-butoxycarbonyl (Boc) derivative of [“C]glycine was prepared as described by Itoh etal. (8).The nonapeptide corresponding to residues 1-9 of C3d-K was synthesized according to established Merrifield solid-phase synthetic procedures (9, 10). The first amino acid, Boc-N8-tosyl-L-arginine, was esterified to chloromethylatedpolystyrene resin (Bio-Rad, 1%cross-linked, 1.37 meq/g) as previously described (11).Subsequent amino acids were double coupled, initially through a diisopropylcarbodiimide-mediatedreaction; and secondly, utilizing a preformed l-hydroxybenzotriazole ester in dimethyl formamide. The Boc group was removedby treating the protected peptide resin for 20 min with 50% trifluoroacetic acid in dichloromethane followed by a 5-min neutralization with 5% diisopropylethylamine in dichloromethane. The resin was washed with appropriate solvents before and after each deprotection, neutralization, and coupling step. Boc group removal and completeness of coupling was monitored by the quantitative ninhydrincolor test of Sarin et al. (12). The nonapeptide was cleaved from the resin, and side-chain protecting groups were simultaneously removed byexposure to anhydrous hydrogen fluoride (HF) for 40 min at 0-4 “C in the presence of a 5-fold molar excess of anisole. After removing HF by water aspiration, the resin was washed

2597

M. Hobbs, P. D. Hoeprich, Jr., and A. E. Davis, unpublished data.

2598

FIG. 1. A model of the human C3 molecule showing the currently accepted degradation products and the sites of proteolytic cleavage which give riseto these catabolic fragments. The nonapeptide moiety is indicated by a block box. The various fragment sizes a:e based on the DNA sequence of the entirehuman C3 molecule (16). K , kallikrein; T,trypsin; E, elastase; I, C3b inactivator.

with anhydrousethyl etherto remove anisole. The peptide was Endotoxin Assay-The presence of endotoxin was measured by the washed fromthe resin with 10% aqueous acetic acid and thecombined Limulus test in accordance with the E-Toxate kit instructions (Sigma, washings were lyophilized. Bulletin No. 210). Purification of the nonapeptide was accomplished by reverse-phase high pressure liquid chromatography. Using a semipreparative (0.9 X RESULTS 50-cm) octadecylsilane column (Waters), the peptide eluted as the Peptide Structure-The nonapeptide representing the predominant peak 12 min after injection. Gradient elution was used that ranged from 82% solvent A (0.1% trifluoroacetic acid/water) and amino-terminal region of C3d-K was synthesized as described 18%solvent B (0.1% trifluoroacetic acid/acetonitrile) to 60% A and above. A small portion of purified material was hydrolyzed 40% B over 30min. Prior to in vivo testing, all traces of trifluoroacetic and evaluated by amino acid analysis; the observed values are acid were removed by desalting the high pressure liquid chromatog- in excellent agreement with theoretical (Table I). raphy-purified peptide over a P-10column (1.3 X 120cm) equilibrated Leukocytosis Activity-The response observed in rabbits to in distilled water; the peptide containing fractions were pooled and lyophilized. Anoverall yield of 60% relative to startingBoc-Ng-tosyl- intravenous administration of nonapeptide is mobilization of leukocytes, specifically, an increase in the number of circulatL-arginine resin was realized. Enzymatic Hydrolysis of the Nowpeptide-Incubation of the non- ing polymorphonuclear leukocytes (Fig. 2). The nonapeptide apeptide with carboxypeptidase B produced the des-Arg derivative. causes an increase in the number of circulating PMNs some Approximately 3 mg of nonapeptide was dissolved in 2% NaHC03, 2.5 times normal levels. The magnitude of the response is and 15 pg of carboxypeptidase B-DFP (Worthington) was added. The mixture was incubated for 1h a t 37 "C.The octapeptide was desalted maximal after 4 h and no significant leukopenia occurs in the and separated from free arginine by gel filtration over P-10 (1 X 8 course of the response. In contrast, human C 5 b A , elicits a cm, equilibrated in pyrogen-free water). Peptide containing fractions severe leukopenia within 5 min followedby a rapid and were pooledand lyophilized. somewhat prolonged rise in total circulating PMNs. Injection Tryptic scission of the nonapeptide was accomplished as follows. of human C3a fails to induce a significant change in leukocyte Approximately 3 mg of peptide was dissolved in 1 ml of0.01 M level compared with the effect of normal saline or phosphateNH4HC03at pH8.0 and 0.14 mg of trypsin-~-l-tosylamido-2-phenylethyl chloromethyl ketone (Worthington) was added; the weight buffered saline. These experiments establish that the time ratio of peptide to trypsin was 20/1. Following incubation a t 37 "C course of leukocytosis induced by nonapeptide is maximal 4 h after injection and subsides to normal levels after 8 h. It for 2.5 h, the digestion was terminated by the addition of 100 pg of soybean trypsin inhibitor (Sigma) accompanied by an additional 10- should be noted that the leukocytotic response may be a min incubation at 37 "C. The two peptides resulting from tryptic consequence of a variety of stimuli including infection, stress, cleavage were filtered through an Amicon PM-10 filter. Filtration immune response, exercise, etc. (6). Because of this compliaccomplished a separationof peptide from enzyme-inhibitor complex, since the latterwas retained by the filter; the filtrate was lyophilized. cation, we chose to minimize stress from multiple bleeds by Leukocytosis Assay-The test for leukocytosis was performed in continuing to test the effect of the nonapeptide with sammale New Zealand White rabbits of uniform age and size (2.5-3 kg). plings a t only three time points,i.e. 0,4 and 8 h. Fig. 3 shows In general, the peptide was dissolved in either sterilesaline or sterile the results from multiple experiments using the less traumatic phosphate buffered saline a t a final concentration of0.6mM.An procedure. These data substantiate the preliminary observainitial bleed from an ear arterywas made to establish normal (0 time) tion that thesynthetic nonapeptide from C3d-K is capable of leukocyte levels. The solubilized peptide (1 ml) was then injected intravenously through an ear vein. Control animals received injec- increasing the number of circulating leukocytes by a factor of tions (1ml) of sterile saline or phosphate-buffered saline. Subsequent nearly 2.5-fold at 4 h. The des-Arg form of the peptide bleedings from the ear were made a t specific time points up to 8 h. (TLDPERLG) has a biologic effect identical to that of the In order to facilitate multiple bleeds over short intervals, the ear nonapeptide. Additionally, C ~ Cwhich , contains the nonapepartery was cannulated for the first hour. A ZBI Coulter counter was tide sequence at the carboxyl terminus of the M , 27,000 used to determine total leukocytes. In general, duplicate 2 0 4 aliquots of blood were added to 10 ml of Isoton; red blood cells were lysedand TABLE I the number of leukocytes was measured. Additionally, duplicate blood Amino acid analysis of synthetic nonapeptide smears of each sample were made and stained with Wright's stain. A differential count of 200 cells/slide was obtained in which granuloAmino acid Found" Theoretical cytes (PMN) were tabulated separately from other leukocytes. 2.0 Arg 2 Bioassay for Vascular Permeability-The skin test to measure Asx 1.1 1 enhanced vascular permeability in guinea pig skin followed the proThr 0.9 1 cedure of Cochrane and Muller-Eberhard (13). Briefly, 50-p1 aliquots Glx 1.1 1 of normal saline containing the nonapeptide (20 nmol) or other C3Pro 1.0 1 derived fragments (1 nmol) were injected subcutaneously into the GlY 1.1 1 shaved back of a guinea pig. The animal, 5-10 min previously, had Leu 2.0 2 been given 0.5 ml of a solution of Evans blue dye ( l % ,w/v) by cardiac "Values expressed as moles/mole of peptide and represent an puncture. After 30 min, the animal was killed and skinned and the average of four determinations. area of dye infiltration was measured.

2599

Structure-Function Studiesof C3d-K 300

400-

r

350300c

E

a

z

c

250-

0 0 3

h -

501

2

I

5

dc 200.#

Nonapeptide 4x10" M

150-

4

1

a

2501

100 Nonspeprids 4 ~ 1 0 % f

50

0

1

I

I

I

I

I

1

I

2

3

4

5

6

1

8

4

a

Hours

FIG.2. Time course of leukocytosis in rabbits induced by the synthetic nonapeptide (O),classical CSa (X), and C3a (0). The latter two fragments were tested along with the synthetic molecule to serve as positive and negative controls, respectively. A baseline for normal saline ( N S S ) or phosphate-buffered saline ( P B S ) containing no solute is shown (A). fragment of the CY chain, also causes a leukocytotic response whereas C3d,g generated from the same Factor I cleavage that produces C3c (Fig. 1) is not leukocytotic. Concern that low levels of endotoxin may account for the observed leukocytosis (14) was addressed after a sample was shown to contain this ubiquitous contaminant following routine screening with Limulus lysate (15). Two approaches were taken toassess the influence of endotoxin in the leukocytosis assay. An aliquot of nonapeptide was digested with trypsin and cleavage at arginine-6 generated a hexapeptide and a tripeptide. Hydrolysis was complete as determined by/high pressure liquid chromatography analysis of the digest on a C1s reverse-phase system. The digest, whether Limulus-positive or not, was administered to rabbits as described above and proved to be inactive as shown in Fig. 3 (middlepanel). Since the leukocytosis-inducing capability of the nonapeptide is abolished after trypsin treatment,we conclude that endotoxin plays no role in theleukocytosis response. Secondly, a portion of contaminated sample was purified by gel filtration through a P-10 column to remove endotoxin. Pyrogen-free water, sterile tubes, and a sterile 25-ml plastic column comprised the gel filtration apparatus used to separate nonapeptide from endotoxin. The nonapeptide was applied to the column and 1-ml fractions were collected. Fractions containing peptide were pooled and lyophilized. This material was negative in the Limulus test. Subsequent intravenous injection and leukocytosis measurements resulted in a profile over 8 h that was identical to that previously observed (Fig. 2). Skin Test-As shown in Table 11, the nonapeptide is capable of enhancing vascular permeability. Other fragments assayed simultaneously, i.e. in the same animal, included C3dK, C3c-K, C3d,g, C ~ CC3a, , and C5a. The latter two polypep-

250

x

50 OO

4

a

Hours

FIG.3. These panels show the results of multiple leukocytosis assays in rabbits. The upper profile indicates that both the nonapeptide and thedes-Arg octapeptide cause an equivalent increase in the number of circulating polymorphonuclear leukocytes. The middle p a n e l shows that theleukocytosis response did not arise from endotoxin. The nonapeptide free of endotoxin elicits a response identical to that shown in the upper p a n e l . The bottom p a n e l shows that C ~ Cwhich , contains the nonapeptide sequence, gives rise to leukocytosis, and C3d,g did not cause a significant increase in PMN levelsover the saline ( N S S ) or phosphate-buffered saline (PBS) control.

tides served as positive controls and sterile normal saline as the negative control. DISCUSSION

Intravenous injection of the synthetic nonapeptide

TLDPERLGR in rabbits causes a significant increase in the number of circulating polymorphonuclear leukocytes. Typically, a 2-3-fold increase in absolute number of PMNs is observed 4 h after intravenous administration of the peptide solution. This response is qualitatively identical to that reported for C3e. In the latter study, a maximal 2-3-fold elevation in leukocyte number was observed approximately 3.5 h following injection (1). On a molar basis, the amount of nonapeptide used was 200 times greater than the required amount ofC3e. Similarly, vascular permeability studies yielded analogous results with only 20 times the amount of nonapeptide needed to elicit an equivalent response relative

Structure-Function Studies of C3d-K

2600

TABLE I1 Biological activity of synthetic mnapeptide and C3-derived fragments permea-Factor

Nonapeptide C3d-K C~C-K C3d,g

Vascular LeukoSuppress Contains T-cell nonapeptide bility cytosis proliferation

+

+

+ +-

-

++

Reference

+ ++ +"

(5) (5) Footnote 2, (7) Footnote 2, c3c (7) (1) C3e a Antiserum to C3e reacted weakly to a nonapeptide-bovine serum albumin conjugate.

-

+ +

+ +

to C3e. Dose-response studies were not undertaken because meaningful comparison of the activity of a small peptide relative to thelarger parent molecule in vivo cannot be made without extensive and detailed pharmacological work up. Our interest has been to identify a region, i.e. active site, within a larger molecular entity that may account for an observed biological function. The intactnonapeptide orthe des-Arg octapeptide appears to be necessary to cause leukocytosis since tryptic scission of the nonapeptide into TLDPER and LGR abrogates activity. It is possible that a defined conformation of the nonapeptide is vital for functional expression and treatment with trypsin destroys this structure. Conceivably, the proline residue at position 4 could bend the peptide into a loop that isstabilized by ionic interaction between Asp-3 and Arg-6 or hydrophobic interaction between Leu-2 and Leu-7. Studies incorporating residue replacements in the nonapeptide are now underway to explore both the sequence and conformational requirements for activity. Overall, the exact nature of C3-derived leukocytosis factors has become considerably clearer with the biochemical characterization of C3d-K (5). Since C3, C3b, C3a, C3c-K(3), and C3d,g (5) (Fig. 3) fail to cause leukocytosis in rabbits and the NHs-terminal nonapeptide portion of C3d-K is all that differs between C3d,g and C3d,K, this unique sequence is strongly implicated in promoting the leukocytotic response. Moreover, C ~ Cthe , larger fragment released from iC3b during C3d,g generation by C3b inactivator (Factor I), promotes leukocytosis because it contains the functional nonapeptide segment at the carboxyl-terminal end of the M,27,000 a-chain fragment (Fig. 1 and Table 11). This observation portends the essential character of the nonapeptide sequence for leukocytosis and suggests that the active conformation depends on neither a free amino nor carboxyl terminus, butthat cleavage of C3b is apparently required to expose the active segment. Additionally, this structural information can be used to ascertain the general location of C3e in theC3 molecule. This fragment, presumed to contain the nonapeptide sequence, spansportions of C3c and C3d-K as proposed in Fig. 1.

Immunologic studies have assigned C3e to the carboxyl-terminal portionof C3c (3). As such, leukocyte mobilizing factor, another C3-derived leukocytosis factor described in the literature (6),must also contain the nonapeptide and presumably maps at a location similar to the C3e fragment. The mechanism(s) of action for the nonapeptide in viuo was not examined in this report. It is not yet known whether the nonapeptide has a direct effect on neutrophils resulting in movement of bone marrow PMN reserves to the blood stream orif the peptide simply enhances permeability of bone marrow microvasculature. From a mechanistic viewpoint, the combination of leukocytosis and enhanced vascular permeability is of particular interest. Enhancement of permeability in the skin may only represent an epiphenomenon, while induced changes in the microvasculature of the bone marrow by a C3-derived factor could lead to enhanced leukocyte mobilization. A synthetic peptide that mimics activities of a natural factor is a new tool to investigate the mechanism(s) underlying these biologic responses. Acknowledgments-We are grateful to EllyeLukaschewskyfor expert preparation of the manuscript and we thank Kevin Ferreri for providing amino acid analyses. REFERENCES 1. Ghebrehiwet, B., and Muller-Eberhard, H. J. (1979) J. Immuml. 123,616-621 2. Muller-Eberhard, H. J. (1978) in Molecular Basis of Biological Degradative Processes (Berlin, R.,Hermann, H., Lepow, I., and Tanzer, J., eds) pp. 65-114, Academic Press, New York 3. Hugli, T. E. (1984) in Progress in Immunology V, Fifth International Congress of Immunology, Kyoto, Japan (Yamamura, Y., and Tada, T., eds) pp. 419-426, Academic Press, New York 4. Hugli, T. E., and Morgan, E. L. (1984) in Contemporary Topics in Immunobiology 14 (Snyderman, R., ed) pp. 109-153, Plenum, New York 5. Meuth, J. L., Morgan, E.L., DiScipio, R. G., and Hugli, T. E. (1983) J. Zmmunol. 130,2605-2611 6. Rother, K. (1972) Eur. J. Immuml. 2,550-558 7. Davis, A. E., Harrison, R. A., and Lachman, P. J. (1984) J. Immuml. 132,1960-1966 8. Itoh, N., Najiwana, D., and Kamiya, T. (1975) Tetrahedron Lett, 4393-4396. 9. Erickson, B.W., and Merrifield, R.B. (1976) in The Proteins (Neurath, H., and Hill, R. L., eds) 3rd Ed. Vol. 2, pp. 256-527, Academic Press, New York 10. Stewart, J. M., and Young, J. D. (1969) Solid Phase Peptide Synthesis, W . H. Freeman, San Francisco 11. Hoeprich, P. D., Jr., and Doolittle, R.F. (1983) Biochemistry 22, 2049-2055 12. Sarin, V. K., Kent, S. B.H., Tam, J. P., and Merrifield, R. B. (1981) Anal. Biochem. 117, 147-157 13. Cochrane, C. G., and Muller-Eberhard, H. J. (1968)J. Exp. Med. 127,371-386 14. Bennett, I. L., and Cluff, L. G. (1957) Pharmacal. Rev. 9, 427475 15. Tomasulo, P. A., Levin, J., Murphy, P. A., and Winkelstein, J. A. (1977) J. Lab. Clin. Med. 89,308-315 16. de Bruijn, M., and Fey, G. (1985) Proc. Natl. Acad. Sci. U. S. A., in press