Antibacterial and Immunostimulatory Properties of Chemotactic N ...

2 downloads 0 Views 1MB Size Report
AND PAUL WILLIAMS*. Department ofPharmaceutical Sciences, The ..... Drugs Exp. Clin. Res. 10:703-711. 17. LeDuc, L. E., G. R. Marshall, and P. Needleman.
ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Sept. 1989, p. 1516-1521 0066-4804/89/091516-06$02.00/0 Copyright © 1989, American Society for Microbiology

Vol. 33, No. 9

Antibacterial and Immunostimulatory Properties of Chemotactic N-Formyl Peptide Conjugates of Ampicillin and Amoxicillin BARRIE W. BYCROFT, PETER M. LOCKEY, AUDREY PENROSE, RAYMOND J. GROUT, AND PAUL WILLIAMS* Department of Pharmaceutical Sciences, The University, Nottingham NG7 2RD, United Kingdom Received 16 February 1989/Accepted 13 June 1989

N-Formyl dipeptide conjugates of ampicillin and amoxicillin related to the chemotactic peptide Nfotmyl-L-methionyl-L-leucyl-L-phenylalanine were synthesized and assessed for antibacterial activity and affinity for the chemotactic peptide receptor of differentiated human promyelocytic leukemia (HL-60) cells. The conjugates and parent P-lactam antibiotics showed similar antibacterial activities against Escherichia coi and Staphylococcus aureus. The affinity of each conjugate for the chemotactic peptide receptor was determined in a competitive binding assay, using 3H-labeled N-formyl-L-methionyl-L-leucyl-L-phenylalanine. All conjugates bound to the receptor, but with affinities ranging from 1/3 to 1/100 that of the tritiated substrate. There was good correlation between receptor affinity and stimulation of chemotaxis. The peptide-antibiotic conjugates also stimulated the oxidative metabolism of the HL-60 cells by inducing the production of superoxide and hydrogen peroxide as determined by Luminol- and Lucigenin-enhanced chemiluminescence. These conjugates, based on N-formyl-L-methionyl-L-leucyl-L-phenylalanine, thus combine both potent antibacterial and immunostimulatory properties within the same molecule.

The ability of cells such as polymorphonuclear leukocytes to phagocytose and kill microorganisms constitutes the first line of host defense against invading bacteria (31, 33). However, the susceptibility of microorganisms to host defense mechanisms may be increased as a result of antibioticinduced damage to the bacterial cell envelope (1, 16, 20, 35). In addition, some antibiotics may modulate the phagocytic process directly by influencing leukocyte cheemotaxis, phagocytic uptake, and killing (2, 3, 6, 19, 20, 34). Such antibiotics exert a much greater therapeutic effect than that predicted by their in vitro bactericidal activity (2, 16, 20). Thus, the interaction between phagocytic cells and antimicrobial agents is of crucial importance in the eradication of bacterial infections. Consequently, the influence of antibiotics on the effectiveness of the immune response has attracted considerable interest (19-21). Antibiotics which stimulate the immune system should offer considerable advantages in aiding the elimination of infecting organisms, especially in

immunocompromised patients. Neutrophils and monocytes respond chemotactically to low-molecular-weight peptides present in bacterial culture supernatants. These peptides, which possess a blocked amino terminus but a free carboxyl terminus, are thought to be derived from newly synthesized proteins since procaryotic cells initiate protein synthesis with N-formyl-Lmethionine (18, 24, 29). N-Formylated peptides such as N-formyl-L-methionyl-L-leucyl-L-phenylalanine (FMLP) induce leukocyte chemotaxis and, at higher concentrations, secretion of enzymes such as lysozyme from cellular organelles. They also stimulate the release of toxic intermediates of oxygen reduction (9, 12, 24) and promote the phagocytosis anid intracellular killing of bacteria (14). These responses are initiated by binding of the chemotactic peptides to specific receptors on the leukocyte cell membrane (9, 12, 18, 24, 36). Linkage of N-formyl peptides to a suitable antibiotic should, providing the biological activity of the peptide is conserved, result in a compound combining both antimicro-

bial and immunostimulatory properties. Since N-formylated peptides are internalized following receptor binding (15), such peptide conjugates may offer a means of intracellular delivery of J-lactam antibiotics or indeed other peptide conjugates of anti-infective or antineoplastic agents. This paper reports the synthesis and in vitro evaluation of N-formyl peptide conjugates of ampiciflin and amoxicillin. To investigate the immunostimulatory properties of these conjugates, the human promyelocytic cell line HL-60, which has the potential to differentiate along either a granulocyte or macrophage lineage, was used (18, 22, 23). Dimethyl sulfoxide (DMSO) or retinoic acid induces the differentiation of HL-60s along the myeloid lineage, while la,25-dihydroxyvitamin D3 or 12-O-tetradecanoyl-13-phorbol acetate induces a monocyte or macrophage lineage (22). In this study, DMSO induction was used since DMSO-treated HL-60 cells differentiate into mature granulocytes which express functional receptors for N-formyl peptides (10, 12, 18, 25, 28), respond chemotactically, and are capable of phagocytosing and killing bacteria (23). (This work was presented in part at the 20th European Peptide Symposium, Tilbingen, Federal Republic of Germany, 4 through 10 September 1988.) MATERIALS AND METHODS Synthesis of N-formylated peptide-antibiotic conjugaies. The detailed synthesis of N-Form-L-Met-L-Leu-ampicillin is described below; similar methods were used to prepare other

N-formyl peptide-antibiotic conjugates. A solution of N-Form-L-Met-L-Leu (1.5 g) and N-hydroxysuccinimide (0.6 g) in dry dimethoxyethane (50 ml) was cooled to 0WC, and a solution of dicyclohexylcarbodiimide (1.06 g) in dimethoxyethane (10 ml) was added. The mixture was stirred at 0°C for 3 h, and then a solution of ampicillin sodium (2.0 g) in water (20 ml) and dimethoxyethane (5 ml) was added; stirring, without further cooling, was continued for 1.5 h. Insoluble material was removed, and the filtrate was reduced to 10 ml. The solution was adjusted to pH 5.0 with 1 M HCl and extracted with ethyl acetate (twice, 20 ml

* Corresponding author; 1516

VOL. 33, 1989

EVALUATION OF CHEMOTACTIC PEPTIDE-ANTIBIOTIC CONJUGATES

1517

TABLE 1. Physical and spectroscopic data for N-formyl peptide-antibiotic conjugates

8H (J, HZ)b"/

Compound'

)

J H)

6H

1 2 3 4

5.50 5.52 5.52 5.54

Vmax'

7H

(dd, J67 4.2; J6.NH 7.5) (dd, J6.7 4.1; J6.NH 7.7) (dd, J6.7 4.0; J7,NH 7.7) (mi, J67 4.0)

5.38 (d, J67 4.2) 5.40 (d, J67 4.0) 5.39 (d, J6.7 4.0) 5.42 (d, J6,7 4.1)

/

(M +

1,777, 1,739, 1,642 1,776, 1,732, 1,640 1,778, 1,733, 1,642

1,785, 1,735, 1,655

1)±d

622 622 638 584

Compounds 1 to 4 are as in Fig. 1. b Chemical shift for the 6H and 7H protons of the penam nucleus. Infrared stretching frequency for the carbonyl groups in each conjugate. d Molecular ion (M+) data obtained by fast atom bombardment mass spectrometry. For compounds 1 and 3, [M + Na]+ ions were also observed.

each). The organic layer was washed with 1 M citric acid (10 ml), and on evaporation to half the original volume, the conjugate crystallized (1 g; 31%). The structure of each conjugate was confirmed by mass spectrometry (molecular ion), proton nuclear magnetic resonance (6 and 7 protons of the penam nucleus), and infrared spectroscopy (characteristic carbonyl stretches) (Table 1). Purity was assessed by preparative and analytical highperformance liquid chromatography. For the latter, samples were applied to a Spherisorb ODS (250- by 4.6-mm column) and chromatographed by using a solvent system consisting of 0.1 M citric acid (pH 4.5)-methanol (1:1). The retention times for compounds 1 to 4 (as given in Fig. 1) were 6.0, 5.7, 5.5, and 4.4 min, respectively. N-Form-L-Met-L-Leu-ampicillin was also applied to a preparative column (C18 Dynamax column; 250 by 21.4 mm) and chromatographed with 0.1 M citric acid (pH 4.5)-methanol (3:7). Only the conjugate showed both receptor-binding and antibiotic activity. MIC determination. MICs of each compound were determined against Escherichia coli lAll and Staphylococcus aureus F/21 by tube dilution after overnight growth in nutrient broth at 37°C. Maintenance and differentiation of HL-60 cells. HL-60 cells were kindly supplied by J. Traynor, University of Loughborough, Loughborough, Leicester, United Kingdom, and were maintained at 37°C in 95% air-5% CO2 in RPMI 1640 medium supplemented with 10% (vol/vol) calf serum, 2 mM L-glutamine, penicillin (50 IU/ml), and streptomycin (50 jig/ml) (GIBCO Ltd., Paisley, Scotland) in 75-cm2 tissue

culture flasks. Differentiation of HL-60 cells was induced by culturing in fresh medium containing 1.25% (vol/vol) DMSO (18, 22, 23). Cells were examined after 3, 4, 5, 6, 7, 8, 9, 10, 11, and 12 days for expression of the chemotactic peptide receptor and oxidative burst. Analysis of [3H]FMLP binding. Ten-day differentiated HL60 (d-HL-60) cells were harvested by centrifugation (600 x g), washed twice with assay buffer (phosphate-buffered saline, pH 6.75, containing 0.1% [wt/vol] bovine serum albumin), and suspended in assay buffer at a density of 2.5 x 106 cells per ml. Cell suspensions (3 ml) were incubated with [3H]FMLP (5 nM; 1 ml) and 1 ml of assay buffer or 1 ml of unlabeled FMLP (5 FLM) at 22°C for 30 min. Total specific binding was defined as the difference in disintegrations per minute between tubes containing FMLP alone and those containing a 1,000-fold excess of FMLP. By using a range of concentrations, the potencies of the peptide-antibiotic conjugates in displacing [3H]FMLP were then determined. The results were calculated as a percentage of the total specific binding. All tubes were cooled in ice water for 5 min before rapid filtration through Whatman GF/A glass-fiber filters and washing with 10 ml of assay buffer (4°C). The filters were added to scintillation vials containing 5 ml of LKB Optiphase scintillation fluid and counted in an LKB 1219 liquid scintillation counter (counting efficiency, 40 to 45%). Respiratory burst. Production of superoxide anion (°2-) and hydrogen peroxide (H202) by d-HL-60 cells was determined by chemiluminescence assay. The chemiluminergenic

SCH3

CH2 H'

CCH2

0

O=CHNH NH' 0 R2 R3 C

EOC H 3 0

CH3

Compound 1

H

CH2CH(CH3)2

H

H

HO

4

H

H

CH2CH(CH3)2

CH2CH(CH3) 2

H

CH 3

CH3

( N-Form-L-Met-L-Leu-Amp*)

(N-Form-L-Met-D-Leu-Amp*) (N-Form-L-Met-D-Leu-Amox*)

(N-Form-L-Met-Aib-Amp*)

FIG. 1. Structures of FMLP and N-formyl peptide conjugates of ampicillin (Amp*) and amoxicillin (Amox*).

1518

BYCROFT ET AL.

ANTIMICROB. AGENTS CHEMOTHER.

TABLE 2. MICs of N-formyl peptide conjugates of ampicillin and amoxicillin MIC (,uM) for: Compound

Ampicillin Amoxicillin

N-Form-L-Met-L-Leu-L-Phe N-Form-L-Met-L-Leu-ampicillin N-Form-L-Met-D-Leu-ampicillin N-Form-L-Met-D-Leu-amoxicillin

N-Form-L-Met-Aib-ampicillin

E. coli

S. aureus

0.36 0.32 >>128.0 1.61 1.61 6.28 6.74

0.36 0.32 >>128.0 6.44 1.61 3.14 1.68

probes Luminol (5-amino-2,3-dihydro-1,4-phthalazinedione; Sigma) and Lucigenin (N,N'-dimethyl-10,10'-biacridinium dinitrate; Sigma) were used at a concentration of 10-4 M each to estimate the production of H202 and 02-, respectively (7, 8). Briefly, 106 cells in 200 ,ul of assay buffer (phosphatebuffered saline, pH 7.4, 37°C) were added to cuvettes containing Luminol or Lucigenin (200 p.1; 300 ,uM) and various concentrations of peptide or peptide-antibiotic conjugate (200 p.l), maintained at 37°C. The time course and magnitude of oxidative metabolite production were monitored with a luminometer (LKB model 1250) linked to a chart recorder. Responses were expressed as percentages of the maximum response produced by a saturating concentration of FMLP (1 p.M). Chemotaxis. Estimation of the chemotactic response of d-HL-60 to peptides and peptide-antibiotic conjugates was performed by using a Boyden chamber. Some 106 cells in 200 p.l of assay buffer (Gey's balanced salt solution with 0.2% [wt/vol] bovine serum albumin) were separated from the chemotactic agent in 200 p.l of assay buffer by a 5-p.m filter (Sartorius). After a 90-min incubation at 37°C, the filters were removed, fixed in glutaraldehyde, and stained with hematoxylin. Chemotaxis was quantified by counting the number of cells which had migrated completely through the filter in four low-power (x200) microscopic fields. RESULTS The structure of each of the four peptide-antibiotic conjugates synthesized is shown in Fig. 1. From high-performance

liquid chromatography experiments, no compound was found to contain >0.5% impurity, and only the conjugates showed both receptor-binding and antibiotic activity. MICs. All of the conjugates were antibacterial, with MICs against S. aureus and E. coli ranging from 1.61 to 6.74 p.M (Table 2). The MICs were, however, greater than those of ampicillin and amoxicillin, which were 0.36 and 0.32 ,uM, respectively. FMLP exhibited no antimicrobial activity up to a concentration of 128 p.M. Interestingly, replacement of the L-leucine residue by D-leucine or ct-aminoisobutyric acid (Aib) increased the activity of the ampicillin conjugates against S. aureus (Table 2). Against E. coli, however, the MIC was not affected by the leucine stereochemistry, although N-Form-L-Met-DLeu-amoxicillin was considerably less antibacterial than the corresponding ampicillin derivative (Table 2). Receptor binding. HL-60 cells transferred to medium containing DMSO were examined daily for the ability to bind FMLP and to exhibit an oxidative burst. Maximal expression of the receptor and respiratory burst was observed after 10 days of incubation in the presence of DMSO. Cells induced for 10 days were therefore chosen for all subsequent experiments. [3H]FMLP did not bind (or stimulate the oxidative burst) in undifferentiated HL-60 cells. All of the conjugates exhibited affinity for the formyl peptide receptor on d-HL-60 cells, as shown by the ability to displace [3H]FMLP from the cells (Fig. 2). Concentrations of peptide required to produce 50o of the maximum effect, as determined by the concentration curve (ED50s), for the conjugates varied from 4.0 to 126 nM compared with 1.3 nM for FMLP (Table 3). Thus, the potency of the compounds varied from approximately 1/3 that of FMLP for N-FormL-Met-L-Leu-ampicillin to approximately 1/100 that of FMLP for N-Form-L-Met-Aib-ampicillin (Table 3). Respiratory burst. Figure 3 shows the kinetics of luminolenhanced chemiluminescence response of d-HL-60 cells exposed to a range of FMLP concentrations. A rapid response (within 20 s), reaching a maximum of 15.9 + 0.3 mV, was observed, which quickly decayed. Undifferentiated HL60 cells showed no response. Stimulation of .02- and H202 production by the compounds was dose dependent, with ED50s varying from 160.0 to 3,310 nM and from 183 to 3,800 nM, respectively (Fig. 4). Furthermore, the rank order of potency for the conjugates in stimulating the respiratory burst was the same as that for

z 0

\|\&

X4

60F:4

40 C2.,

20

O

2 103 10 CONCENTRATION (nM) FIG. 2. Displacement of [3H]FMLP from HL-60 cells by FMLP (O), N-Form-L-Met-L-Leu-ampicillin (U), N-Form-L-Met-D-LeuL-ampicillin (0), N-Form-L-Met-L-Leu-amoxicillin (0), and N-Form-L-Met-Aib-ampicillin (A). Points represent means of at least three

0.1

determinations ± standard errors.

1

EVALUATION OF CHEMOTACTIC PEPTIDE-ANTIBIOTIC CONJUGATES

VOL. 33, 1989

1519

TABLE 3. ED50s for N-formyl peptides for receptor binding, chemotaxis, and stimulation of respiratory burst

ED50 (nM; mean

N-Formyl peptide

N-Form-L-Met-L-LeU-L-Phe

Receptor binding

Chemotaxis

H202

*02

1.3 0.1 4.0 0.3 7.9 ± 0.8 45.7 ± 8.1 126.0 + 417.0

3.3 13.2 38.0 190.0 417.0

36.3 3.3 160.0 ± 17.4 288.0 ± 33.2 1,230.0 ± 182.0

38.9 + 2.7 183.0 ± 16.7 339.0 ± 35.0 1,260.0 ± 118.0 3,800.0 ± 317.0

+ +

N-Form-L-Met-L-Leu-ampicillin N-Form-L-Met-D-Leu-ampicillin N-From-L-Met-D-Leu-amoxicillin

N-Form-L-Met-Aib-ampicillin

± SEM)

receptor affinity (Table 3). ED50s for 02- and H202 production were approximately 30 to 40 times that for receptor binding. The antibiotics ampicillin and amoxicillin produced no response in this system at concentrations up to 200 FxM. Chemotaxis. By the Boyden chamber technique, each conjugate was shown to possess chemotactic activity (Table 3). However, a dual effect was observed in that the chemotactic agent produced an increase in the numbers of cells migrating into the filters up to a critical concentration; at higher concentrations, ligand diffused so rapidly across the membrane that no concentration gradient was established

(Fig. 5).

ED50s for the conjugates ranged from 13.2 to 417.0 nM compared with an ED50 of 3.3 nM for FMLP. The rank order of chemotactic activity of the conjugates was the same as that for receptor binding, but the concentrations required for chemotactic activity were approximately three to four times higher (Table 3).

+

3,310.0

Phe-L-Phe retain biological activity (11, 24). Esters and amides of N-formyl peptides are active, the benzyl ester analog of FMLP, for example, being more active than the parent compound (11). The phenylalanine residue of FMLP can be replaced with D-phenylglycine without significant loss of activity (30); substitution into the 4-position of phenylalanine (e.g., tyrosine) is not well tolerated (24). Interestingly, the monocyclic 1-lactam antibiotic nocardicin A, a compound with only modest in vitro antibacterial activity, has immunostimulatory activity (2). This antibiotic has been reported to enhance the intracellular killing of Pseudomonas aeruginosa. In doses producing levels in blood well below the minimum bactericidal activity, this antibiotic successfully controlled an otherwise fatal pseudomonas infection in z 0

DISCUSSION 0 Ligand binding to the chemotactic peptide receptor on 40 leukocytes is highly selective (5, 11, 24). Structural requirements for optimal binding are reported to be (i) an N-formyl terminus substitution, (ii) a methionine at position 1, and (iii) 99 j X hydrophobic amino acids at positions 2 and 3. Leucine and phenylalanine~~~~~~~~a are the preferred amino acids at positions 2 ~~~~0. and 3, respectively (11, 24). Some degree of structural 100 110 0f diversity is possible, with retention of biological activity (5, 11, 13, 24, 30, 32). Isomers of FMLP containing D-amino CONCENTRATION (nM) acids and tetrapeptides such as N-Form-L-Met-L-Leu-Ltpl 0

0

j

1 PM 0

q

P4~~~~~e

SO_

ioonM 50 nM

4 b .

10

25 nM

123nM

2

10

.

3

10

14 10

CONCENTRATION (nM) FIG. 4. Production of superoxide anion (a) and hydrogen peroxide (b) by HL-60 cells in response to FMLP (O), N-Form-LMet-L-Leu-ampicillin (-), N-Form-L-Met-D-Leu-L-ampicillin (0),

FIG. 3. Time traces of chemiluminescence emitted from d-HL60 cells in the presence of luminol (10-4 M) when exposed to increasing concentrations of FMLP. Abscissa, Time; ordinate,

N-Form-L-Met-L-Leu, amoxicillin (0), and N-Form-L-Met-Aibampicillin (A). Results are expressed as percentage values of the maximal response elicited by a saturating dose (1 ,uM) of FMLP. Points represent means of at least three determinations + standard

chemiluminescence.

errors.

BYCROFT ET AL.

1520

ANTIMICROB. AGENTS CHEMOTHER.

200

9 C)

0.

I

10

on ro2 -

--,0.

CONCENTRk!TION (,3 FIG. 5. Migration of HL-60 cellIs into micropore filters in response to increasing concentratioris of FMLP N-Form-LMet-L-Leu-ampicillin (U), N-Form--L-Met-D-Leu-L-ampicillin (0), N-Form-L-Met-L-Leu-amoxycillin (EI), and N-Form-L-Met-Aib-ampicillin (A). Points represent means of at least two determinations.

(O),

mice (2). Nocardicin A shares some structural similarities with FMLP (30). Its D-phenylglIycine moeity and 1-lactam ring correspond to the phenylala nine and leucine residues of FMLP. However, in our hands, inocardicin A did not bind to the chemotactic peptide receptor or stimulate the production of toxic oxygen radicals by d1-HL-60 cells (unpublished observation). Since changes in the C-terminius phenylalanine of FMLP are tolerated, its replacement by an antibiotic with a suitable amino acid residue may result in the formation of a molecule which combined good antibacteirial activity with the immunostimulatory properties of FML,P. Ampicillin and amoxicillin with D-phenylglycine and D-p -hydroxyphenylglycine side chains appeared, therefore, to b(e good candidate antibiotics for conjugation (30). Several N-formyl peptide conjugates of ampicillin and amoxicillin were synthesized and evaluated for antibacterial and immunostinnulatory properties. The peptide-antibiotic conjugaLtes displayed good antibacterial activity against both E. coi1i and S. aureus, with MICs ranging from 1/4 to 1/20 those of the parent antibiotics ampicillin and amoxicillin. The stereochemistry at the leucine residue appeared to be important for activity against S. aureus. N-Form-L-Met-D-Leu-ampicillin was four times more active than the L-Leu isomer against this organism. This finding may reflect either differences in their affinity for the penicillin-binding proteins or the preferential penetration of the D-Leu isomer through the cell wall. Experiments are under way to investigate this phenomenon. The receptor binding activity of the conjugates correlated with their structural similarity to the parent peptide FMLP. Thus, N-Form-L-Met-L-Leu-ampicillin, the nearest structural analog, had the greatest affinity for the chemotactic peptide receptor of the four ligands. Inversion of the stereochemistry at leucine reduced receptor binding by a factor of 2. N-Form-L-Met-D-Leu-amoxicillin, which possesses a Dp-hydroxyphenyl substituent, considerably reduced activity, a finding paralleled in the replacement of phenylalanine in FMLP by tyrosine (24). The greatest reduction in receptor affinity was observed when the leucine residue was exchanged for Aib. The rationale for replacement with Aib was based on previous proposals that this residue locks the molecule into the biologically active receptor-binding conformation (13, 32). However, this compound, which probably contains a p or y turn (13), was about 100 times less active than FMLP. There was also good correlation between the receptor-

binding activity of the compounds and their potency as agonists. ED50s for chemotaxis were 3 to 4 times those for receptor binding, while ED5Os for stimulation of respiratory burst were 30 to 40 times those for receptor binding. The D-amino acid side chains of ampicillin and amoxicillin should also increase the stability of the peptides to serum peptidases. N-Form-L-Met-Aib-ampicillin should be particularly stable since methyl-substituted amino acid residues greatly reduce lability to protease hydrolysis (17). The in vivo activity of the conjugates may be greater than suggested by their in vitro potencies since serum half-lives should be prolonged. A further possible advantage of these peptide-antibiotic conjugates might be in the treatment of intracellular bacteria. j-Lactam antibiotics in general do not accumulate in phagocytic cells, which may explain their poor activity against organisms such as Legionella pneumophila, Listeria monocytogenes, and also some strains of Salmonella spp. and S. aureus which survive intracellularly following ingestion (26, 27). N-Formyl peptide-receptor complexes are internalized following receptor binding, the peptide is transported to the cytosol, and the unoccupied receptor is recycled back to the cell surface (15, 24). Additional work is under way to determine whether the peptide-antibiotic conjugates are also internalized. In summary, the four N-formyl peptide-penicillin conjugates synthesized were shown to exhibit good antibacterial and immunostimulatory activity in in vitro assays. The in vitro activity demonstrated by these compounds suggests that they warrant in vivo investigation. However, in vivo, their potential usefulness may be limited since they may stimulate chemokinesis rather than chemotaxis and, in addition, cause tissue damage at sites distant from the locus of infection through the widespread release of toxic oxygen radicals. For in vivo use, structural modifications may therefore be necessary such that the immunostimulant portion of the molecule requires activation by bacterial metabolism at the infection site. We are investigating the potential of such molecules. LITERATURE CITED 1. Atkinson, B. A., and L. Amaral. 1982. Sublethal concentrations of antibiotics, effects on bacteria and the immune system. Crit. Rev. Microbiol. 9:101-138. 2. Banks, R. M., and F. O'Grady. 1983. Therapeutic significance of nocardicin A, stimulation of phagocytic function in experimental Pseudomonas aeuruginosa infection. Br. J. Exp. Pathol. 64:231-237. 3. Belsheim, J. A., and G. H. Gnarpe. 1981. Antibiotics and granulocytes: direct and indirect effects on granulocyte chemotaxis. Acta Pathol. Microbiol. Scand. Sect. C 89:217-221. 4. Bodey, G. P., E. Middleman, T. Umsawadi, and V. Rodriguez. 1972. Infections in cancer patients: results with gentamicin sulphate therapy. Cancer 29:1697-1701. 5. Bonora, G. M., C. Toniolo, R. J. Freer, and E. L. Becker. 1986. Retro-all-D and retro isomers of a formyl-methionyl peptide chemoattractant: an insight into the mode of binding at the receptor on rabbit neutrophils. Biochim. Biophys. Acta 884: 545-549. 6. Briheim, G., and C. Dahigren. 1987. Influence of antibiotics on

formylmethionyl-leucyl-phenylalanine-induced leukocyte che-

miluminescence. Antimicrob. Agents Chemother. 31:763-767. 7. Briheim, G., 0. Stendahl, and C. Dahlgren. 1984. Intra- and extracellular events in luminol-dependent chemiluminescence of polymorphonuclear leukocytes. Infect. Immun. 45:1-5. 8. Dahigren, C., H. Aniansson, and K. E. Magnusson. 1985. Pattem of formyl-methionyl-leucyl-phenylalanine-induced luminoland lucigenin-dependent chemiluminescence in human neutrophils. Infect. Immun. 47:326-328.

VOL. 33, 1989

EVALUATION OF CHEMOTACTIC PEPTIDE-ANTIBIOTIC CONJUGATES

9. De Togni, P., P. Bellavite, V. Della Bianca, M. Grzeskowiak, and F. Rossi. 1985. Intensity and kinetics of the respiratory burst of human neutrophils in relation to receptor occupancy and rate of occupation by formylmethionylleucylphenylalanine. Biochim. Biophys. Acta 838:12-22. 10. Goldman, D. W., D. M. Olson, D. G. Payan, L. A. Gifford, and E. J. Goetzel. 1986. Development of receptors for leukotriene B4 on HL-60 cells induced to differentiate by la,25-dihydroxyvitamin D3. J. Immunol. 136:4631-4636. 11. Freer, R. J., A. R. Day, N. Muthukumaraswamy, D. Pinon, A. Wu, H. J. Showell, and E. L. Becker. 1982. Formyl peptide chemoattractants: a model of the receptor on rabbit neutrophils. Biochemistry 21:257-263. 12. Heiman, D. F., J. P. Gardner, W. J. Apfeldorf, and H. L. Malech. 1986. Effects of tunicamycin on the expression and function of formyl peptide chemotactic receptors of differentiated HL-60 cells. J. Immunol. 136:4623-4630. 13. Iqbal, M., P. Balaram, H. J. Showell, R. J. Freer, and E. L. Becker. 1984. Conformationally constrained chemotactic peptide analogs of high biological activity. FEBS Lett. 165:171-174. 14. Issekutz, A. C., K.-Y. Lee, and W. D. Biggar. 1979. Enhancement of human neutrophil bactericidal activity by chemotactic factors. Infect. Immun. 24:295-301. 15. Jesaitis, A. J., J. R. Naemura, R. G. Painter, L. A. Sklar, and C. G. Cochrane. 1983. The fate of an N-formylated chemotactic peptide in stimulated human granulocytes. J. Biol. Chem. 258:1968-1975. 16. Lam, C., G. Laber, J. Hildebrandt, A. Wenzel, F. Turnowsky, and E. Schutze. 1984. Therapeutic relevance of antibioticinduced augmentation of host defences in experimental infections. Drugs Exp. Clin. Res. 10:703-711. 17. LeDuc, L. E., G. R. Marshall, and P. Needleman. 1978. Differentiation of bradykinin receptors and of kinases with conformation analogs of bradykinin. Mol. Pharmacol. 14:413-421. 18. Malech, H. L., J. P. Gardner, D. F. Heiman, and S. A. Rosenzweig. 1985. Asparagine-linked oligosaccharides on formyl peptide chemotactic receptors of human phagocytic cells. J. Biol. Chem. 260:2509-2514. 19. Mandell, L. A. 1982. Effects of antimicrobial and antineoplastic drugs on the phagocytic and microbicidal function of the polymorphonuclear leucocyte. Rev. Infect. Dis. 4:683-697. 20. Milatovic, D. 1983. Antibiotics and phagocytosis. Eur. J. Clin. Microbiol. 2:414-425. 21. Miller, T. E. 1984. Host defence mechanisms, antibiotics and immunosuppressive agents-a complex relationship. Eur. J. Clin. Microbiol. 3:285-287. 22. Momoi, T., H. Senoo, and H. Yoshikura. 1986. Two stages of the differentiation of HL-60 cells induced by la,25-Dihydroxyvitamin D3; commitment by la,25-Dihydroxyvitamin D3 and

1521

promotion by DMSO. Biosci. Rep. 6:95-101. 23. Newburger, P. E., M. E. Chovaniec, J. S. Greenberger, and H. J. Cohen. 1979. Functional changes in human leukemic cell line HL-60. A model for myeloid differentiation. J. Cell. Biol. 82:315-322. 24. Niedel, J. E., and P. Cuatrecasas. 1980. Formyl peptide chemotacic receptors of leukocytes and macrophages. Curr. Top. Cell Regul. 17:137-170. 25. Niedel, J. E., I. Kahane, L. Lachman, and P. Cuatrecasas. 1979. A subpopulation of cultured human promyelocytic leukemia cells (HL-60) displays the formyl peptide chemotactic receptor. Proc. Natl. Acad. Sci. USA 77:1000-1004. 26. Prokesch, R. C., and W. L. Hand. 1982. Antibiotic entry into human polymorphonuclear leucocytes. Antimicrob. Agents Chemother. 21:373-380. 27. Renard, C., H. J. Vanderhaeghe, P. J. Claes, A. Zenebergh, and P. M. Tulkens. 1987. Influence of conversion of penicillin G into a basic derivative on its accumulation and subcellular localization in cultured macrophages. Antimicrob. Agents Chemother. 31:410-416. 28. Roberts, P. J., A. R. Cross, 0. T. G. Jones, and A. W. Segal. 1982. Development of cytochrome b and an active oxidase system in association with maturation of a human promyelocytic (HL-60) cell line. J. Cell Biol. 95:720-726. 29. Schiffmann, E., B. A. Corcoran, and S. M. Wahl. 1975. NFormylmethionyl peptides as chemoattractants for leucocytes. Proc. Natl. Acad. Sci. USA 72:1059-1062. 30. Semus, S. F., B. W. Bycroft, and R. J. Grout. 1984. Synthesis and immunostimulant activity of modified oligopeptides. J. Pharm. Pharmacol. 36:1-68. 31. Solberg, C. 0. 1984. Phagocytic cells in host defence. Acta OtoLaryngol. Suppl. 407:5-13. 32. Sukumar, M., P. Antony Raj, P. Balaram, and E. L. Becker. 1985. A highly active chemotactic peptide analog incorporating the unusual residue 1-aminocyclohexanecarboxylic acid at position 2. Biochem. Biophys. Res. Commun. 128:339-344. 33. Thomas, E. L., R. I. Lehrer, and R. F. Rest. 1988. Human neutrophil antimicrobial activity. Rev. Infect. Dis. 10(Suppl. 2):450-456. 34. Welch, W. D., D. Davis, and L. D. Thrupp. 1981. Effect of antimicrobial agents on human polymorphonuclear leukocyte microbicidal function. Antimicrob. Agents Chemother. 20:1520. 35. Williams, P. 1987. Sub-MICs of cefuroxime and ciprofloxacin influence the interaction of complement and immunoglobulins with Klebsiella pneumoniae. Antimicrob. Agents Chemother. 31:758-762. 36. Zigmond, S. H., and A. W. Tranquillo. 1986. Chemotactic peptide binding by rabbit polymorphonucler leukocytes. J. Biol. Chem. 261:5283-5288.