News & Notes

DNA as the Intracellular Secondary Target for Antibacterial Action ofHuman Neutrophil Peptide-I AgainstMycobacterium tuberculosisH37Ra

Sudhir Sharma, GK Khuller

Department of Biochemistry, Postgraduate Institute of Medical Education and Research, Chandigarh 160 012, India

Received: 25 October 2000/Accepted: 10 January 2001

Abstract. The secondary intracellular target of human neutrophil peptide-1 has been examined inM.tuberculosisH37Ra. Binding studies with radioiodinated HNP-1 revealed biphasic equilibrium bindingkinetics with respect to time. The major site of HNP-1 binding was found to be plasma membrane/cellwall whereas the cytosol appears to be a secondary site. Among the different macromolecules examined,maximum inhibition (75%) was observed in DNA biosynthesis during treatment with HNP-1. Theinteraction of HNP-1 with mycobacterial genomic DNA on the basis of gel retardation assay revealedHNP-1 binding to DNA. These results indicate that HNP-1 has DNA as the secondary intracellular targetfor antibacterial action against mycobacteria.

In recent years, antimicrobial peptides (AMPs), whichare a truly novel class of antimicrobial agents, havegained the attention of researchers all over the world.These peptides possess antimicrobial activity againstgram positive [13], gram-negative bacteria [8], fungi[11], enveloped viruses [7] and mycobacterial species[9]. Regarding the mechanism of these antimicrobialpeptides, two hypotheses have been proposed, i.e., theymight disturb the ion gradients through their interactionwith membranes thus resulting in the loss of microbialviability, or, alternatively, the AMPs might enter thetarget cells through disrupted membranes and interactwith some intracellular target to bring about their action[4]. Among antimicrobial peptides, human neutrophilpeptides (HNPs) 1–4 are antimicrobial peptides stored inazurophilic granules of polymorphonuclear neutrophils[3]. Recently, in M. tuberculosis, the major target ofHNP-1 action has been reported to be cell wall/plasmamembrane [12]. In the present study, it was hypothesizedthat DNA may be the secondary target of HNP-1 and thisaspect has been explored.

Materials & Methods

M. tuberculosisH37Ra (obtained from NCTC London) was grownin Youman’s modified medium [14] containing 0.05% Tween 80 on anorbital shaker (100–120 rpm) at 37°C. Chemically synthesized HNP-1with the same primary structure and the same disulfide linkages (i.e.,between cys-2 and cys-30, cys-4 and cys-19, and cys-9 and cys-29 asthat of native HNP-1) was obtained from Peptide Institute (Japan). Itwas dissolved in 0.01% acetic acid and stored as a stock solution of 100mg/ml at 220°C. HNP-1 was radioiodinated using Na125I by thechloramine-T method [6]. The binding kinetics of labeted HNP-1 toM.tuberculosiscells were studied as described earlier by Lehrer et al [7].To investigate the binding site of HNP-1M. tuberculosisH37Ra cellswere treated with radioiodinated HNP-1 for 24 h. Cells were harvestedand disrupted by sonication for 30 min and components were separatedby differential centrifugation [5]. Radioactivity was measured in cellwall, plasma membrane and cytosol fractions respectively.

M. tuberculosisH37Ra cells 33108 (organisms/ml) were sus-pended in Krebs Ringer Buffer containing HNP-1 at IC90 concentra-tion, pulsed for 90 min with radiolabelled precursors viz. [14C]-asparticacid for proteins (1mCi), [14C]-acetate for lipids, (2mCi) [3H]-thymi-dine (8mCi) for DNA and [14C]-uracil (1mCi) for RNA, respectively,and processed as described earlier [12]. To check whether [3H]-thymi-dine is incorporated into DNA, the mycobacterial genomic DNA wasisolated [1] from thymidine-labeled mycobacteria, its purity waschecked by absorption at 260/280nm and 1% agarose electrophoresiswas performed. The radioactivity was monitored in the DNA.

The gel retardation experiment was performed by the method ofPark et al. [10]. 100ng ofM. tuberculosisH37Ra genomic DNA wasCorrespondence to:G.K. Khuller; email: khullergk@usa.net

CURRENT MICROBIOLOGY Vol. 43 (2001), pp. 74–76DOI: 10.1007/s002840010263 Current

MicrobiologyAn International Journal© Springer-Verlag New York Inc. 2001

incubated for 2 h with increasing amount of HNP-1 in binding bufferto give different ratios of peptide: DNA. To this reaction mixture, 4mlof native loading buffer was added and an aliquot of 12ml was appliedto 1% agarose gel electrophoresis. Histone and syntide were used aspositive and negative controls respectively.

Results and Discussion

To monitor the interaction of HNP-1 with theM. tuber-culosisH37Ra cells, binding kinetics was performed byincubating the radiolabelled HNP-1 with mycobacterialcells. The binding was rapid and biphasic, consisting ofprimary and secondary phases separated by a shoulder at1.5 min (Fig. 1). Further, when the binding was per-formed at 0°C, where the cells are relatively inactive, thebinding was slow and the extent of binding was less ascompared to that at 37°C. Earlier, inCandida albicans,Lehrer et al. [7] reported a shoulder at 2.5 min betweenprimary and secondary phases. The primary rapid phaseindicates binding of HNP-1 to externally located acces-sible binding sites, whereas the secondary phase of bind-ing may reflect the interaction of HNP-1 with the other-wise inaccessible intracellular target, which becomesaccessible only after membrane permeabilization. Thiscontention is being supported by our earlier observationswherein HNP-1 has been demonstrated to increase thepermeability of the plasma membrane ofM. tuberculosisH37Ra [12].

The maximum radioactivity (43.5% of bound cpm)was associated with plasma membrane, whereas 21.34%of bound cpm was recovered from cell wall. However, alower amount of radioactivity was also present inside thecytosol (12.07% of the bound cpm). These results indi-cated plasma membrane/cell wall as major binding siteof HNP-1 and presence of HNP-1 inside the cytosol

suggested that HNP-1 may have an intracellular target incytosol. The available information based on biochemical,structural and electron microscopic studies have shownthe membrane to be the primary target of defensins [8,13]. DNA has been hypothesized as a secondary intra-cellular target of HNP-1 [12] and buforin II, anotherantimicrobial peptide [10]. Among various macromole-cules, maximum inhibition was observed in DNA bio-synthesis (74.46 2.4%), followed by lipids (45%), RNA(42%) and protein (39%) in HNP-1-treated cells, ascompared to control. Further, [3H]-thymidine was foundto be specific for DNA, as more than 80% radioactivitywas recovered in the DNA after thymidine incorporation,while 93% of the [14C]-uracil was found to be incorpo-rated in RNA suggesting the specificities of these radio-labelled precursors.

The results of macromolecular biosynthesis sug-gested the interaction of HNP-1 with mycobacterialDNA. To further substantiate this hypothesis, the bindingof HNP-1 to mycobacterial DNA was monitored by gelretardation assay. The migration of mycobacterialgenomic DNA was completely inhibited at HNP-1 toDNA weight ratio of 50 (Fig. 2, lane 3). No visibleinhibition in the migration of DNA at peptide to DNAweight ratio of 10 was observed (Fig. 2, lane 4). Histone,

Fig. 1. Binding kinetics of HNP-1 inMycobacterium tuberculosisH37Ra: 33 108 cells/ml were incubated with [125I] HNP-1 (3.03 104

cpm) in 1 ml of 10mM phosphate buffer (pH 7.4) up to 60 min at 37°C.Values are the mean of three independent experiments.

Fig. 2. Gel retardation assay: HNP-1 was incubated with mycobacterialgenomic DNA at room temperature for 2 h, at various ratio 1:10 (lane4), 1:50 (lane 3), DNA alone (lane 5), Histone 1:10 (lane 2) and syntide1:10 (lane 1). The reaction mixture was applied to 1% agarose gelelectrophoresis. Inhibition in the migration of DNA was monitored.

S. Sharma and G.K. Khuller: DNA andM. tuberculosisH37 Ra 75

a known DNA binding protein, was used as a positivecontrol. It also resulted in the complete inhibition of themigration of DNA (Fig. 2, lane 2). Syntide, a positivelycharged peptide used as a negative control, did not affectthe mobility of DNA (Fig. 2, lane 1), thus indicating thatthe binding of HNP-1 to DNA was specific and notmerely due to charged interactions. Earlier, buforin IIhad also been shown to bind to DNA by gel retardationassay [2]. A shift in the absorption spectrum of myco-bacterial DNA on HNP-1 treatment and smearing ofDNA isolated from HNP-1-grownM. tuberculosisH37Ra cells indicate oligonucleosomal damage to DNAby HNP-1 (data not shown) thus, substantiating the find-ing of gel retardation assay. These results clearly suggestthe interaction between HNP-1 and the DNA ofM.tuberculosisin accordance with earlier reports for pro-line arginine rich peptide PR-39 [2], tachyplesin-1 [15]and buforin II [10].

It can be concluded from this study that HNP-1 killsM.tuberculosisH37Ra cells through a cascade of events in-volving its binding to a primary target, i.e., plasma mem-brane / cell wall, followed by permeabilization, and its entryinto cytosol where it inhibits the cellular functions by bind-ing electrostatically to polyanionic molecules like DNA.Our data suggest DNA to be one of the intracellular targetsof HNP-1 action, leading to ultimate cell death.

ACKNOWLEDGMENTS

This work was financed by a grant from the Indian Council of MedicalResearch, New Delhi to G.K.K. Authors also thank Dr. N. Siva Prasadof Radiopharmaceutical and Labelled Compounds (RPLC), BRIT,Mumbai, India for providing facilities to label HNP-1 with Na125I.

Literature Cited1. Ausubel FM, Brant R, Kingston RE, Moore DD, Seidman JG,

Smith JA, Strohl K (1988) Current protocols in molecular biology:Wiley, NY

2. Boman HG, Agerberth SB, Boman A (1993) Mechanism of actionon Escherichia coliof cecropin PI and PR 39 two antibacterialfrom pig intestine. Infect Immun 61:2978–2984

3. Ganz T (1987) Extracellular release of antimicrobial defensins bypolymorphonuclear leukocytes. Infect Immun 55:568–571

4. Ganz T, Lehrer RI/(1999) Antibiotic peptides from higher eu-karyotes: Biology and applications. Mol Med Today 5:292–297

5. Goldman DS (1970) Subcellular localization of individual man-nose containing phospholipids inMycobacterium tuberculosis. AmRev Respir Dis 102:543–555

6. Hunter WM, Greenwood FC (1962) Preparation of I–131 labelledhuman growth hormone of high specific activity. Nature 194:495–496

7. Lehrer RI, Szklarek D, Ganz T, Selsted ME (1985) Correlation ofbinding of rabbit granulocyte peptides toCandida albicanswithcandidacidal activity. Infect Immun 49:207–211

8. Lehrer RI, Barton A, Daher KA, Harwig SSL, Ganz T, Selsted ME(1989) Interaction of human defensins withEscherichia coli.Mechanism of bactericidal activity. J Clin Invest 84:553–561

9. Ogata K, Lizer BA, Zuberi RI, Ganz T, Lehrer RI, Catanzaro A(1992) Activity of defensins from human neutrophilic granulocytesagainstMycobacterium avium–Mycobacterium intracellulare. In-fect Immun 60:4720–4725

10. Park CB, Kim HS, Kim SC (1998) Mechanism of action ofantimicrobial peptide buforin II: Buforin II kills microorganismsby penetrating the cell membrane and inhibiting cellular functions.Biochem Biophys Res Common 244:253–257

11. Selsted ME, Szklarek D, Ganz T, Lehrer RI (1985) Activity ofrabbit leukocyte peptides againstCandida albicans. Infect Immun49:202–206

12. Sharma S, Verma I, Khuller GK: (1999) Biochemical interaction ofhuman neutrophil peptide -1 withMycobacterium tuberculosisH37Ra. Arch Microbiol 171: 338–342

13. Shimoda M, Ohki K, Shimamoto Y, Kohashi O (1995) Morphol-ogy of defensin treatedStaphylococcus aureus. Infect Immun63:2886–2891

14. Subrahmanyam D (1964) Studies on the polyglycerophosphatidesof Mycobacterium tuberculosis. Can J Biochem 42:1195–1201

15. Yonizawa A, Kuwahara J, Fuji N, Sugiura Y/(1992) Binding oftachyplesin 1 to DNA revealed by foot printing analysis: Signifi-cant contribution of secondary structure to DNA binding andimplication for biological action. Biochemistry 31: 2998–3004

76 CURRENT MICROBIOLOGY Vol. 43 (2001)