- Email: [email protected]
Inhibition of nuclear import by backbone cyclic peptidomimetics derived from the HIV-1 MA NLS sequence
Biochimica et Biophysica Acta 1594 (2002) 234^242 www.bba-direct.com
Inhibition of nuclear import by backbone cyclic peptidomimetics derived from the HIV-1 MA NLS sequence Elana Hariton-Gazal a , Dorit Friedler b , Assaf Friedler a;1 , Nehama Zakai b , Chaim Gilon a , Abraham Loyter b; * b
a Department of Organic Chemistry, Institute of Chemistry, The Hebrew University of Jerusalem, Jerusalem 91904, Israel Department of Biological Chemistry, The Alexander Silberman Institute of Life Sciences, The Hebrew University of Jerusalem, Jerusalem 91904, Israel
Received 30 August 2001; received in revised form 25 October 2001; accepted 31 October 2001
Abstract In the present work we have constructed a series of backbone cyclic peptides, which differed in the amino acid residues located at the C-terminal position of the previously described BCvir peptide (A. Friedler, N. Zakai, O. Karni, Y.C. Broder, L. Baraz, M. Kotler, A. Loyter, C. Gilon, Biochemistry 37 (1998)). BCvir is a cyclic peptide, derived from the nuclear localization signal (NLS) of the human immunodeficiency virus type 1 matrix protein. The majority of the cyclic peptides described here inhibited nuclear import in vitro. The most potent inhibitors were those bearing bulky hydrophobic amino acids such as Leu, Phe or Nal (naphthyl Ala) at the C-terminus. On the other hand, peptides bearing polar amino acid residues such as Asn, Cys or a reduced amide bond were not inhibitory. The present studies demonstrate the importance of a bulky hydrophobic C-terminal side chain and an exocyclic amide bond preceding it, to the inhibitory activity of the NLSderived BC peptides. Being only inhibitory, these BC peptides resemble classic receptor antagonists. ß 2002 Elsevier Science B.V. All rights reserved. Keywords: Antagonist; Backbone cyclization; Importin K; Matrix protein; Nuclear localization signal; Nuclear import
1. Introduction Abbreviations: BC, backbone cyclic; Boc, t-butyloxycarbonyl; BSA, bovine serum albumin; Bzl, benzyl; DCM, dichloromethane; DIEA, diisopropylethylamine; DMF, dimethylformamide; FCS, fetal calf serum; Fmoc, £uorenylmethoxycarbonyl; HIV-1, human immunode¢ciency virus type 1; IC50 , inhibitory concentration at 50%; MA, matrix protein; MBHA, 4-methyl benzhydrylamine; MeLeu, methyl leucine; Nal, naphthyl alanine; NLS, nuclear localization signal; NMP, 1-methyl-2-pyrrolidinone; NMR, nuclear magnetic resonance; Trt, trityl; Z, benzyloxycarbonyl * Corresponding author. Fax: +972-2-6586-448. E-mail address: [email protected] (A. Loyter). 1 Present address: Center for Protein Engineering, Hills Road, Cambridge CB2 2QH, UK.
Nuclear import is a crucial step in the life cycle of many viruses, among them the HIV-1 [1^4]. The preintegration complex of the HIV-1 (whose nuclear entry is mediated by the viral matrix proteins (MA), Vpr and the Integrase proteins [5^8]) and several viral auxiliary proteins has to be imported into the nuclei of infected cells in order to allow completion of the viral life cycle [9,10]. Thus, inhibitors which will block entry of the viral karyophilic proteins should be of therapeutic potential. Indeed, linear peptides derived from the HIV-1 MA ^ which uses the importin K nuclear import pathway ^ have
0167-4838 / 02 / $ ^ see front matter ß 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 8 3 8 ( 0 1 ) 0 0 3 0 6 - 5
BBAPRO 36530 7-3-02
E. Hariton-Gazal et al. / Biochimica et Biophysica Acta 1594 (2002) 234^242
been shown to inhibit HIV-1 replication in cultured cells [11^15]. However, being conformationally £exible, linear nuclear localization signal (NLS) peptides are highly susceptible to proteolysis and therefore their therapeutic use is limited. Introduction of conformational constraints, e.g. via cyclization, has been shown to increase their resistance to proteolysis as well as to improve their selectivity. `Backbone cyclization' [16,17] of peptides is the method used in our lab for the generation of potent selective and metabolically stable peptido- and proteinomimetics [18,19]. In order to select the most active backbone cyclic (BC) peptide derived from a given sequence, a library of BC peptides is generated and screened in the appropriate biological assay. Previously we have demonstrated that a BC peptide derived from the HIV-1 MA NLS and designated by us `BCvir', e¤ciently inhibited nuclear import of NLS-bovine serum albumin (BSA) conjugates in an in vitro assay system [20]. Being conformationally constraint, the BC NLS mimetic peptides give us a unique opportunity to study how modi¢cation of the side chain as well as peptide conformation a¡ect the inhibitory activity. Such studies may be of substantial contribution for the development of potent inhibitors of nuclear uptake. In the present work we have studied, in detail, the in£uence of the amino acid residue at the C-terminal position of the NLS within the BCvir peptide [20] on the inhibitory activity of the BC MA NLS mimetic peptides. We have discovered a repertoire of BC peptides, which inhibited and did not mediate, nuclear import in vitro, exhibiting classical antagonistic properties. 2. Materials and methods 2.1. Chemicals Protected amino acids, 4-methyl benzhydrylamine (MBHA) resin and coupling reagents were purchased from Nova Biochem (Laufel¢ngen, Switzerland). Other chemicals were purchased from Sigma (St. Louis, MO, USA) or Merck (Darmstadt, Germany). Solvents for peptide synthesis were purchased from Baker (Phillipsburg, NJ, USA). Transport bu¡er contained of 20 mM HEPES (pH
235
7.3), 110 mM potassium acetate, 5 mM sodium acetate, 0.5 mM ethylene glycol-bis(K-aminoethyl) N,N,NP,NP-tetraacetic acid, 2 mM dithiothreitol, 1 mg/ml leupeptin, 1 mg/ml pepstatin, 1 mg/ml aprotinin and 0.1 mM PMS. 2.2. Cultured cells (a) Colo-205 (human colon adenocarcinoma cells (ATcc CCL 222)) were maintained in RPMI 1640 growth medium, supplemented with 10% fetal calf serum (FCS), 0.3 g/l L-glutamine, 100 U/ml penicillin and 100 U/ml streptomycin (Beit Haemek, Israel). (b) Monolayers of a culture of HeLa cells were grown in DMEM growth medium supplemented with 10% FCS, 0.3 g/l L-glutamine, 100 U/ml penicillin and 100 U/ml streptomycin (Beit Haemek). 2.3. Synthesis of BC peptides BC peptides were prepared as described before [20]. Brie£y, peptides were synthesized on MBHA resin (loading 0.56 mmol/g) [21]. The peptide series was synthesized by the solid phase multiple peptide synthesis `tea bags' methodology [22], with 300 mg resin portion in each bag. The resin was sealed in 4U5 cm polypropylene fabric bags, which were placed in polypropylene boxes and shaken with a BigBill shaker. The synthesis of peptide E13 (see below) was performed in a manual solid phase peptide synthesis vessel (Supelco), shaken by a MilliGen504 shaker (Millipore). All amino acids were £uorenylmethoxycarbonyl (Fmoc)-protected on the NK except for Val, which was t-butyloxycarbonyl (Boc)-protected. The side chain protecting groups were Lys(benzyloxycarbonyl (Z)), Ser(benzyl (Bzl)), Cys(MeOBzl), Tyr(Bzl), and Asn(trityl (Trt)). All couplings were performed as described before [20], except for coupling of Asn(Trt), which was performed as follows: 3 eq. Fmoc-Asn(Trt) and 5 eq. hydroxybenzotriazole were dissolved in dichloromethane (DCM):dimethylformamide (DMF) (1:1), and 5 eq. diisopropylcarbodiimide were added to the mixture. The preactivation reaction took place at 0³C for 30 min and then the coupling took place for another hour. The Fmoc-NK -[g-(Boc amino)-alkyl]Gly (n = 6) building unit was prepared as described before [23]. Preparation of the reduced amide bond in pep-
BBAPRO 36530 7-3-02
236
E. Hariton-Gazal et al. / Biochimica et Biophysica Acta 1594 (2002) 234^242
Table 1 Analytical data for the BC peptides Peptide No. E1 E2 E3 E4 E5 E6 E7 E8 E9 E10 E11 E12 E13
Amino acid analysis
MW
Found
Calc.
Found
Calc.
Lys:Tyr 3.8:1 Lys:Ser 4.2:1 Lys:Gly 3.4:1 Lys:Ala 3.5:1 Lys:DLeu 3.6:1 Lys:Phe 3.8:1 Lys:Val 4:2.6 Lys:Val 4.6:1 Lys:Asn 4:1 NDa NDa NDa NDa
Lys:Tyr 4:1 Lys:Ser 4:1 Lys:Gly 4:1 Lys:Ala 4:1 Lys:DLeu 4:1 Lys:Phe 4:1 Lys:Val 4:2 Lys:Val 5:1 Lys:Asn 4:1 ^ ^ ^ ^
1162.9 1086 1055.9 1069.8 1111.9 1145.9 1098 1126.9 1112.9 1195.9 1125.9 998.9 994
1162.5 1086.4 1056.4 1071.4 1112.5 1146.5 1098 1127 1113.3 1195.9 1126.5 999.3 994.7
For experimental details of peptide synthesis and peptide structure see Section 2 and Fig. 1. a ND, not determined.
tide E13 was performed on the solid phase as described below. All other experimental conditions, including removal of protecting groups, synthesis of building unit, cleavage of the peptide from the resin, characterization and puri¢cation of the peptides were performed as described before [20]. Analytical HPLC, amino acid analysis and time of £ight mass spectrometry data of the puri¢ed peptides are shown in Table 1. 2.4. Preparation of peptide E13 (Lys5 8(CH2 -NH)Leu6 ) BCvir
water (2U150 ml), dried over magnesium sulfate and concentrated in vacuo, resulting in FmocLys(Z)-N(Me)OMe which separated as a colorless oil [24]. Yield 3.98 g (93%) 1 H-NMR (CDCl3 ): N = 7.2^7.6 (m, 13H, Ar); N = 5.6 (d, 1H, NH urethane); N = 5.1 (s, 2H, CH2 benzyl); N = 4.38 (d, 2H, £uorenyl-CH2 ); N = 4.15 (t, 1H, CH Fmoc); N = 3.7 (s, 3H, OCH3 ); N = 3.1 (s, 3H, NCH3 ); N = 1.4^1.6 (m, 8H, HbbP, HggP, HddP, HeeP). 2.6. Preparation of Fmoc-Lys(Z)-al
Fmoc-Lys(Z)-al was prepared using the modi¢ed procedure of Guichard et al. [24]. 2.5. Synthesis of Fmoc-Lys(Z)-N(Me)OMe Fmoc-Lys(Z)-OH (3.765 g, 7.5 mmol) and benzotriazol-1-yl-oxy-tris-pyrrolidinophosphate (BOP) (3.315 g, 7.5 mmol) were dissolved in 30 ml DMF containing N,O-dimethylhydroxylamine hydrochloride (0.8 g, 8.25 mmol). Then diisopropylethylamine (DIEA) (2.6 ml, 8.25 mmol) was added to the above solution. After 3 h, the reaction mixture was treated with ethyl acetate (180 ml) followed by saturated NaHCO3 solution (360 ml). The organic phase was washed with saturated NaHCO3 solution (2U150 ml), water (2U150 ml), 1M KHSO4 (2U150 ml),
Fmoc-Lys(Z)-N(Me)OMe (3.98 g, 7.2 mmol) was dissolved in 50 ml of dry pre-distilled tetrahydrofuran. The reaction mixture was cooled down to 0³C and stirred under nitrogen. LiAlH4 (532 mg, 8.54 mmol) was added portion-wise and the solution was stirred for 1 h. Then, ethyl acetate (100 ml) and 1 M KHSO4 solution (200 ml) were added and the mixture was stirred for another 30 min. The organic layer was collected and washed with 1 M KHSO4 solution (2U100 ml) followed by a saturated NaCl solution (2U100 ml), dried over magnesium sulfate and the solvent was evaporated in vacuo, resulting in Fmoc-Lys(Z)-al, which was separated as a yellow oil [24]. Yield 2.91 g (75%) 1 H-NMR (CDCl3 ): N = 9.6 (s, 1H, CHO); N = 7.3^7.8 (m, 13H, Ar); N = 5.1
BBAPRO 36530 7-3-02
E. Hariton-Gazal et al. / Biochimica et Biophysica Acta 1594 (2002) 234^242
(s, 2H, CH2 benzyl); N = 4.38 (d, 2H, £uorenyl-CH2 ); N = 4.5 (s, 1H, Ha); N = 4.35 (d, 2H, £uorenyl-CH2 ); N = 4.15 (t, 1H, CH Fmoc); N = 1.2^1.7 (m, 8H, HbbP, HggP, HddP, HeeP). FAB-MS 487 g/mole. 2.7. Preparation of peptide E13 (Lys5 8(CH2 -NH)Leu6 ) BCvir Fmoc-Lys(Z)-al (0.906 g, 4 eq.) was dissolved in a mixture of methanol:1-methyl-2-pyrrolidinone (NMP) (1:1) and was allowed to react with the MBHA peptidyl resin. Acetic acid (1%) was added and following 5 min shaking, NaCNBH3 (0.07 g, 4 eq.) was added and the resulting mixture was shaken for another 3 h. The reduction reaction using NaCNBH3 was repeated under the same conditions. The resin was washed with NMP, DMF, DCM, ethanol and DCM (2U2 min). The reaction progress was monitored by Kaiser and Chloranil tests [25]. 2.8. Protecting the Lys5 8(CH2 -NH)Leu6 bond Z-Cl (120 Wl, 3 eq.) and DIEA (292 Wl, 6 eq.) were dissolved in DMF and added to the peptidyl resin. The mixture was shaken for 1 h. The reaction progress was monitored by Kaiser and Chloranil tests [25]. After Z protection, the Fmoc protecting group was removed and the elongation of the peptide, cyclization and deprotection were performed as described in Section 2.3. 2.9. Quantitative analysis of nuclear import in an in vitro system Nuclear import was quantitatively determined by the ELISA-based assay system, using digitonin permeabilized colon-205 cells, as described before [20,26]. Biotinylated BSA-SV40 T-antigen NLS was used as transport substrate [10]. The results given in the present work are an average of triplicate ELISA determination, whose standard deviation never exceeded þ 20%. 2.10. Estimation of nuclear import by £uorescence microscopy observations HeLa cells were cultivated on 10 mm coverslips to
237
subcon£uent density and then permeabilized with digitonin as described before [27]. The BC peptides BCvir, E6 and E10 and the linear SV40-NLS peptide (PKKKRKVC-NH2 ) were covalently attached to £uorescently (lissamine rhodamine) labeled BSA molecules as described before [28]. Translocation of the resulting £uorescently labeled peptide-BSA conjugates into nuclei of digitonin permeabilized HeLa cells was followed by £uorescent microscopy as described [27]. 2.11. Susceptibility of the peptides to proteolytic digestion A solution containing 0.05 ml trypsin solution (8% w/v in 0.001 M HCl), 0.5 ml TEA bu¡er at pH 7.8 [29] and 0.2 mg of peptide E10 solution in water (initial concentration 2.5 mg/ml) was subjected to HPLC (RP-4 column, Vydac) immediately after addition of the peptide (t = 0). The gradient was 5^90% ACN:triple distilled water (TDW) (containing 0.1% tri£uoroacetic acid) in 40 min. The mixture was incubated at room temperature and samples were injected every 70 min into the HPLC. The same procedure was performed with peptides E13, E6 and SV40-NLS peptide. The susceptibility of the peptides was calculated as described before [20]. 3. Results 3.1. Design and synthesis of a BC peptide series derived from the HIV-1 MA NLS Based on the sequence of the HIV-1 MA NLS (Fig. 1a) we have designed a BC MA NLS mimetic series as described in Fig. 1b, resulting in the peptides whose list is described in Fig. 1c. The Leu residue at the C-terminus of the MA NLS (Fig. 1a) was replaced by the following amino acids: (i) D-Leu, in order to investigate the in£uence of chirality (peptide E5); (ii) polar amino acids: Ser (peptide E2), Lys (peptide E8) and Asn (peptide E9); (iii) non-polar and hydrophobic amino acids: Tyr (peptide E1), Gly (peptide E3), Ala (peptide E4), Phe (peptide E6), Val (peptide E7), naphthyl alanine (Nal; peptide E10) and methyl leucine (MeLeu; peptide E11). (iv) In addition, the peptide bond between the exocyclic
BBAPRO 36530 7-3-02
238
E. Hariton-Gazal et al. / Biochimica et Biophysica Acta 1594 (2002) 234^242
Fig. 1. A general scheme for the design of the NLS mimetic series. (a) The sequence of the HIV-MA NLS. (b) The ¢gure describes schematically the various modi¢cations which were introduced into the MA NLS-derived series. (c) A detailed description of all the peptides in the series. AA refers to panel a. Peptide No. 12: AA was omitted. Peptide No. 13 contains a reduced bond between Lys5 and Leu6 .
Lys5 and the Leu6 residues was replaced by a methylene amino bond (peptide E13) and (v) the Leu residue was omitted and the terminal Cys was directly connected by a peptide bond to the Lys (peptide E12). 3.2. BC peptides with bulky hydrophobic residues possess high inhibitory activity Each of the peptides within the series was screened
for its ability to inhibit nuclear import in permeabilized cells using SV40 NLS-BSA as a transport substrate and the ELISA-based assay system. The SV40 NLS-BSA was used as the transport substrate since similar to the MA NLS, it employs the importin K nuclear transport pathway [5]. As can be seen in Fig. 2, most of the peptides were found to be inhibitory. Peptides E6 (AA = Phe) and E10 (AA = Nal) (see Fig. 1c), which contain bulky hydrophobic residues, were the most potent inhibitors, reaching inhibition values
BBAPRO 36530 7-3-02
E. Hariton-Gazal et al. / Biochimica et Biophysica Acta 1594 (2002) 234^242
239
3.3. The BC peptides fail to mediate nuclear import in vitro
Fig. 2. Screening of the inhibitory activity of the peptides from the BC series. Relative inhibition was determined by the ELISA-based assay in permeabilized cells using SV40 NLS-BSA as a transport substrate (concentration of the BC peptides: 1 mg/ ml). Nuclear import in the absence of peptides was considered 100% (0% inhibition). For further experimental details see Section 2.
of 70% and 90%, respectively, considering the transport observed in the absence of peptides as 100% transport namely zero inhibition. On the other hand, peptides E9 (AA = Asn) and E13 (reduced peptide bond) were not inhibitory at all. All the other peptides displayed intermediate inhibitory values between 20% and 50% (Fig. 2). The IC50 values of the most inhibitory peptide from this series, E10 (Fig. 3), and of E6 (not shown) were found to be 50 WM and 100 WM respectively.
Following the observation that most of the BC peptides obtained by the method described in the present work inhibited nuclear import, it was of interest to ¢nd out whether these peptides will also be able to promote nuclear import, namely act as functional NLS. The most potent inhibitors from this series, E6 and E10, were conjugated to £uorescently labeled BSA via their Cys residues using Sulfo SMCC as a cross-linker [27]. As can be seen (Fig. 4b) the SV40 NLS-BSA conjugate was e¤ciently translocated in the nuclei of permeabilized cells and its translocation nuclear import was characterized by all the features that characterize active nuclear import, namely it was ATP dependent, inhibited by WGA and by free SV40-NLS peptide (data not shown). On the other hand, no nuclear import was observed when the BC-BSA conjugates were studied. Our results clearly show that peptides E10 (Fig. 4a) and E6 (not shown) failed to mediate nuclear import of their BSA conjugates. 3.4. Resistance of the BC peptides to digestion by trypsin The susceptibility of the peptides E6, E10 and E13 to proteolytic degradation by trypsin can re£ect their metabolic stability. As can be seen, E13, the peptide bearing the reduced amide bond, was not hydrolyzed
Fig. 3. Inhibition of nuclear transport of the SV40 NLS-BSA conjugate by BC peptide E10: quantitative studies. The inhibitory concentration at 50% (IC50 ) was determined as described in Section 2.
BBAPRO 36530 7-3-02
240
E. Hariton-Gazal et al. / Biochimica et Biophysica Acta 1594 (2002) 234^242
and remained intact even after 10 h of incubation with trypsin (Fig. 5). Peptides E10 (Fig. 5) and E6 (not shown) were found to be relatively stable, exhibiting a t1=2 of degradation value of 4 h and 3 h respectively. This should be compared to the t1=2 of degradation of about 1.5 h (Fig. 5) for the SV40NLS linear peptide. 4. Discussion In the present work we have described the synthesis and the selection of several BC peptides, all of which are based on the NLS of the HIV-1 MA protein. Most of the newly synthesized peptides were found to be potent inhibitors of nuclear import, but have failed to promote it, namely did not function as conventional NLSs. The extent of nuclear import inhibition was highly dependent on the spe-
Fig. 4. Inhibition of nuclear import: £uorescent microscopy studies. Nuclear import was studied using £uorescently labeled peptide-BSA as described in Section 2. (a) The inability of BC peptide E10 to mediate nuclear import. Similar pictures were obtained with £uorescently labeled BCvir and E6 (not shown). (b) SV40-NLS mediates nuclear import of BSA.
Fig. 5. Susceptibility of peptides E13 and E10 to tryptic digestion. Treatment of E13 (R), E10 (8) and linear SV40-NLS (F) with trypsin was performed as described in Section 2.
ci¢c amino acids located at the C-terminus of the BC peptides. Insertion of a bulky hydrophobic side chain, such as Phe or Nal, resulted in a peptide whose IC50 value reached 50 WM. Introduction of short and hydrophilic side chains, such as Asn, drastically decreased the inhibitory activity. Also replacement of the Leu residue of HIV-1 MA NLS by a D-Leu residue resulted in a signi¢cant decrease of its inhibitory ability. Thus, a hydrophobic L-amino acid residue at the C-terminus is essential for the inhibitory activity of the BC peptides. Introduction of a reduced exocyclic amide bond between the Lys5 and Leu6 residues within one of the BC peptides (E13) resulted in a greater stability to proteolytic digestion but, unexpectedly, caused a total loss of its inhibitory activity. The resulted reduced peptide E13 is identical in its sequence, ring size and ring position to the rest of the inhibitory peptides, except for the replacement of one carbonyl group by a methylene group. This was su¤cient to cause the complete loss of the inhibitory activity of this peptide. It is reasonable to assume therefore that the reduced peptide bond failed to recognize the nuclear import cellular receptor, namely importin K [30,31], since it lacks a critical carbonyl group, which was shown to establish a hydrogen bond with the cellular receptor. Based on the crystal structure [32], it appears that this hydrogen bond keeps the NLS backbone in an extended conformation stage within its binding pocket. This extended conformation is required for the activation of importin K [33]. It appears therefore that in addition to the requirement for hydrophobic amino acid residues, also an
BBAPRO 36530 7-3-02
E. Hariton-Gazal et al. / Biochimica et Biophysica Acta 1594 (2002) 234^242
intact amide bond between the Lys5 and the Leu6 is crucial for obtaining inhibitory activity. The relative importance of the hydrophobic side chain vs. the carbonyl group of the amide bond preceding this side chain can be inferred from the relative inhibitory activities of peptides E11, E12 and E13. Peptide E11 bears an N-methylated amide bond, and peptide E12 has an amide bond but does not possess a hydrophobic side chain. Peptides E11 and E12 were inhibitory, whereas peptide E13, bearing a hydrophobic side chain but not an amide bond, was totally inactive. It may therefore be speculated that the exocyclic amide bond is crucial for possessing inhibitory activity, while the presence of a hydrophobic side chain potentiates this activity. All the BC peptides described in the present work, although being inhibitory, failed to mediate nuclear import, namely to function as active NLSs. Similar results were obtained by us before with the lead peptide BCvir as well as with a peptide derived from the basic region of HIV-1 Vif protein (Vif88-98) [32]. We have de¢ned such sequences as nuclear transport inhibitory signals [34]. Being inhibitory without being able to activate the import pathway, the BC NLSderived peptides function as classical antagonists. Antagonists are considered molecules that bind to the corresponding receptor thus inhibiting its function, but their attachment does not lead to the formation of an active ligand^receptor complex. It is our speculation that the inhibitors described in the present work interact with the appropriate cellular receptors (possibly importin K [30,31]) resulting in the formation of an inactive complex. The question whether the inhibitory BC peptides indeed interact with the cellular receptor importin in the same manner as conventional NLSs and thus showing a typical antagonist behavior is being currently studied in our laboratory. Acknowledgements This work was supported by the Horowitz Foundation and by the German Israeli Foundation (grant No. I-590-105.09/98) (to A.L. and C.G.), and by the DA'AT consortium (to C.G.). The authors wish to thank Carina Hazan for computer modeling and to Gil Fridkin for helpful suggestions.
241
References [1] P. Gallay, V. Stitt, C. Mundy, M. Oettinger, D. Trono, J. Virol. 70 (1996) 1027^1032. [2] P. Gallay, T. Hope, D. Chin, D. Trono, Proc. Natl. Acad. Sci. USA 94 (1997) 9825^9830. [3] E.A. Nigg, Nature 386 (1997) 779^787. [4] J. Garcia-Bustos, J. Heitman, M.N. Hall, Biochim. Biophys. Acta 1071 (1991) 83^101. [5] M.I. Bukrinsky, S. Haggerty, M.P. Dempsey, N. Sharova, A. Adzhubel, L. Spitz, P. Lewis, D. Goldfarb, M. Emerman, M. Stevenson, Nature 365 (1993) 666^669. [6] M.I. Bukrinsky, H.O. K, Front. Biosci. 4 (1999) 772^ 781. [7] S. Popov, M. Rexach, G. Zybarth, N. Reiling, M.A. Lee, L. Ratner, C.M. Lane, M.S. Moore, G. Blobel, M. Bukrinsky, EMBO J. 17 (1998) 909^917. [8] M. Bouyac-Brtoia, J.D. Dvorin, R.A. Fouchier, Y. Jenkins, B.E. Mayer, L.I. Wu, M. Emerman, Mol. Cell 7 (2001) 1025^1035. [9] B.R. Cullen, Cell 93 (1998) 685^692. [10] F. Melchior, L. Gerace, Curr. Opin. Cell Biol. 7 (1995) 310^ 318. [11] R. Lanford, P. Kanda, R. Kennedy, Cell Aug. 15 (1986) 575^582. [12] R.E. Lanford, R.G. White, R.G. Dunham, P. Kanda, Mol. Cell. Biol. 8 (1988) 2722^2729. [13] I. Dworetzky, C.M. Feldherr, J. Cell Biol. 106 (1988) 575^ 584. [14] S.A. Adam, L. Gerace, Cell 66 (1991) 837^847. [15] U. Von-Schwedler, R.S. Kornbluth, D. Trono, Proc. Natl. Acad. Sci. USA 91 (1994) 6992^6996. [16] C. Gilon, I. Zeltser, V. Rashti-Behar, D. Muller, G. Bitan, D. Halle, G. Bar-Akiva, Z. Selinger, G. Byk, in: N. Yanaihara (Ed.), Peptide Chemistry 1992, ESCOM, Leiden, 1993, pp. 482^485. [17] C. Gilon, D. Halle, M. Chorev, Z. Selinger, G. Byk, Biopolymers 31 (1991) 745^750. [18] G. Bitan, I. Sukhotinsky, Y. Mashriki, M. Hanani, Z. Selinger, C. Gilon, J. Pept. Res. 49 (1997) 421. [19] G. Byk, D. Halle, I. Zeltser, G. Bitan, Z. Selinger, C. Gilon, J. Med. Chem. 39 (1996) 3174. [20] A. Friedler, N. Zakai, O. Karni, Y.C. Broder, L. Baraz, M. Kotler, A. Loyter, C. Gilon, Biochemistry 37 (1998) 5616^ 5622. [21] R.B. Merri¢eld, J. Am. Chem. Soc. 85 (1963) 2149^2154. [22] R. Houghten, Proc. Natl. Acad. Sci. USA 82 (1985) 5131^ 5135. [23] R.N. Zuckermann, J.M. Kerr, S.B.H. Kent, W.H. Moos, J. Am. Chem. Soc. 114 (1992) 10646^10647. [24] G. Guichard, J.P. Briand, M. Friede, Pept. Res. 6 (1993) 121^124. [25] E. Kaiser, D.C. Colescold, C.D. Bossinger, P. Cook, Anal. Biochem. 34 (1970) 595. [26] F. Melchior, B. Paschal, J. Evance, L. Gerace, J. Cell Biol. 123 (1993) 1649^1659.
BBAPRO 36530 7-3-02
242
E. Hariton-Gazal et al. / Biochimica et Biophysica Acta 1594 (2002) 234^242
[27] Y.C. Broder, A. Stanhill, N. Zakai, A. Friedler, C. Gilon, A. Loyter, FEBS Lett. 412 (1997) 535^539. [28] D.S. Goldfarb, J. Gariepy, G. Schoolnik, R.D. Kornberg, Nature 322 (1986) 641^644. [29] R. Geiger, H. Fritz, in: H.U. Bergmeyer, J. Bergmeyer, M. Grabl (Eds.), Methods of Enzymatic Analysis, Verlag Chemie, Weinheim, 1984, pp. 119^129. [30] D. Gorlich, E. Vogel, A.D. Mills, E. Hartmann, R.A. Laskey, Nature 377 (1995) 246^248.
[31] D. Gorlich, I.W. Mattaj, Science 271 (1996) 1513^1518. [32] A. Friedler, I. Blumenzweig, L. Baraz, M. Steinitz, M. Kotler, C. Gilon, J. Mol. Biol. 287 (1999) 93^101. [33] E. Conti, M. Uy, L. Leighton, G. Blobel, J. Kuriyan, Cell 94 (1998) 193^204. [34] A. Friedler, N. Zakai, O. Karni, D. Friedler, C. Gilon, A. Loyter, J. Mol. Biol. 289 (1999) 431^437.
BBAPRO 36530 7-3-02
Inhibition of nuclear import by backbone cyclic peptidomimetics derived from the HIV-1 MA NLS sequence Elana Hariton-Gazal a , Dorit Friedler b , Assaf Friedler a;1 , Nehama Zakai b , Chaim Gilon a , Abraham Loyter b; * b
a Department of Organic Chemistry, Institute of Chemistry, The Hebrew University of Jerusalem, Jerusalem 91904, Israel Department of Biological Chemistry, The Alexander Silberman Institute of Life Sciences, The Hebrew University of Jerusalem, Jerusalem 91904, Israel
Received 30 August 2001; received in revised form 25 October 2001; accepted 31 October 2001
Abstract In the present work we have constructed a series of backbone cyclic peptides, which differed in the amino acid residues located at the C-terminal position of the previously described BCvir peptide (A. Friedler, N. Zakai, O. Karni, Y.C. Broder, L. Baraz, M. Kotler, A. Loyter, C. Gilon, Biochemistry 37 (1998)). BCvir is a cyclic peptide, derived from the nuclear localization signal (NLS) of the human immunodeficiency virus type 1 matrix protein. The majority of the cyclic peptides described here inhibited nuclear import in vitro. The most potent inhibitors were those bearing bulky hydrophobic amino acids such as Leu, Phe or Nal (naphthyl Ala) at the C-terminus. On the other hand, peptides bearing polar amino acid residues such as Asn, Cys or a reduced amide bond were not inhibitory. The present studies demonstrate the importance of a bulky hydrophobic C-terminal side chain and an exocyclic amide bond preceding it, to the inhibitory activity of the NLSderived BC peptides. Being only inhibitory, these BC peptides resemble classic receptor antagonists. ß 2002 Elsevier Science B.V. All rights reserved. Keywords: Antagonist; Backbone cyclization; Importin K; Matrix protein; Nuclear localization signal; Nuclear import
1. Introduction Abbreviations: BC, backbone cyclic; Boc, t-butyloxycarbonyl; BSA, bovine serum albumin; Bzl, benzyl; DCM, dichloromethane; DIEA, diisopropylethylamine; DMF, dimethylformamide; FCS, fetal calf serum; Fmoc, £uorenylmethoxycarbonyl; HIV-1, human immunode¢ciency virus type 1; IC50 , inhibitory concentration at 50%; MA, matrix protein; MBHA, 4-methyl benzhydrylamine; MeLeu, methyl leucine; Nal, naphthyl alanine; NLS, nuclear localization signal; NMP, 1-methyl-2-pyrrolidinone; NMR, nuclear magnetic resonance; Trt, trityl; Z, benzyloxycarbonyl * Corresponding author. Fax: +972-2-6586-448. E-mail address: [email protected] (A. Loyter). 1 Present address: Center for Protein Engineering, Hills Road, Cambridge CB2 2QH, UK.
Nuclear import is a crucial step in the life cycle of many viruses, among them the HIV-1 [1^4]. The preintegration complex of the HIV-1 (whose nuclear entry is mediated by the viral matrix proteins (MA), Vpr and the Integrase proteins [5^8]) and several viral auxiliary proteins has to be imported into the nuclei of infected cells in order to allow completion of the viral life cycle [9,10]. Thus, inhibitors which will block entry of the viral karyophilic proteins should be of therapeutic potential. Indeed, linear peptides derived from the HIV-1 MA ^ which uses the importin K nuclear import pathway ^ have
0167-4838 / 02 / $ ^ see front matter ß 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 8 3 8 ( 0 1 ) 0 0 3 0 6 - 5
BBAPRO 36530 7-3-02
E. Hariton-Gazal et al. / Biochimica et Biophysica Acta 1594 (2002) 234^242
been shown to inhibit HIV-1 replication in cultured cells [11^15]. However, being conformationally £exible, linear nuclear localization signal (NLS) peptides are highly susceptible to proteolysis and therefore their therapeutic use is limited. Introduction of conformational constraints, e.g. via cyclization, has been shown to increase their resistance to proteolysis as well as to improve their selectivity. `Backbone cyclization' [16,17] of peptides is the method used in our lab for the generation of potent selective and metabolically stable peptido- and proteinomimetics [18,19]. In order to select the most active backbone cyclic (BC) peptide derived from a given sequence, a library of BC peptides is generated and screened in the appropriate biological assay. Previously we have demonstrated that a BC peptide derived from the HIV-1 MA NLS and designated by us `BCvir', e¤ciently inhibited nuclear import of NLS-bovine serum albumin (BSA) conjugates in an in vitro assay system [20]. Being conformationally constraint, the BC NLS mimetic peptides give us a unique opportunity to study how modi¢cation of the side chain as well as peptide conformation a¡ect the inhibitory activity. Such studies may be of substantial contribution for the development of potent inhibitors of nuclear uptake. In the present work we have studied, in detail, the in£uence of the amino acid residue at the C-terminal position of the NLS within the BCvir peptide [20] on the inhibitory activity of the BC MA NLS mimetic peptides. We have discovered a repertoire of BC peptides, which inhibited and did not mediate, nuclear import in vitro, exhibiting classical antagonistic properties. 2. Materials and methods 2.1. Chemicals Protected amino acids, 4-methyl benzhydrylamine (MBHA) resin and coupling reagents were purchased from Nova Biochem (Laufel¢ngen, Switzerland). Other chemicals were purchased from Sigma (St. Louis, MO, USA) or Merck (Darmstadt, Germany). Solvents for peptide synthesis were purchased from Baker (Phillipsburg, NJ, USA). Transport bu¡er contained of 20 mM HEPES (pH
235
7.3), 110 mM potassium acetate, 5 mM sodium acetate, 0.5 mM ethylene glycol-bis(K-aminoethyl) N,N,NP,NP-tetraacetic acid, 2 mM dithiothreitol, 1 mg/ml leupeptin, 1 mg/ml pepstatin, 1 mg/ml aprotinin and 0.1 mM PMS. 2.2. Cultured cells (a) Colo-205 (human colon adenocarcinoma cells (ATcc CCL 222)) were maintained in RPMI 1640 growth medium, supplemented with 10% fetal calf serum (FCS), 0.3 g/l L-glutamine, 100 U/ml penicillin and 100 U/ml streptomycin (Beit Haemek, Israel). (b) Monolayers of a culture of HeLa cells were grown in DMEM growth medium supplemented with 10% FCS, 0.3 g/l L-glutamine, 100 U/ml penicillin and 100 U/ml streptomycin (Beit Haemek). 2.3. Synthesis of BC peptides BC peptides were prepared as described before [20]. Brie£y, peptides were synthesized on MBHA resin (loading 0.56 mmol/g) [21]. The peptide series was synthesized by the solid phase multiple peptide synthesis `tea bags' methodology [22], with 300 mg resin portion in each bag. The resin was sealed in 4U5 cm polypropylene fabric bags, which were placed in polypropylene boxes and shaken with a BigBill shaker. The synthesis of peptide E13 (see below) was performed in a manual solid phase peptide synthesis vessel (Supelco), shaken by a MilliGen504 shaker (Millipore). All amino acids were £uorenylmethoxycarbonyl (Fmoc)-protected on the NK except for Val, which was t-butyloxycarbonyl (Boc)-protected. The side chain protecting groups were Lys(benzyloxycarbonyl (Z)), Ser(benzyl (Bzl)), Cys(MeOBzl), Tyr(Bzl), and Asn(trityl (Trt)). All couplings were performed as described before [20], except for coupling of Asn(Trt), which was performed as follows: 3 eq. Fmoc-Asn(Trt) and 5 eq. hydroxybenzotriazole were dissolved in dichloromethane (DCM):dimethylformamide (DMF) (1:1), and 5 eq. diisopropylcarbodiimide were added to the mixture. The preactivation reaction took place at 0³C for 30 min and then the coupling took place for another hour. The Fmoc-NK -[g-(Boc amino)-alkyl]Gly (n = 6) building unit was prepared as described before [23]. Preparation of the reduced amide bond in pep-
BBAPRO 36530 7-3-02
236
E. Hariton-Gazal et al. / Biochimica et Biophysica Acta 1594 (2002) 234^242
Table 1 Analytical data for the BC peptides Peptide No. E1 E2 E3 E4 E5 E6 E7 E8 E9 E10 E11 E12 E13
Amino acid analysis
MW
Found
Calc.
Found
Calc.
Lys:Tyr 3.8:1 Lys:Ser 4.2:1 Lys:Gly 3.4:1 Lys:Ala 3.5:1 Lys:DLeu 3.6:1 Lys:Phe 3.8:1 Lys:Val 4:2.6 Lys:Val 4.6:1 Lys:Asn 4:1 NDa NDa NDa NDa
Lys:Tyr 4:1 Lys:Ser 4:1 Lys:Gly 4:1 Lys:Ala 4:1 Lys:DLeu 4:1 Lys:Phe 4:1 Lys:Val 4:2 Lys:Val 5:1 Lys:Asn 4:1 ^ ^ ^ ^
1162.9 1086 1055.9 1069.8 1111.9 1145.9 1098 1126.9 1112.9 1195.9 1125.9 998.9 994
1162.5 1086.4 1056.4 1071.4 1112.5 1146.5 1098 1127 1113.3 1195.9 1126.5 999.3 994.7
For experimental details of peptide synthesis and peptide structure see Section 2 and Fig. 1. a ND, not determined.
tide E13 was performed on the solid phase as described below. All other experimental conditions, including removal of protecting groups, synthesis of building unit, cleavage of the peptide from the resin, characterization and puri¢cation of the peptides were performed as described before [20]. Analytical HPLC, amino acid analysis and time of £ight mass spectrometry data of the puri¢ed peptides are shown in Table 1. 2.4. Preparation of peptide E13 (Lys5 8(CH2 -NH)Leu6 ) BCvir
water (2U150 ml), dried over magnesium sulfate and concentrated in vacuo, resulting in FmocLys(Z)-N(Me)OMe which separated as a colorless oil [24]. Yield 3.98 g (93%) 1 H-NMR (CDCl3 ): N = 7.2^7.6 (m, 13H, Ar); N = 5.6 (d, 1H, NH urethane); N = 5.1 (s, 2H, CH2 benzyl); N = 4.38 (d, 2H, £uorenyl-CH2 ); N = 4.15 (t, 1H, CH Fmoc); N = 3.7 (s, 3H, OCH3 ); N = 3.1 (s, 3H, NCH3 ); N = 1.4^1.6 (m, 8H, HbbP, HggP, HddP, HeeP). 2.6. Preparation of Fmoc-Lys(Z)-al
Fmoc-Lys(Z)-al was prepared using the modi¢ed procedure of Guichard et al. [24]. 2.5. Synthesis of Fmoc-Lys(Z)-N(Me)OMe Fmoc-Lys(Z)-OH (3.765 g, 7.5 mmol) and benzotriazol-1-yl-oxy-tris-pyrrolidinophosphate (BOP) (3.315 g, 7.5 mmol) were dissolved in 30 ml DMF containing N,O-dimethylhydroxylamine hydrochloride (0.8 g, 8.25 mmol). Then diisopropylethylamine (DIEA) (2.6 ml, 8.25 mmol) was added to the above solution. After 3 h, the reaction mixture was treated with ethyl acetate (180 ml) followed by saturated NaHCO3 solution (360 ml). The organic phase was washed with saturated NaHCO3 solution (2U150 ml), water (2U150 ml), 1M KHSO4 (2U150 ml),
Fmoc-Lys(Z)-N(Me)OMe (3.98 g, 7.2 mmol) was dissolved in 50 ml of dry pre-distilled tetrahydrofuran. The reaction mixture was cooled down to 0³C and stirred under nitrogen. LiAlH4 (532 mg, 8.54 mmol) was added portion-wise and the solution was stirred for 1 h. Then, ethyl acetate (100 ml) and 1 M KHSO4 solution (200 ml) were added and the mixture was stirred for another 30 min. The organic layer was collected and washed with 1 M KHSO4 solution (2U100 ml) followed by a saturated NaCl solution (2U100 ml), dried over magnesium sulfate and the solvent was evaporated in vacuo, resulting in Fmoc-Lys(Z)-al, which was separated as a yellow oil [24]. Yield 2.91 g (75%) 1 H-NMR (CDCl3 ): N = 9.6 (s, 1H, CHO); N = 7.3^7.8 (m, 13H, Ar); N = 5.1
BBAPRO 36530 7-3-02
E. Hariton-Gazal et al. / Biochimica et Biophysica Acta 1594 (2002) 234^242
(s, 2H, CH2 benzyl); N = 4.38 (d, 2H, £uorenyl-CH2 ); N = 4.5 (s, 1H, Ha); N = 4.35 (d, 2H, £uorenyl-CH2 ); N = 4.15 (t, 1H, CH Fmoc); N = 1.2^1.7 (m, 8H, HbbP, HggP, HddP, HeeP). FAB-MS 487 g/mole. 2.7. Preparation of peptide E13 (Lys5 8(CH2 -NH)Leu6 ) BCvir Fmoc-Lys(Z)-al (0.906 g, 4 eq.) was dissolved in a mixture of methanol:1-methyl-2-pyrrolidinone (NMP) (1:1) and was allowed to react with the MBHA peptidyl resin. Acetic acid (1%) was added and following 5 min shaking, NaCNBH3 (0.07 g, 4 eq.) was added and the resulting mixture was shaken for another 3 h. The reduction reaction using NaCNBH3 was repeated under the same conditions. The resin was washed with NMP, DMF, DCM, ethanol and DCM (2U2 min). The reaction progress was monitored by Kaiser and Chloranil tests [25]. 2.8. Protecting the Lys5 8(CH2 -NH)Leu6 bond Z-Cl (120 Wl, 3 eq.) and DIEA (292 Wl, 6 eq.) were dissolved in DMF and added to the peptidyl resin. The mixture was shaken for 1 h. The reaction progress was monitored by Kaiser and Chloranil tests [25]. After Z protection, the Fmoc protecting group was removed and the elongation of the peptide, cyclization and deprotection were performed as described in Section 2.3. 2.9. Quantitative analysis of nuclear import in an in vitro system Nuclear import was quantitatively determined by the ELISA-based assay system, using digitonin permeabilized colon-205 cells, as described before [20,26]. Biotinylated BSA-SV40 T-antigen NLS was used as transport substrate [10]. The results given in the present work are an average of triplicate ELISA determination, whose standard deviation never exceeded þ 20%. 2.10. Estimation of nuclear import by £uorescence microscopy observations HeLa cells were cultivated on 10 mm coverslips to
237
subcon£uent density and then permeabilized with digitonin as described before [27]. The BC peptides BCvir, E6 and E10 and the linear SV40-NLS peptide (PKKKRKVC-NH2 ) were covalently attached to £uorescently (lissamine rhodamine) labeled BSA molecules as described before [28]. Translocation of the resulting £uorescently labeled peptide-BSA conjugates into nuclei of digitonin permeabilized HeLa cells was followed by £uorescent microscopy as described [27]. 2.11. Susceptibility of the peptides to proteolytic digestion A solution containing 0.05 ml trypsin solution (8% w/v in 0.001 M HCl), 0.5 ml TEA bu¡er at pH 7.8 [29] and 0.2 mg of peptide E10 solution in water (initial concentration 2.5 mg/ml) was subjected to HPLC (RP-4 column, Vydac) immediately after addition of the peptide (t = 0). The gradient was 5^90% ACN:triple distilled water (TDW) (containing 0.1% tri£uoroacetic acid) in 40 min. The mixture was incubated at room temperature and samples were injected every 70 min into the HPLC. The same procedure was performed with peptides E13, E6 and SV40-NLS peptide. The susceptibility of the peptides was calculated as described before [20]. 3. Results 3.1. Design and synthesis of a BC peptide series derived from the HIV-1 MA NLS Based on the sequence of the HIV-1 MA NLS (Fig. 1a) we have designed a BC MA NLS mimetic series as described in Fig. 1b, resulting in the peptides whose list is described in Fig. 1c. The Leu residue at the C-terminus of the MA NLS (Fig. 1a) was replaced by the following amino acids: (i) D-Leu, in order to investigate the in£uence of chirality (peptide E5); (ii) polar amino acids: Ser (peptide E2), Lys (peptide E8) and Asn (peptide E9); (iii) non-polar and hydrophobic amino acids: Tyr (peptide E1), Gly (peptide E3), Ala (peptide E4), Phe (peptide E6), Val (peptide E7), naphthyl alanine (Nal; peptide E10) and methyl leucine (MeLeu; peptide E11). (iv) In addition, the peptide bond between the exocyclic
BBAPRO 36530 7-3-02
238
E. Hariton-Gazal et al. / Biochimica et Biophysica Acta 1594 (2002) 234^242
Fig. 1. A general scheme for the design of the NLS mimetic series. (a) The sequence of the HIV-MA NLS. (b) The ¢gure describes schematically the various modi¢cations which were introduced into the MA NLS-derived series. (c) A detailed description of all the peptides in the series. AA refers to panel a. Peptide No. 12: AA was omitted. Peptide No. 13 contains a reduced bond between Lys5 and Leu6 .
Lys5 and the Leu6 residues was replaced by a methylene amino bond (peptide E13) and (v) the Leu residue was omitted and the terminal Cys was directly connected by a peptide bond to the Lys (peptide E12). 3.2. BC peptides with bulky hydrophobic residues possess high inhibitory activity Each of the peptides within the series was screened
for its ability to inhibit nuclear import in permeabilized cells using SV40 NLS-BSA as a transport substrate and the ELISA-based assay system. The SV40 NLS-BSA was used as the transport substrate since similar to the MA NLS, it employs the importin K nuclear transport pathway [5]. As can be seen in Fig. 2, most of the peptides were found to be inhibitory. Peptides E6 (AA = Phe) and E10 (AA = Nal) (see Fig. 1c), which contain bulky hydrophobic residues, were the most potent inhibitors, reaching inhibition values
BBAPRO 36530 7-3-02
E. Hariton-Gazal et al. / Biochimica et Biophysica Acta 1594 (2002) 234^242
239
3.3. The BC peptides fail to mediate nuclear import in vitro
Fig. 2. Screening of the inhibitory activity of the peptides from the BC series. Relative inhibition was determined by the ELISA-based assay in permeabilized cells using SV40 NLS-BSA as a transport substrate (concentration of the BC peptides: 1 mg/ ml). Nuclear import in the absence of peptides was considered 100% (0% inhibition). For further experimental details see Section 2.
of 70% and 90%, respectively, considering the transport observed in the absence of peptides as 100% transport namely zero inhibition. On the other hand, peptides E9 (AA = Asn) and E13 (reduced peptide bond) were not inhibitory at all. All the other peptides displayed intermediate inhibitory values between 20% and 50% (Fig. 2). The IC50 values of the most inhibitory peptide from this series, E10 (Fig. 3), and of E6 (not shown) were found to be 50 WM and 100 WM respectively.
Following the observation that most of the BC peptides obtained by the method described in the present work inhibited nuclear import, it was of interest to ¢nd out whether these peptides will also be able to promote nuclear import, namely act as functional NLS. The most potent inhibitors from this series, E6 and E10, were conjugated to £uorescently labeled BSA via their Cys residues using Sulfo SMCC as a cross-linker [27]. As can be seen (Fig. 4b) the SV40 NLS-BSA conjugate was e¤ciently translocated in the nuclei of permeabilized cells and its translocation nuclear import was characterized by all the features that characterize active nuclear import, namely it was ATP dependent, inhibited by WGA and by free SV40-NLS peptide (data not shown). On the other hand, no nuclear import was observed when the BC-BSA conjugates were studied. Our results clearly show that peptides E10 (Fig. 4a) and E6 (not shown) failed to mediate nuclear import of their BSA conjugates. 3.4. Resistance of the BC peptides to digestion by trypsin The susceptibility of the peptides E6, E10 and E13 to proteolytic degradation by trypsin can re£ect their metabolic stability. As can be seen, E13, the peptide bearing the reduced amide bond, was not hydrolyzed
Fig. 3. Inhibition of nuclear transport of the SV40 NLS-BSA conjugate by BC peptide E10: quantitative studies. The inhibitory concentration at 50% (IC50 ) was determined as described in Section 2.
BBAPRO 36530 7-3-02
240
E. Hariton-Gazal et al. / Biochimica et Biophysica Acta 1594 (2002) 234^242
and remained intact even after 10 h of incubation with trypsin (Fig. 5). Peptides E10 (Fig. 5) and E6 (not shown) were found to be relatively stable, exhibiting a t1=2 of degradation value of 4 h and 3 h respectively. This should be compared to the t1=2 of degradation of about 1.5 h (Fig. 5) for the SV40NLS linear peptide. 4. Discussion In the present work we have described the synthesis and the selection of several BC peptides, all of which are based on the NLS of the HIV-1 MA protein. Most of the newly synthesized peptides were found to be potent inhibitors of nuclear import, but have failed to promote it, namely did not function as conventional NLSs. The extent of nuclear import inhibition was highly dependent on the spe-
Fig. 4. Inhibition of nuclear import: £uorescent microscopy studies. Nuclear import was studied using £uorescently labeled peptide-BSA as described in Section 2. (a) The inability of BC peptide E10 to mediate nuclear import. Similar pictures were obtained with £uorescently labeled BCvir and E6 (not shown). (b) SV40-NLS mediates nuclear import of BSA.
Fig. 5. Susceptibility of peptides E13 and E10 to tryptic digestion. Treatment of E13 (R), E10 (8) and linear SV40-NLS (F) with trypsin was performed as described in Section 2.
ci¢c amino acids located at the C-terminus of the BC peptides. Insertion of a bulky hydrophobic side chain, such as Phe or Nal, resulted in a peptide whose IC50 value reached 50 WM. Introduction of short and hydrophilic side chains, such as Asn, drastically decreased the inhibitory activity. Also replacement of the Leu residue of HIV-1 MA NLS by a D-Leu residue resulted in a signi¢cant decrease of its inhibitory ability. Thus, a hydrophobic L-amino acid residue at the C-terminus is essential for the inhibitory activity of the BC peptides. Introduction of a reduced exocyclic amide bond between the Lys5 and Leu6 residues within one of the BC peptides (E13) resulted in a greater stability to proteolytic digestion but, unexpectedly, caused a total loss of its inhibitory activity. The resulted reduced peptide E13 is identical in its sequence, ring size and ring position to the rest of the inhibitory peptides, except for the replacement of one carbonyl group by a methylene group. This was su¤cient to cause the complete loss of the inhibitory activity of this peptide. It is reasonable to assume therefore that the reduced peptide bond failed to recognize the nuclear import cellular receptor, namely importin K [30,31], since it lacks a critical carbonyl group, which was shown to establish a hydrogen bond with the cellular receptor. Based on the crystal structure [32], it appears that this hydrogen bond keeps the NLS backbone in an extended conformation stage within its binding pocket. This extended conformation is required for the activation of importin K [33]. It appears therefore that in addition to the requirement for hydrophobic amino acid residues, also an
BBAPRO 36530 7-3-02
E. Hariton-Gazal et al. / Biochimica et Biophysica Acta 1594 (2002) 234^242
intact amide bond between the Lys5 and the Leu6 is crucial for obtaining inhibitory activity. The relative importance of the hydrophobic side chain vs. the carbonyl group of the amide bond preceding this side chain can be inferred from the relative inhibitory activities of peptides E11, E12 and E13. Peptide E11 bears an N-methylated amide bond, and peptide E12 has an amide bond but does not possess a hydrophobic side chain. Peptides E11 and E12 were inhibitory, whereas peptide E13, bearing a hydrophobic side chain but not an amide bond, was totally inactive. It may therefore be speculated that the exocyclic amide bond is crucial for possessing inhibitory activity, while the presence of a hydrophobic side chain potentiates this activity. All the BC peptides described in the present work, although being inhibitory, failed to mediate nuclear import, namely to function as active NLSs. Similar results were obtained by us before with the lead peptide BCvir as well as with a peptide derived from the basic region of HIV-1 Vif protein (Vif88-98) [32]. We have de¢ned such sequences as nuclear transport inhibitory signals [34]. Being inhibitory without being able to activate the import pathway, the BC NLSderived peptides function as classical antagonists. Antagonists are considered molecules that bind to the corresponding receptor thus inhibiting its function, but their attachment does not lead to the formation of an active ligand^receptor complex. It is our speculation that the inhibitors described in the present work interact with the appropriate cellular receptors (possibly importin K [30,31]) resulting in the formation of an inactive complex. The question whether the inhibitory BC peptides indeed interact with the cellular receptor importin in the same manner as conventional NLSs and thus showing a typical antagonist behavior is being currently studied in our laboratory. Acknowledgements This work was supported by the Horowitz Foundation and by the German Israeli Foundation (grant No. I-590-105.09/98) (to A.L. and C.G.), and by the DA'AT consortium (to C.G.). The authors wish to thank Carina Hazan for computer modeling and to Gil Fridkin for helpful suggestions.
241
References [1] P. Gallay, V. Stitt, C. Mundy, M. Oettinger, D. Trono, J. Virol. 70 (1996) 1027^1032. [2] P. Gallay, T. Hope, D. Chin, D. Trono, Proc. Natl. Acad. Sci. USA 94 (1997) 9825^9830. [3] E.A. Nigg, Nature 386 (1997) 779^787. [4] J. Garcia-Bustos, J. Heitman, M.N. Hall, Biochim. Biophys. Acta 1071 (1991) 83^101. [5] M.I. Bukrinsky, S. Haggerty, M.P. Dempsey, N. Sharova, A. Adzhubel, L. Spitz, P. Lewis, D. Goldfarb, M. Emerman, M. Stevenson, Nature 365 (1993) 666^669. [6] M.I. Bukrinsky, H.O. K, Front. Biosci. 4 (1999) 772^ 781. [7] S. Popov, M. Rexach, G. Zybarth, N. Reiling, M.A. Lee, L. Ratner, C.M. Lane, M.S. Moore, G. Blobel, M. Bukrinsky, EMBO J. 17 (1998) 909^917. [8] M. Bouyac-Brtoia, J.D. Dvorin, R.A. Fouchier, Y. Jenkins, B.E. Mayer, L.I. Wu, M. Emerman, Mol. Cell 7 (2001) 1025^1035. [9] B.R. Cullen, Cell 93 (1998) 685^692. [10] F. Melchior, L. Gerace, Curr. Opin. Cell Biol. 7 (1995) 310^ 318. [11] R. Lanford, P. Kanda, R. Kennedy, Cell Aug. 15 (1986) 575^582. [12] R.E. Lanford, R.G. White, R.G. Dunham, P. Kanda, Mol. Cell. Biol. 8 (1988) 2722^2729. [13] I. Dworetzky, C.M. Feldherr, J. Cell Biol. 106 (1988) 575^ 584. [14] S.A. Adam, L. Gerace, Cell 66 (1991) 837^847. [15] U. Von-Schwedler, R.S. Kornbluth, D. Trono, Proc. Natl. Acad. Sci. USA 91 (1994) 6992^6996. [16] C. Gilon, I. Zeltser, V. Rashti-Behar, D. Muller, G. Bitan, D. Halle, G. Bar-Akiva, Z. Selinger, G. Byk, in: N. Yanaihara (Ed.), Peptide Chemistry 1992, ESCOM, Leiden, 1993, pp. 482^485. [17] C. Gilon, D. Halle, M. Chorev, Z. Selinger, G. Byk, Biopolymers 31 (1991) 745^750. [18] G. Bitan, I. Sukhotinsky, Y. Mashriki, M. Hanani, Z. Selinger, C. Gilon, J. Pept. Res. 49 (1997) 421. [19] G. Byk, D. Halle, I. Zeltser, G. Bitan, Z. Selinger, C. Gilon, J. Med. Chem. 39 (1996) 3174. [20] A. Friedler, N. Zakai, O. Karni, Y.C. Broder, L. Baraz, M. Kotler, A. Loyter, C. Gilon, Biochemistry 37 (1998) 5616^ 5622. [21] R.B. Merri¢eld, J. Am. Chem. Soc. 85 (1963) 2149^2154. [22] R. Houghten, Proc. Natl. Acad. Sci. USA 82 (1985) 5131^ 5135. [23] R.N. Zuckermann, J.M. Kerr, S.B.H. Kent, W.H. Moos, J. Am. Chem. Soc. 114 (1992) 10646^10647. [24] G. Guichard, J.P. Briand, M. Friede, Pept. Res. 6 (1993) 121^124. [25] E. Kaiser, D.C. Colescold, C.D. Bossinger, P. Cook, Anal. Biochem. 34 (1970) 595. [26] F. Melchior, B. Paschal, J. Evance, L. Gerace, J. Cell Biol. 123 (1993) 1649^1659.
BBAPRO 36530 7-3-02
242
E. Hariton-Gazal et al. / Biochimica et Biophysica Acta 1594 (2002) 234^242
[27] Y.C. Broder, A. Stanhill, N. Zakai, A. Friedler, C. Gilon, A. Loyter, FEBS Lett. 412 (1997) 535^539. [28] D.S. Goldfarb, J. Gariepy, G. Schoolnik, R.D. Kornberg, Nature 322 (1986) 641^644. [29] R. Geiger, H. Fritz, in: H.U. Bergmeyer, J. Bergmeyer, M. Grabl (Eds.), Methods of Enzymatic Analysis, Verlag Chemie, Weinheim, 1984, pp. 119^129. [30] D. Gorlich, E. Vogel, A.D. Mills, E. Hartmann, R.A. Laskey, Nature 377 (1995) 246^248.
[31] D. Gorlich, I.W. Mattaj, Science 271 (1996) 1513^1518. [32] A. Friedler, I. Blumenzweig, L. Baraz, M. Steinitz, M. Kotler, C. Gilon, J. Mol. Biol. 287 (1999) 93^101. [33] E. Conti, M. Uy, L. Leighton, G. Blobel, J. Kuriyan, Cell 94 (1998) 193^204. [34] A. Friedler, N. Zakai, O. Karni, D. Friedler, C. Gilon, A. Loyter, J. Mol. Biol. 289 (1999) 431^437.
BBAPRO 36530 7-3-02