New Human Papilloma Virus E2 Transcription Factor Mimics: A Tripyrrole-Peptide Conjugate with Tight and Specific DNA-Recognition

Diana E. Wetzler1., Maria J. Comin2., Krzysztof Krajewski3, Mariana Gallo4*

1 Department of Biological Chemistry, FCEN, University of Buenos Aires, Buenos Aires, Argentina, 2 Center for Research and Development in Chemistry, INTI, Buenos Aires, Argentina, 3 Department of Biochemistry and Biophysics, School of Medicine, University of North Carolina, Chapel Hill, North Carolina, United States of America, 4 Fundacio´ n Instituto Leloir, Buenos Aires, Argentina



Abstract

Background: Human papillomavirus (HPV) is the main causative agent of cervical cancer, particularly high risk strains such us HPV-16, -18 and -31. The viral encoded E2 protein acts as a transcriptional modulator and exerts a key role in viral DNA replication. Thus, E2 constitutes an attractive target for developing antiviral agents. E2 is a homodimeric protein that interacts with the DNA target through an a-helix of each monomer. However, a peptide corresponding to the DNA recognition helix of HPV-16 E2 binds DNA with lower affinity than its full-length DNA binding domain. Therefore, in an attempt to promote the DNA binding of the isolated peptide, we have designed a conjugate compound of the E2 a-helix peptide and a derivative of the antibiotic distamycin, which involves simultaneous minor- and major-groove interactions.

Methodology/Principal Findings: An E2 a-helix peptide-distamycin conjugate was designed and synthesized. It was characterized by NMR and CD spectroscopy, and its DNA binding properties were investigated by CD, DNA melting and gel shift experiments. The coupling of E2 peptide with distamycin does not affect its structural properties. The conjugate improves significantly the affinity of the peptide for specific DNA. In addition, stoichiometric amounts of specific DNA increase meaningfully the helical population of the peptide. The conjugate enhances the DNA binding constant 50-fold, maintaining its specificity.

Conclusions/Significance: These results demonstrate that peptide-distamycin conjugates are a promising tool to obtain compounds that bind the E2 target DNA-sequences with remarkable affinity and suggest that a bipartite major/minor groove binding scaffold can be a useful approach for therapeutic treatment of HPV infection.

Citation: Wetzler DE, Comin MJ, Krajewski K, Gallo M (2011) New Human Papilloma Virus E2 Transcription Factor Mimics: A Tripyrrole-Peptide Conjugate with Tight and Specific DNA-Recognition. PLoS ONE 6(7): e22409. doi:10.1371/journal.pone.0022409

Editor: Christopher B. Buck, National Cancer Institute, United States of America

Received March 17, 2011; Accepted June 21, 2011; Published July 25, 2011

Copyright: ß 2011 Wetzler et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: This work was supported by Consejo Nacional de Investigaciones Cientifico y Tecnicas (CONICET), the University of Buenos Aires (UBA), and Agencia Nacional de Promocion Cientifica y Tecnica (ANPCyT). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing Interests: The authors have declared that no competing interests exist.

* E-mail: mgallo@leloir.org.ar

. These authors contributed equally to this work.



Introduction

Cervical cancer is a major cause of cancer-related death in women. Human papillomavirus (HPV) is the main causative agent of cervical cancer, in particular high risk strains such us HPV-16 and -18 followed by HPV-31 and -45 [1]. Despite recent advances in preventive HPV vaccine development, such strategies are unlikely to reduce the prevalence of HPV infections within the next few years, due to their high cost and limited availability in developing countries. In addition, preventive HPV vaccines may not be capable of treating established HPV infections and HPVassociated lesions, which account for high morbidity and mortality worldwide. Thus, it is crucial to develop therapeutic strategies for controlling HPV infection and associated malignancies [2]. HPV E2 protein acts as a transcriptional modulator and participates in viral DNA replication and episomal copy number maintenance of the viral genome [3]. Furthermore, E2 represses the transcription of E6



and E7 oncoproteins, and this ability is lost upon viral DNA integration into the host genome. This event often occurs with the disruption of the E2 open reading frame, leading to the deregulation of the expression of E6 and E7, essential for the transformation process [4]. Hence, E2 constitutes an ideal target for HPV therapy. In particular, molecules capable of binding the HPV E2 DNArecognition sequences with high specificity and affinity may interfere with one key event of the viral life cycle: the interaction between the E2 transcription factor and its target DNA sequences. It is expected that these molecules will act with similar efficiency on all the high-risk strains, as all these viruses have in common highly conserved E2 DNA binding sequences [5]. Furthermore, targeting a conserved DNA sequence, instead of a potentially mutable protein, can overcome the problem of resistant virus generation, a major drawback associated with antiviral agents.

There is a great deal of interest in obtaining small synthetic mimics of transcription factors capable of reproducing their DNA-



PLoS ONE | www.plosone.org



1



July 2011 | Volume 6 | Issue 7 | e22409



HPV E2 Mimics: A Distamycin-Peptide Conjugate



binding properties but also presenting good bioavailability [6,7]. E2 consists of an N-terminal transactivation domain and a Cterminal DNA binding and dimerization domains (E2C), separated by a nonconserved hinge domain [8]. E2C is a symmetric dimer that interacts with the major groove of the target DNA through two a-helices, one of each monomer [8,9,10]. A peptide corresponding to the DNA recognition helix of HPV-16 E2 displays only residual structure and binds DNA with low affinity [11,12]. Recently, it was demonstrated in various biological systems that the appropriate tethering of a monomeric DNAbinding helix to a distamycin (Dst)-like tripyrrole, which binds with moderate affinity the minor groove adjacent to the helix target site, provides for specific binding [13,14,15]. The natural cognate DNAs of HPV-16 and -18 E2 present, in the region adjacent to the DNA-sequence recognized by E2 (CGGT), the sequence AAAT, which is one of the sequences with the highest binding affinities for Dst [16]. Therefore, we decided to explore whether conjugation of a peptide corresponding to the DNA-binding a-helix of the E2 protein from HPV-16 with a Dst derivative could improve the affinity of the peptide for specific DNA. Here we report the design, synthesis, characterization and DNA-binding properties of a conjugate between a peptide corresponding to the a-helix of the HPV-16 E2 protein and a Dst tripyrrole derivative carrying a primary amino function in the tether between both moieties. These results show the potential utility of this kind of hybrids as a new class of molecules for the treatment of HPV infections and are encouraging for the design of E2 conjugates with alternative linkers that can bind DNA with higher specificity and affinity.

Materials and Methods

In this study we have worked with the following compounds: aE2: peptide corresponding to the DNA-binding a-helix of the E2 protein from HPV-16; aE2-Ala: the aE2 peptide but with alanine in positions 299, 302 and 306; aE2-conj: the aE2-Ala peptide Dst conjugate (see Figure 1B below).

Computer modelling The model for the simultaneous interaction of Dst and a peptide

corresponding to the DNA-binding a-helix of the E2 protein with DNA was constructed using the program Insight II and was based on the crystal structures of the complexes of E2C from HPV-18 (PDB 1F9F) [5] and Dst (PDB 1JTL) [17] with their cognates DNAs. Both structures were superimposed by phosphate alignment to get an appropriate arrangement for connecting the two moieties. The linker was designed by using the module Builder of this program. We considered only one of the two possible binding modes of Dst to DNA for designing the conjugate [17].

Synthesis of aE2-conj Linear peptides were synthesized by an automated SPPS (ABI

433A Peptide Synthesizer) using Fmoc chemistry (coupling with HBTU/HOBt/DIPEA in NMP). Purification of the crude conjugate was performed by preparative RP-HPLC (Vydac C18 column). Purified products were further characterized by Kompact Axima-CFR MALDI-TOF mass spectrometer (see Text S1).

DNA Synthesis Single-stranded oligonucleotides were purchased, SDS-PAGE

purified, from Primm srl (Milano, Italy). Double-stranded oligonucleotides were annealed by mixing equal amounts of each strand in 10 mM Bis-Tris-HCl buffer, pH 7.0 and 50 mM NaCl, incubating 5 minutes at 95uC, and slowly cooling to 25uC for



16 hours. Completeness of hybridization was checked by native polyacrylamide gel electrophoresis.

NMR Spectrosocopy The conjugate aE2-conj (3 mM) and the corresponding

isolated peptide, aE2-Ala (KGDANTAKCARYRAKKHA), (7 mM) were dissolved in a 40% TFE-d3 - 60% 20 mM phosphate buffer solution pH 6.5 containing 2 mM DTT and 0.01% NaN3. The NMR experiments were performed at 25uC using a Bruker Avance 700 and Avance 400 spectrometers equipped with triple resonance probes incorporating self-shielded gradient coils. All the heteronuclear correlation experiments were carried out at natural abundance. Pulsed field gradients were appropriately employed to achieve suppression of the solvent signal and spectral artefacts. Quadrature detection in the indirectly detected dimensions was obtained using the States-TPPI method [18] or the echo-antiecho method, and the spectra were processed on Silicon Graphics workstations by the NMRPipe software [19] and analyzed using NMRView [20]. All proton, nitrogen and carbon resonances were assigned using 2D spectra. TOCSY (mixing time 60 ms) was used to identify the spin systems. Natural abundance [1H-13C] HMQC optimized for observing aliphatic or aromatic regions and [1H-13C] HMQC-TOCSY (with mixing times of 40, 60 and 80 ms) spectra were used to assist with cross-peaks assignment. NOESY spectra (with mixing times of 0.150 s and 0.250 s) were used for establishing the sequential assignment of the spin systems. 15N chemical shifts were measured from a natural abundance [15N-1H] HSQC experiment.

CD Spectroscopy CD measurements were carried out on a Jasco J-810

spectropolarimeter using a Peltier temperature-controlled sample holder and a 0.1 cm path length cell. Near UV experiments were performed with 2 mM hemisite E2-Dst site alone and mixtures containing 10 mM aE2-conj or Dst at 25uC. Far UV spectra and KD determination experiments were performed at 5uC in solutions containing 5 mM peptide after incubating 20 minutes at room temperature with a stoichiometric amount of the corresponding hemisite.

CD titration experiments were done with 5 mM aE2-conj and increasing concentrations of hemisite E2-Dst. The change in helical population was detected following the ellipticity change at 222 nm and fitted to a 1:1 binding model using the following equation:



f DI~ 0:5DIt : ½peptideŠ



ð½peptideŠz½DNAŠzKDÞ{



qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi



ð½peptideŠz½DNAŠzKDÞ2{4½pepetideŠ½DNAŠ



where DI is the signal change at each DNA concentration, DIt is the signal difference between the free peptide and the complex, KD is the equilibrium dissociation constant and [peptide] and [DNA] are the respective total concentrations.

All CD experiments were done in 10 mM phosphate pH 7.0, 1 mM DTT, 100 mM NaCl buffer.

DNA melting experiments UV DNA melting experiments were conduced with the

oligonucleotide (2 mM) in the absence and in the presence of an excess of peptide (10 mM) in the same buffer conditions used in CD experiments. Absorbance at 260 nm was measured as a function of temperature using the absorbance mode of the Jasco



PLoS ONE | www.plosone.org



2



July 2011 | Volume 6 | Issue 7 | e22409



HPV E2 Mimics: A Distamycin-Peptide Conjugate



Figure 1. Design of aE2-conj. A. Hypothetic model for the simultaneous binding of the aE2 peptide and Dst to DNA. DNA is depicted in cyan and Dst in magenta. The peptide positions involved in backbone DNA recognition are displayed in blue; in red are shown residues involved in specific DNA bases interactions; while in yellow are the hydrophobic residues mutated to alanine to minimize hydrophobic interactions. The distance between the N2 of Dst and the N-atoms of the guanidinium group of R296 is indicated. The lower panel shows the alignment of the DNA binding helices of E2 proteins from HPV-16 and -18 strains, and the sequence of the peptide used for synthesizing aE2-conj. The colour coding is the same used in the model. Position 296 chosen to attach the Dst moiety is boxed. B. Chemical structure of aE2-conj. C. Qualitative model of aE2-conj bound to DNA. doi:10.1371/journal.pone.0022409.g001



spectropolarimeter with its Peltier temperature-controlled sample holder. The melting temperature (Tm) was determined as the midpoint of the curve. Each Tm is the mean value obtained from three independent measurements.

Electrophoretic Mobility Shift Assay (EMSA) Reaction mixtures contained 2 mM fluorescein-labelled oligo-

nucleotides and increasing amounts of the studied peptides. Mixtures were incubated 60 min at room temperature in a final



volume of 10 mL in 25 mM phosphate pH 7.0, 1 mM DTT. Mixtures were loaded continuously into running non-denaturing 10% polyacrylamide gels containing TBE 0.56 (0.1 M Tris-HCl pH 8.0, 0.15 M sodium borate, 4 mM EDTA). The gels were resolved at 4 V/cm, 10uC for 2 h. Fluorescein bands corresponding to the free oligonucleotide were detected by UV and fitted to a 1:1 binding model (the same equation than in the CD titration but with a fixed concentration of peptide and increasing concentrations of DNA). Because of the peptide’s high isoelectric point



PLoS ONE | www.plosone.org



3



July 2011 | Volume 6 | Issue 7 | e22409



PLoS ONE | www.plosone.org



HPV E2 Mimics: A Distamycin-Peptide Conjugate



Figure 2. Spectroscopic interaction evidence of aE2-conj and DNA. A. Near-UV CD spectrum at 25uC of the oligonucleotide hemisite E2-Dst (2 mM) alone, after the addition of 10 mM Dst or aE2-conj. B. FarUV CD spectrum of aE2-conj alone, in the presence of hemisite E2-Dst, hemisite E2-non Dst, hemisite non E2-Dst or hemisite non E2-non Dst. C. Far-UV CD spectrum of aE2-conj alone and in the presence of hemisite E2-Dst (upper panel), and of aE2-Ala alone and in the presence of hemisite E2-Dst (lower panel). CD measurements in B and C were performed at 5uC in solutions containing 10 mM phosphate buffer pH 7.0, 100 mM NaCl, and 5 mM peptide after incubating 20 minutes at room temperature with 5 mM ds-oligonucleotide, when present. B and C show the spectrum of the mixture (peptide and DNA) minus the spectra of the corresponding DNA alone. doi:10.1371/journal.pone.0022409.g002

(pI,11), free DNAs and aE2-DNA complex migrate in opposite directions [11], so the complex band was not observed in the gels and only free DNA was quantified in the experiments.

Results and Discussion

Design of the conjugate aE2-conj In order to design our conjugate, we constructed a model for the

simultaneous interaction of Dst and the E2 helix peptide with adjacent DNA sites based on the structures of the complexes of the HPV-18 E2C domain [5] and Dst [17] with their cognates DNAs (Figure 1A). First, we changed hydrophobic residues in the peptide, which are not involved in DNA recognition, into alanines to reduce unfavourable hydrophobic interactions, to augment the water solubility, and possibly to increase the helix tendency of the peptide [21]. Second, side chain of R296 is very close to the N2 of Dst (,8 A˚ between N2 of Dst and the N-atoms of the guanidinium group of R296). Therefore, these positions were chosen for tethering the peptide and the tripyrrole derivative. As R296 is not involved in the interaction with DNA bases [10,22] and, since lysine was successfully used to attach linkers between peptides and tripyrrole units [14,23,24], we introduced in position 296 a lysine. Finally, the selection of the linker is crucial to have high affinities hybrids [14,15]. We designed a Dst-peptide conjugate, named aE2-conj (Figure 1B). In this compound, the Dst bound to a pentylamine is connected to the nitrogen-f of the side chain of the K296 by a urea functionality [14]. According to our model, this linker length with a pentylene between the urea and the pyrrolic nitrogens should be adequate to allow the simultaneous binding of both moieties to DNA, the Dst in the minor groove and the peptide in the major groove (Figure 1C). The additional primary propylamine, which is protonated at the working pH, may also favour the affinity towards the negatively charged nucleic acid [15].

Synthesis and NMR characterization of aE2-conj The tripyrrole unit was prepared following reported procedures

[13]. Selective functionalization at K296, required to attach the minor groove binding moiety, was performed through the introduction of an Alloc-protected side chain lysine residue [23]. Removal of the Alloc-protecting group and attachment of the tripyrrole unit was performed with the protected peptide still bound to the resin. Coupling reaction was carried out by sequential addition of the bifunctional conjugating agent disuccinimidyl carbonate and the amino pyrrole. After standard cleavage-deprotection and purification steps, we obtained the desired conjugated aE2-conj in approximately 20% yield considering peptide synthesis (see Text S1).

aE2-conj was characterized by NMR spectroscopy (see Text S2, Figure S1 and Tables S1 and S2). The coupling with the Dst



4



July 2011 | Volume 6 | Issue 7 | e22409



HPV E2 Mimics: A Distamycin-Peptide Conjugate



moiety does not affect significantly the conformational properties of the peptide moiety respect to the isolated peptide (Figure S1C). Analysis of the secondary chemical shifts in the aE2-conj peptide indicates that in a 4:6 TFE-water solution it adopts an a-helical conformation from K296 to H309 (Figure S2). In contrast, in aqueos solution the peptide moiety is mainly unstructured as shown by CD spectroscopy (Figure 2B). Additionally, the secondary chemical shift analysis shows that the a-helix population is mainly the same as that found for the isolated HPV-16 E2 helix peptide, aE2 (Figure S2). This result points out that the substitution of three hydrophobic residues for alanines was not sufficient for augmenting the helix propensity of the peptide, at least in the TFE: water mixture. Similar helical propensity of aE2, aE2-conj and aE2-Ala was confirmed by CD spectroscopy in TFE titration experiments following molar ellipticty at 222 nm (Text S3, Figure S3).

DNA binding properties of aE2-conj A first evidence of the interaction of aE2-conj with DNA was

obtained from CD spectroscopy (Figure 2, and see Table 1 for sequences of oligonucleotides used in this work). The appearance of an intense positive band around ,330 nm is observed in the spectrum of the oligonucleotide hemisite E2-Dst upon addition of an excess of aE2-conj (Figure 2A). This is an indication of the tripyrrole moiety of aE2-conj inserting into the DNA minor groove. The fact that only a slight increase in the ellipticity at 330 nm is induced in the spectrum of hemisite E2-Dst after adding Dst shows that both moieties of aE2-conj must be present for a tight binding to DNA. On the other hand, the far-UV CD spectrum of aE2-conj in aqueous solution corresponds to that of a mostly unfolded peptide (Figure 2B). However, a significant increase in the negative CD signal at 222 nm in the spectrum of aE2-conj was induced when added hemisite E2-Dst, indicating the folding in a-helix of the peptide moiety upon DNA binding. Control oligonucleotides, lacking the E2 or the Dst recognition sequences, or both, produced a minor increase of the 222 nm band, suggesting some nonspecific but weaker binding. Additionally, the CD signal at 222 nm is not significantly augmented in the spectrum of the isolated peptide aE2-Ala when incubated with hemisite E2-Dst oligonucleotide (Figure C, lower panel). These



results demonstrate that interaction of both peptide and Dst are necessary in aE2-conj for a tight binding.

Supplementary evidence of specific binding of aE2-conj to DNA was obtained by ds-DNA melting temperature (Tm) measurements (Figure 3 and Table 2). When some compound binds ds-DNA, the duplex is stabilized by the interaction and the Tm of the nucleic acid is increased. Tighter the binging is, higher the melting temperature. The oligonucleotide hemisite E2-Dst melts at 58.5uC (Figure 3A). In the presence aE2 or aE2-Ala the melting is slightly augmented, due to the weak interaction of the isolated peptides with specific DNA [11]. In contrast, in the presence of Dst the increase in the melting temperature of hemisite E2-Dst is higher (7.0uC vs ,2uC). This augment reflects the



Table 1. Sequences of duplex oligonucleotides used in this study.



Oligonucleotidesa hemisite E2-Dstb hemisite E2-non Dst hemisite non E2-Dst hemisite non E2-non Dst Iset non E2-non Dst



Sequencea

GTA GCA Caa atC GGT TGA TCA ACC Gat ttG TGC TAC GTA GCA CAG CTC GGT TGA TCA ACC GAG CTG TGC TAC GTA GCA Caa atT GCG TGA TCA CGC Aat ttG TGC TAC GTA GCA CAG CTT GCG TGA TCA CGC AAG CTG TGC TAC ACA TGG ACC TGT CAA GTA TAC TTG ACA GGT CCA TGT



aThe E2 hemisite is underlined and the Dst binding site is indicated in italics and

in lower-case letters. bThe hemisite E2-Dst corresponds to half E2 target site 35.

doi:10.1371/journal.pone.0022409.t001



Figure 3. DNA melting curves. A. Melting curves of the oligonucleotide hemisite E2- Dst alone, in the presence of aE2, aE2Ala, Dst plus aE2-Ala, Dst and aE2-conj. B. Melting curves of the oligonucleotide Iset non E2-non Dst alone and in the presence of aE2conj. Table 2 summarizes these results. doi:10.1371/journal.pone.0022409.g003



PLoS ONE | www.plosone.org



5



July 2011 | Volume 6 | Issue 7 | e22409



HPV E2 Mimics: A Distamycin-Peptide Conjugate



Table 2. DNA melting temperatures.



hemisite E2-Dst



Tm (6C)a



DTm (6C)



alone +Dst +aE2-Ala +aE2 +aE2-conj +aE2-Ala +Dst



58.560.3 65.560.4 61.060.3 60.560.5 70.060.3 64.560.4



– 7.060.7 2.560.6 2.060.8 11.560.6 6.060.7



hemisite non E2-non Dst



alone +aE2-conj



64.460.4 70.060.4



– 6.060.8



Iset non E2-non Dst



alone +aE2-conj



60.260.2 66.260.3



6.060.5



amean values 6 standard deviation between three independent measurements. doi:10.1371/journal.pone.0022409.t002



tripyrrole binding in the DNA minor groove, as was observed in the CD measurements. Finally, when aE2-conj is added the Tm is the highest measured (70.0uC), showing that a tight binding is obtained only when both moieties of aE2-conj are present. Interestingly, the Tm in the simultaneous presence of aE2-Ala and Dst was even lower than in the presence of only Dst, indicating that the effect is not additive. On the other hand, aE2-conj produces a higher increase in the Tm of the hemisite E2-Dst than

in the Tm of two ds-oligonucleotides not containing the E2 nor the Dst binding sites: hemisite non E2-non Dst and Iset non E2-non Dst (DTm of 6.0uC for both control oligonucleotides vs. 11.5uC for hemisite E2-Dst, Figure 3B and Table 2), suggesting that the binding of aE2-conj is specific.

The full E2C domain in solution binds DNA with high affinity (KD in the low nanomolar range when determined in solution) and specificity (,1000 fold) [9,25]. However, when the affinity was measured by gel shift experiments, the KD for a complex between E2C and a specific ds-oligonucleotide was in the mM order [9]. Dissociation constants of aE2-conj and aE2 to specific and unspecific oligonucleotides at 25uC were determined by gel shift experiments (Table 3 and Figure 4A). The complex between the



aE2 peptide and hemisite E2-Dst gave a KD of ,6 mM, in agreement with previous measurements [11,12]. The KD for a complex between aE2 and the randomnized oligonucleotide Iset non E2-non Dst was of ,50 mM, also in perfect agreement with previous measurements. On the contrary, the complex between the conjugate aE2-conj and hemisite E2-Dst presented a KD of ,0.13 mM. Thus aE2-conj displays an increment of affinity for the target DNA of one order and a half (KD (aE2)/KD (aE2-



Table 3. Dissociation constants.



Complex



KD (mM)

gel shift experiments



aE2/hemisite E2-Dst aE2/Iset non E2-non Dst aE2-conj/hemisite E2-Dst aE2-conj/Iset non E2-non Dst E2C/hemisite E2-Dst



6.362.6 ,50 0.1360.09 1.2260.75 ,1b



an.d.: not determined. bRef. 9.

doi:10.1371/journal.pone.0022409.t003



CD titration n.d.a n.d. 0.1060.15 n.d. n.d.



PLoS ONE | www.plosone.org



Figure 4. Equilibrium dissociation constant determination. A. Electrophoretic mobility shift assay at room temperature, complex fraction as a function of peptide/DNA ratio: aE2-conj/hemisite E2-Dst, aE2/hemisite E2-Dst, aE2-conj/Iset non E2-non Dst. The inset shows a picture of the corresponding gel assays. B. Molar ellipticity change at 222 nm as a function of hemisite E2-Dst/aE2-conj ratio, experiment performed at 5uC. Both titration experiments were carried out in 10 mM phosphate pH 7.0, 100 mM NaCl, 1 mM DTT buffer. Lines are fits to a quadratic equation corresponding to 1:1 binding model. Dash lines in figure B are added to show 1:1 stoichiometry. doi:10.1371/journal.pone.0022409.g004



6



July 2011 | Volume 6 | Issue 7 | e22409



HPV E2 Mimics: A Distamycin-Peptide Conjugate



conj) = 48). Regarding E2C, aE2-conj exhibits an augment of affinity for specific DNA of one order of magnitude, when measured in the same conditions [9]. The specificities of aE2 and aE2-conj are similar (KD (Iset non E2-non Dst)/KD (Iset non E2non Dst) is 8 and 9 for aE2 and aE2-conj, respectively).

The KD of aE2-conj hemisite E2-Dst complex was also determined by CD titration at 222 nm at 5uC (0.1060.15 mM, Table 2 and Figure 4B). As expected, the value thus obtained is slightly lower than the value obtained by gel shift at 25uC. The observed differences are not related with the incubation buffer conditions but with dissociating conditions of the EMSA assay [26]. Moreover, the value might be overestimated because this measurement was performed at high concentrations to stress the helical signal in the CD spectrum. The CD experiment additionally shows that the stoichiometry of the complex is 1:1.

Conclusions In this study we have demonstrated that it is possible, by

properly conjugating the peptide to a Dst-like tripyrrole unit, to increase the affinity of the monomeric aE2 peptide for specific binding to its cognate DNA. We have designed a conjugate carrying a primary amino function in the tether linking the peptide and the Dst moieties to facilitate interaction with the phosphate DNA-backbone. Our conjugate displays a ,50-fold increase of the affinity for specific DNA respect to the peptide alone and a ,10fold increase when compared with the full-length domain, also measured by gel shift experiments [9]. These results are encouraging for the design of new E2 conjugates with alternative linkers with higher specificities and better affinities capable of acting as antiviral agents in HPV infections.

Work to design new more potent conjugates and to structurally characterize them is currently in progress, in order to possibly further understand the factors that govern sequence-specific protein-DNA interactions.

Supporting Information

Figure S1 NMR chemical shift assignments of aE2-conj. A. Chemical structure and numbering of aE2-conj. Inset: An alternative chemical connection between the peptide and the Dst derivative moieties. B. 1H-monodimensional spectrum of aE2-conj. C. [13C,1H]-HMQC spectra of aE2-Ala and aE2-conj (right and left panels, respectively). The spectra were recorded in a 700 MHz spectrometer at 25uC in 4:6 TFE:aqueous solution. Residue chemical shift assignment of the peptide moiety is indicated in black and signals corresponding to the linker and Dst portions of the conjugate are indicated in red. In bold letters

References

1. Mun˜ oz N, Bosch FX, de Sanjose S, Herrero R, Castellsague X, et al. (2003) Epidemiologic classification of human papillomavirus types associated with cervical cancer. N Engl J Med 348: 518–527.

2. Su JH, Wu A, Scotney E, Ma B, Monie A, et al. (2010) Immunotherapy for cervical cancer: Research status and clinical potential. BioDrugs 24: 109–129.

3. Thierry F (2009) Transcriptional regulation of the papillomavirus oncogenes by cellular and viral transcription factors in cervical carcinoma. Virology 384: 375–379.

4. Dell G, Gaston K (2001) Human papillomaviruses and their role in cervical cancer. Cell Mol Life Sci 58: 1923–1942.

5. Kim SS, Tam JK, Wang AF, Hegde RS (2000) The structural basis of DNA target discrimination by papillomavirus E2 proteins. J Biol Chem 275: 31245–31254.

6. Koh JT, Zheng J (2007) The new biomimetic chemistry: artificial transcription factors. ACS Chem Biol 2: 599–601.

7. Vazquez ME, Caamano AM, Mascarenas JL (2003) From transcription factors to designed sequence-specific DNA-binding peptides. Chem Soc Rev 32: 338–349.



are shown the crosspeaks corresponding to residue in position 296 in both compounds, for the isolated peptide 296 is an alanine while is a lysine in the conjugate. (Imp: Impurity). (TIF)

Figure S2 Secondary chemical shifts for aE2-conj and aE2. Plot of the chemical shift differences between the observed resonances and values found in a random coil conformation, Dd = d(obsd) - d(random coil), vs. the position along the peptidic

N sequence for 13Ca, 13Cb and 1Ha on the top, medium, and lower

panel, respectively, for aE2-conj ( ) and aE2 (#). Positive deviations of the shifts of Ca and negative of Cb and Ha are indicative of a-helical conformation. (TIF)

Figure S3 TFE titration for aE2-Ala, aE2 and aE2-conj. Molar ellipticity at 222 nm for aE2-Ala, aE2, aE2-conj as a function of TFE concentration. Line corresponds to the fit to a two-state coil-helix equilibrium proposed. (TIF)

Table S1 1H, 13C and 15N Chemical shifts assignments of the peptide moiety of aE2-conj. (DOCX)

Table S2 1H, 13C, and 15N chemical shifts assignments of the Dst moiety of aE2-conj. (DOCX)

Text S1 aE2-conj purification. (DOC)

Text S2 NMR chemical shit assignments of aE2-conj. (DOC)

Text S3 TFE titrations. (DOC)

Acknowledgments

MG and DEW especially thank Daniel Cicero and Gonzalo de Prat Gay for generously allowing us to perform the experiments in their respective labs. The authors also thank Daniel Cicero for critical reading of the manuscript and helpful discussions.

Author Contributions

Conceived and designed the experiments: DEW MJC KK MG. Performed the experiments: DEW MJC KK MG. Analyzed the data: DEW MJC KK MG. Contributed reagents/materials/analysis tools: DEW MJC KK MG. Wrote the paper: DEW MJC MG.

8. Hegde RS (2002) The papillomavirus E2 proteins: structure, function, and biology. Annu Rev Biophys Biomol Struct 31: 343–360.

9. Ferreiro DU, Lima LM, Nadra AD, Alonso LG, Goldbaum FA, et al. (2000) Distinctive cognate sequence discrimination, bound DNA conformation, and binding modes in the E2 C-terminal domains from prototype human and bovine papillomaviruses. Biochemistry 39: 14692–14701.

10. Cicero DO, Nadra AD, Eliseo T, Dellarole M, Paci M, et al. (2006) Structural and thermodynamic basis for the enhanced transcriptional control by the human papillomavirus strain-16 E2 protein. Biochemistry 45: 6551–6560.

11. Wetzler DE, Gallo M, Melis R, Eliseo T, Nadra AD, et al. (2009) A strained DNA binding helix is conserved for site recognition, folding nucleation, and conformational modulation. Biopolymers 91: 432–443.

12. Faber-Barata J, Mohana-Borges R, Lima LM (2006) Specificity in DNA recognition by a peptide from papillomavirus E2 protein. FEBS Lett 580: 1919–1924.

13. Vazquez ME, Caamano AM, Martinez-Costas J, Castedo L, Mascarenas JL (2001) Design and Synthesis of a Peptide That Binds Specific DNA Sequences through Simultaneous Interaction in the Major and in the Minor Groove. Angew Chem Int Ed Engl 40: 4723–4725.



PLoS ONE | www.plosone.org



7



July 2011 | Volume 6 | Issue 7 | e22409



HPV E2 Mimics: A Distamycin-Peptide Conjugate



14. Blanco JB, Vazquez ME, Castedo L, Mascarenas JL (2005) Sequence-specific DNA recognition by monomeric bZIP basic regions equipped with a tripyrrole unit on the N-terminal side. Towards the development of synthetic mimics of Skn-1. Chembiochem 6: 2173–2176.

15. Blanco JB, Vazquez O, Martinez-Costas J, Castedo L, Mascarenas JL (2005) High affinity, sequence specific DNA binding by synthetic tripyrrole-peptide conjugates. Chemistry 11: 4171–4178.

16. Chen FM, Sha F (1998) Circular dichroic and kinetic differentiation of DNA binding modes of distamycin. Biochemistry 37: 11143–11151.

17. Uytterhoeven K, Sponer J, Van Meervelt L (2002) Two 1 : 1 binding modes for distamycin in the minor groove of d(GGCCAATTGG). Eur J Biochem 269: 2868–2877.

18. Marion D, Wuthrich K (1983) Application of phase sensitive two-dimensional correlated spectroscopy (COSY) for measurements of 1H-1H spin-spin coupling constants in proteins. Biochem Biophys Res Commun 113: 967–974.

19. Delaglio F, Grzesiek S, Vuister GW, Zhu G, Pfeifer J, et al. (1995) NMRPipe: a multidimensional spectral processing system based on UNIX pipes. J Biomol NMR 6: 277–293.

20. Johnson BA (2004) Using NMRView to visualize and analyze the NMR spectra of macromolecules. Methods Mol Biol 278: 313–352.



21. Lyu PC, Sherman JC, Chen A, Kallenbach NR (1991) Alpha-helix stabilization by natural and unnatural amino acids with alkyl side chains. Proc Natl Acad Sci U S A 88: 5317–5320.

22. Ferreiro DU, Dellarole M, Nadra AD, de Prat-Gay G (2005) Free energy contributions to direct readout of a DNA sequence. J Biol Chem 280: 32480–32484.

23. Vazquez O, Vazquez ME, Blanco JB, Castedo L, Mascarenas JL (2007) Specific DNA recognition by a synthetic, monomeric Cys2His2 zinc-finger peptide conjugated to a minor-groove binder. Angew Chem Int Ed Engl 46: 6886–6890.

24. Vazquez O, Blanco-Canosa JB, Vazquez ME, Martinez-Costas J, Castedo L, et al. (2008) Efficient DNA binding and nuclear uptake by distamycin derivatives conjugated to octa-arginine sequences. Chembiochem 9: 2822–2829.

25. Dellarole M, Sanchez IE, Freire E, de Prat-Gay G (2007) Increased stability and DNA site discrimination of ‘‘single chain’’ variants of the dimeric beta-barrel DNA binding domain of the human papillomavirus E2 transcriptional regulator. Biochemistry 46: 12441–12450.

26. Hill JJ (1997) Fluorescence approaches to study of protein-nucleic acid complexation. Methods Enzymol 278: 390–416.



PLoS ONE | www.plosone.org



8



July 2011 | Volume 6 | Issue 7 | e22409