Newly tailored peptide nucleic acids (PNA) and PNA-modified magnetic nanoparticles for DNA targeting. PEG branched polymer chains functionalization of nano-systems for biocompatibility

Abstract 1) Newly tailored peptide nucleic acids (PNA) and PNA-modified magnetic nanoparticles for DNA targeting. 2) PEG branched polymer chains functionalization of nano-systems for biocompatibility. Newly tailored peptide nucleic acids: azaPNA1 Peptide nucleic acid (PNA) is an artificial DNA mimic introduced by Nielsen in 1991, characterized by a pseudopeptide backbone (figure 1) replacing the sugar-phosphate chain. The backbone is made of N-(2-aminoethyl)glycine (aeg) units joint in a polyamide structure, and the purine (A, G) and pyrimidine (C, T) nucleobases are linked to the β-nitrogen atom of the amino acid unit through methylene carbonyl residues. The repeating unit is made of six atoms, exactly as in DNA and RNA. The PNAs suffer of some drawbacks such as low water solubility, the tendency to selfaggregate and low cell uptake. It is possible to change in many ways the monomer structure of the classic PNA in order to improve their physical-chemical properties. Our strategy consisted in introducing polar groups, such as nitrogen atom, in the backbone of aegPNA. Our hypothesis was that this would increase their solubility in aqueous medium by increasing their hydrophilicity. At the same time, this would improve their binding affinity towards DNA thanks to more favorable interactions (e.g. through possibly additional hydrogen bonds). Therefore, first we focused our attention on the synthesis of new PNA monomers that we called azaPNA (Figure 1). In these new molecules, the substitution of the glycine CH2 with an NH group confers them new chemical-physical properties. NH N O OR O B aegPNA B = nucleobase NH N NH O OR O B azaPNA PG PG Figure 1. Aminoethylglycine and azaPNA (aeg- and azaPNA) monomers. Retrosynthetic analysis of new azaPNA monomers 4-7 and 8-11 led us to the synthesis of the two backbones 12 and 13, on with we performed the coupling of nucleobases 14- 17 (Scheme 1). X Boc N H HN NH O OCH3(Fmoc) Boc N H N NH O OCH3(Fmoc) B O HO O B 4-7; (8-11) 12; (13) 14-17 4, 8: B NH N O O 5, 9: B N N N N NH N N NH O NH N N N O NH Cbz Cbz Cbz 6, 10: B 7, 17: B + Scheme 1. Retrosynthetic scheme for azaPNA monomers. The backbone 12 (with the COOMe group on the nitrogen atom of hydrazine) was prepared following the synthetic sequence shown in Scheme 2. Boc N H HN NH O OCH3 H2N OH NH Boc OH a > 98% 21 22 NH Boc O 23 Boc N H N NH O OCH3 b 83% c 90% d 69% 12 24 Scheme 2. Reagents: (a) (Boc)2O, EtOH; (b) (i) Dess-Martin periodinane, CH2Cl2; (ii) Na2S2O3, NaHCO3, H2O, CH2Cl2; (C) H2NNHCO2CH3, PhCH3; (d) NaBH3CN, MeOH, CH3COOH. Compounds 12 and 24 are new, and were completely characterised by means of spectroscopic data. Similarly, the azaPNA backbone 13 was synthesised following the strategy shown in Scheme 3. Boc N H HN NH H2N OH NH Boc OH a > 98% 21 22 NH Boc O 23 Boc N H N NH Fmoc b 85% c 90% d >98% Fmoc 13 25 Scheme 3. Reagents: (a) (Boc)2O, EtOH; (b) (i) Dess-Martin periodinane, CH2Cl2; (ii) Na2S2O3, NaHCO3, H2O, CH2Cl2; (C) H2NNHFmoc, PhCH3; (d) NaBH3CN, MeOH, CH3COOH. This time, we have used hydrazine-Fmoc to obtain the carbazone 25, that was then reduced by means of NaBH3CN to give the backbone 13. XI Finally, the target azaPNA monomers 4-7 and 8-11 were obtained in good yield by introducing the nucleobases onto the nitrogen atom of 12 and 13, using standard coupling condition (Schema 4). Boc N H HN NH O OCH3 Boc N H N NH O OCH3 B O HO O B 12 14-17 + a Boc N H HN NH Boc N H N NH Fmoc B O HO O B 13 14-17 + b Fmoc 4: B = Thymine 70% 5: B = Cytosine (Cbz) 57% 6: B = Adenine (Cbz) 61% 7: B = Guanine (Cbz) 68% 8: B = Thymine 63% 9: B = Cytosine (Cbz) 69% 10: B = Adenine (Cbz) 89% 11: B = Guanine (Cbz) 80% Scheme 4. (a) DhbtOH, DIPEA, EDC.HCl, DMF, 30 h, rt; (b) EDC.HCl, DMF, 5-10 h, rt. The new PNA monomers 4-7 and 8-11 are the building blocks necessary for the construction of azaPNA oligomers. First, it was necessary to find the appropriate conditions for methyl ester hydrolysis for the monomers 4-7 and Fmoc deprotection for the monomers 8-11. The ester group in compound 1 was hydrolysed very efficiently but in very strong conditions by means of aq lithium hydroxide at reflux to give the corresponding decarboxylate product 1 in good yield. (scheme 5). The new azaPNA 1 is stable, and it was completely characterised by means of spectroscopic data. NH Boc N NH OCH3 O O NH N NH OH O O B Boc NH N NH2 O B Boc -CO2 4, 8-11 1: B= Thymine 18: B= Cytosine 19: B= Adenine 20: B= Guanine B Scheme 5. Compounds 5-7 can not be deprotected at the ester group, because the unexpectedly strong basic hydrolysis conditions remove first the Cbz-group from nucleobase. The Fmoc group in compounds 8-11 was hydrolysed very efficiently and easily by means of a reaction with piperidine to give the corresponding decarboxylate monomers 18-20 in high yield. Once synthesized the azaPNA monomers 1, 18-20, we proceeded by introducing one or more azaPNA monomers in a oligomer of aegPNA. First we decided to introduce one monomer of aza in the middle of sequence of a decamer aegPNA. In order to do this, XII we synthesized the tetramer of aegPNA on the resin 51. Then we coupled the aza monomer 1 to the tetramer. We tried many different reaction conditions to form the ureic bond between the aza monomer and the tetramer 51. We obtained the best results through isocyanate. We preformed the isocyanate on the γNH2 of aegPNA 51 with COCl2 in toluene and then we added 1 in THF, to obtain the pentamer 25. The synthesis was completed by adding the last five aegPNA monomers by automated synthesizer, to afford the decamer 3 after the cleavage (scheme 6). NH O N H3 COCl2 N Toluene 15 min O NH N O O H3C CF3COONH O N NH O NH N O O H3C NH O N O NH Boc NH N O O H3C NH2 N O NH Boc NH N O O H3C NH O N O NH N O O H3C NH O N O NH N O O H3C NH O N O NH N O O H3C NH O N O NH N O O H3C NH O N O NH N O O H3C NH O N O NH N O O H3C NH O N N O NH N O O H3C NH O N O NH N O O H3C NH O N O NH N O O H3C NH O N O NH N O O H3C C O (aeg(T)PNA)4 (aza(T)PNA)-(aeg(T)PNA)4 THF 12 h, DIEA NH O N NH O NH N O O H3C NH O N O NH NH N O O H3C NH O N O NH N O O H3C NH O N O NH N O O H3C NH2 O N O NH N O O H3C (aeg(T)PNA)5-(aza(T)PNA)-(aeg(T)PNA)4 NH O N NH O NH N O O H3C O N O NH NH N O O H3C NH O N O NH N O O H3C NH O N O NH N O O H3C O N O NH N O O H3C H3C O 51 53 3 1 Scheme 6 Similarly, we synthesized the decamer (not shown) with the sequence (TCACTAGATG), containing a monomer of azaPNA 1 in the middle of the sequence. The decamers were purified by reverse phase HPLC. The purity was checked by LCMS. Then we checked the solubility of the decamer (TCACTAGATG) and the ability to recognise a complementary sequence of DNA by measuring the melting temperature. PNA-modified magnetic nanoparticles for DNA targeting 2. The first generation of PNAs suffers from some drawbacks including low cell uptake and low solubility in physiological media. To overcome these problems, and to improve their physical–chemical properties, many modified PNAs have been synthesized in recent years following different strategies. In this context, superparamagnetic iron oxide nanoparticles (MNP) attracted our attention because of their unique magnetic properties, which can be controlled rigorously and activated easily by applying an external magnetic field. In this thesis we have set up an effective synthetic platform for the development of monomer and decamer PNA-nanoconjugates, starting from synthetic PNA and nanometer-sized maghemite. The sequence-selective DNA recognition and sequestration ability of the resulting magnetic PNA (MPNA) were assessed according XIII to their capability of enhancing the T2 relaxation response in aqueous solutions under conventional hybridization conditions with complementary DNA. We have used nearly homogeneous commercial γ-Fe2O3 nanoparticles (10 ± 4 nm), approximately spherical in shape. We chose the thymine monomer as a model PNA building unit. First, we designed a small library of PNA monomer derivatives endowed with three different appropriate linkers useful for MNP conjugation. The trialkoxysilane, the carboxyl, and the propargylic groups are all suitable linkers to stably connect biomolecules to MNP through the formation of different kinds of linkages. Such linkers, however, have never been applied to PNA for MNP conjugation. aegPNA HN Si Si N3 + + + 72, 81 O O 74, 83 73, 82 Si aegPNA O HN Si EtO EtO OEt aegPNA O NH 77, 87 78, 90 79, 89 aegPNA in: NH N O NH N O O Me OMe O NH N O NH N O O Me NH2 O 10 72-74, 77-79 81-83, 87, 89-90 O HO O 37 NH N O NH N O O Me Boc 37, 76 NH O aegPNA O O aegPNA O O aegPNA 76 + aegPNA O O HO O aegPNA N N N gFe2O3 gFe2O3 gFe2O3 gFe2O3 Scheme 7 Scheme 7 illustrates the conjugation strategies exploited for the preparation of MPNA monomers and decamers. Monomers 37 and 72 were able to interact directly with nanoparticles through their carboxylate linkers. Similarly, compound 74 could be anchored to MNP effectively as a consequence of the high affinity of siloxane group for iron oxide. The functionalized nanoparticles were isolated by centrifugation and carefully washed affording 76-78. The propargyl-terminating monomer 73, was instead conjugated to the nanoparticles via a Cu(I)-catalyzed azide-alkyne click reaction with XIV a MNP-silylpropylazide adduct, affording 79. MNP-silylpropylazide was prepared according to our recently developed procedure. The organic structures of magnetic products 76-79 were completely characterized by FT-IR (bulk in KBr) and by high-resolution magic angle spinning (HR-MAS) NMR. Inductively coupled plasma spectroscopy (ICP-OES) analysis gave the content of iron isotopes in a sample of functionalized MNP, whereas EA provided quantitative data on the amount of organic material in the sample. In the same way we have prepared homo-thymine 10-mer PNA supported on NPs to afford the 87, 89 and 90. MPNA nanohybrids 87, 89 and 90 were characterized by FTIR, demonstrating the presence of PNA on nanoparticles, while EA and ICP-OES provided information on the average amount of PNA decamers 81-83 on the nanoparticles. Once a reliable, multiple-approach synthetic platform for the production of MPNA was optimized, the next objective was to determine the capability of these MPNA to recognize and bind the complementary DNA. On the basis of the fact that the formation of magnetic aggregates would result in a detectable increment in the T2 relaxation time of water protons, we decided to determine whether a variation of T2 of MPNA could occur under MPNA/DNA hybridization conditions, and could be sensitively detected by relaxometric analysis. Our preliminary results demonstrate that these MPNA maintain excellent performances in PNA/DNA hybridization events, which can be evidenced in a new way, through the measurement of variation of the T2 relaxation time of water dispersions in the presence of complementary DNA and by measurement of Tm. As mentioned, the NPs-PNA described above, is weakly soluble in water. A good solubility in water is necessary for the NPs-PNA conjugates to be used in practical domains, such as the MRI imaging or in vivo applications. In order to increases the hydrophilicity we conjugated NP-PNA with the biocompatible water soluble polyvinyl pyrrolidone carboxilic acid (PVPCOOH) polymer (4.500 Da). In particular, we used the nanoparticles cover with 3-aminopropyl silane (NPs-NH2), witch can be treated with an excess of PVPCOOH and sonicated for 30 minutes in presence of EDC chloridrate, in order to connect the polymer to the construct NP-NH2 by a covalent bond. In this way we achieved a very stable and high concentrated solution of NPs in water. We tried to repeat the reaction of functionalization of NPs-NH2 with PVP and PNA at the same time. After several attempts, we managed to find the right ratio of PVP/PNA in order to obtain a stable solution of NPs-PNA-PVP. PEG branched polymer chains functionalization of nano-systems for biocompatibility3-4. Introduction: I spent my II year Ph.D. thesis at Stanford University in the lab of Professor Hongjie Dai. I have worked on: Various novel nanomaterials have been actively pursued for biogolical and medical applications in recent years. XV Poly-γ-glutamic acid (γPGA) is a naturally occurring bio-material, produced by microbial fermentation. γPGA is water soluble, biodegradable, nontoxic, and even edible. As a result, it is promising for various applications, and has recently attracted considerable interest for biomedical applications such as drug delivery. We thus envisioned γPGA as a polymeric backbone in the synthesis of a new amphiphilic polymer capable of suspending nanostuctures of many shapes and aspect ratios. In particular, we envisioned that the free carboxylic acid of γPGA would first be used to attach lipophilic groups for robust particle interaction, while the remaining carboxylic acids would be conjugated with PEG, providing enhanced aqueous solubility and better biocompatibility. Pyrene-containing and phosfo-lipds-coating moieties have been used extensively for suspension of carbon nanotubes, gold nanoparticles and semiconductor quantum dots due to strong surface interactions via Van der Waals forces, π-π overlap, charge transfer, and/or hydrophobic interactions. Thus, in the first synthetic step (see Method), we used the free carboxylic acid of γPGA to couple 1-methylaminopyrene via EDC amidation, and in a second step, we used the remaining carboxylic acid groups of γPGA to attach primary amine-terminated poly(ethylene glycol) methyl ethers (mPEG-NH2, MW 5000). This synthetic strategy proved to be quite general. Long aliphatic amines (C18, C12) were coupled via the same procedure to obtain additional amphiphilic gPGA-based polymers. Moreover, by varying the amount of hydrophobic molecule and PEG, the properties of the polymer could be tuned to optimize aqueous stability and protein adsorption resistence. After much experimentation, the optimal polymer to obtain stable suspensions was found to contain 30% pyrene, while the remaining 70% of carboxylic acids were loaded with PEG to yield a water soluble, amphiphilic polymer. In the case of PGA with DSPE, the optimal polymer to obtain stable suspensions was found to contain 10% of DSPE, while the remaining 60% of carboxylic acids were loaded with PEG and 30% of free carboxylic acids, to yield a water soluble, amphiphilic polymer. O NH O NH O O O n HN NH O NH OO O NH O O n O 30% HN NH HN NH O O NH OO O O NH O O O n O 10% 30% O P OO O O O O O O O O n n HN 70% O 60% OH PMHC18-mPEG (118) PGA-DSPE-mPEG (114) PGA-Py-mPEG (109) Figure 2 XVI We also use the amphiphilic polymer base on poly(maleic anhydride-alt-1-octadecene) (PMHC18) like a backbone. In this case the polymer itself contein the idrofobic unit, the C18 chain, so in this case we need to do only the PEGylation step. In particular, in the first synthetic step, we used to open the anhydride with primary amine-terminated poly(ethylene glycol) methyl ethers (mPEG-NH2, 5KDa), then we used the second carboxylic acid of PMHC18 to couple another moleculer mPEG-NH2 via EDC amidation to obtain a full PEGylated polymer very high soluble in water. As Figure 3 demonstrat, following sonication, it is possible obtain direct suspension of bulk NTs stabilized in water with the polymer PGA-Py-mPEG, DSPE-PGA-mPEG and PMHC18-mPEG, even after removal of excess polymer by vacuum filtration. NT suspensions demonstrated excellent stability at pH’s ranging from 1 to 12, at 70°C overnight, and in 50% fetal calf serum for 48 h. The AFM (figure 3b) image inset in the figure shows mostly dispersed, single NTs. The UV/visible absorbance spectrum of methoxyPEGylated pyrene-γPGA NTs, methoxyPEGylated DSPE-γPGA NTs and methoxyPEGylated PMHC18 NTs are typical of well dispersed SWNTs, demonstrating van Hove singluarity resonances, as well as pyrene absorption below 400 nm. In the casa of pyrene moiety has a strong tendency to adsorb on SWNTs by π-stacking and hydrophobic interactions in aqueous media. In this way, we obtiain three robust polymers coating SWNTs. Figure 3 Also, very good suspensions of gold NPs in water were obtained through sonication for 10 min in presence of excess pyrene/PEG-γPGA or DSPE/PEG-γPGA to displace citrate. As with NTs, this suspension was observed to be stable to conditions ranging pH , at 70°C overnight, and in 50% serum for 48 h. In contrast, thiol-mPEG(5KDa), a strong and covalent passivator of gold nanoparticles, showed less stability. In particular, as shown in figure 4c, the NPs-thiol-PEG(5KDa) are stable only with the excess of thiol-PEG, indeed if the excess of the thiol-PEG is remuved by centrifugation the solution of NPs become anstable and they form agragate (purpe solution). The UV/visible spectrum in Figure 4b shows the absorbance of gold nanoparticles at 550 nm, as well as the peaks of pyrene. XVII Figure 4 We also used our polymeric amphiphile to suspend gold nanorods. The procedure gave mostly dispersed nanorods, as shown by TEM and UV. Suspensions of gold nanorods in pyrene/PEG-γPGA were stable at neutra base pHs at 70°C overnight, and in serum for 48 h. This result is important because nanorods with covalent thiol-based passivation are anstable. The UV/visible spectrum in figure 3b shows the transverse and longitudinal adsorbance of gold nanorods at 520 nm and 860 nm respectively. To further demostrate the versatility of pyrene/PEG-γPGA and DSPE/PEG-γPGA, we successfully suspended InAs/InP/ZnSe core/shell/shell quantum dots. We found that the lability of the cysteamine capping layer allowed it to be easily exchanged for pyrene/PEG-γPGA or DSPE/PEG-γPGA. InAs/InP/ZnSe core/shell/shell quantum dots were synthesized by previously reported methods in organic solvent. The particles were then transferred from chloroform to water using cysteamine, which was then replaced by γPGA surfactant by dialyzing against a 3500 MWCO membrane. This suggests that the pyrene/PEG-γPGA and DSPE/PEG-γPGA are able to displace the cysteamine coating to stabilize the quantum dots. Again, these suspensions were were stable at differents neutral-base pH , at 70°C overnight, and in serum for 48 h. This is an important result as quantum dots are susceptible to surface oxidation and instability in harsh environments. Figure 6b shows the absorption of InAs/InP/ZnSe quantum dots with excess removed and in 50% serum. No significant loss of quantum yield was observered in the quantum dots upon addition to serum. 1 Giuseppe Prencipe; Paolangelo Cerea; Anna Daghetti; Sergio Dall’Angelo; Clelia Giannini; Emanuela Licandro and Stefano Maiorana. Aza-peptide nucleic acid (azaPNA) monomers: building blocks for the construction of nitrogenenriched PNA oligomers. Manuscript in preparation. 2 Giuseppe Prencipe; Stefano Maiorana; Paolo Verdelio; Miriam Colombo; Paola Fermo; Enrico Caneva; Davide Prosperi and Emanuela Licandro. Magnetic Peptide Nucleic Acids for DNA targeting. ChemComm (2009), 40, 6017 – 6019. 3 Giuseppe Prencipe; Scott Tabakman; Kevin Welsher; Zhuang Liu; Andrew Goodwin; Li Zhang; Joy Henry and Hongjie Dai. PEG Branched Polymer for Functionalization of Nanomaterials with Ultralong Blood Circulation. Journal of the American Chemical Society (2009), 131 (13), 4783-4787. XVIII 4 Andrew P. Goodwin; Scott M. Tabakman; Kevin Welsher; Sarah P. Sherlock; Giuseppe Prencipe and Hongjie Dai. Phospholipid−Dextran with a Single Coupling Point: A Useful Amphiphile for Functionalization of Nanomaterials. Journal of the American Chemical Society (2009), 131, 289-296.