WO2006027244A2 - Paramagnetic metal polyamino complexes for mri visualization of internalized polynucleotides - Google Patents

Abstract

The invention relates to paramagnetic ion-based contrast agents that are positively charged at physiological pH and that bind non covalently to a polynucleotide chain for use in the preparation of a diagnostic formulation for the 'in vivo' visualization of polynucleotides internalized into human or animal body cells by use of Magnetic Resonance Imaging techniques.

Description

PARAMAGNETIC METAL COMPLEXES FOR THE MRI VISUALIZATION OF INTERNALIZED POLYNUCLEOTIDES

The present invention relates to the field of diagnostic imaging and, in particular, to paramagnetic metal complex compounds for use in the Magnetic Resonance Imaging (MRI) of polynucleotides internalized into human or animal body cells as well as the compositions thereof. BACKGROUND OF THE INVENTION

Nowadays there is an increasing interest in dealing with relevant diseases by means of therapies that rely on the cellular uptake of DNA, RNA and related materials. These therapies have first to front the fact that the phospholipid bilayer of the plasma membrane has a hydrophilic exterior and a hydrophobic interior and, consequently, any polar molecule, including DNA, RNA and proteins, is unable to freely pass through the membrane. A number of routes for delivering polynucleotides containing materials to cells have been provided including, for example, the use of cationic lipids, inactivated viruses, targeted liposomes and other vesicles-based systems. One emerging way of introducing large molecules directly into cell cytoplasm is the electroporation, a procedure where electrical fields are used to create a transient permeabilization of cell membrane. The "in vivo" electroporation, in particular, represents a rapidly emerging technique applied to delivery enhanced amounts of DNA [1-3] and hydrophilic drugs [4-5] directly to the target organ or tissue (i.e. tumors, epithelia, endothelia, and organs such as brain, heart or liver) in gene therapy or anticancer treatment, respectively. The gene delivery to skeletal muscle fibers, resulting in antigen expression by the host's cell, may be also used to promote the local secretion of therapeutic factors, i.e. angiogenic or neurotrophic factors, and to operate DNA-mediated vaccinations [6] . Clinically important applications of this technique include, for example, the electrochemotherapy wherein the electroporation mediated delivery of the chemotherapeutic agent, directly inside the cancer cells, allows a significant reduction in the effective dose of chemotherapeutic agent to be administered. Hence this technique further allows a significant limitation of the toxic effects caused by the chemotherapeutic agent on normal cells (see, as an example, published studies on the bleomycin delivery [7]). Electroporation-mediated DNA delivery may also be exploited to elicit protective antitumour immunity. In mice developing a mammary invasive carcinoma because of the presence of a transgenic rat HER-2 oncogene, for example, it has been observed that repeated electroporation-mediated administrations of DNA plasmids coding for rat HER-2 receptor have maintained the majority of 1-year-old mice as tumour free [8.9]. In this respect it is worth noting that, when compared with DNA vaccination trough intramuscular plasmid delivery, the electroporation-based route provides a greater, more persistent immunity [10]. Accordingly, DNA vaccine model represents a promising, practical and effective way to elicit immune response and it may be seen as a potential new immunotherapeutic strategy against cancer.

Moreover, gene transfer efficiency provided by the "in vivo" electroporation has been found to be generally superior than that promoted by different DNA injection methods, including the direct injection of naked plasmid DNA into muscle fibers [8], the use of ballistic technology (i.e. "gene gun") [11] and neutral polymers [12], that only resulted in a moderate enhancement of gene transfer into skin and muscle, generally too low to elicit an immune response. Accordingly, even if this technique does not represent the only way of introducing large molecules such as polynucleotides into cell cytoplasm, since the work by Neumann et al. on 1982 [13], the electroporation has been used to efficiently introduce foreign DNA into prokaryotic and eukaryotic cells [14] and in several organs [15].

In view of the advantages this newly emerging technique provides, and of the consequent development of new electroporation-mediated therapeutic strategies, there is a need for optimising electrotransfer protocols to provide an accurate quantisation of the plasmid DNA and, more generally, of a polynucleotide material entered into the targeted cell, as well as the precise localization of the involved area. Nowadays, the success of gene electrotransfer is indirectly evaluated by measuring the expression of a reporter gene and the measurement of the involved tissue area is carried out by analysing histological sections. Recently, Paturneau-Jouas M. et al. [16] proposed the use of magnetic resonance imaging (MRI) to "in vivo" detect the region involved in the electrotransfer process. Current MRI technology displays superb spatial resolution (up to < 100 μm) allowing the "in vivo" observation of up to a small number of cells labelled with suitable contrast agents, so as to render it as a preferential technique for this proposed use [17-22]. The method proposed by Paturneau et al. is based on the simultaneous intraperitoneal or local injection of the plasmid DNA and the commercially available contrast agent Gd-DTPA (trade name: Magnevist®, Schering). According to this method, an increased signal intensity on Tpweighted MR images can be taken as a direct reporter of efficiency and spatial extent of gene electrotransfer.

Now it is known that the electroporation occurs when an applied external field exceeds the capacity of the cell membrane. Upon application of a suitable pulse (i.e. 375 V cm^'1) between two electrodes placed in the region of interest, transient hydrophilic pores are formed. Their formation occurs in the frame of less than a second whereas their resealing may take minutes. It has been seen that for small molecules diffusion alone appears to be the main determinant of intracellular uptake through the hydrophilic pores in the membrane. Moreover, low molecular weight compounds continue to diffuse into electropermeabilized cells for minutes, during all the time required for the cell membrane resealing. On the contrary, large molecules (i.e. DNA and RNA) can enter the cell only if they were already present in the extracellular matrix at the time of the application of the electroporating pulse [1.4] and no gene transfer takes place if the plasmid DNA is added after the application of the said pulse. This means that relatively small molecules, such as Gd-DTPA, and macromolecular systems, such as plasmid DNA behave quite differently as far as their electroporation mediated cellular entrapment. Consequently, the use of the low molecular weight contrast agents as per the cited literature seems unable to provide an accurate delineation of the region of plasmid DNA entrapment and may not provide accurate information concerning the efficiency and spatial extent of gene electrotransfer. SUMMARY OF THE INVENTION It has now been found that an improved method for MRI-visualization of a polynucleotide internalized into a human or animal body organ or tissue cells relies on the use of a class of paramagnetic contrast agents that bind to the polynucleotide.

Accordingly, in a first embodiment, the present invention provides a class of paramagnetic ion-based contrast agents that bind to a polynucleotide chain, for use in the preparation of a diagnostic formulation for the "in vivo" visualization of polynucleotides internalized into human or animal body organ or tissue cells. Unless otherwise provided, the above "in vivo" visualization occurs by use of Magnetic Resonance Imaging techniques. The binding between the said contrast agents and the polynucleotide chain may include any type of binding; the non covalent binding is particularly preferred.

In a second embodiment, the present invention provides a Magnetic Resonance Imaging based method for the " in vivo" visualization of polynucleotides internalized into human or animal body cells, comprising the use of a paramagnetic ion-based contrast agent according to the invention.

The invention also refers to injectable diagnostic compositions comprising an effective amount of a paramagnetic ion-based contrast agent that binds to a polynucleotide. The said composition can be employed to "in vivo" visualize, by use of MRI techniques, a polynucleotide internalized into a human or animal body organ or tissue cells.

The term "polynucleotides" as used in the present description refers to DNA, cDNA, plasmid DNA₅ RNA and PNA sequences. In the present invention, the expression "a polynucleotide internalized into a human or animal body cells or tissue" means a polynucleotide introduced into the said human or animal body organ or tissue cells cytoplasm by way of any possible delivery route known to a skilled person in the art, for example including, but not limited to, the use of "gene gun" technology and of neutral polymers as per the above cited literature as well as the use of cationic lipids (23) targeted liposomes or vesicles (24) and inactivated viruses. In a preferred aspect of invention, the internalization of the polynucleotide into the targeted cells is performed by "in vivo" electroporation.

In the present invention, the term paramagnetic ion-based contrast agent means a chelated complex with a bi- and trivalent paramagnetic metal ion, preferably having atomic number ranging between 20 and 31, 39, 42, 43, 44, 49, and between 57 and 83. Particularly preferred are Fe(²⁺), Fe(³⁺), Cu(²⁺), Cr(3⁺), Eu(3+), Dy(3+), La(³+), Yb(³+), Tm(3+), or Mn(2+) and Gd(³⁺), this latter being even more preferred. Preferred contrast agents for use in the method of the invention include either linear or cyclic paramagnetic chelated complexes that are positively charged at physiological pH.

In the present description, unless otherwise indicated, with the expression "chelated complexes that are positively charged at physiological pH" we intend a chelated complex that at physiological pH is endowed with a residual positive net charge that is higher than 1 and, more preferably, higher than 2. In other words this means that the global sum of the negative charges on the chelating ligand and of the positive charges shown by both the chelated metal ion and the chelating ligand of the invention results in a positive number higher than 1 and, more preferably, higher than 2.

More preferably, the contrast agent for use according to the invention is a Gd(3⁺) chelated complex that bears on its surface one or more positively charged residue(s) at physiological pH and that binds to the negatively charged phosphate groups on the DNA, RNA or PNA polymeric chains.

In the present description, unless otherwise indicated, with the term "positively charged residue at physiological pH" we intend a moiety including at least one amino group that is protonated at physiological pH and/or ammonium groups. In a preferred aspect, the above moiety includes more than one protonated amino and/or ammonium groups that endow the MRI contrast agent of the invention with a positive net charge when present at physiological conditions.

Preferred positively charged residues are polyamino chains including up to 25 carbon atoms and up to 10 amino and/or ammonium groups.

More preferably, the positively charged residue is selected from the group consisting of: polyamines such as, for example, spermidine, putrescine and spermine; amino acids such as lysine, arginine, ornithine and amino acid derivatives such as agmatine [i.e. (4-aminobutyl)guanidine)]. The said polyamino chain may be linked to the polycarboxylic chelating moiety of the chelating ligand either directly or through a spacer group. When the polyamino chain is linked to the polycarboxylic chelating moiety through a direct bond, the formation of the said bond usually involves a reaction between functional groups located on the polycarboxylic moiety and on the polyamino chain. Non limiting examples of chemically reactive functional groups which may be employed to this purpose include amino, hydroxyl, thiol, carboxy, carbonyl, thiocyanide groups and the like. In a preferred embodiment, the direct covalent bond is between a free carboxyl group of the polycarboxylic chelating moiety and an amino group of the polyamino chain so as to provide a carboxamido group. Alternatively, the polyamino chain may be linked to the polycarboxylic chelating moiety of the chelating ligand through a bifunctional spacer group. In this case, said spacer comprises at least two reactive moieties separated by a spacing arm wherein one of the reactive moieties will provide for a covalent binding with the polycarboxylic chelating moiety and the other with the polyamino chain. The spacing arm may consist, as an example, of an alkylidene, alkenylidene, alkynylidene radical at most comprising 15 carbon atoms, that can be optionally substituted and/or interrupted by a cycloalkyl or aryl ring and/or by one or more heteroatoms such as oxygen, nitrogen and sulphur. The reactive moieties in said bifunctional spacer, that may be the same or different, have to be capable of reacting with the functional groups present in the polyamino chain and in the polycarboxylic moiety, i.e. with hydroxyl, thiol, carboxy, carbonyl groups and the like. In a preferred embodiment, the said reactive moieties include: diazo compounds such as diazoacetate esters diazoacetamides, carbodiimides, alkylating agents such as α-haloacetyl compounds, aryl and alkyl halides, α-haloalkyl ether, aldehydes and ketones capable of Schiff s base formation with amino groups, epoxide derivatives, acylating agents such as isocyanates and isothiocyanates, acid anhydrides, acid halides, active esters and, in general, those useful reagents for amide bond formation, widely known by the skilled person in the art.

The contrast agent for the use of the invention is, as formerly said, positively charged at physiological pH. Accordingly, a paramagnetic chelate complex optionally exhibiting a residual negative charge on the coordination cage should be neutralized by salification of the negative charge(s) with physiologically acceptable cations. Preferred cations or inorganic bases suitable for this purpose, if necessary, include the ions of alkali or alkaline- earth metals such as potassium, sodium, calcium or magnesium, including any mixed salt.

Preferred cations of organic bases comprise those obtained by protonation of primary, secondary and tertiary amines such as ethanolamine, diethanolamine, morpholine, glucamine, N-methylglucamine, N₅N- dimethylglucamine, basic amino acids such as lysine, arginine, ornithine.

In a particularly preferred aspect, the complex compounds for use according to the present invention have a zero residual charge. In a most preferred aspect, the complex compounds for use according to the present invention include a chelating ligand selected from the following: Structure 1

DOTA-Spd

The compound of structure 1 is hereinafter named as DOTA-Spd and the chelate complex thereof with Gd(3⁺) is hereinafter referred to as Gd-DOTA-Spd. Structure 2

DOTA-C6-Spd

The compound of structure 2 is hereinafter named as DOTA-C6-Spd and the chelate complex thereof with Gd(3⁺) is hereinafter referred to as Gd-DOTA-C6-Spd.

The compounds of structure 1 and 2 are new and are a further object of the present invention as well as the chelated complexes thereof with a metal ion having atomic number ranging between 20 and 31, 39, 42, 43, 44, 49, and between 57 and 83, and the physiologically acceptable salts thereof.

Structure 3

DTPA-Spd

The compound of structure 3 is hereinafter named DTPA-Spd. Structure 4

DTPA-Arg The compound of structure 4 is hereinafter named DTPA-Arg. Structure 5

DOTA-ph-Spd The compound of structure 5 is hereinafter named DOTA-ph-Spd.

The compounds of structure 3, 4 and 5 are known. For a general reference about their preparation see, as an example, WO 03/103722.

The paramagnetic metal complexes of the invention may be prepared according to synthetic methods well known by a skilled person in the art. In general, for the preparation of the paramagnetic metal complexes of the invention it is possible to conjugate first the spacer, if any, with the polyamino chain, then conjugate the obtained intermediate product with the polycarboxylic moiety of the ligand and subsequently metallating the obtained chelating ligand. Alternatively, it is also possible to conjugate first the polycarboxylic chelating moiety with the spacer, if any, then conjugate the obtained intermediate with the polyamino chain and then metallating the obtained chelating ligand. However, when the polyamino chain is directly linked to the polycarboxylic chelating structure of the chelating ligand, a direct condensation reaction is also possible between the polyamino chain (suitable protected and/or activated according to conventional methods) and the polycarboxylic chelating residue.

Among the various possible synthetic approaches, preferred routes for the preparation of Gd-DOTA-Spd, including a direct bond between the ligand and the polyamino residue, and of DOTA-C6-Spd, wherein the said linking is through a spacer group, are reported, as non limiting examples, in the following Scheme land Scheme 2, respectively: Scheme 1

ney

DOTA- Spd in which:

1) intermediate 2 is prepared by alkylation of the l,6-bis(ferr- butyloxycarbonyl)-l,6,10-triazadecane with acrylonitrile,

2) free secondary amine on intermediate 2 is protected by use of BOC-ON (di-tert-butyl dicarbonate), to give intermediate 3

3) the cyanide group on intermediate 3 is reduced by use of hydrogen under pressure to give intermediate 4, 4) the intermediate 4 is the alkylated with bromoacetyl bromide to give intermediate 5,

5) DO3AtBu macrocycle (l,4,7-tris-(tert-butyloxycarbonylmethyl)- 1,4,7,10-tetraazacyclododecane) is reacted with intermediate 5 to give the condensation product 6 in a suitably protected form,

6) amino and carboxylic groups in 6 are deprotected with TFA to give the desired product as hydrochloride salt.

Scheme 2

H₂ Pd/C ' ' MeOH

TFA I CHO₃

DOTA-C6-Spd in which:

1) the spacer arm suitably protected is prepared by benzylation of the 6-aminohexanoic acid and subsequent reaction of the benzylated derivative with bromoacetyl bromide in presence of K₂CO₃ to give the intermediate (2), 2) the intermediate (2) is conjugated with the polycarboxylic chelating moiety (DO3A) suitably protected as tris- fert-butyl ester derivative to give the intermediate (3),

3) the intermediate 3 is debenzylated by use of hydrogen under pressure to give intermediate (4),

4) the carboxy group of the intermediate 4 is suitably protected by use of NHS (N-hydroxysuccinimide) to give the intermediate (5),

5) the intermediate 5 is reacted with suitably protected spermidine to give the condensation product 6 in a suitably protected form, 6) amino and carboxy lie groups in (6) are deprotected with TFA to give the desired product.

The paramagnetic complexes of the invention are characterized by a residue that is positively charged at physiological pH. This positive charge promotes a strong binding interaction between the complex and the negatively charged groups such a as the phosphate groups onto the polynucleotide chain. For example, the above Gd-DOTA-spd includes a tripositively charged residue on the surface of the neutral Gd-complex cage. The linear spermidine residue can wrap around the DNA chain by setting electrostatic interactions with adjacent, negatively charged, phosphate groups. In a preferred aspect of the invention, the association constant Ka between the positively charged chelated complex of the invention and the negatively charged polynucleotide chain is equal to, or even higher than, 10³ M^"1.

As a first result of this binding interactions, a stable supramolecular adduct between the polynucleotide chain and the positively charged complexes of the invention is formed, wherein the most of the complex is bound to the polynucleotide chain. The stability of the said supramolecular adduct is high enough to ensure the co-localization of the polynucleotide and the paramagnetic complexes. In the present invention with the term co-localization we intend the simultaneous localization of either the contrast agent or the polynucleotide it binds, in the same area of an organ or tissue or in the same cellular compartment. This is to say that the localization of the first component results, in turn, in the localization of the second one. As a first important result, this co-localization makes it possible the advantageous use of a contrast agent of the invention as MRI reporter of the localization of an injected polynucleotide. Accordingly, the monitoring, by use of Magnetic Resonance Imaging technique, of the Gd-complex cellular uptake allows to provide either the precise assessment of those cells that have successfully entrapped the plasmid trough the electroporation process, or it may provide accurate information concerning the efficiency and spatial extent of an optional gene electrotransfer. Furthermore, from the extent of signal enhancement in spin eco Tl -weighted images, it is possible to estimate the local concentration of Gd chelate in the given ROI. Having established the number of Gd-DOTA-spd chelates bound to each DNA plasmid chain, based on the determination of the association constant Ka between chelated complex and the negatively charged polynucleotide chain, the MRI method allows one to give an estimate of the actual concentration of the entrapped plasmid DNA. On the basis of the uptaken plasmid DNA, it could be possible to anticipate whether the gene expression in the transfected cell will be high enough, for instance, for pursuing an immunogenic response in DNA vaccination protocols. In a further aspect, the MRI contrast agent of the invention may also be used to "in vivo" atraumatically visualize the efficiency and spatial extent of the plasmid DNA electrotransfer, without the need of any traumatic biopsy, as per today practice. In a different aspect, the use of the agents of the invention makes possible to set up experimental or therapeutic electrotransfer protocols aimed to optimize the dose of plasmid DNA to be delivered so as to elicit the desired immune response. Another important result of the above referred strong binding interaction existing between the positively charged agents of the invention and a polynucleotide chain is that each polynucleotide chain binds a great number of the positively charged contrast agents. As an example, the quite strong (K_a = 2.2 ± 1 x 10³ M^"1) binding interaction between the tripositive Gd-DOTA-spd complex of the invention and DNA results in that each plasmid chain can bind up to 2500 ± 500 Gd-DOTA-spd complexes. This, from one side, assures that the most of the agents involved in the determination is bound to the polynucleotide chain. From a different view, the binding yields a 50% increase in the relaxivity (R_b = 11.0 ± 0.4 s^mM^'1) of the Gd(III) complex, likely as a consequence of the lengthening of its molecular reorientation time. This latter aspect represents a further significant improvement the positively charged MR contrast agents of the invention provide over the prior art.

Accordingly, it is a further object of the present invention a method of DNA vaccination comprising the use of a contrast agent according to the invention. In another aspect of the invention, the positively charged Gd(III) chelated complex of the invention may be used to set up experimental or therapeutic electrotransfer protocols aimed, as an example, to identify an optimized area to be transfected and the dose of plasmid DNA to be delivered to elicit the desired immune response. According to an additional aspect, the invention also provides a kit separately including: (i) a DNA vaccine and (ii) a contrast agent according to the invention; as well as the pharmaceutical compositions comprising a supramolecular adduct between the chain of a DNA vaccine and the Gd(III)-complex compounds according to the invention. In this latter case, the number of Gd(III)-complex molecules linked to the vaccine polynucleotidic chain is preferably up to 2000.

To support the efficacy of the compounds of the invention as well as the improvement they provide over the prior art, in vivo MRI tests have been performed in mice. In particular, in vivo MRI tests have been performed in electroporated areas of mice muscles administered with Gd-DOTA-Spd complex mixed with ECTM plasmid and, also, devoid of the said plasmid. The obtained results shown that in both cases the amount of internalised Gd-complex is large enough to obtain a significant effect on MRI signal intensity, but the extension of the hyperintense region resulted markedly different whether the contrast agent was injected alone or bound to the DNA. In particular, the hyperintense region corresponding to the uptake of this small-sized molecule (Gd-DOTA-Spd alone) is much wider than that depicted by the internalization of (Gd-DOTA-Spd)_nDNA constructs. This result clearly reflects the occurrence of differences in the internalisation pathways and thus confirms that the entrapment of low- weight molecules is mainly governed by diffusion mechanisms consenting a more extended internalization in space and in time. Analogous experiments were carried out by replacing Gd-DOTA-Spd, with Gd-HPDO3A (commercial name: ProHance®), a non specific contrast agent similar to the Gd-DTPA compound of the prior-art. In this case, the extension of the hyperintense regions depicted by internalization of the Gd- HPDO3A with and without DNA plasmid are almost the same. These results confirm that the use of contrast agents such as Gd-HPDO3 A, a neutral contrast agent, or Gd-DTPA, both of them unable to provide any binding interaction with the negative groups on the polyamino chain, did not lead to any significant delineation of the region of plasmid DNA entrapment.

The contrast agents according to the present invention may be administered to the patients for the MRI based imaging of polynucleotides in an amount sufficient to give the desired information concerning the localization and quantisation of electroporation-mediated entrapped polynucleotides. Generally, the dosage of from about 0.001 to about 5.0 mmoles of paramagnetic contrast agent per kg of body weight is sufficient to obtain the desired visualization. For most applications, however, preferred dosage of MRI contrast agent will be in the range of from 0.01 to 2.5 mmoles per kg of body weight.

The compounds of the invention can be employed for the manufacture of a contrast medium for use in a method of diagnosis by MRI involving administering said contrast medium to a human or animal body organ or tissue cells and generating an image of at least a part of the said body organ or tissue.

Accordingly, in a further aspect, the present invention relates to the use of a paramagnetic ion-based contrast agent comprising at least one positively charged residue at physiological pH that binds to a polynucleotide chain for the manufacture of a contrast medium for use in a method for the MRI-visualization of a polynucleotide internalized into cells by "in vivo" electroporation, said method involving administering the contrast medium and the polynucleotide to a human or animal body organ or tissue, electroporating at least a part of the administered human or animal body organ or tissue, and detecting the uptake of said paramagnetic contrast agent by the human or animal body organ, or tissue cells through MRI techniques.

For this use the paramagnetic complex compounds according to the present invention may be formulated with conventional pharmaceuticals aids, such as emulsifiers, stabilisers, antioxidants agents, osmolality adjusting agents, buffers, and the like, all of which are suitably selected in order to avoid any possible interaction with the positively charged groups of the complex compound. Thus, the paramagnetic complex compounds according to the present invention may be in conventional administration forms such as solutions, suspensions, dispersions in physiologically acceptable carriers media, such as a water for injection. Parenterally administrate forms, e.g. i.v. solutions, should be sterile and free from unacceptable agents. The parenterally administrable solutions can be prepared as customarily done with injectable solutions. In particular, they may include the paramagnetic complex compounds alone or they may further include the polynucleotide to be internalized by "in vivo" electroporation. In the first case, the solution containing the polynucleotide should be mixed with that containing the paramagnetic complex before the administration. In this last case the two separate solutions may be provided in the form of a kit. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 A. Proton relaxation rate of a 84 μM solution of Gd-DOT A- spd as a function of the plasmid (MW= 6.4 Kb) concentration measured at 20 MHz, 25°C, PBS, pH=7.4.

Figure 1 B. Proton relaxation rate of a 0.28 μM solution of the same plasmid as a function of the Gd-DOTA-spd concentration measured at 20 MHz₅ 25°C, PBS, pH=7.4.

Figure 2. In vivo MRI of quadriceps muscles treated with Gd-DOTA- spd. T₁ weighted spin echo images images (TR/TE/NEX 260/4.4/6, FOV 3.6 cm, 1 slice 1 mm, panel A, and TR/TE/NEX 200/3.2/4, FOV 3.1 cm, 1 slice 1 mm, panel B) three days after the electroporation. 0.3 μmoles of Gd-DOTA- spd were injected with (right leg) or without (left leg) plasmid DNA (0.065 mg). Ti weighted spin echo images (TR/TE/NEX 200/3.2/4, FOV 3.1 cm, 1 slice 1 mm) of an untreated leg (panel C).

Figure 3. In vivo MRI of quadriceps muscles treated with Gd- HPDO3A. T₁ weighted spin echo image (TR/TE/NEX 260/3.2/3, FOV 2.9 cm, 1 slice 1 mm) three days after the electroporation. 0.3 μmoles of Gd-HPDO3A were injected with (right leg) or without (left leg) plasmid DNA (0.065 mg).

Figure 4. MRI and confocal analysis of a muscle electroporated with Gd-DOTA-spd and GFP plasmid. The localized distribution of Gd-DOTA-spd (panel A) correspond to areas of greater GFP expression (panel B, magnification 10X), whereas in the other regions, the GFP signal is weakly (panel C, magnification 10X) or completely absent (panel D₅ magnification 10X). The invention is illustrated in more detail in the following, non limiting, examples.

EXPERIMENTAL SECTION Materials and Methods.

¹H-NMR and ¹³C-NMR spectra were obtained on a JEOL EX-400 (400 and 100.4 MHz respectively) spectrometer. Elemental analyses were performed with a Perkin-Elmer 240 apparatus.

The longitudinal water proton relaxation rate was measured on the

Stelar Spinmaster spectrometer (Stelar, Mede (PV) Italy) operating at

20 MHz, by means of the standard inversion-recovery technique (16 experiments, 2 scans). A typical 90° pulse width was 3.5 μs and the reproducibility of the T₁ data was ± 0.5%. The \IT_\ nuclear magnetic relaxation dispersion profiles of water protons were measured over a continuum of magnetic field strength from 0.00024 to 0.5 T (corresponding to

0.01-20 MHz proton Larmor frequency) on the fast field-cycling Stelar Spinmaster FFC 2000 relaxometer equipped with a silver magnet. The relaxometer operates under complete computer control with an absolute uncertainty in the \IT_\ values of ± 1%. The typical field sequences used were the NP sequence between 40 and 8 MHz and PP sequence between 8 and

0.01 MHz. The observation field was set at 13 MHz. 16 experiments of 2 scan were used for the T₁ determination for each field.

Variable-temperature ¹⁷O NMR measurements were recorded on a JEOL EX-90 (2.1 T) spectrometer, equipped with a 5 mm probe, by using D₂O as external lock. Experimental settings were: spectral width 10000 Hz, pulse width 7 μs, acquisition time 10 ms, 1000 scans and no sample spinning. Solutions containing 2.6% of ¹⁷O isotope (provided by Yeda Institute, Israel) were used. The observed transverse relaxation rates (R§_obs) were calculated from the signal width at half height. l,6-bis(/er/-butyloxycarbonyl)-l,6,10-triazadecane was prepared as described in literature [25]. All starting materials were obtained from Sigma- Aldrich Co. and were used without further purification. EXAMPLE 1 - preparation of Gd-DOTA-spd a) Synthesis of 7,10-bis-(te/Ϋ-butyloxycarbonyl)~3,7,10- triazatridecanenitrile (2).

Acrylonitrile (0.285 g, 5.34 mmol, 1.3 eq.) was added to l,6-bis(tert- butyloxycarbonyl)-l,6,10-triazadecane (1) (1.43 g, 4.13 mmol) and the resulting mixture was heated to 7O⁰C and stirred for 5h. The excess of solvent was removed at the rotary evaporator and the yellowish oil obtained was dried in vacuum (1.50 g, 3.76 mmol, yield 91.0%).

¹H NMR (400 MHz, 298 K, CDCl₃): S_n 1.42, 1.43 (CH₃, 18Η, s), 1.50 (NCH₂CH₂, 4Η, m), 1.67 (CH₂CH₂NH, 2H, q, J= 6.7 Hz), 2.50 (CH₂CN, 2 Η, t, J= 6.6 Hz), 2.60 (CH₂NH, 2 H, t), 2.90 (NHCH₂CH₂CN₅ 2 H, t, J= 6.6 Hz), 3.05-3.29 (BOcNCH₂, 6Η, m), 4.55 (BocNH, 1 Η, s). b) Synthesis of 3,7,10-tris-(te/*/~butyloxycarbonyl)-3,7,10- triazatridecanenitrile (3).

BOC-ON (1.08 g, 4.40 mmol) was added in small portions to a solution of (2) (1.67 g, 4.19 mmol) and trietylamine (2.12 g, 0.021 mmol, 5 eq.) in dioxane/Η₂O (9:1 v/v, 30 cm ). The resulting mixture was stirred at room temperature in the dark for 3 days; then the solvent was removed under reduced pressure. The resulting pale yellow oil was dissolved in Et₂O (40 cm³) and washed with IM NaOH (3 x 25 cm³) and brine (2 x 20 cm³). The organic fraction was dried over Na₂SO₄ and filtered. After removal of the solvent, a pale yellow oil was obtained which was dried under reduced pressure (1.79 g, 3.60 mmol, yield 85.7%).

¹H NMR (400 MHz, 298 K, CDCl₃): <5fc 1.42, 1.43, 1.45 (CH₃, 27Η, s), 1.50 (NCH₂CH₂, 4Η, m), 1.75 (CH₂CH₂N, 2H, q, J = 6.7 Hz), 2.55-2.61 (CH₂CN, 2 Η, m), 3.05-3.27 (BocNCH₂, 10Η, m), 4.62 (BocNH, 1 Η, s). c) Synthesis of l,6,10-tris-(^r^-Butyloxycarbonyl)-l,6,10,14- tetraazatridecane (4).

The intermediate (3) (1.79 g, 3.60 mmol) and NaOH (0.4 g, 0.01 mol) were dissolved in 94% EtOH (30 cm³). After addition of Ni-Raney as catalyst, the resulting mixture was hydrogenated under pressure overnight (20 bar). After filtration through celite, the solvent volume was reduced to 5 cm³, H₂O (30 cm³) was added and the product was extracted with CHCl₃ (5 x 50 cm³). The collected organic fractions were dried over Na₂SO₄ and filtered. After removal of the solvent, a pale yellow oil was obtained which was dried under reduced pressure (1.61 g, 3.20 mmol, yield 89.0%).

¹H NMR (400 MHz, 298 K, CDCl₃): δa 1.42, 1.43, 1.44 (CH₃, 27Η, s), 1.50, 1.62, 1.72 (NCH₂CH₂, 8Η, m), 2.67 (CH₂NH₂, 2 H, t, J= 6.9), 3.05-3.27 (BoCNCH₂, 10Η, m), 4.62 (BocNH, 1 Η, s). d) Synthesis of N-(4,8,13-tris-(tert-butyloxycarbonyl)-4,8,13- triazatridecane)-2-bromoacetamide (5).

To a suspension of K₂CO₃ (1.7 g, 12.3 mmol) and of (4) (1.33 g, 2.65 mmol) in CH₃CN (40 cm³) cooled to 0⁰C and kept under N₂ atmosphere, a solution of bromo acetyl bromide (0.642 g, 3.18 mmol, 1.2 eq.) in CH₃CN (20 cm³) was added in Ih. The mixture was allowed to warm to room temperature, stirred overnight and then filtered. The solvent was removed at the rotary evaporator to yield a yellow oil which was dissolved in CH₂Cl₂ (20 cm³) and washed with H₂O (2 x 10 cm³) and 5% NaHCO₃ (10 cm³). The organic solution was then dried over Na₂SO₄ and filtered. After removal of the solvent, a pale yellow oil was obtained which was dried under reduced pressure (1.26 g, 2.02 mmol, yield 76.3%).

¹R NMR (400 MHz₅ 298 K₅ CDCl₃): fa 1.42, 1.43, 1.45 (CH₃, 27Η, s), 1.50, 1.72 (NCH₂CH₂, 8H₅ m), 3.05-3.30 (NCH₂, 12Η, m), 3.84 (CH₂Br, 2H₅ s)₅ 4.62 (BocNH, 1 Η, s). e) Synthesis of l-(7,U,16-tris-(re/^</-butyloxycarbonyl)-3,7,ll,16- tetraaza-2-oxo-exadecan)-4,7,l O-triscarboxy methyl-11 ,4,7,10- tetraazacyclododecane (6).

To a solution of DO3AtBu.HBr (1.204 g, 2.02 mmol) and of diisopropylethylamine (2.6 g, 20.2 mmol) in CH₃CN (40 cm³) heated to reflux temperature and kept under N₂ atmosphere, a solution 5 (1.26 g, 2.02 mmol) in

CH₃CN (30 cm³) was added in Ih. The mixture was stirred for 2Oh and then the solvent was removed at the rotary evaporator to yield a yellow oil which was dissolved in Et₂O (30 cm³) and washed with H₂O (3 x 20 cm³). The organic solution was then dried over Na₂SO₄ and filtered. After removal of the solvent, a pale yellow oil was obtained which was dried under reduced pressure (1.29 g, 1.22 mmol, yield 60.4%).

¹H NMR (400 MHz₅ 298 K, CDCl₃): fa 1.41-1.45 (CH₃, 54Η, m), 1.50- 1.72 (NCH₂CH₂, 8Η, m), 2.5O₅ 2.70 (NCH₂ ring, 8H, br) 2.82, 2.87 (NCH₂ ring, 8H, br), 3.01-3.30 (NCH₂, 20Η, m), 4.73 (BocNH, 1 Η, s). f) Synthesis of l-(3,7,ll,16-tetraaza-2-oxo-exadecan)-4,7,10- triscarboxymethyl-11, 4,7,10-tetraazacyclododecane. 5ΗCI.

The intermediate 6 (1.29 g, 1.22 mmol) was dissolved in a solution of TFA/CHC1₃ (30 cm³, 1 :1 v/v) and stirred at room temperature for 2h. After removal of the solvents the reaction was repeated for further 2h. The solvents were removed at the rotary evaporator and the oil obtained was dissolved in EtOH (15 cm³). After addition of some drops of a solution of cone. HCl in EtOH (1:1 v/v) a white solid precipitated from the solution which was filtered and dried under reduced pressure (0.742 g, 0.96 mmol, 78.9% yield).

HPLC: Waters Atlantis RPC18, H₂O TFA 0.1%, CH₃CH TFA 0.1%, retention time 4.65 min, flux 1 ml/min.

Elem. Anal.: found (calc. for C₂₆H₅₇Cl₅N₈O₇) C, 40.35 (40.50); H, 7.21 (7.45); N, 14.23 (14.53%).

¹R NMR (400 MHz, 298 K, D₂O): Sa 1.68 (CH₂CH₂CH₂NH₂, 4H, m),

1.85 (CONHCH₂CH₂, 2Η, J = 6.9 Hz, q), 2.04 (NHCH₂CH₂, 2Η, J = 7.7 Hz, q), 2.96 (CH₂NH₂, 2H, J = 6.9 Hz, t), 3.03 (NHCH₂(CH₂)₃, 2Η, J = 6.6 Hz, t),

3.05-3.10 (CH₂NHCH₂ and CH₂ ring, 10 Η, m), 3,17 (CONHCH₂, 2Η, J = 6.6 Hz, t), 3.35 (NCH₂ ring, 12Η, br), 3.51 (NCH₂CONH, 2H, s), 3.74

(CH₂COOH, 6H, s). ¹³C NMR (100.4 MHz, 298 K, D₂O): δ_c 22.8

(NHCH₂CH₂), 22.9 and 24.1 (CH₂CH₂CH₂NH₂), 24.3 (CONHCH₂CH₂), 36.4

(CONHCH₂), 38.9 (CH₂NH₂), 44.5, 44.8, 47.2 (CH₂NHCH₂), 48.5

(NHCH₂(CH₂)₃), 48.9, 50.4, 52.3, 55.9 (CH₂ ring), 56.4 (CH₂COOH), 57.8 (NCH₂CONH).

The elemental analysis is consistent with the formation of the pentahydrochloride salt of DOTA-Spd. Furthermore, analytical HPLC chromatography confirmed the purity higher than 95% of the obtained final product. g) Synthesis of Gd-DOT A-Spd.

The DOTA-Spd ligand (73.2 mg, 0.095 mmol) was dissolved in water (1.5 ml) and the pH was adjusted to 7 by adding NaOH IM. GdCl₃.6H₂O (40 mg, 0.108 mmol) was dissolved in 0.5 ml of water and slowly added to the first solution maintaining the pH value at 6.7 with NaOH. The mixture was then stirred at room temperature for 16h. The pH was then increased to 9, and the solution was stirred for 2h. The suspension was centrifuged at 10000 rpm and filtered over 0.2 μ syringe filter. The free Gd³⁺ still present in solution was quantified by UV measurement in presence of Xylenol Orange, by determining the absorbance ratio between 573 and 433 nm; the overall Gd contents was determined by ¹H NMR Ti measurement of the mineralized complex solution (in HCl 6M at 120⁰C for 16h). The excess free Gd³⁺ was then complexed with a stoichiometric amount of ligand as mentioned before. The final free Gd³⁺ was 0.25%. Finally, the solution was lyophilised and a white solid was obtained. h) Characterisation of Gd-DOTA-Spd.

The relaxivity (the proton relaxation enhancement of water protons in the presence of the paramagnetic complex at 1 mM concentration) of Gd-DOTA-Spd, measured at 20 MHz and 298 K, is 5.6 mM^"1 s^'1, i.e. a value slightly higher than that reported for the parent Gd-DOTA complex (ri_p= 4.7 mM^'1 s^"1). Moreover, Gd-DOTA-Spd displays a constant ri_p value up to pH 1, to suggest an overall good stability as far as the release of free Gd³⁺ ions is concerned. EXAMPLE 2 -preparation of Gd-DOTA-C6-Spd a) Synthesis of 6-aminohexanoic acid benzyl ester (1).

In a round bottom flask, equipped with a Marcusson device, 47.9 g (0.366 mol) of 6-aminohexanoic acid and 56.7 mL (0.55 mol) of benzylic alcohol were mixed with 400 mL of toluene. Then 73 g (0.38mol) of p-toluensulfonic acid were added. The reaction was carried out under magnetic stirring at reflux temperature until the amount of water collected into the Marcusson device was equal to 0.677 mol. The solution was then concentrated to 150 mL and cooled at 4°C. The precipitate was collected, washed with 50 mL of toluene and then two times with 100 mL of diethyl ether. After drying in vacuo, 144g (yield 95.5%) of a white solid was collected. b) Synthesis of 7-aza-9-bromo-8-oxo-nonanoic acid benzyl ester (2).

In a round bottom flask, 71 g (0.18 mol) of intermediate (1) and 49.90 g (0.36 mol) Of K₂CO₃ were mixed with 1.1 L of acetonitrile. The mixture was stirred at 0-5⁰C for one hour, then a solution of Bromoacetyl bromide (15.75 mL, 0.18 mol in 200 mL of acetonitrile) was slowly dropped in 4 hours. After further 40 minutes, the solid was removed and the organic phase was evaporated under reduced pressure. The residue was dissolved in 540 mL of dichloromethane and washed with a 5% solution Of Na₂CO₃ (2 x 250 mL), with H₂O (250 mL), with HCl 0.01 N (2 x 250 mL), and finally two times with 250 mL of H₂O. The organic phase was then dried with sodium sulfate, filtrated, and evaporated under reduced pressure. The product was then dried under vacuum and 123g of a white solid was obtained (yield 80%). c) Synthesis of l-(3-aza~2,9-dioxo-10-oxa-ll-phenyl-undecyl)- 4,7,10-tris(/er^butyloxycarbonylmethyl)-l,4,7,10-tetraazacyclododecane

(3).

51.44 g (0.1 mol) of D03A-tris-te/ϊ-butylester and 16.58 g Of K₂CO₃ (0.12 mol) were mixed with 1.25 L of acetonitrile. The temperature of the mixture was maintained at 0-5⁰C and a solution of (2) (36.76 g, 0.107 mol in 350 mL of acetonitrile) was slowly dropped over a period of 5 hours, then the reaction was left stirring for another hour. The resulting mixture was filtered and the solvent was evaporated in vacuo. The residue was treated with diethyl ether (150 ml) to give a white solid. After filtration, the solution was evaporated in vacuo and the desired product was collected as colourless oil (75 g, yield 96.5%). d) Synthesis of l-(3-aza-8-carboxy-2-oxo-octyl)-4,7,10-tris(tert- butyloxycarbonylmethyl)-l ,4,7,10-tetraazacyclododecane - DOTAC6OH (4).

75 g of the intermediate (3) were mixed with 100 ml of methanol and 7 g of Pd/C-10% was added, then the reaction was carried out under atmospheric hydrogen pressure and room temperature for 4 hours. The mixture was then filtered and the solvent evaporated; the solid obtained was dissolved in CH₂Cl₂:MeOH 15: 1 v/v (150 ml) and purified by flash chromatography on a 7.5 x 30 cm silica column with the same elution mixture. 33g of white solid was obtained (Yield = 50%). e) Synthesis of l-(3-aza-8-carboxy-2-oxo-octyl)-4,7,10-tris(tert- butyloxycarbonylmethyl)-l ,4,7,10-tetraazacyclododecane N-

Hydroxysuccinimide ester - DOTAC6-NHS (5).

A solution of the DotaC6-OH (4) (0.50 g, 0.923 mmol) and of N-hydroxysuccinimide (0.127 g, 1.11 mmol) in CH₂Cl₂ (30 ml) was cooled to 0⁰C and a solution of DCC (0.229 g, 1.11 mmol) in CH₂Cl₂ (10 cm³) was added in l/2h. The mixture was stirred for 2Oh at room temperature, filtered and then the solvent was removed at the rotary evaporator to yield a yellow oil which was dissolved in Et₂O (10 cm³) and filtered. After removal of the solvent, a pale yellow oil was obtained which was dried under reduced pressure and used without further purifications. f) Synthesis of l-(2,9-dioxo-3,10,14,18,23-pentaaza-

3,10,14,18,23-ter/-butyloxycarbonyl-triicosanyl)-4,7,10-tris(tert- butyloxy carbonylmethyl)-l ,4,7,10-tetraazacyclododecane (6).

A solution of the intermediate Spd-NH₂ (0.464 g, 0.923 mmol) and of diisopropylethylamine (0.3 cm³, 1.85 mmol) in CH₂Cl₂ (20 cm³) was added in l/2h to a solution of the intermediate DOTAC6-NHS (5) in CH₂Cl₂ (10 cm³). The mixture was stirred for 3 days and then the solvent was removed at the rotary evaporator to yield a yellow oil which was dissolved in Et₂O (30 cm³) and washed with H₂O (3 x 20 cm³). The organic solution was then dried over Na₂SO₄ and filtered. After removal of the solvent, a yellow oil was obtained which was purified by flash chromatography (eluent EtOAc/MeOH 9:1 > 8:2) to obtain the final product as a pale yellow oil (0.48 g, 0.47 mmol, yield 51.0%). g) Synthesis of l-(2,9-dioxo-3,10,14,18,23-pentaaza-triicosanyl)- 4,7,10-tris(carboxymethyl)-l,4,7,10-tetraazacyclododecane (7) - DOTA- C6-Spd.

The intermediate (6) (0.48 g, 0.47 mmol) was dissolved in a solution of TFA/CHCI₃ (30 cm³, 2:1 v/v) and stirred at room temperature for 16h. The solvents were removed at the rotary evaporator and the oil obtained was washed with Et₂O (3 x 15 cm³). A very hygroscopic solid was obtained after purification by HPLC (Waters Atlantis column, H2O/TFA 0.1%; CH3CN/TFA 0.1%). (0.258 g, 0.37 mmol, 78.7% yield). h) Synthesis of Gd-DOTA-C6-Spd.

The Gd complex of the DOTA-C6-Spd ligand has been prepared by use of the same reagents and analogous procedure that have been used for the preparation of the above Gd-DOTA-Spd compound.

EXAMPLE 3 - MRI-visnalization of DNA internalized by "in vivo" electroporation a) Plasmids and DNA preparation. pcDNA3 vector (Invitrogen, San Diego, CA) coding the extracellular (EC) and transmembrane (TM) domains of rpl85 neu (ECTM) was produced and used as described [13]. Green Fluorescent Protein (GFP) plasmid was commercially obtained (Invitrogen, San Diego, CA). Escherichia coli strain DH5α was transformed with ECTM and GFP and then grown in Luria-Bertani medium (Sigma, St. Louis, MO). Large-scale preparation of the plasmids was conducted by alkaline lysis using Endofree Qiagen Plasmid-Giga kits (Qiagen, Chatsworth, CA). DNA was precipitated, suspended in sterile saline water and stored in aliquots at -20⁰C for use in electroporation protocols. b) DNA Injection and Electric-Pulse Delivery.

BALB/c mice (Charles River, Calco, Italy) were anesthetized with 1.15 mg sodium pentobarbital by intraperitoneal injection. 20 μl of a solution containing 0.3 μmoles of the Gd-DOTA-Spd complex mixed with 0.065 mg of ECTM plasmid was injected directly into the quadriceps muscles of the right posterior leg of a first group of five animals (group 1) with a 28-gauge syringe needle. The same amount of Gd-DOTA-Spd devoid of ECTM plasmid was also injected into the left posterior leg of the same animal. A second group of mice (Group 2) (n = 3) received 0.3 μmoles of the Gd-HPDO3A (commercial name: ProHance®, Bracco, Milan, Italy) in the left leg and 0.3 μmoles of the Gd-HPDO3A mixed with 0.065 mg of ECTM plasmid in the right leg. Another group of four animals (group 3) received 0.3 μmoles of the Gd-HPDO3A mixed with 0.065 mg of GFP plasmid injected into both the right and the left legs. One minute after DNA injection, transcutaneous electric pulses were applied by two stainless steel plate electrodes placed at each side of the leg. Electrical contact with the leg skin was ensured by shaving each leg and applying a conducting gel. Two square- wave 25 ms, 375 V/cm pulses were generated by a T820 electroporator (BTX, San Diego, CA). In the group 1 and 2 MRI examinations were performed 3 days after injection and group 1 mice were followed until 21 days. Group 3 treated animals were sacrificed with a lethal dose of sodium pentobarbital, their quadriceps muscles removed and analysed by MRI and immunohistochemistry 1, 2 and 3 days after treatments. All experiments were performed according to recommendations of the

National Institutes of Health (NIH) (Bethesda, MD) Guide for the Care and Use of Laboratory Animals. c) MRI. All MR images were acquired on a Bruker Avance 300 (7T) equipped with a microimaging probe. The system is endowed with two birdcage resonators with 30 and 10 mm inner diameter, respectively. Imaging of the mice legs was performed using a Ti -weighted, fat suppressed, multi slice multi echo protocol (TR/TE/NEX 200/3.2/4, 1 slice = 1 mm). Fat suppression was performed by applying a presaturation pulse (90° BW= 1400 Hz) at the precession frequency of fat (-1100 Hz from water). Statistical analysis was performed using GraphPad Prism Software. Probability values < 0.05 were considered statistically different. "In vivo" Ti has been measured using a SNAP (multi slice gradient echo for fast applications) sequence, by applying a preparatory 180° inversion pulse. Hyperintense ROIs were defined including all pixels with a signal intensity (SI) 30% higher than those measured on the untreated legs. Two untreated legs were analysed as control. d) Immunohistochemical Procedures. Group 3 treated animals were sacrificed by intramuscular injection of a lethal dose of sodium pentobarbital on day 1, 2 and 3 after injection with 20 μl of a solution containing 0.3 μmoles of the Gd-DOTA-Spd alone or mixed with 0.065 mg of the plasmid codifying for GFP. Muscle were then removed, fixed in 4% paraformaldehyde 154 mmol/L Piperazine-N,N'-bis(2-ethanesulfonic acid) (PIPES) pH 7.5 overnight at 4°C and imaged at 7 Tesla. The pieces were washed in rinse water and treated with 0.1 mole Tris, pH 7.5 with 0.2% glycine, then dewaxed and embedded in paraffin wax. Serial sections were taken and finally mounted on Glass Microscope Slides (Bacto Laboratories). GFP fluorescence at 488 excitation and 520 nm emission wave lengths was analyzed with a confocal laser scanning microscopy system (CLSM) equipped with an argon-ion laser (LSM510; Zeiss, Germany). All images were prepared at the same contrast and brightness under the same magnification (10X). Images of 512 x 512 were employed. e) Binding of Gd-DOTA-Spd to plasmid DNA. The assessment of the interaction strength between the complex and

ECTM plasmid (coding for the extracellular (EC) and transmembrane (TM) domain of rpl85neu, M. W =6 A kb) and of the number of the binding sites on the polymeric chain was carried out using the Proton Relaxation Enhancement (PRE) method [26]. In a first titration, the longitudinal water proton relaxation time of a 0.084 mM solution of Gd-DOTA-Spd complex was measured at the fixed frequency of 20 MHz and in the presence of an increasing concentration of DNA (figure IA). For the equilibrium nGdDOTA - spd + DNA < ^Ka >(θd - DOTA - spd)_n DNA

the product (K_a ' n) and ri_p ^b is obtained, K_a being the thermodynamic association constant, n the number of the equivalent binding sites, and ri_p ^b the millimolar relaxation rate of the macromolecular adduct, respectively. By means of a second titration at constant DNA concentration (0.28 μM), the number of binding sites was determined. Figure IB shows the result of the second titration linearized as a Scathchard Plot. The interaction between the tripositive Gd-DOTA-spd complex and DNA results to be quite strong (K_a = 2.2 ± 1 x 10³ M^"1) and each plasmid chain can bind up to 2500 ± 500 Gd-DOTA-Spd complexes. The binding yields a 50% increase in the relaxivity (R_b = 11.0 ± 0.4 s^''mM^'') of the Gd(III) complex likely as a consequence of the lengthening of its molecular reorientation time. f) "In vivo" MRI of the electroporated area.

The evaluation of DNA electrotransfer was performed by injecting, into quadriceps muscle on the right posterior leg of BALB/c mice (group 1) 20 μl of a solution containing 0.3 μmoles of the Gd-DOTA-Spd complex mixed with 0.065 mg of the ECTM plasmid. On the basis of the K_a value, it can be assessed that, under these conditions, about 70% of the Gd-complex is bound to the DNA chain. The same amount of Gd-DOTA-Spd devoid of DNA was also injected in the left posterior leg on the same animal. One minute after injection, transcutaneous electric pulses were applied. After three days, when the elimination of the non-internalized Gd-complex has been completed, fat- suppressed T₁ weighted MR multi slice multi echo images (7 Tesla) were recorded. Figure 2A and 2B clearly shows hyperintensity in both posterior legs with respect the corresponding image of an untreated mouse (figure 2C). The amount of internalised Gd-complex is large enough to obtain a significant effect on MRI signal intensity as a consequence of the drastic decrease of the longitudinal relaxation time (Ti) in the region of cellular entrapment. However, the extension of the hyperintense region results markedly different whether the contrast agent is injected alone (0.28 ± 0.097 cm²) or bound to the plasmid (0.13 ± 0.051 cm²), clearly reflecting the occurrence of differences in the internalisation pathways. The mean signal intensity (SI) values measured on legs treated only with Gd-DOTA-Spd resulted about 40-50% higher than those obtained on legs treated with the same amount of Gd-DOTA-Spd co- injected with DNA (p < 0.003 (see Table 1 below). The hyperintensity was detectable at day 7, 11 and 16 whereas disappeared at day 21 in all mice.

Table 1: Comparison of ROI signal intensify (SI) on Ti weighted images of group 1 mice.

Signal Intensity (x 10⁵)

GdDOTA-Spd + DNA Gd-DOTA-Spd mouse 1 1.7 ± 0.2 2.8 ± 0.2 mouse 2 1.5 ± 0.15 2.9 ± 0.3 mouse 3 1.3 ± 0.25 3.4 ± 0.25 mouse 4 1.5 ± 0.3 3.5 ± 0.15 mouse 5 0.74 ± 0.1 1.2 ± 0.1

The mean SI integrals measured on legs treated only with Gd-DOTA- Spd (left column) resulted about 40-50% higher that those obtained on legs treated with the same amount of Gd-DOTA-Spd co-injected with DNA (right column) (p < 0.003). Data are the mean SI ± standard deviation.

Analogous experiments were carried out in mice of group 2 by replacing Gd-DOTA-Spd with the aspecific Gd-HPDO3A (commercial name: ProHance®). Figure 3 shows that, as expected, also using this contrast agent the region involved in the electropermeabilization appears easily detectable by Ti weighted spin echo image. Conversely to the result obtained with Gd-DOTA-Spd. the extension of the area in the two legs treated with 0.3 μmoles of GdHPDO3A with and without DNA appears very similar. The differences of SI integrals measured in the regions of interest were < 10% and reflected the normal variability of these determinations (p > 0.91) (see Table 2 below). Table 2. Comparison of ROI signal intensity (SI) on Ti weighted images of group 2 mice.

Signal Intensity (x Kr)

ProHance® + DNA ProHance® mouse 1 2.5 ± 0.2 2.7 ± 0.15 mouse 2 2.3 ± 0, 15 2.5 ± 0.3 mouse 3 0.80 ± 0.1 0.73 ± 0.15

The differences of SI integrals measured in the regions of interest were < 10% (p > 0.91). Finally, in order to estimate the amount of the Gd-complex internalized in the skeletal muscle fibers, the T₁ of treated legs (n=3) were measured "in vivo" using the SNAP sequence. Assuming that the "in vivo" relaxivity of the adduct formed by Gd-DOTA-Spd and DNA is the same than that measured at 7T "in vitro", the determination of the Gd-complex concentration in the tissue was carried out applying the following equation:

[Gdcomplex] mM = (Rl_treated - Rl_untreate_d )/ r_lp ^(Gd/DNA) where Ri t_reated is the relaxation rate of protons in the selected ROI of the muscle treated with Gd-DOTA-Spd and DNA, Re_treated is the relaxation rate of the same area measured in the control, and _rip ^(Gd/DNA) j_s the relaxivity of the Gd-DOTA- Spd/DNA adduct at 7T in water (r_lp^7.6). By using this method, it was estimated that a concentration of 0.10 ± 0.07 mmoles/1 of residual Gd-complex is present in the ROI three days after the electroporation. On the basis of the presence of 2500 binding sites on the polymeric chain for the Gd complex, the corresponding amount of ECTM plasmid successfully transfected into muscle fibers was ca. 4.5 ± 1.5 x 10^"13 moles. g) Relationship between histological analysis and MRI. To confirm the co-localization between Gd-DOTA-Spd complex and DNA plasmid, histologic analyses were compared with ex vivo MRI visualization of excised treated muscles. MR images at day 3 in all mice of group 3 revealed a strong signal intensity in the electroporated area (figure 4, panel A). After fixation, dissection and slicing of the treated quadriceps muscles, the intrinsic fluorescence of the constitutive Green Fluorescent Protein (GFP) expression was studied by confocal microscopy. To this purpose, a plasmid coding for GFP was co-injected in the quadriceps muscle with Gd-DOTA-Spd and then electroporated as described previously. Very strong GFP-induced fluorescence was observed in the electroporated area (panel B), in good agreement with the hyperintense areas detected in the MR images. As expected, the GFP signal was weak (panel C) or completely absent in the rest of tissue (panel D). The histologic analysis revealed GFP protein only in animals that received both pulses and intra-muscle plasmid injection; the simple injection of GFP-coding plasmid did not result in a detectable expression of the protein. REFERENCES

[I] Gehl J (2003), Acta Physiol Scand 177, 437-447.

[2] Jaroszeski MJ, et al, (1999), Adv Drug Deliv Rev 35, 131-137. [3] RoIs MP, et al, (1998), Nat Biotechnol 16, 168-171. [4] Neumann E, et al, (2000), Bioelectrochemistry 51, 125-132.

[5] Kamate C₅ et al, (2002), Int J Cancer 100, 571-579. [6] Montgomery DL, et al, (1997), Pharmacol Ther 74, 195-205. [7] Heller R, et al, (1998), Cancer 83, 148-157. [8] Mir LM, et al, (1999), Proc Natl Acad Sci USA 96, 4262-4267. [9] Aihara H, and Miyazaki J (1998), Nat Biotechnol 16, 867-870.

[10] Rovero S, et al, (2000), J Immunol 165, 5133-5142.

[I I] Baranov V, et al, (1998), Gene Therapy, North Atlantic Treaty Organization Advanced Science Institute series, Eds. Xanthopoulos, K. G. (Springer, Berlin), Vol. H105, pp219-223. [12] Mumper RJ, et al, (1996), Pharm Res 13, 701-709.

[13] Neumann E, et al, (1982), EMBO J l, 841-845.

[14] Potter H (1988), Anal Biochem Il A, 361-373.

[15] Gilbert RA, et al, (1997), Biochim Biophys Acta, 1334, 9-14.

[16] Paturneau-Jouas M, et al, (2003), Radiology 228, 768-775. [17] Hoehn M, et al, (2002), Proc Natl Acad Sci USA 99, 16267-16272.

[18] Lewin M₅ et al, (2000), Nat Biotechnol 18, 410-414.

[19] Bulte JW, et al, (2001), Nat Biotechnol 19, 1141-1147.

[20] Geninatti Crich S, et al, (2004), Magn Reson Med 51, 938-944.

[21] Louie AY₅ et al₅ (2002), Nat Biotechnol 18, 321-5. [22] Gillies RJ, (2002), J Cell Biochem Suppl, 39, 231-238.

[23] Mahato RL, (2005), Adv Drug Deliv Rev., 57, 699-712.

[24] Felnerova D., et al, (2004), Curr Opin Biotechnol, 15, 518-529.

[25] Doll MK, et al, (1996), HeIv Chim Acta, 79, 1373-1386.

[26] Merbach AE, and Toth E (2001). Contrast Agents in Medical Magnetic Resonance Imaging. John Wiley & Sons, Chichester, England.

Claims

1. A paramagnetic ion-based chelated complex that binds non-covalently to a polynucleotide chain for use in the preparation of a diagnostic formulation for the Magnetic Resonance Imaging based "in vivo" visualization of polynucleotides internalized into human or animal body cells.

2. A chelated complex of claim 1 with a paramagnetic metal ion selected from Fe(2+), Fe(3+), Cu(2+), Cr(3+), Eu(3+), Dy(3+), La(^⁺), Yb(3+), Tm(³⁺), Mn(²⁺) and Gd(3+).

3. A chelated complex of claim 2 wherein the paramagnetic metal ion is Mn(2+) or Gd(3+).

4. A chelated complex of claim 1 that is positively charged at physiological pH.

5. A chelated complex of claim 4 comprising a positively charged polyamino chain that binds non-covalently to the negatively charged phosphate groups on the polynucleotide chain.

6. A chelated complex of claim 5 wherein the association constant Ka of the said non covalent binding interaction is equal to or higher than 10³ M^"1.

7. A chelated complex of claim 5 wherein the polyamino chain includes up to 25 carbon atoms and up to 10 amino groups and, optionally, one or more ammonium groups.

8. A chelated complex of claim 7 wherein the polyamino chain is selected from the group consisting of polyamines and suitable amino acids or derivatives thereof.

9. A chelated complex of claim 8 wherein the polyamino chain is selected from the group of: spermidine, putrescine, spermine, lysine, arginine, ornithine and agmatine.

10. A chelated complex of any previous claim wherein the chelating ligand is selected from the group consisting of: DOTA-Spd, DOTA-C6-Spd, DTPA-Spd, DTPA-Arg and DTPA-ph-Spd.

11. A compound selected from the group consisting of: DOTA-Spd; DOTA-C6-Spd; any chelated complex thereof with a paramagnetic metal ion selected from the group consisting of Fe(²⁺)₅ Fe(³⁺), Cu(²⁺), Cr(3+), Eu(³⁺), Dy(3+), La(S⁺), Yb(3+), Tm(3+), or Mn(²⁺) and Gd(³+); and any physiologically acceptable salt thereof.

12. A chelated complex of claim 11 with a paramagnetic metal ion selected from Mn(²⁺) and Gd(3⁺), and any physiologically acceptable salt thereof.

13. Use of a chelated complex of claim 11 as a diagnostic agent.

14. An injectable diagnostic composition comprising an effective amount of a chelated complex of claim 11.

15. A chelated complex according to claim 11 for use in the preparation of a diagnostic formulation for the Magnetic Resonance Imaging based "in vivo" visualization of polynucleotides internalized into human or animal body cells.

16. A chelated complex according to any one of claims from 1 to 10 or 15 wherein the polynucleotide is selected from the group consisting of: DNA, cDNA, plasmid DNA, RNA and PNA sequences.

17. A chelated complex of claim 16 wherein the polynucleotide is internalized by "in vivo" electroporation.

18. A method for the "in vivo" MRI- visualization of a polynucleotide internalized into human or animal body cells which method comprises administering both a paramagnetic ion-based chelated complex that binds non- covalently to a polynucleotide chain, and the polynucleotide, to a human or animal body organ or tissue, electroporating at least part of the administered human or animal body organ or tissue, and detecting the cellular uptake of said paramagnetic contrast agent by use of MRI techniques.

19. The method of claim 18 wherein the paramagnetic ion-based chelated complex is as defined in any one of claims from 2 to 12.

20. The method of claim 18 wherein the polynucleotide is as defined in claim 16.

21. A kit of parts comprising: (i) a physiologically acceptable solution of a paramagnetic ion-based chelated complex that binds non-covalently to a polynucleotide chain, and (ii) a physiologically acceptable solution of the said polynucleotide.

22. The kit of claim 21 wherein the paramagnetic ion-based chelated complex is as defined in any one of claims from 2 to 12.

23. An injectable diagnostic composition comprising a stable supramolecular adduct between a paramagnetic ion-based chelated complex that binds non-covalently to a polynucleotide chain and the said polynucleotide.

24. The injectable diagnostic composition of claim 23 wherein the said supramolecular adduct includes up to 2000 molecules of chelated complex linked to the polynucleotide chain.

25. The injectable diagnostic composition of claims 23 or 24 wherein the paramagnetic ion-based chelated complex is as defined in any one of claims from 2 to 12 and the polynucleotide is as defined in claim 16.

26. The injectable diagnostic composition of claims 25 wherein the polynucleotide is a DNA plasmid.

27. The injectable diagnostic composition of claim 26 wherein the polynucleotide is a DNA vaccine.

28. A DNA vaccination method comprising administering to a human or animal subject in need of it, a pharmaceutically effective amount of an injectable diagnostic composition as defined in claim 27.

29. A kit of parts comprising (i) a paramagnetic ion-based chelated complex that binds non-covalently to a polynucleotide chain wherein the polynucleotide is a DNA vaccine and (ii) a DNA vaccine.