DESCRIPTION

The present invention relates to a complex product comprising a cellular carrier and a cytoxic chemotherapeutic drug, and a procedure for preparing said complex product.

It is well known that classic pharmacological therapy, notwithstanding the considerable progress made, still has major limits which regard above all the capacity to be selective and thus not cause side effects. Pharmacokinetic data also indicate that the quantities of drug able to reach the target in effective concentrations can often only be obtained at the expense of systemic or organ- specific toxicity, which may even be considerable at times. In some cases the pathological tissue does not receive adequate therapeutic concentrations due to anatomo-physiological barriers. This aspect is particularly relevant in the realm of tumor chemotherapy, where the chemotherapeutic drug used is often extremely toxic and thus produces considerable harmful side effects on oncology patients. As is well known, in order to increase the selectivity/specificity of pharmacological therapies, for some time reliance has been made on parenteral administration of carriers capable of transporting molecules, such as, for example, systems based on the use of antibodies, nanoparticles of different materials, liposomes or on genetic manipulation of cells through viral carriers capable of encoding the production of substances that are directly cytotoxic or of transforming pro-drugs into active drugs. Disadvantageously, however, all these solutions which use molecules or cells as "carriers" are not devoid of collateral toxicity. In particular, those obtained with viral transfections can pose serious safety problems. The technical task of the present invention is therefore to provide a complex product comprising a cellular carrier and a cytoxic chemotherapeutic drug, and a procedure for preparing said complex product which enables the aforementioned drawbacks of the prior art to be overcome.

Within the scope of this technical task, one object of the invention is to realize a complex product of the type referred to above in which the carrier is capable of slowly releasing a high concentration of a molecule or molecular compound in a localized manner, prevalently at the target cell.

Another object of the invention is to provide a complex product of the type referred to above in which the carrier is capable of slowly releasing a concentration of a cytotoxic and/or pharmacologically active molecule and/or molecular compound in a localized manner, prevalently at the target cell.

Another object of the invention is to provide a complex product of the type referred to above in which the carrier enables the delivery of a pharmacologically active molecule and/or molecular compound in such a way as to increase the therapeutic specificity thereof and in parallel reduce or even eliminate any side effects.

Another object of the invention is to realize a complex product of the type referred to above for diagnostic, prognostic and therapeutic purposes.

The technical task, as well as these and other objects, according to the present invention, are achieved by realizing a complex product characterized in that it comprises a cellular carrier including physiological mesenchymal cells isolated from a mammal and possessing at least one marker selected from among CD 13, CD90, CD73, CD 105 and CD 14, and a cytoxic chemotherapeutic drug internalized by said physiological mesenchymal cells in a physiological manner in a high concentration which is pharmacologically efficacious against a sensitive cellular target.

Preferably the cytotoxic drug comprises Paclitaxel (PTX).

The present invention also discloses a procedure for preparing such a complex product, in which after the step of isolating said mesenchymal cells, the mesenchymal cells are exposed to high doses of drug so as to be used therapeutically or cryopreserved for subsequent use.

The present invention also discloses a mouse model characterized in that it is a carrier of neoplastic pathologies of human or mouse origin and is treated with such a complex product.

The present invention likewise discloses the use of such a complex product per diagnosis, prognosis, therapy and therapeutic monitoring.

Finally, the present invention discloses a method for the targeted controlled release of a pharmacologically efficacious concentration of a cytoxic chemotherapeutic drug, characterized in that it delivers to sensitive target cells a population of physiological mesenchymal cells internalizing in a physiological manner a high concentration of said cytoxic chemotherapeutic drug for the slow, targeted release thereof.

Further characteristics and advantages of the present invention will become more apparent from the description that follows, in which it is illustrated by way of non- restrictive example with reference to the appended figures 1 to 17:

Figure 1 shows the phenotypic characterization of mesenchymal stem cells (MSCs): the photos show a fresh MSC culture observed under a microscope and their multipotency expressed by the capacity to differentiate into osteoblasts, chondroblasts and adipocytes. The figure confirms the mesenchymal character evaluated on the basis of the phenotypic expression of CD analyzed by FACS . Figure 2 shows the resistance of MSCs to the cytotoxic activity of Paclitaxel: the histogram indicates cell viability determined by MTT assay after 24 hours of PTX priming. Each point represents the mean +/- sd of 5 experiments. The % is calculated on the O.D. of unprimed cells (considered as 100% viability).

Figure 3 shows the time-dependent kinetics of Paclitaxel release by MSCs-PTX: the presence of PTX in the conditioned medium was evaluated by assaying its antitumor activity against Molt-4 leukemia. The histogram indicates release at different times of culture (with a total change of culture medium at each time studied). The curve indicates the accumulation of PTX in the conditioned medium. Each value indicates the mean +/- sd of 5 experiments.

Figure 4 shows PTX internalization by the MSCs: internalization was evaluated using fluorescein-conjugated PTX. The cells were both observed under a microscope (A) and analyzed by FACS (B).

Figure 5 shows the results of HPLC analysis: the figure shows the chromatograms of pure Paclitaxel (top) and of conditioned medium obtained from cells primed with PTX, trypsinized and recultured 24 hours (bottom). The chromatogram confirms the presence of PTX in the conditioned medium.

Figure 6 shows the antiproliferative activity of the conditioned medium from MSCs-PTX against Molt-4, DU145 and T98G tumor cells: the activity of the conditioned medium (B) was evaluated with 1 :2 serial dilutions up to 1 : 128 by MTT assay at day 7. The activity of pure PTX was evaluated with 1 :2 serial dilutions starting from a concentration of 50 ng/ml. The figure demonstrates that the conditioned medium is active against neoplastic growth with kinetics similar to those of pure taxol. The PTX Equivalent Concentration (PEC) , was calculated as described in 3.5. The table shows the IC50 values of the PTX for the three tumors. Figure 7 shows the effect of growth in adherent/suspension culture and of freezing on PTX release by MSCs-PTX: the figure shows the quantity of PTX released by a single cell in the different experimental conditions:

CTRL= Standard culture with treatment and release evaluated on adherent cells. Co/Sc= Adherent primed cells subsequently frozen. Cultured under adherent conditions after thawing.

A/S= Adherent primed cells, PTX release evaluated in cultures with cells in suspension.

Figures 8-9-10 show the direct antitumor activity of MSCs-PTX against human tumor cell proliferation in vitro: the histogram indicates the mean +/- sd of tumor proliferation determined in vitro in the presence of normal MSCs and MSCs-PTX in different ratios. Proliferation is expressed as % of proliferation of tumor cells that did not receive MSCs in an MTT assay at day 7. Figure 8 shows the results for MOLT-4 leukemia cells. Figure 9 for T98G glioblastoma cells. Figure 10 for DU145 prostate carcinoma cells.

Figure 1 1 shows how PTX inhibits the proliferation of human endothelial cells (ECs): HUVECs and HMECs (2 x 10 ⁵ ) were cultured for 72 hrs in the presence of different concentrations of PTX. The numbers in the figure are means ± SD of three experiments in triplicate. Note that at a concentration of 10 ng/ml PTX completely blocks the growth both of HUVECs and HMECs; at higher doses it is e toxic. **p<0.01 PTX treated vs. controls.

Figure 12 shows the inhibition of growth of ECs co-cultured with hMSCsPTX: HUVECs (panel A) and HMECs (panel B) were mixed with hMSCsPTX or with the control hMSCs and cultured for. 72 hours. As the figure shows, the addition of MSCsPTX to ECs in ratios of 1 : 1 and 1 :5 is cytotoxic both for HUVECs and HMECs. At a 1 : 10 ratio hMSCsPTX continue to inhibit EC proliferation. The mixed co-cultures of HUVECs or HMECs with the unprimed control hMSCs do not inhibit EC proliferation. Shown in panel C are images of (20x magnification) HMECs mixed 5: 1 with hMSCsPTX fixed and stained at different incubation times after seeding. HMSCsMTX already kill HMECs at early times (24 hours). The white arrows in the pictures show the endothelial islands which are still present in the mixed culture. Note that at 72 hours practically all the HMECs seeded had been killed by the hMSCsPTX, which alone remained in the culture. The numbers in (A) and (B) are the means ± SD of three separate experiments * p<0.05, **p<0.01 vs. unprimed control hMSCs.

Figure 13 shows that the conditioned culture medium (CM) of hMSCsPTX (hMSCsPTX-CM ) is as efficacious as free PTX in inhibiting the growth both of HUVECs and HMECs: HUVECs (panel A and C) and HMECs (panel B and D) were cultured for 72 hours in the presence of different dilutions of CM derived both from cultures of hMSCsPTX (hMSCsPTX-CM) and unprimed control MSCs (hMSCs-CM). The addition of hMSCsPTX-CM induces a strong inhibition of EC growth at all dilutions tested. In particular, at the 1 :2 dilution hMSCsPTX-CM is even cytotoxic for ECs, at 1 :4 it is as efficacious as PTX at 5ng/ml. Shown in panels C and D are photographic images of the cultures of HUVECs and HMECs exposed to different dilutions of hMSCsPTX-CM. Note that the photographic image of the cultures in the presence of hMSCsPTX-CM at 1 :2 shows how only a few ECs survive the treatment, confirming the numerical data of the graph. The arrows indicate the few surviving ECs which remain anchored to the plastic (photo 10x magnification). The values in graphs A and B are means ± SD of three different experiments. *p<0.05, **p<0.01 vs. CTRL control medium or hMSCs- CM at the same dilution.

Figure 14 shows that HMSCsPTX-CM inhibits the outgrowth of microvessels from aorta rings and facilitates their regression: the rat aorta ring assay was used to test the efficacy of hMSCsPTX-CM in inhibiting vessel growth in vitro. Aorta rings were placed in collagen gel and then exposed to different dilutions of hMSCsPTX- CM or control hMSCs-CM. 20ng/ml VEGFa was added in some ring cultures as an angiogenetic stimulus and positive control to evaluate the efficiency of the aorta rings prepared. Twelve days later the capillaries newly formed from the rings were counted under a microscope. As shown by panel A in the figure, the addition of HMSCsPTX-CM at 1.2 and 1 :4 produces a strong reduction in the number of vessels sprouting from the rings if compared to the control rings or those treated with hMSCs-CM. Shown m panel B are photos of the effect of hMSCsPTX-CM and hMSCs-CM added to the aorta rings after 7 days of culture, i.e. when the capillaries are already present. The photos (20x magnification) were taken at day 10. Note that the addition of hMSCsPTX-CM at 1 :2 or 1 :4 induces regression of already formed capillaries, whereas the control medium CTRL or the addition of hMSCs-CM is not efficacious. The arrows indicate the capillaries in an advanced stage of regression with evident necrotic zones. The numerical values in (A) are the means ± SD of two separate experiments. **p< 0.01 vs. hMSCs-CM or CTRL medium alone.

Figure 15 shows that in vivo co-injection of hMSCsPTX with DU145 human prostate carcinoma cells delays their growth in Nod-Scid mice: to evaluate the efficacy of hMSCsPTX against tumor growth in vivo, approximately 2xl0 ⁶ DU145 cells were injected s.c. (right flank) in 8-week-old male Nod/Scid mice, alone or mixed in a ratio of 5: 1 (tumor cells/hMSCs) with hMSCsPTX (0.4xl 0 ⁶ ) or with control hMSCs. Tumor appearance was evaluated by palpation, whereas the tumor volumes were calculated by measuring the tumor diameters with calipers at different times after injection. The figure shows that co-injection of DU145 cells with hMSCsPTX significantly reduces tumor volume as well as delaying initial tumor appearance. Co-injection of DU145 cells and control hMSCs has no effect when compared with the growth of DU145 cells injected alone. The mice receiving DU145 cells mixed with free PTX (100 times more concentrated) likewise do not differ from the control. The mice were sacrificed 40 days after cell injection and the tumors were excised and weighed. In the box there is an image of a DU145 control tumor treated with hMSCsPTX and hMSCs after removal from the flank of the mouse. *p<0.05 and **p< 0.01 vs. DU145 alone or treated with hMSCs.

Figure 16 shows that co-injection of hMSCsPTX with B16 mouse melanoma cells reduces tumor growth in nude mice: the capacity of hMSCsPTX to inhibit tumor growth in vivo was also evaluated using the experimental B16 mouse melanoma tumor. 8-week-old nude mice were injected s.c. with 2xl0 ⁶ B 16 cells alone or mixed in a 5: 1 ratio (tumor cells/MSCs) with hMSCsPTX (0.4xl0 ⁶ ) or with control hMSCs. The tumor volumes were calculated by measuring the tumor diameters with calipers at different times after injection. As the figure shows, the addition of hMSCsPTX to B16 cells significantly reduces B16 growth as compared with B16 cells alone or B16 cells mixed with control hMSCs. At day 17, the mice were sacrificed and the tumors were excised and weighed. In the box in the figure there is a photographic image of a B16 control tumor treated with hMSCsPTX or with hMSCs. * p<0.05 vs. control B16 alone ** p<0.01 vs. control B16 alone or treated with hMSCs.

Figure 17 shows that co-injection of mouse mesenchymal MMSCsPTX with B16 melanoma cells inhibits tumor growth in syngeneic C57B16 mice: the capacity of mouse MSCs to inhibit B16 tumor growth in syngeneic mice was also tested. The figure shows the effect of 0.4 x 10 ⁵ MMSCs isolated from BM of C57B16 mice, primed or not primed with PTX and mixed with 2xl0 ⁵ B16 s.c. in a 5 : 1 ratio (B 16 tumor /MSCs) and then injected into 8-week-old male C57BL6 mice. The tumor volumes were calculated by measuring the tumor diameters with calipers at different times after injection. The Figure shows that treatment with MMSCsPTX inhibits the growth of B16 melanoma also in the syngeneic mouse strain. In this case, however, the control MMSCs injected with B16 cells reduce the growth of B16 cells when compared to mice transplanted with B 16 control cells alone. The box in the figure shows the photo of a B16 control tumor and B 16 tumors treated with MMSCsPTX or with MMSCs * p<0.05 and ** p<0.01 vs. B16 control.

Figure 18 contains Table 1, which shows the antitumor activity of hMSCsPTX and MMSCsPTX.

At the basis of this invention there is experimental evidence that in the body of mammals there exist cells contained in the connective stroma of various tissues and organs which are capable of incorporating molecules and drugs and subsequently releasing them, both in their original form and in the form of metabolites.

In previous studies it was shown that in the adult body there exists a cell population (preferentially contained in the mesenchymal, i.e. connective tissue of all organs, including blood), within a broader population identified with the generic term "stromal cells", or with the more specific term "mesenchymal stem cells" (MSCs), which is particularly resistant to high concentrations of cytotoxic drugs. In fact, such resistant MSCs (MSCs-R) can avoid suffering biological damage even in the presence of drug concentrations 100-1000 times greater than those that are capable of killing many other cells in our body. This property of theirs is initially expressed through the capacity to internalize toxic drugs, thus giving rise to a phenomenon of acute detoxification (removing/sequestering the drugs from the organ or tissue microenvironment and thus preventing them from acting upon more sensitive cells). Subsequently, however, thanks to an active secretion mechanism (probably regulated by a specific protein called P glycoprotein), these MSCs-R can release the internalized toxic drug into the microenvironment, more or less rapidly, both in the original form and in the form of metabolites. Obviously, the secretion of the drug in its original form can still cause phenomena of toxicity ("delayed acute toxicity" or chronic toxicity in the case of "slow release" of the drug by the cells) to adjacent cells sensitive to the drug. In any case, the cells most sensitive to the drug avoid direct damage that could be mortal, but would nonetheless suffer damage, depending precisely on the quantity and the rate at which the MSCs-R release the original active drug. If the MSCs-R release not the original drug but rather its "metabolites", there could arise either a condition of complete neutralization of toxicity (inefficacious metabolites) or of increased toxicity (more toxic metabolites) of the original drug. These observations were published and discussed by Pessina et al. (1999).

The principle on which this invention is based is thus the experimental evidence that in our body there exist MSCs-R capable of internalizing exogenous molecules and subsequently releasing them. The invention consists in the practical application of such a mechanism and in the use of these cells for therapeutic, diagnostic and prognostic purposes, which means defining a new way of using drugs in humans. In particular, for example, the application of our invention may serve to enhance the efficacy of chemotherapeutic drugs in the realm of cancer therapy, thanks to the capacity of MSCs-R to transport and concentrate the chemotherapeutic drug almost exclusively in neoplastic tissue, or where the application of the chemotherapeutic is advisable. To this end, after verifying the validity of the biological principle and investigating the phenotype of the MSCs-R, we standardized a procedure that provides for the production of an MSC-drug product (MSCs-D) and the therapeutic use thereof, which is defined by the type of drug used.

The procedure consists of various steps: 1 ) isolating MSCs from tissues, in particular from bone marrow and adipose tissue, which seem to be the richest; 2) expanding them in vitro according to standard culture methods; 3) selecting MSCs- R through exposure to high concentrations of the drug it is desired to have transported/internalized by the MSCs-R; 4) evaluating whether uptake of the drug has taken place in the cells with formation of the MSCs-D product and evaluating, by means of biological assays {in vitro and in vivo), the therapeutic efficacy of the MSCs-D product; 5) cryopreserving the MSCs-D product for subsequent therapeutic use.

Thanks to our procedure the MSCs of our body (but also of other species of mammals) thus become carriers of drugs (or molecules) which subsequently, thanks to their innate expulsion capacity, can be released into the tissue where it is desired to have the drug act. To do this, the MSCs-D can be either directly injected into the diseased tissue concerned, or administered systemically (via the bloodstream), also exploiting their capacity to concentrate mainly in the diseased tissue (homing). This new approach to pharmacological therapy should thus contribute to increasing the therapeutic specificity of the drug and in parallel to reducing, or even eliminating, any unwanted side effects.

The procedure was standardized for the pharmacological application of a chemotherapeutic drug called Paclitaxel (PTX), already used in cancer therapy for humans. The protocol described is however also applicable to other chemotherapeutics, and potentially also extendible to molecules with other than chemotherapeutic activity. The procedure described in detail below leads to the preparation of a pharmaceutical product composed of MSCs and PTX (MSCs- PTX) which can be used immediately at the end of preparation, or may envisage the preservation thereof (by freezing in liquid nitrogen at -196°C ) for subsequent use as necessary. In fact, the MSCs-PTX product thus obtained from this procedure does not lose its efficacy on being thawed some time (months/years) after its initial preparation.

The MSCs-PTX product was prepared with MSCs isolated from human and mouse bone marrow and adipose tissue, but can also be reproduced using MSCs/ stromal cells isolated from other organs and tissues, provided that they have morphological and phenotypical characteristics similar to those used in this study.

However, as both bone marrow and in particular adipose tissue, are tissues easily accessible for harvest in humans, it is foreseeable that this procedure can be carried out prevalently with MSCs isolated from these tissue sources. The easy accessibility of these tissues may moreover allow MSCs to be successfully prepared from practically every donor. This technical aspect is important in order to successfully arrive at the preparation of an MSCs-D product of an autologous type. The procedure thus applies both for MSC samples of heterologous (patient- donor different from the recipient) and autologous (MSCs of the same patient- donor) origin. This makes it possible to develop an MSCs-D product for implementing a personalized pharmacological therapy.

It is possible to envisage the cryopreservation of donor MSCs in biobanks without prior manipulation. This does not prevent the MSCs of a donor from being utilized, in case of need, to produce the MSCs-D product after they have been thawed.

The procedure in general is safe because it does not involve any "genetic manipulation" of the MSCs used. The procedure exploits an innate physiological behavior of the MSCs and thus does not pose any safety problems related to the introduction of new genes (as in gene therapy) or of nanoparticles, which are synthetic elements foreign to cells. Several safety problems may in any case arise with MSCs-D. First of all, MSCs have per se a spontaneous secretory activity and are capable of differentiating (being immature cells) once injected into various cellular phenotypes. However, they are already therapeutically used in clinical practice and to date no particular "toxic" activities have been documented in relation to their use. The safety problem of the MSCs-D product is potentially tied more to the type of drug that it is desired to have transported. For example, if it is desired to use MSCs as a carrier of drugs that act directly or indirectly on the DNA of a cell, when incorporated such drugs can also induce modifications in the DNA of the carrier MSCs themselves, potentially inducing genome modifications in the MSCs themselves.

Materials and methods MSC isolation, expansion and phenotypic characterization.

The procedure for isolating and expanding MSCs in vitro applied to human and mouse bone marrow and human adipose tissue is as described by Pittenger et al, (1999). The stabilized SR-4987 line generated and maintained in our laboratory through serial passages (Pessina et al, 1992 ) was also used as a mouse MSC model. The MSCs, both mouse and human, show similar morphology in vitro. MSCs grow as plastic-adherent cells in culture and appear with a fibroblast-like morphology. When cultured in NHEM medium (Miltenyi, Germany), MSCs generally double in number in approximately 72-96 hours and their cellular cycle appears to be fairly slow, with a high percentage of cells in the G0/G1 phase (70- 90 %), a modest fraction in phase S (8-25 %) and only a small fraction in the G2/M phase (2-4%). As shown in figure 1, the phenotypic characterization reveals that the MSCs used display the classic markers present on mesenchymal cells, namely, Cdl 3, Cd90, Cdl 05, Cd29, Cd44, Cd73 and Cdl 66, whereas mouse endothelial and hematopoietic markers Cd31 , Cd34, Cd45 and Cdl 33 are absent. The markers Cd54, Cdl46, Cdl 17 and Stro-1 are barely present. Finally, the class I histocompatibility antigen (HLA-I) is present, whereas class II (HLA-II) is absent. Preparation of the drug.

In this study, as previously illustrated above, use was made of a chemotherapeutic drug, Paclitaxel (PTX), already known for its antitumor activity in humans, in order to realize the final MSCs-PTX product. The PTX stock solution (Serva, Germany) is prepared by dissolving 5 mg/ml in DMSO and stored in ΙΟμΙ aliquots at -20°C. The working solution is freshly prepared by 1 :25 dilution in DMEM medium (Lonza, Usa) so as to obtain a PTX solution of 200 micrograms/ml. Procedure for formation of the product MSCs-PTX and verification of PTX uptake in MSCs

Confluent cultures of MSCs prepared and characterized as previously illustrated above are treated with trypsin-EDTA (0.05% /0.02%), washed in PBS and counted in a Burker chamber.

10 cells are seeded in 25 cm plastic flasks (Costar) in 5 ml of NHEM medium and incubated at 37°C in an atmosphere of air + 5% C0 ₂ after 96- 120 hours. When confluence is reached (3-4 x 10 ⁵ cells), 50 microliters of freshly prepared 200 μg/ml PTX solution (as described in 3.2) is added to the culture medium to obtain a final PTX concentration of 2^ig/ml. In order to generate the MSCs-R-PTX, the MSCs are maintained at 37°C in air + 5% C0 ₂ for at least 24 hours. At the end of incubation the supernatant of the cell culture is aspirated and discarded, the adherent cells are washed twice with 5 ml of HBSS and then detached by adding 1 ml of a trypsin-EDTA solution (0.05 % + 0.02 % weight/vol.) for 5'. When about 80 % of the cells in the flask are detached from the plastic, 2 ml of a complete DMEM medium with 2 mM L-glutamine plus 20% FCS is added.

After vigorous stirring (to favor detachment of all the cells), the detached cells are aspirated and transferred into a 15 ml conical tube. The flask is further washed with 10ml of HBSS and the wash is added to the conical tube containing the initially collected cells. This is followed by 10' centrifugation at 400xG at room temperature to recover the cells. At the end of centrifugation, the supernatant is discarded and the MSCs-PTX are resuspended in 2.0 ml of HBSS. A second 10' centrifugation at 400xG is carried out to completely eliminate any PTX not incorporated by the MSCs. Finally, the MSCs-PTX are resuspended in 1 ml of NHEM and counted in the presence of a vital stain (trypan blue) to evaluate their viability. At this point the MSCs-PTX are ready to be used directly in a therapeutic application (for example, as an antitumor agent) or frozen and stored until the time of use. PTX uptake in the MSCs can be confirmed, in the specific case of PTX, via parallel use of PTX marked with a fluorescent substance (FITC) and subsequently analyzed by cytofluorimetry (with FACS). One or more aliquots of MSCs-PTX produced are used to establish the time of release of PTX from the MSCs and its therapeutic efficacy after release.

Evaluation of antitumor activity of conditioned medium (CM) from MSC- PTX cultures in vitro.

PTX release by the MSCs-PTX produced can be analyzed by evaluating the antiproliferative and cytotoxic activity of MSCs-PTX. This evaluation was performed by initially testing their conditioned medium (CM). To prepare it, approximately 3-5 xl O ⁴ MSCsPTX /ml (viability > 90%) are seeded in 5 ml of NHEM in a new 25 cm ² flask.

After 24 hours of cell culture (37°C in air+5%C0 ₂ ) as previously described, the CM is collected and replaced with fresh medium. In order to study the release kinetics this operation is repeated at different times (48, 72 , 96 and 144 hours) after seeding. The collected CM is centrifuged at 2,500xG for 15 minutes to eliminate any dead cells. The CM is then divided into aliquots which can be tested immediately or frozen at -70°C until the time of use.

PTX release by the MSCs-PTX in the CM was verified by evaluating the antiproliferative or cytotoxic activity against a reference tumor model, Molt-4 (human acute lymphoblastic leukemia; Minowada et al 1972), which is highly sensitive to PTX. The proliferation test on Molt-4 was performed by MTT assay as previously described (Mossman 1983; Pessina et al., 1994). The antitumor activity of MSCs-PTX in the CM was evaluated by making various dilutions so as to establish an IC50 and IC90 concentration. The same type of procedure was also applied on other human tumor lines such as T98G glioblastoma (Stain et al, 1979) and DU- 145 prostate carcinoma (Mickey et al.,1977) and also on a line of mouse melanoma, B 16 (Riley, 1963). The Molt 4, DU145 and B 16 cells were cultured and expanded in vitro according to previously published culture methods (Pessina et al 1994, Mickey et al 1977, Riley et al 1963). The CM derived from MSCs cultured under the same conditions but not primed with PTX was used as the reference conditioned medium (negative control). The IC50 and IC90 value for each line is determined according to the Reed and Muench formula (1938)

Evaluation of antitumor activity of MSCs-PTX co-cultured with tumor cells. The antitumor activity of MSCs-PTX in vitro was also evaluated through mixed cultures with tumor cells. The MSCs-PTX were co-cultured with the reference Molt-4 tumor cells. The MSCs-PTX obtained as described in 3.3 were mixed in different ratios with the Molt-4 cells (MSCs-PTX/Molt-4 1 : 1 , 1 :5, 1 : 10 etc.) in order to define the IC50 and IC90 for Molt-4. The proliferation test was performed as described in 4.1 via MTT assay. The antitumor activity of MSCs-PTX was also evaluated through mixed cultures with other tumor lines described in the previous paragraph. In this case as well the control cells were MSCs not primed with PTX and mixed with tumor cells in the same ratios as used for the MSCs-PTX.

Evaluation of anti-angiogenic activity of MSCs-PTX in vitro

PTX is also considered a chemotherapy drug with an anti-angiogenic capacity and thus capable of inhibiting the growth of capillaries. Cultures of endothelial cells (ECs) were used to evaluate the anti-angiogenic action of MSCs-PTX in vitro. Two different types of ECs were used: HUVECs, which are ECs derived from human umbilical cord veins and ECs derived from human skin microvessels (HMECs). The HUVECs and HMECs were isolated and cultured as described by Alessandri et al (2005). The antiproliferative effect of the MSCs-PTX-CM on the ECs was evaluated using the same methods as already described for the tumor lines described previously. Similarly, the co-culture between MSCs-PTX and ECs took place with the same methods as described in the previous paragraph. The anti-angiogenic activity of MSCs-PTX was also evaluated by means of the rat aorta ring assay. With this test, described by Nicosia et al (1990), it is possible to study microvessel growth in vitro. The MSCs-PTX derived CM was assayed at various dilutions. The microvessels arising from the rings were counted according to the methods described by Nicosia et al ( 1990). The controls of all these experiments were either the CM derived from MSCs not primed with PTX, or normal MSCs in the co-culture assays with ECs.

Biological assay of PTX release in vitro

The quantity of PTX released by MSCs-PTX was calculated by biological assay of the antitumor activity of their CM. The calculation was based on the activity of a known quantity of pure PTX. The concentration was then expressed as a paclitaxel equivalent concentration (PEC) according to the following algorithm:

PEC(ng/ml) = FD50mc x PTXc/ DF50PTX

where FD50mc and FD50ptx are the dilution factors at which CM and pure PTX respectively cause 50% inhibition of Molt-4 proliferation and PTXc is the starting concentration of pure PTX. The PTX release by a single cell (Rptx), expressed in pg/cell, is calculated with the formula Rptx = PEC (ng/ml) x CM Volume (ml) / number of cells seeded. PTX release in the CM from MSCs-PTX was also evaluated by HPLC according to the method suggested by Gianni et al., 1995.

Evaluation of antitumor activity of MSCs-PTX in vivo.

The capacity to inhibit tumor growth in vivo was tested both with MSCs-PTX of human (hMSCs-PTX) and mouse (MMSCs-PTX) origin. The control cells were MSCs of the two species not primed with PTX. The hMSCs-PTX and MMSCs- PTX were prepared according to the previously described procedure. The tumor cells used for the in vivo study were the human DU145 line and mouse B16 line. The two tumor lines were cultured in vitro under the standard culture conditions described in the paragraph regarding the antitumor activity in vitro of the conditioned media derived from MSCs-PTX cultures. To evaluate the antitumor activity of hMSCs-PTX on DU145 cells (human prostate tumor), 60-day-old male NOD/SCID mice were used. The experimental groups were thus composed: 1) control mice which each received DU145 cells alone at a concentration of 2xl 0 ⁶ /0.2ml saline injected subcutaneously (s.c.) into the right flank; 2 ) mice each injected s.c with 2xl0 ⁶ DU145 cells and 0.4x 10 ⁶ hMSCs-PTX (ratio 1 :5) in 0.2 ml mixed 5' prior to injection; 3) mice each injected s.c. with 2x10 DU145 cells and 0.4 xlO ⁶ control hMSCs (i.e. not primed with PTX) (ratio 1 :5) in 0.2 ml mixed 5' prior to injection: 4) mice each injected s.c. with 2 x 10 ⁶ DU145 cells resuspended in 0.2 ml saline containing 100μg/ml of pure PTX 5' prior to injection. To evaluate the antitumor activity of hMSCs-PTX on B 16 mouse melanoma, 60-day-old male nude mice were used. The experimental groups were thus composed: 1) control mice each injected s.c. with 2x10 ⁶ B 16 cells in 0.2 ml; 2) mice each injected s.c. with 2xl0 ⁶ B16 cells and 0.4 xl O ⁶ hMSCs-PTX (ratio 1 :5) in 0.2 ml mixed 5 ' prior to injection; 3) mice each injected s.c. with 2χ10 ^ό B16 cells and 0.4x10 ⁶ hMSCs (ratio 1 :5) control mixed 5' prior to injection; 4) mice each injected s.c. with 2x10 ⁶ B 16 cells resuspended 5' prior to injection in 0.2 ml saline containing 10C^g/ml of pure PTX. To evaluate the antitumor activity of MMSCs-PTX on B 16 mouse melanoma, 45-day-old male C57/B16 syngeneic mice were used for both cell preparations. The experimental groups were thus composed: 1) control mice each injected s.c. with 2x10 ⁵ B16 melanoma cells in 0.2 ml; 2) mice each injected s.c. with 2xl0 ⁵ B 16 cells and 0.4 xl O ⁵ MMSCs-PTX (ratio 1 :5) in 0.2 ml mixed 5' prior to injection; 3) mice each injected s.c. with 2xl0 ⁵ B 16 cells and 0.4 xl O ⁵ control MMSCs (ratio 1 :5) in 0.2ml mixed 5' prior to injection; 4) mice each injected s.c. with 2 xl O ⁵ B16 cells resuspended 5' prior to injection in 0.2 ml saline containing 100μg/ml of pure PTX. Tumor appearance was monitored by palpation of the flank of the mouse injected s.c. with cells, whereas the tumor growth rate was determined by measuring the tumor diameter with calipers at various times after the transplant. The mice were sacrificed and the subcutaneous tumors were excised and weighed.

Summary of parallel procedures for verification of some experimental conditions

1. The passive release of PTX (not internalized by the cells, hence specifically bonded to the cells) adsorbed onto the cell membrane was evaluated by treating confluent MSCs with PTX for 24 hours; the MSCs were pre-fixed with 10% formalin for 30 minutes and washed three times with bidistilled water prior to addition of the drug. The rest of the procedure is as described above.

2. PTX internalization was also evaluated with MSCs treated with the drug under conditions of forced growth in suspension. For this purpose 3x 10 ⁵ StCs were resuspended in a conical tube with 5 ml of their culture medium containing 2μg/ml of PTX. The cells in the tube were incubated for 24 hours under continuous stirring to prevent their adhesion. At the end of incubation the MSCs-PTX were collected by centrifugation and washed twice, again by centrifugation, to eliminate any PTX not incorporated. The subsequent release of the PTX internalized by the MSCs-PTX was evaluated by determining the cytotoxic activity of their CM after the MSCs-PTX had been cultured for 24 hours under adherent conditions in a new culture flask. In another series of experiments we also evaluated the capacity of MSCs- PTX, treated under adherent conditions, to release PTX if forced to remain in suspension. Briefly, the procedure for incorporating PTX into MSCs was carried out as previously described. After detachment with trypsin the MSCs-PTX were incubated for 24 hours suspended in their medium. At the end of incubation the CM was collected after centrifugation and tested for its cytotoxic activity as already described.

3. Release after freezing was verified with cells treated as previously described and frozen in liquid nitrogen (-196° C) after trypsinization. The cells were thawed after one month and seeded in a flask. The cytotoxic activity was evaluated as previously described.

Experimental results

As shown in figure 2, high concentrations of PTX (up to 10,000 ng/ml) do not affect the viability of MSCs, demonstrating them to be particularly resistant. The analysis of MSC resistance to PTX was also confirmed by analyzing their cellular cycle, which is not substantially modified by priming with PTX (data not shown). As shown in figure 3, MSCs-PTX (treated 24 hours with PTX), trypsinized, washed and recultured in vitro release the drug with time-dependent kinetics at concentrations which are pharmacologically efficacious against neoplastic proliferation.

As shown in figure 4, PTX internalization by MSCs was verified by treating the MSCs with fluorescent PTX (Invitrogen, USA). Thereafter, according to the above-described procedure, the cells were trypsinized, washed, suspended in PBS (Phosphate Buffered Saline) and then analyzed with a Fluorescence Activated Cell Sorter (FACS) and observed under a microscope ( Ex :496nm; Em: 524 nm).

This demonstrated that the majority of the cells contained PTX, which was subsequently released as shown by the fluorescence time decay curve (2, 4, 8, 24, 48 hours) as determined by FACS and shown in figure 4 as a ratio to unmarked PTX.

The presence of PTX in the conditioned medium of the primed cells was confirmed by HPLC analysis as shown in figure 5.

As shown in figure 6, the "in vitro" antitumor activity of the cell conditioned medium collected 24 hours after treatment evaluated in three different human tumor models, namely, Molt-4 lymphatic leukemia, T98G glioblastoma and DU- 145 prostate carcinoma, is comparable to that of pure PTX. On the basis of this activity the Paclitaxel equivalent concentration (PEC) is calculated as described in the paragraph regarding the biological assay of PTX release in vivo.

As shown in figure 7, MSCs-PTX maintained in suspension release PTX. After priming the same cells can be frozen in liquid nitrogen and subsequently thawed and cultured without losing their drug release capacity. As shown in figures 8, 9 and 10, trypsi ized, washed MSCs-PTX are efficacious against in vitro proliferation of the three human tumor models. These (drug loaded) cells mixed with tumor cells (in different proportions, 1 : 1 - 1 : 10 - 1 : 100) show a significantly greater antiproliferative activity compared to the control MSCs not primed with PTX.

The same analysis of MSCs-PTX efficacy in inhibiting tumor growth in vitro was also carried out to evaluate anti-angiogenic potential. Initially the efficacy of pure PTX to inhibit the growth of ECs was determined. As Figure 1 1 shows, the addition of PTX at concentrations greater than l Ong/ml is cytotoxic both in cultures of HUVECs and of microvascular HMECs. The concentration of PTX that inhibits EC growth by 50% is around 4.69 ng/ml.

The capacity of MSCs-PTX to inhibit EC growth was then evaluated. In a first series of experiments, MSCs-PTX of human origin were tested by being mixed with ECs in varying ratios. Co-culture of hMSCs-PTX with ECs in ratios of 1 : 1. 1 :5 and 1 : 10 produce strong, dose-dependent inhibition of the growth both of HUVECs and of HMECs, as is shown in figure 12. The 1 :5 ratio of hMSCs- PTX/ECs at 72 hours of co-culture is sufficient to cause the complete elimination of ECs from the culture, as is shown by the image in Figure 12 (panel C). Co- culture with control hMSCs, not primed with PTX, in the same ratios, did not bring about any particular inhibitory effect on the growth either of HUVECs or HMECs, as can be seen from panels A and B of Figure 12.

The strong inhibitory action of hMSCs-PTX on the proliferation of the ECs was also confirmed by the analysis of their CM. Up to 1 :8 dilutions hMSCs-PTX-CM is capable of significantly inhibiting the growth both of HUVECs and HMEC, as figure 13 shows. At 1 :4 dilutions or greater the activity of hMSCs-PTX-CM produces a cytotoxic effect on ECs that is even greater than the effect obtained with 5ng/ml of pure PTX, as can be seen from panels A and B of Figure 13. The cytotoxic action of 1 :2 and 1 :4 dilutions of MSCs-PTX-CM is so great that only a few ECs remain alive after a 72 h treatment, as shown by the images in panels C and D of Figure 13. In this case as well the addition of control hMSCs-CM to ECs at the same dilutions shows no efficacy.

Finally, the capacity of hMSCs-PTX-CM to inhibit microvessel growth ex-vivo was analyzed with the aorta ring test. Different dilutions of hMSCs-PTX-CM were applied to aorta ring cultures and dose-dependent inhibition of microvessel growth was observed. 1 :2 and 1 :4 dilutions of hMSCs-PTX-CM inhibit the sprouting of new capillaries from the ring to a highly significantly degree, as can be noted in panel A of figure 14. It is interesting to note that after around 7-10 days of ring culture, i.e. when the vessels are already formed, the addition of hMSCs-PTX-CM at 1 :2 and 1 :4 dilutions induces vascular regression, as is shown in panel B of Figure 14. The addition of hMSCs-CM to aorta rings does not have an inhibitory effect, but rather seems to increase their number, as is shown in panel A of figure 14 .

The addition of hMSCs-CM after seven days likewise does not induce capillary regression and seems even to increase their vascular stability (panel B of Figure 14).

The above-described experiments on the anti-angiogenic activity of hMSCs was confirmed using mouse mesenchymal cells. MMSCs-PTX in co-cultures with ECs and use of their CM were both as effective as the human-derived counterparts (data not shown). To verify that priming MSCs with PTX is efficacious in inhibiting the growth of human tumor cells in vivo, a DU145 prostate tumor line transplanted into NOD/SCID (immunodeficient) mice was used. DU145 cells (at a concentration of 2xl 0 ⁶ per mouse) were co-injected subcutaneously (s.c.) either alone or mixed in a 5: 1 ratio (tumor cells/mesenchymal cells) with hMSCs-PTX (0.4x10 ⁶ ) or with unprimed hMSCs used as control cells. Finally, mice were also injected with DU145 cells resuspended with free PTX at a concentration of lC^g/ml. As shown in figure 15, co-injection of DU145+hMSCs-PTX produces a significant inhibition of tumor growth (measured as a reduction in tumor volume at different times after injection). This inhibition of DU145 growth also resulted in a reduction of tumor weight at the time of sacrifice, which took place around 40 days after the transplant, as shown in Table 1. The co-injection of DU145 with control hMSCs did not produce any significant effect, just as in the case of DU145 previously treated with pure PTX (the curve is not shown in Figure 15 because it is practically identical to the one with DU145 alone). It is interesting to note that co-injection of DU145 with hMSCs-PTX also produces a delay in tumor appearance, as can be seen from figure 1 .

The efficacy of hMSCs-PTX in inhibiting the growth of tumor cells in vivo was also analyzed using an experimental mouse tumor, B 16 melanoma. This tumor model is widely used in the realm of cancer research. In this experiment, B 16 melanoma cells were co-injected s.c. into nude mice at a concentration of 2xl0 ⁶ cells/mouse, again in a 5: 1 ratio (tumor cells/mesenchymal cells) with hMSCs- PTX ( 0.4 x 10 ⁶ ) or with control hMSCs and B16 cells in the presence of pure free PTX at 10μg/ml. The B16 tumor has a much faster growth rate than DU145; despite this, co-injection of B16 + hMSCs-PTX produced a significant reduction in the growth of B16 cells of and in tumor weight at the time of sacrifice of the mice, which took place on day 14 after injection, as shown in figure 16 and in Table 1. In this case as well, a delay was observed in the appearance of the tumor in the mice co-injected with B16-hMSCs-PTX, as was the absence of any effect on tumor weight at the time of sacrifice in the case of co-injection of B 16 with control hMSCs or of B16 injected in the presence of pure PTX (Table 1).

Finally, the capacity of MSCs of mouse origin (MMSCs) to inhibit the growth of B 16 melanoma was studied. The aim of this experiment was to verify the capacity of MMSCs-PTX to function in a syngeneic tumor system. The MMSCs used were derived from bone marrow of C57B16 mice, the strain of mice in which B 16 melanoma grows. Male C57B16 mice were co-injected s.c. with 2xl 0 ⁵ B16 melanoma cells alone, or mixed in a 1 : 1 or 5: 1 ratio with MMSCs-PTX or control MMSCs. In this case as well, a group of mice was treated with B16 cells mixed with pure PTX at the time of injection. As Figure 17 shows, the MMSCs-PTX significantly inhibited B 16 growth in the C57BL6 mice, both at 1 : 1 dilution and at a greater dilution of 5 : 1 dilution (tumor cells/mesenchymal cells); indeed, at the latter ratio the effect on B16 tumor weight seems to have been even more efficacious than the 1 : 1 ratio (Table 1 ). In this experiment a certain efficacy was also observed on the part of control MMSCs in reducing the growth of B 16 cells, particularly at the 1 : 1 ratio (Table 1). The action of free PTX added to B16 had no effect on tumor growth (Table 1 ). Finally, in the syngeneic C57B16 model as well, MMSCs-PTX delays subcutaneous B16 tumor take (data not shown), suggesting that the immune system has little influence at least in the initial phase of tumor take. CONCLUSIONS

The present invention discloses a procedure for preparing a product defined with the generic name "MSCs-D". This product is composed of two elements, one consisting of mesenchymal cells called MSCs, the other element consisting of a specific drug. The end pharmaceutical product will thus be named according to the drug that is used in the therapy. In our case, as we used the drug PTX , the final product was named "MSCs-PTX".

At the basis of this invention there is experimental evidence that in the body of mammals there exist cells contained in the connective stroma of various tissues and organs which are capable of incorporating molecules and drugs and subsequently releasing them, both in their original form and in the form of metabolites These cells, MSCs, cany out a detoxification process to neutralize the activity of molecules having potentially toxic effects on our body.

This property makes these cells utilizable as drug carriers, thus permitting the development of an approach which is an alternative to the classic present-day administration of drugs in humans.

The procedure described here was developed in order to obtain an MSCs-D product with cytotoxic, antitumor and anti-angiogenic activity. In fact, the drug used was Paclitaxel (PTX), a well-known antitumor drug with demonstrated anti- angiogenic capacity. The procedure is not solely applicable for the transport of chemotherapy drugs, but rather may also be applied to deliver drugs and/or molecules other than the ones described.

The pharmacological efficacy (antitumor and anti-angiogenic activity) of the "new" MSCs-PTX drug produced was confirmed both in tumor and endothelial cells in vitro and in human and mouse tumors in vivo. The MSCs-PTX product demonstrated to be active in vitro to the same degree as free PTX, while in vivo it is much more efficacious. In fact, adding free PTX (100 times more concentrated than the PTX uptake in cells) to tumor cells demonstrated to be inefficacious in inhibiting their growth in vivo, as did adding the same "carrier" MSCs not previously primed with PTX.

In conclusion, the experimental data reported here thoroughly support the validity of our discovery. The procedure for obtaining a new pharmaceutical product generically called "MSCs-D", to be used in treating human pathologies, is described in a detailed manner.

7.0 References

Invernici G, Ponti D, Corsini E, Cristini S, Frigerio S, Colombo A, Parati E, Alessandri G. Human microvascular endothelial cells from different fetal organs demonstrate organ-specific CAM expression.Exp Cell Res. (2005) 308 :273-282.

Gianni L, Kearns CM, Giani A, et al. Nonlinear pharmacokinetics of paclitaxel and its pharmacokinetic / pharmacodynamic relationships in humans. J Clin Oncol 1995; 13 : 180.

Mickey DD, Stone KR, Wunderli H, Mickey GH, Vollmer RT, Paulson DF.

Heterotransplantation of a human prostatic adenocarcinoma cell line in nude mice.Cancer Res. 1977 Nov;37(l l):4049-58. Minowada J, Onuma T, Moore GE. Rosette-forming human lymphoid cell lines. I. Establishment and evidence for origin of thymus-derived lymphocytes. J Natl Cancer Inst. (1972) Sep;49(3):891-5.

Mossman T.Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. Journal of Immunological Methods 65 (1983), pp. 55-63.

Nicosia R F and Ottinetti A. Growth of microvessels in serum free matrix culture of rat aorta: a quantitative assay of angiogenesis in vitro. Lab Invest (1990) 63: 1 15-122

Pessina A, Mineo E, Neri MG, Gribaldo L, Colombi R, Brambilla P, Zaleskis G. Establishment and characterization of a new murine cell line (SR-4987) derived from marrow stromal cells. Cytotechnology. 1992; 8 (2) : 93- 102.

Pessina A, Gribaldo L, Mineo E, Neri MG. In vitro short-term and long-term cytotoxicity of fluoroquinolones on murine cell lines. Indian J Exp Biol. 1994 Feb;32(2): 1 13-8.

Pessina A, Piccirillo M, Mineo E, Catalani P, Gribaldo L, Marafante E, Neri MG, Raimondi A. Role of SR-4987 stromal cells in the modulation of doxorubicin toxicity to in vitro granulocyte-macrophage progenitors (CFU-GM).Life Sci.

1999;65(5):513-23. Pittenger MF, Mackav AM. Beck SC, Jaiswal RK, Douglas R, Mosca JD,

Moorman MA, Simonetti DW. Craig S, Marshak DR. Multilineage potential of adult human mesenchymal stem cells. Science. 1999 ;284(541 1): 143-7.

Reed, L.J., Muench, H. A simple method of estimating fifty percent endpoints. The American Journal of Hygiene (1938). 27: 493-497.

Riley ,V. Enzymatic determination of transmessible replicating factors associated with mouse tumors, Ann. N. Y. Acad. Sci.( 1963) 100: 762-790.

Stein GH.. T98G: an anchorage-independent human tumor cell line that exhibits stationary phase Gl arrest in vitro J Cell Physiol. 1979 Apr;99( l ):43-54.