Field of the invention

The invention relates to a method of determining a gene combination that can identify in a very short time the potential of stem cells, typically human stem cells, of generating bone tissue.

Background art

Muscoloskeletal diseases and disorders are more frequent than cardiac and respiratory diseases or tumors (see for instance :ip;// w i)oneandiointburden.orq/).

For this reason, research is focusing on the development of novel stem cell- based therapeutic approaches.

As is known, there are various classes of stem cells, including: embryo stem cells, which are capable of generating any type of tissue, and adult stem cells, which are capable of generating tissues deriving from the same germ layer from which they are originated.

Mesenchymal stem/stromal cells, hereinafter briefly referred to as MSC, are the adult stem cells that are most commonly used in regenerative medicine. The Mesenchymal and Tissue Stem Cell Committee, ISCT has set rules for unique identification of these MSC.

According to a first rule, MSCs must be capable of adhering to plastic, if they are maintained in standard culture conditions.

According to another rule, about 90% of the population must have the following molecules on the cell surface: CD90, CD73 and CD105.

Furthermore, these MSC must be negative to the following molecules:

CD1 1 b, CD14, CD19, CD34, CD45, CD79a and class II HLA.

According to a further rule, MSCs must be capable of differentiating into osteoblasts, adipocites and chondrocites, under particular "ex vivo" stimulating conditions.

The effectiveness of cell therapies was tested using bone marrow, hereinafter briefly referred to as BM, in the treatment of different hematological diseases and solid tumors. Considering the ability of MSCs of differentiating into bone, several animal models of long bone defects (Nather, 2010) and osteogenesis imperfecta (Otsuru, 2012) were successively treated

The first MSCs were isolated in 1974 from bone marrow (Friedenstein, 1 966). In bone tissue regeneration approaches therapeutic doses of about ten millions MSCs are infused per kilogram of patient weight.

In the BM, MSCs are from 0.001 % to 0.01 % the total cell populations.

As a result, in order to obtain suitable therapeutic doses, the amount of MSCs in the BM must be separated and enriched by cell culture (Dominici, 2001 ).

The production of cells required for clinical applications shall comply with existing good manufacturing practices, hereinafter briefly referred to as GMP. Such GMP standards have the purpose of ensuring that the cell product will be: sterile, pure, safe and functional.

Animal derivatives are currently used in MSC isolation and growth media.

In order to eliminate contaminants from the cell culture and maintain compliance with GMP standards, MSCs cultivated for clinical applications must be treated with animal derivative-free media.

In recent years, media have been provided which contain serum or other derivatives of human origin, such as platelet lysate (PL) (Xia, 201 1 ).

Production of cell therapies in GMP-accredited structures shall occur within a period of about three weeks, such that therapy administration will not be delayed and no changes will be introduced in the cell phenotype.

Short cell isolation and growth times are critical for MSCs in autologous cell therapy, i.e. if the cells to be used have been obtained from the same patient to be later treated.

No specific test is known at present, that can predict the ability of MSCs to form bone tissue in the patient.

The key in the creation of such a test consists in identifying an cell potential marker during testing ("ex vivo") that might directly predict the ability of MSCs to generate bone tissue in a living body ("in vivo" system).

The ability of MSCs of differentiating into bone "ex vivo" can be investigated by analyzing different stages of the process.

This is because, during the differentiation process from MSCs to mature osteoblasts, cells express genes that code for factors regulating the production of other genes or acting in the formation of the matrix (that acts as a scaffold for the bone tissue).

During formation of the bone tissue, solidity shall be imparted to the tissue structure, by deposition of minerals onto the matrix: this process is known as mineralization and can be verified by cytochemical stains of MSCs.

Particularly, the so-called Von Kossa staining highlights phosphate and carbonate anions deposited onto the matrix, whereas Alizarin staining reacts with calcium and other cations (Puchtler, 1 969).

Both cytochemical methods can be used in the late stages of osteoblast maturation and have a reduced sensitivity.

Due to the necessity of checking that all the stages of the bone differentiation process actually occur, certain researchers proposed to combine gene analysis or cytochemical staining with bone formation in an animal model, to check whether a correlation exists between the osteogenic potential of "ex vivo" MSCs and "in vivo" bone formation.

A prior art approach consisted in analyzing: alkaline phosphatase activity, the expression of type I procollagen and osteopontin in human MSCs implanted subcutaneously in a nude mouse model (Mendes, 2004).

Nevertheless, these researchers have not been able to solve the problem of highlighting some direct correlation between the expression levels of the markers and the bone formation ability n a murine model.

Further studies focused on the analysis of the correlation between "ex vivo" expression of genes typically expressed by osteoblasts and the ability of these MSCs of forming bone "in vivo".

These studies analyzed the following genes: bone sialoprotein (BSP), osteopontin (OPN) and osterix (OSX) (Jaquiery, 2005).

The researchers found that high expression levels of the 3 genes are directly correlated, with statistically significant values, for "in vivo" bone formation. This kind of analysis was supported by another study that investigated the correlation between the proliferation ability, the ability of differentiating into bone and the "ex vivo" gene expression with "in vivo" bone formation (Janicki, 201 1 ).

Thus, the researchers analyzed the bone generation potential of MSCs by combining the use of a cytochemical test and molecular analysis.

Positive cytochemical tests were found not to be directly correlated with "in vivo" bone formation whereas, during molecular analysis, the researchers found highly variable alkaline phosphatase gene expression levels (ALPL) in MSCs having an "in vivo" bone formation ability.

Another group of researchers studied the correlation between the ability of MSCs of differentiating into bone "ex vivo" and "in vivo" bone formation (Larsen et al., 201 0, Burns et al., 201 0), but this study differs from the present invention by four basic aspects.

The first difference is based on the use of MSCs that are genetically modified with the TERT gene to become immortal.

The second difference consists in that MSCs are treated with media containing animal derivatives.

The last two differences are based on the use of different sequences to identify the expression of genes typical of osteoblasts and on molecular analysis conducted after fourteen days' treatment.

The above results and various pre-clinical tests confirm a high heterogeneity in the osteogenic differentiation potential of BM-MSC populations (Phinney, 1999).

Therefore, considering the short cell expansion times, heterogeneity and the need to obtain an effective cell product, one object of the invention is to provide a potential assessment test that can define the quality of the cell product before administration to the patient and during cell expansion time. Another object of the invention is to improve the prediction of the bone tissue generation potential of MSCs by using a multi-parameter test that accounts for the various basic aspects for "in vivo" bone formation.

These aspects comprise both the ability of expressing critical genes for bone formation and the ability to proliferate and deposit minerals into the matrix of the bone tissue.

Disclosure of the invention

According to the invention a group of genes was identified, named biological markers or biomarkers, which are part of the typical genetic complement of osteoblasts and whose expression levels are found to increase upon treatment of MSCs with an appropriate culture medium.

More in detail, according to the invention an assessment was made to check whether MSCs isolated in compliance with GMP standards and later stimulated toward bone differentiation with or without animal derived reagents, are able to increase the expression of the selected biological markers as compared with the same cells without such treatment, i.e. control cells.

These gene expression levels were assessed both after two weeks' treatment, with cytochemical methods and after one single week, to comply with the times required by the clinical application.

Still according to the invention, MSCs from various BM donors were selected based on their ability to:

• depositing minerals into the matrix, by selecting Alizarin and Van-Kossa stain-positive samples;

· proliferating, by MKI67 gene analysis,

• forming bone "in vivo" in a murine animal model.

Once positive MSCs have been identified for the aforementioned "ex vivo" and "in vivo" tests, the biological markers were later tested for their expression, and were found to be significantly overexpressed in the treated samples as compared with control samples.

This multi-parameter approach could mediate the heterogeneity of the different BM-MSC samples being tested.

At the end of this molecular assessment, five genes were identified, whose expression values were found to be increased in a statistically significant manner in all the positive samples in the various tests.

These five genes were this defined "marker genes", i.e. genes suitable for bone formation potential analysis. The "marker genes" have essential functions in the bone formation process, including:

• coding alkaline phosphatase(ALPL), i.e. coding an essential enzyme for bone matrix mineralization;

· coding protein components of the bone matrix, i.e. Coding for producing collagen 1 A2 (COL1 A2), decorin (DCN) and elastin (ELN);

• regulating the expression of other genes (RUNX2).

Therefore, the invention teaches how the osteogenic potential in a heterogeneous group of six BM-MSC donor can be defined from the analysis of five genetic markers.

In one aspect, the invention relates to a method of determining a gene combination, as defined in the features of claim 1 .

Further aspects of the invention are defined in the dependent claims.

The invention achieves the following advantages: assessing the ability of MSCs to differentiate into bone, by analyzing the expression of five genetic markers, before administration to the patient.

Brief description of the drawings

Further features and advantages of the invention will be more readily apparent upon reading of the detailed description of a preferred non exclusive embodiment of a method of determining a gene combination, which is shown as a non limiting example by the annexed drawings, in which:

FIG. 1 a shows images of BM-MSCs induced into bone and stained by cytochemical staining with Alizarin;

FIG. 1 b shows microscope images of BM-MSCs induced into bone and stained by cytochemical staining with Von Kossa;

FIG. 2 is a chart that represents the expression of the marker MKI67, at one week (white column) and at two weeks (grey column) respectively;

FIG. 3 is a chart that shows the percents of formed bone (dark column), of biomaterial implanted in the animal model (grey column) and of other formed tissue (bright column) according to the individual donors and the control parameter CTL;

FIG. 4 is a chart that represents the number of known and predicted interactions between the protein products of the five marker genes during the three bone differentiation stages;

FIG. 5 is a chart that represents the expression of the five molecular biomarkers in BM-MSCs after one week's induction treatment;

FIG. 6a is a chart that represents the similarities in the expression of the five biomarkers in the six BM-MSC donors;

FIG. 6b is a chart that represents the correlation coefficient "r ² " between the expression of the biomolecular markets at one week and bone formation in the animal model of the six BM-MSC donors;

FIG.7a is a chart that represents the expression of COL1 A2 in the donor #6 after treatment with the inhibitor SB431 542 and interferon-gamma (INF - Y);

FIG. 7b are microscope images of the bone-induced donor #6, before and after treatment to decrease the expression of COL1 A2;

FIG.8a is a chart that represents the correlation coefficient "r ² " between the similarity of the various donors in the expression of the five bio-markers at one week and the "in vivo" bone formation, using a COL1 A2 expression value for donor #6 defined as an average of the test values of the other donors capable of "in vivo" bone formation;

FIG. 8b is a chart that represents the correlation coefficient "r ² " between the similarity of the various donors in the expression of the 5 bio-markers at two weeks and "in vivo" bone formation;

FIG. 8c is a chart that represents the correlation coefficient "r ² " between the similarity of the various donors in the expression of: ALPL, COL1 A2, DCN and ELN, after one week's osteoinduction treatment, and "in vivo" bone formation;

FIG. 8d is a chart that represents the correlation coefficient "r ² " between the similarity of the various donors in the expression of: ALPL, COL1 A2, DCN and RUNX2, after one week's osteoinduction treatment, and "in vivo" bone formation;

FIG. 9a is a chart that represents the correlation coefficient "r ² " between the similarity of the various donors in the expression of: COL1 A2, DCN and RUNX2, after one week's osteoinduction treatment, and "in vivo" bone formation;

FIG. 9b is a chart that represents the correlation coefficient "r ² " between the similarity of the various donors in the expression of: ALPL, COL1 A2, and RUNX2, after one week's osteoinduction treatment, and "in vivo" bone formation;

FIG. 9c is a chart that represents the correlation coefficient "r ² " between the similarity of the various donors in the expression of RNX2 and DCN after one week's osteoinduction treatment and "in vivo" bone formation;

FIG. 9d is a chart that represents the correlation coefficient "r ² " between the similarity of the various donors in the expression of COL1 A2 and RUNX2 after one week's osteoinduction treatment and "in vivo" bone formation.

Detailed description of a preferred embodiment Further characteristics of the invention, concerning the selection of the methods to be used for multi-parameter analysis and the genetic markers are provided in the annexed drawings.

Figures 1 a and 1 b represent the ability of BM-MSCs from six donors to deposit minerals into the matrix of the bone tissue.

Particularly, cells were osteogenically induced for two weeks.

This method includes treatment in an appropriate culture medium consisting of the basal medium (nucleoside-free Alpha Medium added with 8% PL, 1 %

L-glutamine, 1 UI/mL eparin and 10 μg/mL ciprofloxacin) added with: β- glycerophosphate, ascorbic acid and dexamethasone.

Such basal medium was maintained for one week, and replaced every 2-3 days.

On the seventh day, the bone morphogenetic protein (BMP2) was added to the medium.

After 14 days' culture in total, the MSCs were tested by cytochemical staining with Alizarin and Von Kossa.

Alizarin identified precipitation of calcium minerals by more intense staining (red color in reality), whereas Von Kossa staining identifies phosphate and carbonate anions, by a typical black stain. Properly induced cells exhibit a stronger signal, which is red reality (Figure 1 a) or a stronger black signal (Figure 1 b), given by the deposition of minerals into the matrix of the differentiated cells.

A sample, herein referred to as "control sample" was treated with the basal culture medium only and was found in both cases negative to staining.

As shown in Figure 2, the proliferative ability of BM-MSCs from six donors was tested by a molecular analysis on the gene MKI67. After one or two weeks' induction treatment, the BM-MSCs that exhibit a higher expression of such marker are those deriving from donors #1 , #3 and #4. The asterisks in the chart represent the statistic significance of these values as compared to those generated by the BM-MSCs of the same donor without no induction treatment.

As mentioned above, Figure 3 is a chart that shows the analysis of bone formation in an animal model.

The dark grey column of the histogram represents the percent of bone in the overall area of the histological slide being tested. The symbol § represents the statistical significance of the percent of bone formed by each donor with respect to the control graft without BM-MSCs.

The mesenchymal stem/stromal cells of bone marrow, briefly BM-MSCs, from donors #1 , #2, #3 and #4 were found to be positive to Alizarin, whereas only BM-MSCs from donors #1 , #2 and #3 were found to be also positive to Von Kossa staining.

The cells from donor #4 gave inconsistent functional results, i.e. were found to be: positive to Alizarin, negative to Von Kossa and negative to bone formation in the animal.

The cells from donor # 6 also gave inconsistent results in terms of bone tissue generating ability, due to the lack of "ex vivo" mineralization and bone forming ability in the mouse.

The cells from donor # 5 were found to be negative to both "ex vivo" and "in vivo" assays. The symbol * represents the statistical significance of the percent of bone formed by the various BM-MSC donors with respect to the BM-MSC graft from donor #5. The lack of a direct correlation between positive "ex vivo" mineralization assays and bone formation in the animal model involves the need to identify a more reliable bone differentiation potential assay.

In recent times, specific molecular markers have been identified to define the "ex vivo" osteoblast differentiation ability (Twine, 2014).

After introducing protein products corresponding to the five molecular markers which, according to the invention, were selected from the networked database known as "String 9.1 ", these protein products were found to have interactive ratios, in the following osteoblast differentiation stages: early (0-24 hours), intermediate (3-6 days) and late (9-1 2 days), as shown in Figure 4 which depicts a histogram indicating the number of interactions between the five molecular markers in the three stages of osteoblast differentiation.

Namely, a greater number of interactions appears in the early stages of osteoblast differentiation.

This result is consistent with the invention, which shows that the biomarkers are essential in the first stages of the differentiation process from MSCs to bone tissue-specific osteoblasts.

Referring to Figure 5, the chart shows that, after one week's induction treatment on the cells, heterogeneous gene expression levels are obtained. This method includes treatment with an appropriate culture medium consisting of the basal medium with: β-glycerophosphate, ascorbic acid, dexamethasone and bone morphogenetic protein (BMP2) for one week, with the medium being replaced every 2-3 days.

It shall be further noted that no marker, as taken individually, has constant gene expression levels that might directly correlate it with "in vivo" bone formation.

Referring to Figure 6 and particularly to Figure 6a, there is shown a chart, known as dendrogram, in which the degree of similarity of gene expression in the five molecular markers from the 6 BM-MSC donors may be verified.

The dendrogram establishes a hierarchy of similarities among the various donors, considering a single link and the euclidean distance.

Particularly, the donors #1 are #2 were found to be most similar, as well as the donor #3, and all are capable of "ex vivo" and "in vivo" bone formation. The analysis further associated the gene expression of donors #6 and #4, and particularly the donor #6 generated bone and the donor #4 did not generate bone "in vivo".

In spite of this apparent inconsistency, the potential assay identified the difference between the donors #1 , #2, #3 and the BM-MSCs of the donor #6. More in detail, the donor #6 could not mineralize the matrix "ex vivo", but formed bone "in vivo".

The dendrogram further shoes that the BM-MSC population from donor #5 is the most different from the others and was the only population that was negative to both "ex vivo" and "in vivo" functional tests.

These results support the concept of the invention, i.e. that the expression of molecular markers is directly correlated with bone formation ability.

This correlation is expressed by the line represented in Figure 6b, which shows a linear relation between the percent of bone tissue in the sections of the animal model and the correlation of molecular marker expression in the six donors.

An exception is represented by the association of the results from donors #4 and #6, as shown by an empty dot, in which inconsistent "ex vivo" mineralization and poor bone formation ability appear in the animal model. The above discussed results prove that the expression levels of the five selected molecular markers can provide predictive information about actual "in vivo" bone formation.

In order to understand the mechanisms that generate the inconsistency between the ability of donor #6 to form bone "in vivo", and its inability to mineralize the matrix "ex vivo", differences were analyzed between the expression of molecular markers between the three positive donors for both bone formation models and donor #6.

Back to Figure 5, it shall be noted that the most different gene expression relates to gene COL1 A2, which is particularly high compared with other samples.

For this reason, the BM-MSCs of donor #6 were treated, before osteogenic induction, with two molecules that can modulate collagen expression.

The first molecule is interferon gamma (INF-Y), which is present in the human body, and the second molecule is a synthetic molecule named SB431 542 which modulates the TGFp signal, which in turn regulates collagen expression.

After 3 days' treatment with the previous molecules, the BM-MSCs from donor #6 were osteogenically induced.

Then, both the expression level of collagen (Figure 7a), and the "ex vivo" matrix mineralization ability (Figure 7b) were assessed.

As shown in Figure 7a, expression levels of collagen considerably decreased after the two treatments, as compared with cells deriving from the same donor #6, not treated before the differentiation.

The symbol * represents the statistical significance of the change of collagen 1 A2 expression in INF-Y- or SB431 542-treated cells as compared with untreated cells from the same donor #6.

Then the INF-Y- or SB431 542-treated BM-MSCs from donor #6, the untreated cells from donor #6 (which are considered as a negative staining control, see Figure 1 a) and the cells from donor #1 (which are considered as positive staining control, see Figure 1 a) were treated for two weeks with a bone differentiation-inducing medium and stained with Alizarin.

As shown in Figure 7b, the INF-Y- or SB431 542-treated BM-MSCs from donor #6 are stained with a color similar to that of donor #1 .

Referring to Figures 8a to 8d and 9a to 9d, it should be noted that the charts were obtained using the so-called "single link-Euclidean distance" method. More in detail, referring to Figure 8a, the x-axis of the Cartesian diagram represents the average percent values of "in vivo" bone formation, whereas the y-axis represents the correlation values of the expression of the four molecular markers from the six donors after standardization of the value of COL1 A2 expression in donor #6, using a middle value of the expression of the same gene in donors capable of forming bone "in vivo".

The line obtained by joining the dots represented in the chart describes the progress of correlation between the molecular data of donors and bone formation in the animal.

More in detail, the slope of such line indicates the absolute value of the correlation coefficient, referenced to as "r ² " whose maximum value coincides with 1 .

As shown, the slope of this line has a value corresponding to 0.9482, and as a result the analysis of markers at one week describes a strong correlation between marker values and the percent of "in vivo" bone formation.

The chart of Figure 8b comprises a Cartesian diagram which shows only the molecular data analyzed after two weeks' bone induction treatment: the chart shows a line whose slope is described by the value 0.5335.

The results show that the expression levels of biomolecular markers can be analyzed after two weeks' induction, but the relation between the percent of formed bone in the animal and the correlation of molecular marker expression in the six donors is decreased.

According to the invention, the results so obtained support the strategy of analyzing expression levels of biomolecular markers after one week's osteogenic induction.

The charts of Figures 8c and 8d show Cartesian diagrams which analyze expression levels after selection of four biomolecular markers, analyzed after one week's osteogenic induction treatment.

More in detail, in the chart 8c the marker RUNX2 was excluded from the analysis, and the slope of the line that joins the represented dots is described by the value 0.5651 , whereas in the chart of Figure 8d the marker ELN was excluded and the slope of the line that joins the represented dots is described by the value 0.9485.

The charts of Figures 9a and 9b show respective Cartesian diagrams in which the expression levels are analyzed after selection of three biomolecular markers, after one week's osteogenic induction treatment.

Particularly, in the chart of Figure 9a the markers COL1 A2, DCN and RUNX2 were analyzed: the slope of the line that joins the represented dots is described by the value 0.9476.

In the chart of Figure 9b the markers APLL, COL1 A2 and RUNX2 were analyzed and the slope of the line that joins the represented dots is described by the value 0.890.

The charts of Figures 9c and 9d show Cartesian diagrams which analyze expression levels of two biomolecular markers, analyzed after one week's osteogenic induction treatment.

In the chart of Figure 9c the markers DCN and RUNX2 were analyzed and the slope of the line that joins the dots is described by the value 0.0259. In the chart of Figure 9d the markers COL1 A2 and RUNX2 were analyzed and the slope of the line that joins the dots is described by the value 0.8906. Therefore, the results obtained according to the invention show that the expression levels of the selected biomolecular markers directly affect the bone differentiation ability of BM-MSCs.

The invention was conceived to use five biomolecular markers and analyze their results after one week's osteogenic treatment.

The same type of analysis applies when analyzing the results after two weeks' induction, but the relation between the percent of formed bone in the animal model and the correlation of molecular marker expression in the six donors is decreased.

The use of a smaller number of biomolecular markers after one week showed that this relation weakens.

Furthermore, using all the five markers, the analysis can be based on more stringent parameters and is thus capable of producing more reliable results.

The invention has been found to fulfill the intended objects.

The invention so conceived is susceptible to changes and variants within the inventive concept.

Also, all the details may be replaced by other technical equivalent elements.

In its practical implementation, any material, shape and size may be used as needed, without departure from the scope as defined by the following claims.