Diagnosis and treatment of neurogenic heterotopic ossification

Abstract

The present invention focuses on an in vitro method for diagnosing or prognosing neurogenic heterotopic ossification (NHO), which comprises assessing the expression level of inducible transforming growth factor beta (TGFBi) in a biological sample containing astrocytes and / or neuronal secretome obtained from a subject. Furthermore, the present invention also relates to a therapeutic strategy for treating NHO based on TGFBi inhibitors.

Description

[Technical Field]

[0001] The present invention relates to the medical field. In particular, the present invention focuses on an in vitro method for diagnosing or prognosing neurogenic heterotopic ossification (NHO). Furthermore, the present invention also relates to a therapeutic strategy for treating NHO. [Background technology]

[0002] The term heterotopic ossification (HO) refers to the presence of mature lamellar bone in extraskeletal soft tissues. Bone growth can occur through the mechanisms of endochondral or intramembranous ossification.

[0003] Clinically, there are no pathognomonic signs to establish early diagnosis of the disease, and therefore the diagnosis is based mainly on nonspecific signs and symptoms.

[0004] Currently, the etiopathology of this disease remains unknown. However, several risk factors contribute to its onset. Among them, neurogenic injury, fracture, or joint replacement should be emphasized. However, it has been explained that ectopic bone formation requires inducers, osteoprogenitor cells, and an environment that promotes osteoblast formation.

[0005] Depending on its etiology, the disease may be classified as hereditary HO and acquired HO. Genetic variants (fibrodysplasia ossificans progressiva and skeletal dysplasia progressiva) are rare and are considered the most severe manifestations of HO. Regarding non-hereditary forms, these represent the most common HO variants. Acquired HO may develop after trauma or neurogenic insult.

[0006] NHO is one of the most common complications of injuries to the central nervous system (CNS), and its incidence has been demonstrated to increase after the simultaneous occurrence of CNS damage, such as traumatic brain injury (TBI) or spinal cord injury, and peripheral injuries, such as long bone fractures (LBF).

[0007] The pathophysiology of NHO remains unclear. Nevertheless, it has been hypothesized that humoral neuroimmune factors released by the brain may play an important role in the development of this disease. After TBI, the blood-brain barrier (BBB) ​​is compromised, allowing some neurogenic factors to leak from the brain into distal tissues. At the same time, distal tissues located near the fracture site secrete other factors that promote bone healing. Therefore, it has been suggested that the concentration of these two factors may enhance abnormal bone growth. Similarly, studies have pointed to osteoprogenitor cells and circulating cells with osteogenic potential as potential factors promoting ectopic bone formation.

[0008] Recently, the peripheral nervous system (PNS) has also been postulated as a promoter of heterotopic bone. Specifically, studies have linked blood-nerve barrier (BNB) disruption to an increased incidence of NHO. Furthermore, the presence of osteoprogenitor cells within nerves has also been suggested.

[0009] Despite the high incidence of this pathology, the therapies used in its management have low efficacy rates, and some are highly invasive and associated with side effects and complications.

[0010] Therefore, there is an unmet medical need to find effective methods for the diagnosis, prognosis, and treatment of NHO. The present invention focuses on solving this problem, and a new strategy is presented herein. Summary of the Invention

[0011] Brief description of the invention The present invention relates to in vitro methods for diagnosing or prognosing NHO, as well as therapeutic strategies for treating NHO.

[0012] First, we developed an in vitro traumatic brain injury (TBI) model to investigate the effects of released unknown factors on osteoblastogenesis and adipogenesis. To do this, we cocultured astrocytes and osteoblasts. RNA and protein were then extracted to analyze the effects of the coculture on osteoblasts. The results revealed that coculture of astrocytes and osteoblasts increased the expression of two key bone markers associated with HO, namely SPP1 and BMP2. Given these results, we investigated whether the observed effects were due to the effects of astrocytes on osteoblasts or vice versa. Therefore, we dissected the coculture. Astrocytes were cultured and allowed to acclimate to culture medium for 72 hours. The astrocyte supernatant was then added to osteoblasts. The results previously observed for the bone markers SPP1 and BMP2 remained consistent, suggesting that astrocytes secrete several factors that may induce the expression of bone-related genes in osteoblasts. Given the similar effects in the coculture and astrocyte-conditioned medium (ACM) models, we performed the remaining experiments using ACM on day 3 to investigate other bone-related bone markers. ACM increased the expression of bone-related markers, such as CD44 (the osteopontin receptor) and LIF, in osteoblasts. Furthermore, ACM induced the expression of AXIN2 and decreased the expression of SOST and DKK1, suggesting activation of the WNT pathway (a major bone anabolic pathway). Interestingly, the RANKL / OPG ratio was decreased by ACM, indicating modulation of bone remodeling toward bone anabolism. Furthermore, ACM-induced cell morphological changes were also observed. Therefore, actin and tubulin staining was performed. Surprisingly, these morphological changes resembled those of osteocytes. Given these findings, we decided to investigate the expression of osteogenic markers. Consistent with the morphological changes, ACM increased the expression of PDPN, the earliest bone cell marker. Nevertheless, other markers were not upregulated by ACM.These results support the idea that ACM may accelerate the bone formation process by accelerating the differentiation of osteoblasts into osteocytes. These results demonstrate that ACM can enhance bone anabolism. It is well known that a certain degree of inflammation is necessary for bone healing. Therefore, we also investigated the effect of ACM on the inflammatory response of osteoblasts. In this context, we investigated the expression of key inflammatory markers (IL6, VCAM, and CCL2). Results revealed that ACM significantly increased these three genes. Therefore, ACM may not only enhance bone anabolism but also increase inflammation in osteoblasts. Considering that astrocyte culture medium can induce inflammation in osteoblasts themselves, we decided to stimulate osteoblasts with inflammatory stimuli such as IL1B to mimic the inflammatory environment associated with TBI and LBF. Furthermore, osteoblasts were stimulated with ACM after 72 hours and then treated with IL1B for another 72 hours. Interestingly, IL1B was observed to synergize with ACM, increasing the expression of the three inflammatory genes studied. Surprisingly, IL1B stimulation also increased several bone-related markers, such as BMP2 and LIF, in ACM-stimulated osteoblasts. This further supports the link between inflammation and bone anabolism. Considering that osteoblastogenesis and adipogenesis are balanced and contrasting processes, we decided to investigate the effect of ACM on MSCs (mesenchymal stem cells) differentiated into adipocytes. To do this, astrocytes were conditioned with the medium for 72 hours as previously described. The supernatant was then centrifuged and filtered to obtain an ACM pellet. Finally, this was resuspended in adipogenic medium, and the MSCs were differentiated into adipocytes for 7 days. The results revealed a significant decrease in three of the four adipogenic markers studied (FABP4, PPARG, and ADIPOQ). To confirm this, lipid droplets were stained, and the results revealed that ACM has the ability to completely blunt lipid droplet accumulation. If MSCs were not differentiated into adipocytes, we decided to study several bone markers in the adipogenic model.The results revealed that SPP1 and RUNX2 were increased in MSCs differentiated into adipocytes. Therefore, ACM not only suppressed adipogenesis but also promoted the expression of bone-related markers in MSCs when they differentiated into adipocytes. Finally, we found that MSCs differentiated with ACM proliferated more than cells treated with differentiation medium (DM). To verify this, we performed MTT for 7 days, and the results were consistent with previous findings: ACM exerted a proliferative effect on MSCs when they differentiated into adipocytes. Overall, these results suggest that ACM alters the osteoblast / adipogenesis balance by promoting osteoprogenitor cells. To more precisely identify the factors responsible for these results, we decided to perform a proteomic experiment on the supernatant. These proteins were analyzed by quantitative (SWATH) and qualitative (DDA) methods. Among the proteins, one of them, TGFBi (inducible transforming growth factor beta) (Unitprot Q15582), was highly expressed in ACM. Specifically, this protein was increased 14-fold in ACM compared with DM. Western blot analysis was performed to verify the proteomic results. Results demonstrated that TGFBi protein expression was higher in osteoblasts treated with ACM compared with differentiation medium. RT-PCR was also performed on osteoblasts to determine the expression levels of TGFBi in these cells. Results revealed that ACM can increase TGFBi expression levels in osteoblasts, suggesting that osteoblasts can also amplify the levels of this factor. To assess the contribution of TGFBi to the effects of ACM on osteoblasts, osteoblasts were stimulated with recombinant TGFBi for 3 days. Results revealed that the expression of SPP1, BMP2, VCAM, and CCL2 partially reproduced the results obtained when osteoblasts were treated with ACM. Regarding the effect of TGFBi on adipogenesis, the effect on one adipogenic marker was reproduced. Specifically, TGFBi stimulation significantly reduced the expression of PPARG, a key regulator of adipogenesis.This may be due to the concentration of TGFBi used, as well as the lack of interaction between TGFBi and other proteins in ACM. Furthermore, differences in cell proliferation were observed in MSCs treated with TGFBi, similar to those observed when cells were treated with ACM. This observation was verified by not only RNA quantification but also cell counting. In summary, TGFBi partially reproduced the effects of ACM on osteoblastogenesis, adipogenesis, and cell proliferation. These results highlight TGFBi as a potential biomarker and therapeutic target for HO. However, it was necessary to know whether this protein is also present in patients at risk for developing this disease. Therefore, we collected serum from patients at high risk for developing NHO (TBI+LBF), as well as from patients at moderate and low risk for developing the disease (TBI and LBF, respectively), and patients at no risk for developing NHO (controls). Blood samples were collected on days 0, 1, 3, and 7 after trauma. Only one sample was collected in the control group. Dual trauma patients are thought to express higher levels of anabolic circulating factors. Therefore, we analyzed serum proteome profiles and correlated these with their effects on bone metabolism. Results demonstrated a greater number of correlations in the dual trauma group (TBI + LBF) compared with the other groups. To further characterize the dual trauma group, pathway enrichment analysis was performed. Consistent with findings assessed in in vitro models, an inflammatory environment was identified in the serum of these patients. Interestingly, significant enrichment of TGF-β signaling was observed. This is consistent with the literature and is also consistent with in vitro findings of TGF-β induction. Finally, the expression of TGF-β in patient serum was assessed by Western blot. Results revealed that patients with dual trauma expressed higher TGF-β levels, which peaked on day 3 (consistent with our in vitro TBI model).Overall, these data point to TGFBi as a potential biomarker and therapeutic target for NHO. In the context of NHO, recent studies have also suggested that peripheral nervous system damage contributes to ectopic bone formation. To elucidate the role of nerves in bone anabolism, we co-cultured nerve explants (NE) and explanted neural cells (neurons) with osteoblasts. We first decided to characterize the secretome profile by proteomics. Results revealed higher TGFBi expression in NE-conditioned medium compared to differentiation medium. Finally, these results were verified by Western blot. We then investigated the gene expression of key bone markers in osteoblasts co-cultured with NE. Results revealed significant increases in SPP1, RUNX2, BMP2, LIF, and some bone remodeling and inflammatory markers, demonstrating that nerves can induce TGFBi expression in osteoblasts. Considering these results, we confirmed a similar profile to that previously detected in astrocytes. Interestingly, when comparing the expression of TGFBi between healthy and congenital HO nerves, we found that this protein was highly produced in HO samples. Following a similar method, we examined the effect of neural cells on osteoblast formation. Although this effect was not as strong as in the coculture with NE, neural cells increased the expression of SPP1, VCAM, and TGFBi. One reason for the observed lower effect may be that the number of cells contained in NE was greater than the number of cells cultured in this experiment. However, these data are quite intriguing because these cells can migrate and target various tissues, where they can recapitulate some of the phenotypes induced by neural induction. Finally, we examined the effect of NE on adipogenesis. Results revealed that in MSCs differentiated into adipocytes, the three adipogenic markers were decreased, while SPP1 and TGFBi were increased. Furthermore, neural cells were also able to increase the expression of inflammatory markers IL6, VCAM, and CCL2.Overall, this data suggests that the central and peripheral nervous systems may regulate the balance between osteoblastogenesis and adipogenesis by enhancing osteoblastogenesis. Furthermore, it has been demonstrated herein that TGFBi may play an important role in bone formation through enhancing osteoblastogenesis and inhibiting adipogenesis. Therefore, TGFBi may be a potential biomarker and a potential therapeutic target for NHO.

[0013] Accordingly, a first embodiment of the present invention relates to an in vitro method for diagnosing or prognosing NHO, comprising assessing the expression level of TGFBi in a biological sample obtained from a subject, wherein determination of an increased expression level of TGFBi relative to a pre-set threshold expression level measured in a healthy control subject is indicative that the subject is likely to be affected with NHO or exhibits a poor prognosis.

[0014] In preferred embodiments, the biological sample is selected from blood, plasma, serum, tissue biopsies from peripheral nerves, including the saphenous nerve, posterior tibial nerve, and sciatic nerve, or synovial fluid.

[0015] A second embodiment of the present invention relates to the in vitro use of TGFBi or a kit comprising reagents for assessing the expression level of TGFBi for the diagnosis or prognosis of NHO.

[0016] A third embodiment of the present invention relates to a TGFBi inhibitor, for example selected from an siRNA or a TGFBi neutralizing antibody, for use in the treatment of NHO.

[0017] In a preferred embodiment, the siRNA comprises: hs.Ri.TGFBi 13.1: SEQ ID NO: 1→5-GUGGCAAAUCAACAGUCAUCAGCTA-3' SEQ ID NO: 2 → 5-UAGCUGAUGACUGUUGAUUUGCCACAG-3' hs.Ri.TGFBi 13.2: SEQ ID NO: 3 → 5-GUUUUCAAAACCAAGUAUCACACTT-3' SEQ ID NO: 4 → 5-AAGUGUGAUACUUGGUUUUGAAAACAU-3' hs.Ri.TGFBi 13.3: SEQ ID NO: 5 → 5-CUACAUUGAUGAGCUACUCAUCCCA-3' SEQ ID NO: 6 → 5-UGGGAUGAGUAGCUCAUCAAUGUAGUG-3' is.

[0018] In a preferred embodiment, the TGFBi neutralizing antibody is Immunogen Catalog Number: Ag0241 GenBank accession number: BC000097 Gene ID(NCBI):7045 Proteintech Catalog Number: 10188-1-AP is.

[0019] In preferred embodiments, treatment involves silencing TGFBi in astrocytes using siRNA and / or sequestering TGFBi (e.g., produced by astrocytes altered as a result of trauma) using a neutralizing antibody against TGFBi. Similarly, the present invention also relates to a method for treating NHO, comprising administering a therapeutically effective dose or amount of a TGFBi inhibitor to a patient. Furthermore, siRNA inhibition of TGFBi can be used to ensure that tissues affected by astrocyte- or neuron-conditioned medium also do not produce TGFBi. Thus, while it is possible to eliminate TGFBi produced by astrocytes in trauma patients, it is equally possible to prevent TGFBi production using siRNA. However, because TGFBi produced upon reaching target tissues is capable of self-induction, both antibodies and siRNA can be used to stop its self-amplification in tissues distant from the brain or CNS. The goal of silencing TGFBi in astrocytes is to prevent the production of this protein. The results presented herein hypothesize that astrocytes are responsible for the release of molecules (including TGFBi) that enhance bone metabolism while damaging adipogenic metabolism. Therefore, it is suggested herein that astrocytes may be partially responsible for and involved in the development of NHO. This also applies to peripheral nerves, because, as shown herein, peripheral nerves secrete TGFBi and increase the expression of this protein during osteoblast differentiation. Silencing TGFBi in astrocytes would result in astrocytes no longer producing TGFBi, which would be expected to reduce the effect of ACM on osteoblast differentiation. Meanwhile, an alternative therapeutic strategy would be to sequester TGFBi (e.g., TGFBi present in the secretome produced by astrocytes altered as a result of trauma) using neutralizing antibodies against TGFBi.Therefore, it is expected that when antibodies sequester TGFBi, the effects of ACM and NCM on osteoblast differentiation will be less potent due to the absence of TGFBi.

[0020] It is well known that the risk of heterotopic ossification is significantly increased when TBI occurs concomitantly with other distal lesions, such as fractures of long bones. Normally, TBI compromises the blood-brain barrier, allowing factors secreted by astrocytes (such as TGFBi) to reach peripheral structures. In this situation, the secretome of these astrocytes synergizes with osteoinductive factors (such as TGFBi) released by peripheral structures (such as bone) to promote abnormal bone anabolism.

[0021] Thus, in a preferred embodiment, a therapeutic agent (e.g., an siRNA or a neutralizing antibody) can be administered systemically, e.g., intravenously, to reduce the amount of TGFBi systemically and reduce the synergistic effect between central and peripheral lesions on bone anabolism.

[0022] Alternatively, the therapeutic agent could be administered topically or intra-articularly in patients who already have bone plaques. Use of the therapeutic agent after or during surgery to remove ectopic bone would reduce the likelihood of recurrence of ectopic bone formation.

[0023] Finally, the present invention also relates to an in vitro method for identifying compounds for treating NHO, comprising: a) determining whether a candidate compound inhibits TGFBi or abolishes the anabolic effect of astrocytes on osteoblasts; and b) where inhibition of TGFBi or abolishment of the anabolic effect of astrocytes on osteoblasts occurs, which is an indication that the candidate compound may be effective in treating NHO.

[0024] In a preferred embodiment, the elimination of the anabolic effect of astrocytes on osteoblasts is determined by confirming the absence of bone-associated markers such as SPP1 and BMP2.

[0025] Meanwhile, the present invention also relates to a method for detecting TGFBi in a test sample from a human subject at risk of developing NHO, the method comprising: (a) contacting the test sample with a reagent specific for TGFBi; (b) amplifying a TGFBi biomarker to generate an amplification product in the test sample; and (c) measuring the expression level of TGFBi by determining the level of the amplification product in the test sample.

[0026] In a preferred embodiment, the present invention is a computer-implemented invention, where the processing unit (hardware) and software are configured as follows: receiving the expression level value of TGFBi; Processing the expression level values ​​of TGFBi to find substantial fluctuations or deviations; and An output of the variation or deviation in expression levels is provided through a terminal display, wherein the variation or deviation in expression levels indicates that the subject may be suffering from NHO.

[0027] In a preferred embodiment, the present invention relates to an in vitro method for diagnosing or prognosing NHO, comprising assessing the expression level of TGFBi in a biological sample comprising astrocytes, neurons, and / or neuronal secretomes obtained from a subject, wherein determining an increased expression level of TGFBi relative to a pre-determined expression level measured in a healthy control subject is indicative that the subject is likely to be affected with NHO or exhibits a poor prognosis.

[0028] In preferred embodiments, the sample is selected from blood, plasma, serum, peripheral nerves including the saphenous nerve, posterior tibial nerve, and sciatic nerve, cerebrospinal fluid, or synovial fluid.

[0029] In vitro use of TGFBi in a biological sample comprising astrocytes, neurons, and / or neuronal secretomes obtained from a subject for the diagnosis or prognosis of NHO.

[0030] In vitro use of a kit comprising reagents for determining the level of expression of TGFBi in a biological sample comprising astrocytes, neurons, and / or neuronal secretomes obtained from a subject for the diagnosis or prognosis of NHO.

[0031] A TGFBi inhibitor for use in a method for treating NHO, comprising inhibiting TGFBi in astrocytes, neurons, and / or neuronal secretomes, wherein the inhibitor is selected from a neutralizing TGFBi antibody, an siRNA that base pairs with TGFBi mRNA, an miRNA, or a combination medication comprising a thiazolidinedione, a corticoid, and a nonsteroidal anti-inflammatory drug.

[0032] In a preferred embodiment, the thiazolidinedione is rosiglitazone or pioglitazone, the corticoid is dexamethasone, and the nonsteroidal anti-inflammatory drug is indomethacin.

[0033] In preferred embodiments, the TGFBi inhibitor is administered intravenously, intrathecally, topically, or intra-articularly.

[0034] 1. An in vitro method for identifying a compound for treating NHO, comprising: a) determining whether a candidate compound inhibits TGFBi in astrocytes, neurons, and / or neuronal secretomes or abolishes the anabolic effect of astrocytes on osteoblasts; and b) wherein inhibition of TGFBi or abolishment of the anabolic effect of astrocytes on osteoblasts occurs, which is an indication that the candidate compound may be effective in treating NHO.

[0035] For purposes of the present invention, the following terms are defined:

[0036] According to the present invention, the reference value can be a "preset threshold" or "cut-off" value. Typically, the "threshold" or "cut-off" value can be determined experimentally, empirically, or theoretically. According to the present invention, the "preset threshold" refers to a value previously determined in subjects not suffering from NHO or in healthy subjects. Thus, for example, if higher expression of TGFBi is confirmed compared to the preset "threshold," the subject is likely to suffer from NHO or exhibit a poor prognosis. As will be recognized by those skilled in the art, the "threshold" may also be arbitrarily selected based on existing experimental and / or clinical conditions. The "threshold" needs to be determined to obtain optimal sensitivity and specificity depending on the function and benefit / risk balance (clinical consequences of false positives and false negatives) of the test. Typically, the optimal sensitivity and specificity (i.e., "threshold") can be determined using a receiver operating characteristic (ROC) curve based on experimental data.

[0037] The term "comprising" means including, but not limited to, what follows the word "comprising." Thus, use of the word "comprising" indicates that the listed elements are required or essential, but that other elements are optional and may or may not be present.

[0038] The term "consisting of" means inclusive of and limited to what precedes the phrase "consisting only of." Thus, the phrase "consisting only of" indicates that the listed elements are required or essential, and that no other elements may be present.

[0039] By "therapeutically effective dose or amount" of a TGFBi inhibitor is intended an amount that, when administered as described herein, results in a positive therapeutic response in a subject suffering from NHO. The exact amount required will vary from subject to subject, depending on the species, age, and general condition of the subject, the severity of the condition being treated, the mode of administration, etc. The appropriate "effective" amount in any individual case can also be determined by one skilled in the art using routine experimentation based on the information provided herein.

[0040] By "inhibition of TGFBi" or "TGFBi inhibitor" in the context of the present invention is understood a total or partial impairment of TGFBi, a reduction in TGFBi levels, and / or a downregulation of TGFBi, which may preferably be achieved by a neutralizing TGFBi antibody, an siRNA that base pairs with TGFBi mRNA, an miRNA, or a pharmaceutical combination comprising a thiazolidinedione, a corticoid, and a non-steroidal anti-inflammatory drug (i.e., a triple therapy). [Brief explanation of the drawings]

[0041] [Figure 1] Figure 1 shows the effect of co-culture of astrocytes and osteoblasts on osteoblast formation. A-C) Expression of osteoblastic differentiation markers SPP1, RUNX2, and BMP2 measured by RT-PCR after 21 days of cell co-culture of astrocytes and osteoblasts. [Figure 2] Figure 1 shows the effect of MCA on bone anabolism during the osteoblastogenesis process. A) Expression of osteoblastic differentiation markers SPP1, RUNX2, and BMP2 measured by RT-PCR in osteoblasts stimulated with ACM for 3 days. B and C) Quantification by RT-PCR of the expression of bone metabolism-related markers BMP4, LIF, GPNMB, CD44, PDGFβ, PDGFRβ, RANKL, OPG, RANKL / OPG ratio, and WNT-related genes AXIN2, SOST, and DKK1 during the differentiation of SaOS2 into osteoblasts for 3 days in the presence or absence of MCA. [Figure 3]Figure 1 shows the effect of MCA on the inflammatory response. A) Expression of the inflammatory markers VCAM and CCL2 was measured by RT-PCR in SaOS2 cells differentiated into osteoblasts for 3 days in the presence or absence of MCA. B and C) Effect of adding IL1β or LPS on the basal effect of MCA on osteoblast formation. The inflammatory and bone turnover markers IL6, VCAM, CCL2, BMP2, LIF, and SPP1 were measured by RT-PCR in SaOS2 cells differentiated into osteoblasts for 3 days in the presence or absence of MCA and then stimulated with IL1β (0.1 ng / mL) or LPS (100 ng / mL) for 48 hours. [Figure 4] Figure 1 shows morphological changes to osteoblasts induced by MCA. A) Tubulin, actin, and nuclear staining of SaOS2 cells differentiated for 3 days in the presence or absence of ACM. B) Gene expression of osteocyte markers PDPN and BGLAP determined by RT-PCR in SaOS2 cells differentiated into osteoblasts for 3 days in the presence or absence of ACM. [Figure 5] Figure 1 shows the effect of ACM on adipogenesis. A) Adipogenic marker study. Gene expression of key adipogenic markers FABP4, PLIN2, PPARG, and ADIPOQ measured by RT-PCR in C3H10T1 / T2 cells differentiated into adipocytes for 7 days in the presence or absence of ACM. B) Oil red staining of lipid droplets during adipocyte differentiation of C3H10T1 / 2 for 7 days in the presence or absence of DM and ACM. C) Anabolic marker study. Gene expression of bone-related markers SPP1, RUNX2, CD44, GPNMB, and AXIN2 measured by RT-PCR in C3H10T1 / 2 differentiated for 7 days in the presence or absence of ACM. D) Inflammatory marker study. Inflammatory markers IL6, VCAM, and CCL2 determined by RT-PCR during the adipogenic process in the presence or absence of ACM. E) Cell viability study using MTT assay of C3H10T1 / 2 differentiated into adipocytes for 7 days in the presence or absence of ACM. [Figure 6]Figure 1 shows the effect of NE on osteoblast formation. A) Gene expression of bone-related markers SPP1, RUNX2, and BMP2 measured by RT-PCR 3 days after NE-SaOS2 coculture during osteoblast differentiation. B and C) RT-PCR determination of the expression of bone-related markers BMP4, LIF, GPNMB, CD44, PDGFβ, PDGFRβ, RANKL, OPG, the RANKL / OPG ratio, and AXIN2 during SaOS2 osteoblast differentiation for 3 days in the presence or absence of NE. D) Gene expression of inflammatory markers VCAM and CCL2 measured by RT-PCR during SaOS2 osteoblast differentiation for 3 days in the presence or absence of NE. [Figure 7] Figure 1 shows the effect of ONC on osteoblast formation. A) RT-PCR quantification of gene expression of SPP1, RUNX2, and BMP2 in SaOS2 cells differentiated into osteoblasts for 3 days in the presence or absence of ONC. B-D) RT-PCR confirmation of gene expression of bone-related and inflammatory markers BMP4, LIF, GPNMB, CD44, PDGFβ, PDGFRβ, RANKL, OPG, RANKL / OPG ratio, AXIN2, VCAM, and CCL2 in SaOS2 cells differentiated into osteoblasts for 3 days in the presence or absence of ONC. [Figure 8] Figure 1 shows the effect of ACM on the co-culture of NE and osteoblasts. A-D) Gene expression of bone markers SPP1, RUNX2, BMP2, BMP4, LIF, GPNMB, CD44, PDGFβ, PDGFRβ, RANKL, OPG, RANKL / OPG ratio, AXIN2, VCAM, and CCL2 was confirmed by RT-PCR in SaOS2 cells co-cultured with NE for 3 days in the presence or absence of ACM. [Figure 9]Figure 1 shows the effect of NCM on adipogenesis. A) Expression of key adipogenic markers, FABP4, PLIN2, PPARG, and ADIPOQ, quantified by RT-PCR during 7-day differentiation of C3H10T1 / 2 into adipocytes in the presence or absence of NCM. B and C) Quantification by RT-PCR of gene expression of SPP1, RUNX2, CD44, GPNMB, AXIN2, IL6, VCAM, and CCL2 in C3H10T1 / 2 cells differentiated into adipocytes for 7 days in the presence or absence of NCM. [Figure 10] Figure 1 shows the characterization of the secretome of ACM and DM. A) Qualitative (DDA) and quantitative (SWATH) analysis of statistically significant proteins detected by proteomic studies in DM and ACM. B) Comparison of total spectral counts obtained by quantitative proteomic studies (SWATH) of TGFβi expression in DM and ACM. C) Protein expression of TGFβi in DM and ACM measured by Western blot. [Figure 11] Figure 1 shows secretome and protein expression analysis of NCM and NM. A) Qualitative (DDA) and quantitative (SWATH) analysis of statistically significant proteins detected by proteomic studies in DM and NCM. B) Comparison of results obtained by quantitative proteomic studies of TGFβi expression in DM and NCM. C) Protein expression of TGFβi in NM and ACM observed by Western blot. [Figure 12]Figure 1 shows TGFβi gene expression during osteoblast formation. A) TGFβi gene expression measured by RT-PCR during the differentiation of SaOS2 into osteoblasts over 3 days in the presence or absence of ACM. B) TGFβi gene expression measured by RT-PCR during the differentiation of SaOS2 into osteoblasts over 3 days in the presence or absence of NE. C) TGFβi gene expression quantified by RT-PCR during osteoblast differentiation in the presence or absence of ONC. D) TGFβi gene expression measured by RT-PCR during the differentiation of NE-SaOS2 into osteoblasts over 3 days in the presence or absence of ACM. [Figure 13] Figure 1 shows TGFβi gene expression during adipogenesis. A) TGFβi gene expression measured by RT-PCR in C3H10T1 / 2 cells stimulated with ACM and differentiated into adipocytes for 7 days. B) TGFβi gene expression measured by RT-PCR in C3H10T1 / 2 cells differentiated into adipocytes for 7 days in the presence or absence of NCM. [Figure 14] Figure 1 shows TGFβi protein expression in an in vivo model of HO. A) Plasma levels of TGFβi for TBI, LBF, and TBI+LBF rats 2 days after injury. Results were compared to controls. B) Plasma levels of TGFβi in animals that developed HO (including TBI+LBF and LBF) 6 weeks after injury. [Figure 15] Correlation between genes related to osteoblastogenesis and proteomic analysis of patient serum. A-C) Correlation between gene expression measured by RT-PCR of SPP1, RUNX2, BMP2, BMP4, LIF, GPNMB, CD44, PDGFβ, PDGFRβ, RANKL, OPG, RANKL / OPG ratio, and AXIN2 during differentiation of SaOS2 into osteoblasts over 3, 7, 14, and 21 days in the presence or absence of serum from G1 (A), G2 (B), and G3 (C) patients, and proteomic data from each patient's secretome. D) Pathway enrichment analysis of G1 patient secretomes versus the other groups. [Figure 16] Figure 1 shows TGFβi protein expression in patients at risk of developing NHO. A) Protein expression of TGFβi quantified by Western blot of serum precipitates from patients with TBI and FHL (G1). B) Confirmation of TGFβi expression by Western blot in precipitates from serum of patients with TBI (G2). C) Quantification of serum TGFβi expression by Western blot in patients with FHL (G3). [Figure 17] Figure 1 shows the effect of TGFβi stimulation on osteoblast formation. A-D) Gene expression of bone-related and inflammatory markers, SPP1, RUNX2, BMP2, BMP4, LIF, GPNMB, CD44, PDGFβ, PDGFRβ, RANKL, OPG, RANKL / OPG ratio, AXIN2, TGFβi, VCAM, and CCL2, measured by RT-PCR, in SaOS2 cells differentiated into osteoblasts for 3 days in the presence or absence of TGFβi (3 μg / mL). [Figure 18] Figure 1 shows the effect of TGFβi stimulation on adipogenesis. A) Gene expression of key adipogenic markers FABP4, PLIN2, PPARG, and ADIPOQ measured by RT-PCR in C3H10T1 / 2 cells differentiated into adipocytes for 7 days in the presence or absence of TGFβi (3 μg / mL). B and C) Gene expression of bone-related and inflammatory markers SPP1, RUNX2, CD44, GPNMB, AXIN2, TGFβi, IL6, VCAM, and CCL2 confirmed by RT-PCR in C3H10T1 / 2 cells differentiated into adipocytes for 7 days in the presence or absence of TGFβi (3 μg / mL). D) Effect of TGFβi stimulation on C3H10T1 / 2 cell proliferation. Determination of cell proliferation based on counting the number of cells per well and RNA quantification during differentiation of C3H10T1 / 2 into adipocytes over 7 days in the presence or absence of TGFβi (3 μg / mL). [Figure 19]Figure 1 shows the effect of TGFβi on chondrogenesis. A-C) Gene expression measured by RT-PCR of major cartilage components COL2A1, COLX, and transcription factor SOX9 in ATDC5 cells differentiated into chondrocytes for 3, 7, and 14 days in the presence or absence of TGFβi (3 μg / mL). D) Expression of bone marker SPP1 quantified by RT-PCR during chondrogenic differentiation of ATDC5 cells for 3, 7, and 14 days in the presence or absence of TGFβi (3 μg / mL). [Figure 20] Figure 1 shows TGFβi blockade on astrocytes.TGFβi protein expression 48 hours after TGFβi silencing by using siRNA technology. [Figure 21] 1 shows the effect of TGFβi blockade in ACM medium on osteoblasts. SPP1 gene expression in SaoOS2 cells differentiated into osteoblasts after TGFβi blockade by adding TGFβi antibody to ACM 1 hour before SaOS2 stimulation. [Figure 22] Figure 1 shows the effect of tyrphostins on TGFBi expression levels during osteoblast formation.After differentiation of primary osteoblasts into osteoblasts for 3 days, the expression levels of TGFBi were measured by RT-PCR. [Figure 23] Figure 16 shows circulating TGFBi levels in patients at risk for NHO (updated from Figure 16). Protein expression levels of TGFBi were quantified by ELISA in serum from TBI and FHL patients (G1), TBI patients (G2), FHL patients (G3), and healthy controls (discontinuous line). [Figure 24]Figure 1 shows the accuracy of TGFBi as a biomarker for predicting NHO. The sensitivity and specificity of TGFBi as a biomarker for predicting NHO were calculated by performing the area under the curve method using R software. TGFBi levels from ELISA were compared between high-risk and low-risk NHO groups (group I vs. groups II, III, and IV). A circulating TGFBi level of 3.159 μg / ml corresponds to the optimal sensitivity and specificity of this biomarker for predicting NHO (AUC 0.838). [Figure 25] Figure 1 shows the accuracy of TGFBi as a biomarker for predicting NHO based on simulation. Based on data obtained from the area under the curve method, a simulation of a population of 200 people was performed using R software to confirm the accuracy of TGFBi as a biomarker for predicting NHO. TGFBi has excellent sensitivity for predicting the development of NHO when a TGFBi threshold of 3.159 μg / ml is used. [Figure 26] Figure 10: TGFBi expression levels in acquired OH. TGFBi expression levels were measured by RT-PCR in healthy bone samples and primary bone samples from HO. [Figure 27] Figure 1 shows the effect of GNAS- / - cells derived from a genetic HO variant on adipogenesis. C3H10T1 / 2 mesenchymal stem cells were differentiated into adipocytes in the presence or absence of GNAS- / - cell supernatant for 7 days. The expression levels of adipogenic markers (ADIPOQ, FABP4), bone marker (SPP1), and TGFBi were measured by RT-PCR. [Figure 28] Figure 1 shows the effect of co-culture of nerve explants with osteoblasts on osteoblast formation. Three days after nerve explant-Saos2 cells differentiated into osteoblasts, TGFBi expression levels were measured by RT-PCR. [Figure 29] Figure 1 shows the effect of triple therapy on TGFBi-induced expression levels by ACM. Saos2 cells were differentiated into osteoblasts for 3 days in the presence or absence of ACM and / or triple therapy (Trytherapy). DETAILED DESCRIPTION OF THE INVENTION

[0042] The present invention is illustrated by the following examples, without intending to limit its scope of protection.

[0043] Example 1. Materials and Methods Example 1.1. Human Samples blood sample According to the inclusion criteria (Table 1), blood samples were collected from 38 patients and classified into four groups based on their risk of developing NHO: Group I (G1), patients with TBI and LBF who are at high risk of developing HO; Group II (G2), patients with TBI who are at lower risk of developing this disease; Group III (G3), patients with LBF who are at low risk of developing HO; and Group IV (G4), control patients who are not at risk of developing HO.

[0044] All blood samples were collected after project approval by the Santiago and Lugo Area Research Ethics Committee (reference code 2017 / 262), in accordance with the Declaration of Helsinki, and after the patients or their relatives subsequently signed an informed consent regarding the handling of the samples and the purpose of the study.

[0045] Blood samples were collected at admission (D0), 24 hours (D1), 72 hours (D3), and 168 hours (D7) after the trauma, except for G4, which was only collected once.

[0046] TIFF2026504991000001.tif56170

[0047] Neuronal samples Peripheral sciatic, posterior tibial, and saphenous nerves were obtained from human donors. The procedure was approved by the Santiago Lugo Regional Research Ethics Committee (registration code: 2017 / 262). Similarly, samples were collected after the donor's relatives duly signed an informed consent regarding the handling of the samples and the purpose of the study, in accordance with the Declaration of Helsinki.

[0048] Example 1.2. Animal Model animal Eighty-five male Sprague-Dawley rats (276.12 ± 15.2 g) were obtained from Monash Research Platform (Clayton, Australia). They were 7 weeks old. All rats were housed individually with a 12-hour light / dark cycle (lights on at 7:00 AM) and food and water available ad libitum throughout the experimental period. All procedures were approved by the Alfred Animal Ethics Committee (E / 1923 / 2019 / B) and were within the guidelines of the Australian National Health and Medical Research Council's "Australian Code of Practice for the Care and Use of Animals for Scientific Purposes."

[0049] Experimental group Rats were assigned to one of four injury groups: polytrauma (POLY; muscle contusion + fracture + TBI; n = 32); peripheral injury only (PERI; muscle contusion + fracture + Sham TBI; n = 18); TBI only (TBI; Sham muscle contusion + Sham fracture + TBI; n = 22); and Sham (SHAM; Sham muscle contusion + Sham fracture + Sham TBI; n = 13). Eighteen POLY rats, five PERI rats, and 11 TBI rats were excluded due to death shortly after injury, euthanasia during the acute recovery period, or comminuted fractures. This left a total of 14 POLY rats, 13 PERI rats, 11 TBI rats, and 13 SHAM rats.

[0050] Damage method Extracranial injury was performed as previously described (24). Briefly, anesthesia was induced using 5% isoflurane in 2 L / min of oxygen and then maintained with 2% isoflurane (flow rate: 500 mL / min). Buprenorphine (dosage: 0.05 mg / kg) in sterile saline was administered subcutaneously, followed by muscle contusion. A 1.2 kg impactor (diameter: 1 cm, depth: 1.5 cm) was launched from a height of 55 cm and guided by two metal rods to impact the right hamstring muscles. Following muscle injury, femoral fracture was performed. First, an incision was made on the medial side of the patella, which was then displaced laterally to expose the femur. For the duration of the experiment, a 1.1 mm Kirschner wire was inserted into the medullary cavity to stabilize the fracture. The patella was returned to its original position (i.e., anterior to the Kirschner wire) and secured with sutures to facilitate ambulation. A 500 g impactor (diameter: 3 mm) was released from a height of 55 cm and guided to strike the mid-shaft of the femur, inducing a non-comminuted transverse fracture of the femur, which was confirmed via X-ray.

[0051] After extracranial injury, TBI was administered using a lateral fluid percussion injury (FPI) model. As previously described (25), after a craniotomy (5 mm diameter, 4.5 mm posterior, 2.5 mm left of bregma), a hollow injury cap was attached over the cranial incision using dental acrylic. The rat was then connected via the injury cap to a fluid percussion device (model FP302, Amscien Instruments, USA), and a fluid pulse (approximately 3 atmospheres) was delivered. For all sham injuries, anesthesia and buprenorphine were administered, followed by incision and suture of the craniotomy, but no weight was released and no FPI was delivered.

[0052] blood sampling Under isoflurane-induced anesthesia, blood was collected from the lateral tail vein using a 23G needle into 500 μl of a K2-EDTA Microtainer (#365975, McFarlane Medical) two days after injury. Immediately after collection, the Microtainer was gently inverted to mix the blood and EDTA, and then centrifuged at 1300 g for 10 minutes at 4°C. 100 μL of supernatant (plasma) was aliquoted into individual 0.5 mL Protein LoBind tubes (#0030108434, Eppendorf) and stored at -80°C.

[0053] Example 1.3. Cell Culture cell culture Human SaOS2 cells and CCF-STTG1 astrocytes were obtained from CLS (CLS, Eppelheim, Germany). C3H10T1 / 2 MSCs were donated by Dra Pardo from the IDIS Institute in Santiago de Compostela. ATDC5 cells were purchased from the RIKEN Cell Bank.

[0054] cell differentiation SaOS2 cells were differentiated for 3 days, C3H10T1 / 2 cells for 7 days, and ATDC5 cells for 14 days. To induce cell differentiation, the medium in SaOS2 and ATDC5 cells was changed three times a week, starting after 24 hours of culture. The same medium was used for all changes. For C3H10T1 / 2 cells, two types of medium were used to differentiate them into adipocytes: induction medium, which was changed after 6 hours of culture, and maintenance medium, which was replaced with induction medium after 4 days of culture. All cells were seeded into 24-well plates (Thermo Fisher Scientific, Waltham, MA, USA).

[0055] Meanwhile, SaOS2 cells were differentiated in the presence of 10% FBS or 10% serum from patients in four groups (G1, G2, G3, and G4) for 0, 3, 7, 14, and 21 days. To differentiate, 58,000 cells were seeded per well in a 24-well plate. The next day, the seed medium was removed and FBS-free SaOS2 differentiation medium (DM) was added. Then, 10% FBS or 10% patient serum was added, and cell differentiation was performed as previously described by the inventors.

[0056] co-culture Cell co-culture was performed in Millicell™ Hanging Cell Culture Inserts (Thermo Fisher Scientific, Waltham, MA, USA) with 0.4 μm pore size and in 24-well plates. Astrocytes (CCF-STTG1) were cultured in the inserts at a density of 10,500 cells per insert. The next day, the astrocyte medium was replaced with DM, and the cells were continued to be cultured for 72 hours. 48 hours after seeding, SaOS2 cells were seeded in 24-well plates at a density of 58,000 cells per well. 24 hours after SaOS2 culture, both cultures were combined to form CCF-STTG1 and SaOS2 co-cultures. At this point, the medium for both cells was replaced with fresh DM, and the cells were continued to be co-cultured for 3, 7, 14, and 21 days.

[0057] For the co-culture of NE and SaOS2, the same procedure as described above for astrocytes was used, the only difference being that NE was placed on the insert instead of astrocytes.

[0058] For co-culture with ONCs, the protocol followed was identical to that previously described for NEs, except that cells were seeded upon reaching confluence in the wells containing NEs. In this case, ONCs were seeded onto inserts at a density of 21,000 cells per insert. Subsequently, SaOS2 cells were seeded to form a co-culture of ONCs and SaOS2s, which was maintained for 3 days.

[0059] Example 1.4. Cell culture using conditioned medium Obtaining conditioned medium Astrocyte-conditioned medium (ACM): Astrocytes were seeded in 6-well plates at a density of 332,500 cells per well. After 24 hours of incubation, the medium was replaced and 2 mL of DM was added. Under these conditions, astrocytes were allowed to acclimate to the medium for 72 hours to obtain ACM.

[0060] Neuronal conditioned medium (NCM): After processing primary neuronal samples, NEs were plated in 6-well plates at a density of 3–4 NEs per well. NCM was collected after 1.5 months and 72 hours of medium conditioning (before changing it).

[0061] Cell culture using conditioned medium Once MCAs were obtained, they were used to stimulate the differentiation of SaOS2 and C3H10T1 / T2 cells as previously described.

[0062] After 24 hours of post-SaOS2 culture, the medium was changed to ACM and these were allowed to differentiate for 3 days.

[0063] For C3H10T1 / 2, due to differences in the differentiation medium and SaOS2 composition, 2 mL of DM and ACM proteins were concentrated and purified using a 3 kDa Amicon™ Ultra-0.5 mL centrifugal filter Ultracel 3K (Millipore, Burlington, MA, USA) filter unit according to the supplier's instructions. Once the process was complete, the pellet was resuspended in the lineage-specific differentiation medium and filtered through a 0.2 μm filter (Thermo Fisher Scientific, Waltham, MA, USA). The cells were differentiated for 7 days as previously described.

[0064] For NCM-stimulated cell differentiation, NCM was concentrated using an Amicon™ Ultra-0.5 mL Ultracel 3K 3 kDa centrifugal filter unit according to the supplier's specifications. They were then resuspended in C3H10T1 / 2 differentiation medium and filtered through a 0.2 μm filter. Finally, C3H10T1 / 2 cells were differentiated into adipocytes for 7 days in the presence or absence of NCM as previously described.

[0065] Example 1.5. Treatment inflammatory stimulus To examine the effects of external inflammatory factors on the ACM-stimulated differentiation process of SaOS2 cells, SaOS2 cells were seeded and stimulated with ACM in the presence or absence of 0.1 ng / mL IL-1β and 100 ng / mL lipopolysaccharide (LPS). To do this, after SaOS2 cells completed 3 days of differentiation with ACM, these cells were treated with inflammatory stimuli and further differentiated for 48 hours.

[0066] Inducible transforming growth factor β (TGFβi) stimulation SaOS2, C3H10T1 / 2, and ATDC5 cells were differentiated in the presence or absence of 3 μg / mL of the human recombinant inducible transforming growth factor β (TGFβi), a concentration similar to the blood level of TGFβi. Note that this protein was added freshly with each medium change for differentiation of each cell lineage.

[0067] Example 1.6. Cell viability RNA quantification and cell counting RNA of differentiated C3H10T1 / 2 was quantified using a NanoDrop One™ spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA).

[0068] Further micrographs were taken of the various conditions. The photographs were then divided into three equally sized rectangles, each placed in the same location within the image. Viable cells were counted and the cell counts were compared with the rest of the sample.

[0069] MTT assay Viability was tested using MTT reagent. Briefly, 6 × 10 cells were cultured per well. 3 The cells were plated in a 96-well plate and treated as described above. The cells were then incubated with MTT reagent for 4 hours. After dissolving the formazan salt, the absorbance was measured at 570 nm in a spectrophotometer.

[0070] Example 1.7. Cell staining Actin, tubulin, and nuclear staining SaOS2 cells were seeded onto glass coverslips in 24-well plates at a density of 58,000 cells per well. These cells were stimulated with ACM and allowed to differentiate into osteoblasts for 3 days. After the differentiation period, the medium was removed from the wells, and staining was performed according to the supplier's instructions. Finally, the cells were visualized and photographed under a confocal microscope.

[0071] Oil Red O staining C3H10T1 / 2 cells were differentiated as previously described. After 7 days of differentiation, the cells were fixed with 4% formaldehyde at room temperature for 10 minutes. The formaldehyde was then removed, and the cells were completely dried. Next, the cells were incubated with 21% Oil Red O for 10 minutes, followed by four washes with distilled water to remove nonspecific staining residues. Finally, 500 μL of distilled water was added, and the cells were photographed under a microscope.

[0072] Example 1.8. Western Blot Supernatants and serum were precipitated using methanol and chloroform. After protein precipitation, pellets were resuspended in Ripa buffer. Immunoblots were performed using Immobilon™ Western (Millipore, Burlington, MA, USA) and visualized using a ChemiDoc MP Imaging System (BioRad Laboratories, Hercules, CA, USA). The data obtained were further verified by densitometric analysis using Image Lab™ software (Image Lab™ 6.0.1, Bio-Rad Laboratories, USA) and ImageJ (ImageJ 1.51s, NIH, USA).

[0073] Example 1.9. TGFβi Enzyme-Linked Immunosorbent Assay (ELISA) Plasma levels of TGFβi were quantified using the Rat beta IG-H3 / TGFBI ELISA kit PicoKine™ (#EK1571, Boster Biological Technology, Pleasanton, CA, USA). All samples and standards were run in duplicate, and the assay was performed according to the manufacturer's instructions.

[0074] Once the reaction was complete, the absorbance of each well was measured at a wavelength of 450 nm using a FLUOstar Omega (BMG Lab Technologies) and processed with MARS Data Analysis software (BMG Lab Technologies) to calculate the average protein concentration for each sample.

[0075] Example 1.10. RNA Extraction and Real-Time PCR (RT-PCR) RNA was extracted using TriReagent™ and the RNA extraction kit EZNA Total RNA Kit I™ (Omega Bio-Tek Inc., Norcross, GA, USA) according to the manufacturer's instructions. Gene expression of bone-related, inflammatory, and adipogenic markers was measured using iTaq Universal SYBR Green Supermix (BioRad Laboratories, Hercules, CA, USA). Relative quantification was performed using the ΔΔCt comparative method. Results are presented as a 2% reduction compared to unstimulated controls or DM. -ΔΔCt It is expressed as:

[0076] Example 1.11. Statistical Analysis The results from this study were expressed as the mean ± standard error of the mean (SEM) of at least three experiments or independent samples and analyzed using the GraphPad Prism statistical program (GraphPad Software Inc. 8, USA) and R (RStudio 1.3.1093, USA). The analysis used parametric and non-parametric Student's t-tests depending on the origin of the samples. A p-value of less than 0.05 (p<0.05) was considered significant. For graphical presentation, results were expressed as positive if p<0.05; negative if p<0.05; * , if p<0.01 ** , if p<0.001 *** , if p<0.0001 **** For correlations, Pearson's non-parametric correlation analysis was used, applying a 5% FDR.

[0077] Example 2. Results Example 2.1. Effect of astrocytes on osteoblast formation Example 2.1.1. Study of the anabolic effects of astrocytes on osteoblasts In the context of TBI, astrocytes play a crucial role in pathophysiological and repair processes through the release of numerous factors. To investigate the potential contribution of these cells in NHO, we cocultured astrocytes (CCF-STTG1) and osteoblasts (SaOS2) and differentiated them into osteoblasts for 3, 7, 14, and 21 days. Interestingly, coculture of the cells promoted osteoblastogenesis by increasing the gene expression of essential markers of bone anabolism, such as SPP1, RUNX2, and BMP2, in osteoblasts (Figure 1A-C).

[0078] To determine whether this effect was due to a direct effect of astrocytes on osteoblasts or the result of feedback between both cell types, astrocytes were acclimated to DM for 3 days to obtain ACM. Osteoblasts were subsequently stimulated with ACM and allowed to differentiate for 3 days. The anabolic effects previously observed in the coculture of cells were maintained, with the exception of the effect on RUNX2. The effect on RUNX2 was unaltered by the presence of ACM (Figure 2A), suggesting the release of anabolic factors by astrocytes. Based on these findings, we investigated a larger number of bone metabolic markers. Results demonstrated that ACM increased the gene expression of bone anabolic markers BMP4, LIF, GPNMB, CD44, PDGFβ and its receptor PDGFRβ (Figure 2B), and OPG (Figure 2C) in osteoblasts. However, ACM did not induce changes in RANKL expression, resulting in a decrease in the RANKL / OPG bone remodeling ratio (Figure 2C).

[0079] Finally, ACM regulated key gene markers of the WNT pathway. Thus, ACM increased the gene expression of AXIN2 in osteoblasts, a gene inducible after activation of the WNT pathway, which subsequently acted as its inhibitor through a negative feedback mechanism, decreasing the gene expression of SOST and DKK1, the main inhibitors of this pathway (Fig. 2C).

[0080] Example 2.1.2. Determining the Inflammatory Effects of Astrocyte-Conditioned Medium on Osteoblasts Considering the role of inflammation in bone anabolism, we investigated the effect of ACM on the inflammatory process in osteoblasts. ACM increased the gene expression of inflammatory markers, such as VCAM, a gene associated with inflammation-associated leukocyte recruitment and vascular adhesion, and CCL2, a gene promoting monocyte chemotaxis and macrophage polarization during the inflammatory process, in SaOS2 cells differentiated into osteoblasts for 3 days (Figure 3A).

[0081] To investigate the contribution of the endogenous inflammatory effects of ACM to the innate immune response-related inflammatory responses described in NHO, SaOS2 cells were differentiated into osteoblasts for 3 days in the presence or absence of ACM and then stimulated with IL1β (0.1 ng / mL) and LPS (100 ng / mL) for 48 hours. The presence of ACM enhanced the inflammatory effects of IL1β, as evidenced by increased gene expression of IL6, VCAM, and CCL2 (Figure 3B). Interestingly, inflammatory stimulation increased the expression of the bone anabolic genes BMP2 and LIF induced by ACM (Figure 3B). However, IL1β did not alter the expression of SPP1 (Figure 3B). LPS, a major Toll-like receptor 4 (TLR4) agonist, did not alter the inflammatory profile or bone metabolism (Figure 3C).

[0082] Example 2.1.3. Determination of cell morphological changes induced by astrocyte-conditioned medium The results demonstrated the ability of astrocytes to increase the expression of bone-related markers in SaOS2 cells. To examine the effect of ACM on osteoblast morphology, tubulin, actin, and nuclear staining were performed. The data revealed that ACM induced osteocyte-like morphological changes in SaOS2 cells (Figure 4A). Taking these findings into consideration, we investigated osteocyte markers in SaOS2 cells by RT-PCR. Consistent with the staining results, ACM increased podoplanin (PDPN) gene expression, an early marker of osteoblast-to-osteocyte differentiation (Figure 4B). However, ACM did not alter BGLAP, another osteocyte marker studied (Figure 4B).

[0083] Example 2.2. Effect of astrocyte-conditioned medium on adipogenesis Previous results indicated that ACM promotes bone anabolism in osteoblasts. Considering that osteoblasts and adipocytes share the same cellular origin, MSCs, and that osteoblastogenesis and adipogenesis are balanced and contrasting processes, we performed experiments to examine the effect of ACM on the differentiation of C3H10T1 / 2 cells into adipocytes over a 7-day period. The obtained data revealed a decrease in the gene expression of adipogenic markers FABP4, PPARG, and ADIPOQ in ACM-treated cells. However, ACM did not alter the expression of PLIN2 (Figure 5A).

[0084] To validate the data, we performed Oil Red O staining on C3H10T1 / 2 cells differentiated into adipocytes for 7 days in the presence or absence of ACM. Consistent with the gene expression results, the data revealed a decrease in fat deposition in DM-treated cells and a complete inhibition of lipid droplet formation in ACM-stimulated cells, confirming its anti-adipogenic effect (Figure 5B).

[0085] Given the ability of MSCs to differentiate into both bone and fat cells, we investigated bone metabolic and inflammatory markers in C3H10T1 / 2 cells differentiated into adipocytes for 7 days, since ACM inhibited adipogenesis. Interestingly, ACM was able to increase gene expression of the markers SPP1, RUNX2, CD44, AXIN2 (Figure 5C), IL6, and VCAM (Figure 5D). In contrast, this medium reduced the expression of GPNMB (Figure 5C) and did not alter the expression of CCL2 (Figure 5D).

[0086] Finally, changes in cell proliferation were observed during the differentiation process of these cells. To examine these changes, MTT assays of C3H10T1 / 2 differentiated into adipocytes were performed over 7 days in the presence and absence of ACM. The results revealed that ACM-stimulated C3H10T1 / 2 cell proliferation increased during their differentiation into adipocytes (Figure 5E).

[0087] Example 2.3. Effect of nerve explants on osteoblast formation HO is a multifactorial disease whose etiology remains unknown. However, certain risk factors exist that increase its prevalence. Among these, neurological damage, such as traumatic brain injury (TBI) and spinal cord injury, is noteworthy. Recent studies have suggested that the PNS may be involved in the pathogenesis of NHO. To investigate the role of peripheral nerves in ectopic bone formation, we cocultured NE and SaOS2 cells and allowed them to differentiate into osteoblasts for 3 days. Our data demonstrated that SaOS2 cells cocultured with NE increased the gene expression of bone markers SPP1, RUNX2, BMP2, LIF, CD44, and RANKL, as well as the RANKL / OPG ratio, as well as the expression of inflammatory markers VCAM and CCL2 (Figures 6A-6D). In contrast, coculture of NE and SaOS2 cells decreased the expression levels of BMP4 and GPNMB (Figure 6B), while the expression levels of PDGFβ, PDGFRβ, OPG, and AXIN2 remained unchanged (Figures 6B and 6C).

[0088] Similarly, ONC and SaOS2 cells were cocultured and allowed to differentiate into osteoblasts over a 3-day period. Although the observed effects were not as strong as those observed in the coculture of NE and SaOS2, ONC was able to increase the expression of SPP1 and VCAM genes in SaOS2 cells. Similarly, the presence of ONC increased the RANKL / OPG ratio and decreased the expression of BMP4 and PDGFRβ. However, this coculture did not induce any changes in the other genes studied (Figure 7).

[0089] Considering the inflammatory and anabolic effects observed in the coculture of NE and SaOS2, we investigated the combined effects of ACM-stimulated NE and SaOS2 coculture. Results revealed that the presence of NE significantly enhanced the effect of ACM on the gene expression of osteoblastic and inflammatory markers SPP1, BMP2, LIF, RANKL, and CCL2 (Figures 8A-8D). Interestingly, ACM synergized the effect of NE coculture with osteoblasts by increasing the gene expression of SPP1, LIF, and CCL2 (Figures 8A, 8B, and 8D). In contrast, NE significantly reduced the gene expression levels of BMP4, GPNMB, CD44, PDGFβ, PDGFRβ, and AXIN2 (Figures 8B and 8C). Similarly, ACM was able to modulate the effects of NE beyond the previously described synergistic induction by significantly enhancing the expression of markers BMP4, GPNMB, CD44, PDGFβ, OPG, and AXIN2 (Figures 8B and 8C), and by decreasing the expression of RUNX2 and RANKL genes and the RANKL / OPG ratio (Figures 8A and 8C).

[0090] Example 2.4. Effect of neural explant conditioned medium on adipogenesis Considering the balance between osteoblastogenesis and adipogenesis, we tested the effect of NCM on C3H10T1 / 2 cells differentiated into adipocytes for 7 days, and investigated key adipogenic markers by RT-PCR. Results revealed a significant decrease in the expression of three of the studied markers: FABP4, PPARG, and ADIPOQ. However, NCM increased the expression of PLIN2 (Figure 9A).

[0091] Similarly, bone metabolism and inflammatory markers were investigated in the adipogenic differentiation model. The data revealed that the expression of SPP1, CD44, IL6, VCAM, and CCL2 was significantly increased in cells stimulated with NCM (Figures 9B and 9C). However, this medium did not alter the expression of RUNX2, GPNMB, and AXIN2 (Figure 9B).

[0092] Example 2.5. Characterization of astrocyte-conditioned medium and neural explant-conditioned medium Based on the obtained results, the secretomes of astrocytes, NE, and DM were characterized by proteomics.

[0093] Data obtained from qualitative proteome analysis (DDA) demonstrated elevated expression of TGFβi in ACM compared with DM (Figure 10A). These results were confirmed by quantitative proteome analysis (SWATH). Thus, SWATH analysis revealed an increase in this protein in ACM, which was 14.52-fold higher in ACM than in MD (Figures 10A and 10B). To validate the proteome data, protein expression studies of cell supernatants were performed using Western blot. The information obtained revealed higher protein expression of TGFβi in ACM compared with DM, which was consistent with previous results obtained by proteomics (Figure 10C).

[0094] Similarly, NCM was characterized as in ACM. Qualitative proteomic analysis (DDA) revealed higher TGFβi expression in NCM compared to DM (Figure 11A). Similarly, quantitative proteomic studies (SWATH) confirmed this increase, revealing a 2.2-fold increase in NCM compared to DM (Figures 11A and 11B). These results were verified by Western blot, revealing increased protein expression of TGFβi in NCM compared to NM (Figure 11C).

[0095] Finally, we compared the proteomes of ACM and NCM. Qualitative analysis revealed two common proteins in both media: TGFβi and heavy chain inter-alpha trypsin inhibitor 4 (ITIH4). Interestingly, TGFβi was the only common protein that remained significant in the quantitative proteome analysis of ACM and NCM.

[0096] Considering these results, we analyzed the expression levels of TGFβi by RT-PCR during the differentiation of SaOS2 cells into osteoblasts and C3H10T1 / 2 cells into adipocytes in the presence or absence of ACM, NE, and ONC. The results demonstrated the ability of ACM to increase TGFβi gene expression during osteoblastogenesis (Figure 12) and adipogenesis (Figure 13). However, the increase in TGFβi in the adipogenesis model was not significant compared to the control. Similarly, the presence of NE and ONC increased TGFβi gene expression during both osteoblast differentiation (Figure 12) and adipogenic differentiation (Figure 13). Similarly, the expression of this gene was studied in SaOS2 cells differentiated in the presence of NE and coculture with ACM. The results revealed how ACM can significantly enhance TGFβi expression compared to NE, and how NE significantly reduced the expression of this gene in cells treated with ACM (Figure 12).

[0097] Example 2.6. Confirmation of TGFβi Expression in an NHO In Vivo Model To verify the previous results, we investigated the expression of TGFβi in blood samples collected from rats in the NHO in vivo model. The results revealed that this protein was significantly increased in the dual trauma group (TBI + LBF) compared with the remaining groups (Figure 14A). Furthermore, TGFβi was more elevated in animals that developed ectopic bone (including animals from the TBI + LBF and LBF groups) compared with the remaining groups (Figure 14B).

[0098] Example 2.7. Confirmation of TGFβi Expression in Patients at Risk for NHO To characterize patients at higher or lower risk for NHO, serum samples from G1, G2, and G3 patients were analyzed by proteomics. The data correlated with changes in gene expression induced by patient serum during SaOS2 differentiation. Results demonstrated a greater number of correlations with bone metabolism in patients with dual trauma (G1) compared with the remaining groups (G2 and G3) (Figures 15A-C). Furthermore, proteome enrichment revealed a greater association with inflammatory cascades and pathways involved in TGFβ signaling in patients at high risk for NHO (G1). Overall, the data correlated with the in vitro results (Figure 15D).

[0099] Since in vitro and in vivo experiments confirmed that TGFβi may contribute to the anabolic and inflammatory processes of NHO, and taking into account the differences in the proteomes of different patient groups, we performed a kinetic study of TGFβi expression in serum. The results revealed that this protein was increased in patients at high risk of developing NHO (G1), reaching its maximum expression on day 3 after trauma (Figures 16 and 23).

[0100] Example 2.8. Identification of TGFβi as a protein potentially involved in ectopic bone formation Example 2.8.1. Effect of TGFβi stimulation on osteoblast formation We previously demonstrated the pro-osteoblastogenic and anti-adipogenic effects of ACM and NCM. These studies also identified TGFβi as a potential biomarker for NHO, given its high expression induced by both ACM and NCM and its presence in the serum of rats and patients at risk for the disease. To clarify the involvement of TGFβi in this process, SaOS2 cells were differentiated for 3 days in the presence or absence of human recombinant protein TGFβi (3 μg / mL). Interestingly, treatment with TGFβi partially reproduced the effects of ACM on osteoblasts. Thus, TGFβi increased the gene expression of SPP1, BMP2, TGFβi, VCAM, and CCL2 (Figure 17). In contrast, stimulation with TGFβi did not alter the gene expression of the other genes studied (Figure 17).

[0101] Example 2.8.2. Effect of TGFβi stimulation on adipogenesis C3H10T1 / 2 cells were differentiated into adipocytes for 7 days in the presence or absence of TGFβi (3 μg / mL). Treatment with recombinant proteins reduced the gene expression levels of all adipogenic markers studied, including FABP4, PLIN2, PPARG, and ADIPOQ (Figure 18A). However, the only marker for which the reduction was statistically significant was PPARG (Figure 18A).

[0102] Similarly, bone- and inflammation-related markers were investigated in C3H10T1 / 2 MSCs differentiated into adipocytes. Results revealed increased gene expression of SPP1, RUNX2, CD44, TGFβi, and VCAM in TGFβi-stimulated C3H10T1 / 2 (Figures 18B and 18C). In contrast, no changes were observed in the gene expression of the remaining genes studied (Figures 18B and 18C).

[0103] Finally, stimulation of cells with the recombinant protein TGFβi promoted cell proliferation, as determined by the increase in cell number and RNA content in stimulated cells (FIG. 18D).

[0104] Example 2.8.3. Effect of TGFβi stimulation on chondrogenesis Chondrocytes play an essential role in bone formation through the endochondral ossification process. Therefore, we evaluated the effect of treatment with recombinant protein TGFβi (3 μg / mL) on ATDC5 differentiation into chondrocytes for 3, 7, and 14 days. The obtained data reflected a decrease in the gene expression of COL2A1, COLX, and SOX9 markers, demonstrating the detrimental effect of TGFβi on chondrocyte differentiation (Figures 19A-C). Furthermore, TGFβi stimulation increased SPP1 expression at an early stage of ATDC5 differentiation (Figure 19D).

[0105] Example 2.9. TGFβi Blockade In this study, TGFβi was identified as a potential therapeutic target for NHO. Therefore, we explored two different approaches to inhibit its action as a potential example of a TGFβi inhibitor. First, we used specific siRNA to silence TGFβi mRNA in astrocytes. The results revealed that this approach was an effective way to blunt TGFβi released by astrocytes into the cell culture medium (Figure 20). Second, we added a TGFβi-neutralizing antibody to ACM to block the action of TGFβi. The data revealed that SaOS2 cells differentiated into osteoblasts for 3 days in the presence of this antibody expressed lower mRNA levels of SPP1, a gene associated with NHO (Figure 21).

[0106] Example 2.10. Validation of TGFBi as a biomarker or therapeutic target in NHO The effect of tyrphostins on primary osteoblasts was evaluated. To study this, primary osteoblast-like cells were differentiated into osteoblasts in the presence or absence of tyrphostins (JAK inhibitors) for 3 days. Results revealed that tyrphostin treatment increased the expression levels of bone-related markers (SPP1, BMP2) and TGFBi (Figure 22).

[0107] Meanwhile, this study demonstrated increased circulating levels of TGFBi in patients at risk of developing HO. To further investigate the involvement of TGFBi in the development of ectopic bone masses, the expression levels of TGFBi in primary bone samples from patients with acquired HO were studied (Figure 26). The results showed higher expression levels of TGFBi in ectopic bone masses when compared with control primary bone samples (Figure 26). Furthermore, GNAS from patients with genetic HO variants was also shown to be significantly higher in GNAS than in control primary bone samples. - / - Studies performed with cells demonstrated that differentiation of MSCs into adipocytes was accompanied by an increase in TGFBi, a bone-related marker (SPP1), and a decrease in adipogenesis-related markers (ADIPOQ, FABP4) (Figure 27). Finally, primary NE from patients with HO genetic variants was also shown to increase TGFBi expression levels in Saos2 cells differentiated into osteoblasts for 3 days (Figure 28).

[0108] As demonstrated by the present inventors, the expression of TGFBi is higher in patients at risk of developing NHO, as well as in patients with acquired HO and patients with hereditary HO. To investigate the accuracy of TGFBi as a biomarker for predicting NHO, area under the curve (AUC) and AUC-based simulation analysis were performed (Figures 24 and 25, respectively). The obtained data showed a significant sensitivity of TGFBi for predicting the development of NHO (sensitivity = 0.82 and 0.853, respectively).

[0109] Overall, this data indicates the involvement of TGFBi in NHO. Thus, TGFBi was significantly increased in primary ectopic bone samples from patients at risk for developing NHO (TBI+LBF) and from patients with HO. Furthermore, it was demonstrated that nerves from patients with HO induce high levels of TGFBi in osteoblasts, which may contribute to the development of NHO. Furthermore, GNAS obtained from primary ectopic bone samples from patients with a genetic variant of HO (predominant intramembranous ossification) showed a significant increase in TGFBi. - / - The cells were determined to secrete higher levels of TGFBi compared to controls. Finally, statistical testing demonstrated that TGFBi has excellent sensitivity for predicting the onset of NHO. Therefore, this data indicates that TGFBi is a biomarker for predicting the onset of NHO, which involves both endochondral and intramembranous ossification processes.

[0110] Finally, it has been demonstrated that bone mass-reducing drugs (tyrphostins) do not necessarily target TGFBi. This finding could explain the shortcomings of current therapeutic approaches and the high rate of bone recurrence after ectopic bone removal. Considering this, using TGFBi as a therapeutic target for NHO may be an excellent tool for preventing and treating this disease.

[0111] In conclusion, this data provides a new, more accurate approach to the management of NHO, including its diagnosis, prognosis, and treatment.

[0112] Example 2.11. Triple Therapy Comprising a Thiazolidinedione, a Corticoid, and a Nonsteroidal Anti-inflammatory Drug in the Treatment of NHO In the context of the present invention, a combination of three compounds (a thiazolidinedione, a corticoid, and a nonsteroidal anti-inflammatory drug, preferably rosiglitazone or pioglitazone, dexamethasone, and indomethacin) was assayed.

[0113] As can be seen in Figure 29, this therapeutic agent significantly reduced the levels of TGFBi induced by ACM during 3 days of Saos2 differentiation.

Claims

1. 1. An in vitro method for diagnosing or prognosing neurogenic heterotopic ossification, comprising assessing the expression level of inducible transforming growth factor beta (TGFBi) in a biological sample comprising an astrocyte and / or neuronal secretome obtained from a subject, wherein determining an increased expression level of TGFBi relative to a pre-determined expression level measured in a healthy control subject is an indication that the subject is likely to suffer from neurogenic heterotopic ossification or exhibits a poor prognosis.

2. 2. The in vitro method of claim 1, wherein the biological sample is selected from blood, plasma, serum, peripheral nerves including the saphenous nerve, the posterior tibial nerve, and the sciatic nerve, cerebrospinal fluid, or synovial fluid.

3. In vitro use of TGFBi in a biological sample comprising an astrocyte or neuronal secretome obtained from a subject for the diagnosis or prognosis of neurogenic heterotopic ossification.

4. In vitro use of a kit comprising reagents for determining the level of expression of TGFBi in a biological sample comprising an astrocyte or neuronal secretome obtained from a subject for the diagnosis or prognosis of neurogenic heterotopic ossification.

5. A TGFBi inhibitor for use in a method for treating neurogenic heterotopic ossification, comprising inhibiting TGFBi in astrocytes and / or neuronal secretomes, wherein the inhibitor is selected from a neutralizing TGFBi antibody, an siRNA that base pairs with TGFBi mRNA, an miRNA, or a combination drug comprising a thiazolidinedione, a corticoid, and a non-steroidal anti-inflammatory drug.

6. The TGFBi inhibitor of claim 5, wherein the thiazolidinedione is rosiglitazone or pioglitazone, the corticoid is dexamethasone, and the nonsteroidal anti-inflammatory drug is indomethacin.

7. The TGFBi inhibitor according to claim 5 or 6, wherein the TGFBi inhibitor is administered intravenously, intrathecally, locally, or intraarticularly.

8. 1. An in vitro method for identifying a compound that treats neurogenic heterotopic ossification, comprising: a) determining whether a candidate compound inhibits TGFBi in astrocytes and / or neuronal secretomes or abolishes the anabolic effect of astrocytes on osteoblasts; and b) wherein, if said inhibition of TGFBi or abolishment of the anabolic effect of astrocytes on osteoblasts occurs, this is an indication that said candidate compound may be effective in treating neurogenic heterotopic ossification.

Images