Analysis of miRNAs in Osteogenesis imperfecta Caused by Mutations in COL1A1 and COL1A2: Insights into Molecular Mechanisms and Potential Therapeutic Targets

Osteogenesis imperfecta (OI) is a group of connective tissue disorders leading to abnormal bone formation, mainly due to mutations in genes encoding collagen type I (Col I). Osteogenesis is regulated by a number of molecules, including microRNAs (miRNAs), indicating their potential as targets for OI therapy. The goal of this study was to identify and analyze the expression profiles of miRNAs involved in bone extracellular matrix (ECM) regulation in patients diagnosed with OI type I caused by mutations in COL1A1 or COL1A2. Primary skin fibroblast cultures were used for DNA purification and sequence analysis, followed by analysis of miRNA expression. Sequencing analysis revealed mutations of the COL1A1 or COL1A2 genes in all OI patients, including four previously unreported. Amongst the 40 miRNAs analyzed, 9 were identified exclusively in OI cells and 26 in both OI patients and the controls. In the latter case, the expression of six miRNAs (hsa-miR-10b-5p, hsa-miR-19a-3p, hsa-miR-19b-3p, has-miR-204-5p, has-miR-216a-5p, and hsa-miR-449a) increased, while four (hsa-miR-129-5p, hsa-miR-199b-5p, hsa-miR-664a-5p, and hsa-miR-30a-5p) decreased significantly in OI cells in comparison to their expression in the control cells. The identified mutations and miRNA expression profiles shed light on the intricate processes governing bone formation and ECM regulation, paving the way for further research and potential therapeutic advancements in OI and other genetic diseases related to bone abnormality management.


Introduction
Osteogenesis imperfecta (OI) is a group of genetically related connective tissue disorders, diverse in terms of both genotypes and phenotypes, characterized by increased bone fragility.Clinically, five types of OI ranging from mild (type I) to moderate (types III and IV) and severe or lethal (type II) are distinguished [1][2][3].So far, numerous mutations in genes encoding collagen type I (Col I) or involved in collagen modification, e.g., cartilage-associated protein (CRTAP) and leucine proline-enriched proteoglycan 1 (LEPRE1) [4,5]; folding, e.g., serpin peptidase inhibitor (SERPINH1) [6]; and processing, e.g., bone morphogenetic protein 1 (BMP1) [7], have been shown to cause various types of OI.Nevertheless, a majority of OI cases inherited in an autosomal-dominant manner are caused by mutations in the collagen type I α1 or α2 encoding genes (COL1A1 and COL1A2, respectively), resulting in a reduced amount or abnormal structure of collagen [8][9][10].
The earliest stage when an OI diagnosis can be made is the prenatal period.It depends on the severity of the OI and is based on family history; genetic tests, i.e., sequencing of genomic DNA and specific mutation identification; and prenatal tests, i.e., a decrease in the femoral length (FL) [11].In the post-natal period, positive family history and pivotal clinical symptom (bone fragility, typical features of blue sclera) notification with bone density measurement (BMD) implementation are routinely used in OI diagnosis [12,13].Additional diagnostic methods include magnetic resonance imaging (MRI), radiography of the uterus, and other genetic tests, such as multiplex ligation-dependent probe amplification (MLPA) or chorionic villus sampling (CVS) [12,14].If the typical clinical symptoms and positive family history are noted by the clinician, routine diagnostic methods are sufficient to diagnose OI; however, about 10-15% of patients suffer from less common variants of the disease.Moreover, any diagnostic method used may be clear-cut and have drawbacks, e.g., employment of the FL carries the risk of other skeletal dysplasia's recognition instead of OI [13,15].In such circumstances, further procedures, mostly based on molecular diagnostics, are needed [11].
MicroRNA (miRNA) constitutes short, about 20-nucleotide, single-stranded, and endogenous RNA molecules that have been observed to negatively regulate gene expression in two manners: i) by attaching to their target messenger RNA (mRNA) at different mRNA sites, most often 100% complementary, and cutting it; or ii) by binding to the 3 -untranslated region (3 UTR) of the target mRNA and inhibiting translation, an action that does not require 100% complementarity [16].The principal role of miRNAs is the regulation of a number of biological processes, such as embryogenesis, apoptosis, hosting immune responses during infections, or cancerogenesis [16].miRNAs fulfill a crucial role in the regulation of gene expression and are involved in bone formation and bone diseases [17][18][19].For instance, miR-217 has been reported to promote cell proliferation and osteogenic differentiation during the development of steroid-associated osteonecrosis in bone marrow mesenchymal stem cells [20,21], whereas miR-200a-3p has been linked with the synthesis of ECM components in fibrotic diseases [22].
The aim of this study was to identify and analyze the expression profiles of miRNAs in osteogenesis imperfecta (OI) type I caused by mutations in the COL1A1 and COL1A2 genes.

Mutation Analysis in the COL1A1 and COL1A2 Genes from OI Type I
Osteogenesis imperfecta type I in the majority of cases (~95%) is caused by mutations in the COL1A1 and COL1A2 genes; hence, in this study, we focused on sequencing analyses of those genes [1,2].Mutations in COL1A1 were found in 8 study subjects and in COL1A2 in 2 out of 10 study subjects diagnosed with OI type I.The missense mutations were detected in both COL1A1 (donor nos.1-4) and COL1A2 (donor nos. 9 and 10).In COL1A1, other mutation types were also found, including a single-nucleotide deletion resulting in a frameshift in three donors (nos.5-7) and an exon skipping in one patient (no.8).The missense mutations detected in the COL1A1 gene resulted in three different substitutions, namely glycine by alanine in exon 17 (patient no.1), glycine by valine in exon 9 (patient no. 2), and a third one also found in exon 17 that resulted in valine-to-phenylalanine alteration (patient no.3).Additionally, in two other patients (nos.6 and 7), two deletions of T located in exon 30 and exon 17 of the COL1A1 gene resulted in the production of an altered protein.In contrast, both missense mutations in the COL1A2 gene resulted in the substitution of glycine by serine in exon 31 (patient no.9) or exon 38 (patient no.10).
Moreover, we identified a series of new mutations in both genes: substitution of guanine for thymine in exon 9 in patient no. 2 and deletion of 10 nucleotides in exon 50 (frameshift) adjacent to a missense mutation (substitution of glutamine by valine) in OI patient no. 4, both in the COL1A1 gene.Another alteration comprised a guanine deletion in exon 2 of the COL1A1 gene and was identified in patient no. 5.A new mutation found in the COL1A2 gene was represented by a substitution of guanine for adenine in exon 31 of patient no. 9.
A brief clinical description of OI patients harboring previously reported mutations as well as the new mutations in COL1A1 and COL1A2 identified in this study are presented in Table 1.

Identification of miRNAs in OI Type I
First, we performed identification analyses of the miRNAs in the OI-derived cells and control cells.As shown in Table 2, the results were categorized into the three following parts: (A) miRNAs detected in OI but not in the control fibroblasts; (B) miRNAs detected in both OI and the control cells; (C) miRNAs not detected in OI nor the control cells.As presented in Table 2, nine specific miRNAs, i.e., hsa-miR-19a-3p, hsa-miR-19b-3p, hsa-miR-133a-3p, hsa-miR-200b-3p, hsa-miR-204-5p, 216a-5p, hsa-miR-377-3p, hsa-miR-449a, and hsa-miR-590-5p, were identified in the fibroblasts of 30% to 70% of OI donors, but not in the control fibroblasts.Other 26 miRNA molecules were detected in both the OI and the control cells, with the percentage of OI donors with a specific miRNA detection ranging from 70% to 100%.Finally, five miRNAs selected for analysis, i.e., hsa-miR-1-3p, hsa-miR-141-3p, hsa-miR-142-3p, hsa-miR-200a-3p, and hsa-miR-217, were not detected in the OI or the control cells.
Considering the type of mutation, it was noted that patients no. 1, 2, and 3 with missense mutations in the COL1A1 gene revealed similar overexpression patterns for several miRNAs, indicating that these mutations may have similar effects on miRNA regulation.Patients no. 4, 6, and 7, with frameshift mutations in the COL1A1 gene, displayed similar overexpression patterns for some miRNAs but also individually dependent differences among particular types of miRNAs, and the profiles of expression also varied from those found in OI with missense mutations.In turn, patient no. 5, with a frameshift mutation in the COL1A1 gene, had a unique expression pattern of overexpressed and downregulated miRNAs.In summary, the miRNA expression patterns observed in the individually analyzed OI patients indicate that particular mutations may influence the regulation of specific miRNAs, potentially leading to downstream effects on genes involved in bone development and maintenance (Table 6).

Discussion
Despite similar and uniform phenotypic symptoms, the molecular mechanisms of pathogenetic processes contributing to OI remain unclear.This disorder is difficult to treat as it can be caused not only by direct defects in the amount of type I collagen produced but also by the influence of various proteins and regulatory RNA molecules, like miRNAs, on the processing of synthesized collagen, including the pool of aberrant proteins.
Around 5% of OI cases have an unknown genetic background.DNA sequencing analyses revealed the presence of mutations in both genes of all OI donors enrolled in this study, including four novel and previously unreported, of which three were identified in the COL1A1 gene and one in the COL1A2 gene.The missense mutations detected in the COL1A1 and COL1A2 genes resulted in the substitution of glycine and other amino acids, leading to disruptions in the collagen chain interactions and destabilization of the collagen triple helix [1,76].Frameshift mutations, including single-nucleotide deletion, ten-nucleotide deletion, and exon skipping in the COL1A1 gene, altered the reading frame, resulting in truncated or absent Col I chains.Additionally, a single-nucleotide deletion in exon 2 of COL1A1 disrupted the formation of the collagen triple helix.These mutations likely contribute to the development of OI by affecting the stability and packing of the collagen molecule [77].The mutations identified in this study align with previous reports on OI, reaffirming that mutations in the COL1A1 and COL1A2 genes are responsible for the majority of OI cases [78,79].
The involvement of miRNAs in ECM regulation has recently become an area of intense research, as their ability to modulate gene expression post-transcriptionally makes them potent regulators of crucial cellular processes [17][18][19].Understanding how these miRNAs affect ECM components will be of great importance in the search for new therapeutic targets for therapies for diseases associated with ECM dysregulation.
Further, the fold change analysis provided robust validation of the observed miRNA expression alterations in OI compared to the controls, displaying a significant multiplication in up-or downregulation.Remarkably, miR-449a's expression increased about 20-fold, while miR-449a was exclusively detected in OI patients, which may suggest the potential relevance of both miRNAs in the pathogenesis of OI.
In contrast, we noted a six-fold expression decrease in miR-664a-5p.Though this miRNA has been implicated in the modulation of tumor suppressor p53 and the expression of p53 target genes and has recently been associated with osteosarcoma [70], this finding adds an intriguing dimension to this study, hinting at potential links between OI and p53-related pathways that warrant further investigation.
The intricate relationships between miRNAs may offer a unique opportunity to establish connections between molecular processes, providing insights into the pathogenesis of diverse diseases.For better clarification of the impact of miRNAs on bone ECM changes in OI, we explored their potential correlations, whether positive or negative.Correlation analyses were carried out within the groups with upregulated and downregulated expressions of miRNAs.The miRNAs overexpressed in OI that strongly correlate with each other are involved in the regulation of osteoblast differentiation and osteogenesis [26,27,[44][45][46]66,67], while the miRNAs presented in the latter group have been previously associated with various biological processes and diseases, including cancer, inflammation, and neuronal differentiation [28,29,[32][33][34][36][37][38].With the exception of miR-129-5p, which showed a moderate negative correlation with four other miRNAs associated with various cancers, both groups of analyzed miRNAs revealed positive correlations with miR-29 family members [49,52,56].The miR-29 family negatively regulates collagen expression by directly targeting COL1A1's and COL1A2's mRNA transcripts, while the other miRNAs are linked to fibrosis and cancers, suggesting their effect on the synthesis of the ECM [30,31,[48][49][50][51].
In the present study, 40 miRNAs in total were analyzed in 10 OI donors in whom mutations, including four unreported ones, were identified in COL1A1/COL1A2.Overexpression of miR-10b-5p and miR-382-5p in some patients suggests their involvement in the pathogenesis of OI.On the other hand, decreased expression of miR-129-5p, miR-199b-5p, miR-29b-3p, and miR-664a-5p in most patients suggests their importance in normal bone development and function (Table 6).Kaneto et al. reported decreased expression of miR-29b during the osteoblastic differentiation of mesenchymal stem cells (MSCs).Likewise, in our research, we observed a modest decrease in the expression of miR-29b in skin fibroblasts from a group of patients diagnosed with OI type I. Upon conducting separate analyses for each OI patient, we observed an upregulation of miR-29b expression in certain individuals, which contrasts with the overall trend observed in the entire OI group.This variability may be attributed to the diverse range of patient demographics, including age and gender.miR-29 family members, including miR-29a, miR-29b, and miR-29c, have been implicated in the regulation of ECM proteins, such as collagen, which are important for bone development and homeostasis [52][53][54][55][56].The aberrant regulation of those miRNAs in patients with OI type I induced by mutations in COL1A1 may contribute to reduced collagen synthesis and impaired bone formation.Regarding OI patients harboring mutations in the COL1A2 gene, we observed overexpression of miRNA-25-3p.Noticeably, this miRNA has been implicated in the regulation of bone mineralization and may play a role in osteoblast differentiation [30,31].We also observed some discrepancies in the expression of certain miRNAs, such as let-7a, which was found to be overexpressed in some patients with missense mutations, but its expression was downregulated in patients with an exon-skipping mutation.Interestingly, one patient (no.5) with a novel identified mutation that resulted in a frameshift in exon 2 of the COL1A1 gene displayed a distinct miRNA expression pattern compared to other examined patients, which may suggest a unique background of OI in this individual.
One of the potential therapies for OI based on targeting the miRNAs involved in the regulation of bone formation and remodeling can be applied by either increasing the expression of beneficial miRNAs or decreasing the expression of harmful ones [80][81][82].However, there are several challenges that need to be addressed before its implementation into clinical practice, including: (i) the identification of disease-specific and most effective targeted miRNAs (as discussed in the current work); (ii) the validation of relevant delivery methods that would not only effectively "reach" the targeted micromolecule but also efficiently deliver miRNAs to the bone; and (iii) the overcoming of the immune response to exogenous miRNAs [83].

Conclusions
Despite similar and uniform phenotypic symptoms in individuals with OI, the molecular basis of this disease seems to be varied and still unclear.Moreover, it is difficult to treat, as it can be caused not only by direct defects in the amount of type I collagen produced but also by the influence of various proteins and regulatory RNA molecules, including miRNAs, on the processing of synthesized collagen and the pool of aberrant proteins.Therefore, understanding the function of miRNAs that can affect collagen or signaling pathways that indirectly regulate the amount or quality of collagen is extremely important and necessary.Overall, the current study underscores the importance of the ECM in the development of OI and other genetic diseases of connective tissue.The identified mutations and miRNA expression profiles shed light on the intricate processes governing bone formation and ECM regulation, paving the way for further research and potential therapeutic advancements in OI and other genetic diseases related to bone abnormality management.

Study Participants and Control Material
This study was carried out in accordance with the Declaration of Helsinki and was approved by the Bioethics Committee of the Jagiellonian University in Krakow, Poland (protocol no.DK/KB/CM/0031/689/2010). Clinical examination and medical histories of the subjects were taken by a clinician from the Chair of Pediatrics, Department of Medical Genetics, UJCM, Krakow, Poland.All subjects (or representing individuals) read and signed a written informed consent form prior to inclusion in this study.
Ten subjects (ranging in age from 0.5 to 29 years, of both sexes, and diagnosed with OI type I) were enrolled in this study.The diagnosis was confirmed on the basis of multiple (from to 20) bone fractures in the post-natal period and/or blue sclera or osteoporosis.The inclusion criterion was a clinically confirmed diagnosis of OI type I. Exclusion criteria included the clinical diagnosis of other types of OI and absence of mutations in the COL1A1 and/or COL1A2 genes.A clinical and molecular description of the OI type I subjects enrolled in this study is presented in Table 1.As a control, the human fibroblast cell line BJ (CRL-2522, ATCC, Manassas, VA, USA) was used.

Isolation Procedure of Skin Fibroblasts from OI Donors
Skin biopsy samples were collected from 10 patients diagnosed with OI type I.To establish a fibroblast culture, skin biopsies of a size of about 3-5 mm 2 were cut into smaller pieces and placed in a Petri dish with 10 mL of Dulbecco's Modified Eagle's Medium (DMEM), supplemented with 4.5 g/L glucose, 10% fetal calf serum (FCS), 50 µg/mL streptomycin, 50 U/mL penicillin, and 25 µg/mL amphotericin B (PAA, Austria).After one week, the remaining parts of the skin explants were removed from the dish, and the fibroblast culture was kept until 90% confluence.Subsequently, the cells were harvested using trypsin/EDTA, plated in fresh DMEM with 4.5 g/L glucose, 10% FCS, streptomycin, penicillin, and amphotericin B in plastic flasks, and kept until homogeneous cultures were obtained.The homogeneity of the fibroblast culture was confirmed after first passage by immunofluorescent staining against vimentin.Cell viability amounted to 98% using Trypan blue.

Fibroblast Cultures
For the experiments, 2 × 105 cells per well were plated in 6-well culture plates and maintained for 7 days in DMEM with 4% FBS supplemented every second day with a fresh L-ascorbic acid 2-phosphate sesquimagnesium salt hydrate (Sigma-Aldrich, St. Louis, MO, USA) at a concentration of 40 µg/mL.The experiments were carried out between the third and sixth passages.Control human fibroblast cell line BJ (CRL-2522, ATCC, Manassas, VA, USA) was cultured and treated identically as the primary skin fibroblasts recovered from the study group.

Fibroblast DNA Isolation and Amplification
Lysis and DNA isolation from primary skin fibroblasts and the control cell line were conducted according to the manufacturer's protocol using DNA purification kit Blood Mini (A&A Biotechnology, Gdansk, Poland).The DNA amplification reactions were conducted using FastStartTaq DNA polymerase kit (Roche, Basel, Switzerland).The composition of the reaction mixture and conditions used are described elsewhere [84,85].Briefly, for each gene, 10 amplification reactions were conducted, and the DNA fragments were visualized in 1% agarose gel (Agarose, LE Analytical Grade, Promega, Madison, WI, USA) in TAE buffer.

Sequence Analyses of COL1A1 and COL1A2 Genes from OI Donors
Sequencing PCR for the modified Sanger's enzymatic method was conducted according to the recommendation provided by the manufacturer of the ABI Prism 3130xl DNA Analyzer.In brief, 20 ng of each purified amplified PCR product was mixed with 5 pM of the appropriate sequencing primer and Big Dye Terminator v.3.1, as previously described [84,85].Before sequencing analysis, DNA fragments were purified employing the BigDye XTerminator Purification Kit (Applied Biosystems, Waltham, MA, USA) according to the manufacturer's manual.Bioinformatic analysis of the DNA sequences obtained was conducted by the Chromas Lite 2.01 and NCBI Nucleotide Blast software against the COL1A1 (NG_007400.1) and COL1A2 (NG_007405.1)genes as the reference sequences.

Fibroblast RNA Isolation
Primary skin fibroblasts and control cell line cultures were harvested with trypsin/EDTA (PAA Laboratories, Colbe, Germany) and washed twice with sterile phosphate buffered saline (PBS) (PAA Laboratories, Colbe, Germany).The cell amount was counted using Trypan Blue (BIO-RAD, Hercules, CA, USA) in the TC-20 Automated Cell Counter (BIO-RAD, Hercules, CA, USA).Cell lysis and purification of total RNA and miRNA were conducted using 2.5 × 106 cells in total.The isolation procedure was performed with the miRNeasy Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions.The purity and concentration of the collected RNA samples were determined using a Nanodrop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA).

Real-Time Quantitative Polymerase Chain Reaction
The miScript II RT Kit (Qiagen, Hilden, Germany) and an RNA concentration of 250 ng were used to synthesize cDNA.The obtained cDNA was used to examine the defined miRNA expression (Table 2).For this purpose, expression analysis of Custom miScript miRNA PCR Array, conf.no.CMIHS02722F (Qiagen, Hilden, Germany), was performed in a thermal cycler, the LightCycler 480 II (Roche, Basel, Switzerland).The quantitative polymerase chain reaction analysis was carried out in quadruplicate with miScript SYBR Green PCR Kit (Qiagen, Hilden, Germany) according to the manufacturer's manual.

Statistical Analyses
The expression analysis of miRNAs was performed by GenEx ver6 software (MultiD Analyses AB).Raw data were normalized to the reference genes (SNORD61, SNORD68, SNORD72, SNORD95, SNORD96A, and RNU6-2) according to the manufacturer's protocol.The Kolmogorov-Smirnov test was employed to determine the distribution of the data (Supplementary Table S1).For miRNA data that presented normal distribution, a parametric, unpaired, one-tailed t-test was conducted.In the case of miRNAs that showed non-normal distribution of data, a one-tailed Mann-Whitney test was used.The results were considered statistically significant at p < 0.05.The correlation analyses of miRNAs were performed using the Spearman method.Comparative expression of miRNAs in the study group and the control cell line was determined by scatter plot analysis.

Supplementary Materials:
The following supporting information can be downloaded at https: //www.mdpi.com/article/10.3390/ph16101414/s1: Figure S1: The relative expression of the tested miRNAs in subjects with mutations in the COL1A1 and COL1A2 genes and the control cells; Table S1

Author
Contributions: M.B. designed the experiments and contributed to data acquisition, manuscript writing, and preparation; A.A.-D.contributed to data acquisition, manuscript writing, and preparation; M.L., Ł.S., A.D.-K., J.W., M.A., and A.M.-T.contributed to data acquisition and interpretation; M.B.-M.and A.G. critically revised the manuscript; A.L.S. designed the experiments and drafted and revised the manuscript; K.G. designed and supervised the experiments and wrote and prepared the manuscript.All authors have read and agreed to the published version of the manuscript.

Table 1 .
Brief clinical summary and characterization of mutations detected in the COL1A1 and COL1A2 genes of OI donors enrolled in this study.

Table 3 .
Fold change in miRNA expression in vs. controls.

Change in Expression in OI vs. Control Cells
Negative values show upregulated miRs, and positive values show downregulated miRs in OI vs. control cells.*

Table 6 .
List of downregulated and overexpressed miRNAs analyzed individually for each OI donor.The table depicts the expression trends of miRNAs vs. controls.