FGFR1 frameshift mutation mediates the dysfunction of mesenchymal stem cells in craniofacial microsomia

Abstract

Craniofacial microsomia represents a congenital craniofacial anomaly characterized by a complex etiology, the precise genetic determinants of CFM have remained largely elusive. Comprehensive exome sequencing analysis performed on a cohort of CFM patients has identified a specific pathogenic frameshift mutation (c.225dupA, p. I76fs) located within the Fibroblast Growth Factor Receptor 1 gene. Extensive functional characterization utilizing human umbilical cord-derived mesenchymal stem cells demonstrated that either the presence of this FGFR1 mutation or the targeted knockdown of FGFR1 significantly attenuated the endogenous expression of the receptor. This genetic perturbation was associated with a marked impairment of osteogenic differentiation potential, a substantial reduction in cellular proliferative and migratory capacities, and an exacerbated apoptotic response. From a mechanistic perspective, the disruption of FGFR1 function exerted a suppressive effect on pivotal signaling cascades essential for skeletal development, including the P38/ERK, Wnt/β-catenin, BMP2/SMAD4, and PI3K/AKT transduction axes. Furthermore, the integrity of critical molecular interactions between FGFR1 and structural co-factors, specifically vimentin, COL1A1, and FGF1, was compromised, resulting in a concomitant downregulation of their protein abundance. Collectively, these findings elucidate that the FGFR1 frameshift mutation is a significant contributor to the pathogenesis of CFM, principally by obstructing MSC-mediated osteogenesis through the dual disruption of essential signal transduction pathways and vital protein–protein interaction networks.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s13287-026-05096-0.

Introduction

Craniofacial microsomia (CFM) comprises a spectrum of congenital malformations, including hemifacial microsomia and Goldenhar syndrome, originating from aberrant morphogenesis of the first and second pharyngeal arches [1]. This condition exhibits pronounced phenotypic heterogeneity, characterized by mandibular hypoplasia, microtia, cutaneous appendages, dermoid cysts, and commissural clefts [2]. Despite its significant prevalence, the precise etiology remains elusive. Prevailing hypotheses implicate disruptions in cranial neural crest cell ontogeny, Meckel’s cartilage formation, and craniofacial vascular dysgenesis [3]. Contemporary management of severe CFM mandates a multidisciplinary surgical strategy, utilizing costal cartilage grafting, distraction osteogenesis, and orthognathic surgery to rectify craniofacial asymmetry [4]. Consequently, elucidating the molecular pathophysiology is imperative for advancing preventive paradigms and precision therapeutics.

Mesenchymal stem cells (MSCs) are extensively investigated in craniofacial bone tissue engineering owing to their multilineage plasticity and inherent osteogenic, chondrogenic, and adipogenic differentiation potential [5]. Concurrently, fibroblast growth factor signaling, transduced via tyrosine kinase receptors, plays a paramount role in the hierarchical regulation of cellular behavior, governing proliferation, apoptosis, and differentiation to influence embryonic morphogenesis and tissue homeostasis [6]. Specific FGF constituents, including ligands FGF8 and FGF10 and receptors FGFR2 and FGFR3, are mechanistically implicated in the etiology of craniofacial anomalies [7–9], while the broader FGF cascade orchestrates neural crest cell migration and specification essential for craniofacial development [10]. At the molecular level, the transcriptional landscape of MSCs during lineage commitment exhibits significant FGF and FGFR1 upregulation, highlighting the pathway’s pivotal role in fate determination [11]. Mechanistically, FGFR1 activation induces proto-oncogene expression, suppressing cyclin-dependent kinase inhibitors p21 and p27 to facilitate cell cycle progression and MSC expansion [12]. Conversely, its inactivation and subsequent ERK attenuation impair osteogenic differentiation in bone marrow-derived MSCs [13]. Furthermore, FGFR1 modulates chondrogenesis by directing MSC differentiation into chondrocytes, often synergizing with the WNT-β-catenin signaling pathway [14].

As a vital downstream effector under the regulatory governance of Notch signaling, the FGF pathway is crucial for the maintenance of cellular phenotypes. Under physiological conditions where FGF signaling remains unimpeded, the resulting secondary NCSCs preserve a dose sensitivity virtually indistinguishable from that of the ancestral founder cells [15]. Furthermore, FGF-2 has been demonstrated to exert a salutary influence on the preservation of perilesional bone stock, concurrently fostering the regeneration of the periodontal ligament and cementum [16]. Within the specific microenvironment, FGF/FGFR signaling facilitates MAPK pathway activation, accelerating early skeletal ossification and development. This mechanism ameliorates the chondrocyte microenvironment and rectifies aberrant skeletal ossification processes [17]. Regarding craniofacial development, FGF/FGFR signaling, predominantly transduced via FGFR3, is indispensable for the normal morphogenesis of mandibular third molars. Pharmacological inhibition of this pathway results in microdontia, characterized by reduced tooth dimensions, and leads to dysmorphic abnormalities in both crown and root architecture [18]. Finally, in experimental models of temporomandibular joint degeneration, the FGF18/FGFR signaling pathway demonstrates significant therapeutic efficacy characterized by potent anti-catabolic and pro-anabolic activities, effectively orchestrating the repair and regeneration of fibrocartilaginous tissues [19].

In this study, we conducted comprehensive exome sequencing of peripheral blood from a cohort of patients diagnosed with CFM. Through rigorous genomic screening, we identified a novel frameshift mutation located within an exon of the FGFR1 gene. To our knowledge, this constitutes the inaugural documentation of an FGFR1 frameshift mutation in CFM, significantly expanding the genetic architecture of the disorder. Subsequently, we performed an in-depth functional characterization to elucidate the biological ramifications of this variant. Our experimental approach systematically evaluated its impact on critical MSC behaviors, specifically osteogenic differentiation, proliferative capacity, migratory potential, and apoptotic regulation, within a controlled in vitro environment. Furthermore, we aimed to dissect the mechanistic contributions of this mutation to CFM pathogenesis. Collectively, our findings posit that FGFR1 genetic variants play a determinative role in the developmental trajectory of CFM, offering novel insights into the molecular etiology of the condition and illuminating potential avenues for targeted therapeutic interventions.

Materials and methods

Sample collection and exome sequencing

The present study recruited a cohort of patients, aged 0–18 years, 56 patients, who received a clinical diagnosis of CFM at the Plastic Surgery Hospital, Chinese Academy of Medical Sciences, and Peking Union Medical College. Before their inclusion in the study, written informed consent was obtained from all participants or, in the case of minors, their designated legal guardians. To ensure the integrity of the genetic analysis, stringent exclusion criteria were implemented; specifically, individuals exhibiting karyotypic abnormalities or presenting with syndromic manifestations characterized by microtia and/or mandibular hypoplasia were disqualified from the study. Peripheral blood specimens were systematically collected from the enlisted subjects, and genomic DNA was isolated following standardized extraction protocols. For comprehensive genomic profiling, exome sequencing was performed utilizing the Agilent SureSelect Human All Exon capture platform. The study protocol was approved by the Medical Ethics Review Board of the Plastic Surgery Hospital, Chinese Academy of Medical Sciences (Reference Number: 2024 (341); Date of Approval: October 24, 2024).

Isolation of HucMSCs

Isolation and culture of HucMSCs

HucMSCs were isolated from fresh umbilical cords obtained from healthy mothers following childbirth. Written informed consent was obtained from all participants prior to sample collection, and the study protocol was approved by the Institutional Review Board of the Plastic Surgery Hospital, Chinese Academy of Medical Sciences, and Peking Union Medical College. Briefly, to remove residual blood, umbilical cord tissues were rinsed with phosphate-buffered saline (PBS) supplemented with streptomycin and penicillin. The cords were then dissected into 3–4 cm segments, and blood vessels were meticulously removed. The remaining tissues were minced into small pieces and cultured in α-MEM (12571063, Gibco) containing 100 μg/mL streptomycin, 100 U/mL penicillin (15140122, Invitrogen), and 10% fetal bovine serum (SH30406.05, Hyclone). The tissues were incubated in a humidified environment with 5% CO2 at 37 °C for 72 h. Following the emergence of fibroblast-like cell colonies, the cord tissue explants were removed. Cells were passaged upon reaching approximately 80% confluence. HucMSCs between passages 2 and 6 were utilized in subsequent experiments.

Osteogenic and adipogenic differentiation

To assess osteogenic differentiation, HucMSCs were incubated with osteogenic differentiation medium (PWL080, Meilunbio) for 7, 14, or 21 days. To validate adipogenic differentiation potential, HucMSCs were incubated with adipogenic differentiation medium (PWL081, Meilunbio) for 7 or 14 days.

Lentiviral packaging and stable cell line generation

Recombinant expression plasmids encoding the wild-type FGFR1 sequence, FGFR1 mutants, and three distinct short hairpin RNAs (designated shFGFR1-1, shFGFR1-2, and shFGFR1-3), alongside a corresponding negative control, were chemically synthesized by General Biol (Anhui, China) and Sangon Biotech (Shanghai, China). High-titer lentiviral particles were generated via an advanced second-generation packaging system to facilitate efficient gene delivery. HEK293T cells, in the logarithmic growth phase, were seeded into 10 cm culture dishes. Upon attaining 70–80% confluence, the cells underwent co-transfection with specific transfer vectors and auxiliary plasmids psPAX2 and pMD2.G, mediated by Lipofectamine 8000 (C0533, Beyotime) to ensure optimal efficiency. The medium was replaced 6 to 8 h post-transfection to alleviate cytotoxicity. Viral supernatants were subsequently harvested at 48 and 72 h, clarified through a 0.45 µm filter to remove cellular debris, and concentrated via ultracentrifugation at 50,000× g for 2 h at 4 °C. Viral titers were precisely quantified using RT-PCR analysis of genomic copy numbers, and aliquots were preserved at − 80 °C for long-term storage. The resultant lentiviral particles were utilized to transduce hUC-MSCs. Preliminary assays established the optimal multiplicity of infection. For formal experiments, target cells seeded in 6-well plates reached 30–40% confluence prior to transduction with varying viral titers, enhanced by 5–8 μg/mL Polybrene.

The culture medium was refreshed 24 h post-transduction. Stable transductants were selected using the vector-encoded antibiotic resistance marker Puromycin (P8230, Solarbio) following a 48–72 h infection period. The minimum lethal concentration was empirically determined to be 2 μg/mL; cells were subjected to continuous selection at this concentration for 7–10 days until complete mortality in the control group was observed. Surviving populations were expanded and harvested, with FGFR1 expression levels assessed via Western Blot and/or RT-qPCR to evaluate the knockdown efficiency of the mutants and shRNAs. The shRNA demonstrating the most profound knockdown efficacy and its respective stable cell line were selected for subsequent functional investigations.

Alizarin red S staining

To evaluate the osteogenic potential of differentiated hUC-MSCs, Alizarin Red S staining (G1450, Solarbio) was employed to visualize extracellular matrix mineralization. Cells were seeded into 24-well plates and, upon reaching 90% confluence, were fixed with 4% paraformaldehyde (or 95% ethanol) for 10 min. This was followed by rinsing with distilled water to remove residual fixative. Subsequently, the cells were incubated with 2% (w/v) Alizarin Red S solution at 37 °C for 30 min to stain calcium nodules. Finally, calcium deposits were visualized under a microscope, and the degree of mineralization was quantitatively analyzed using the acquired images.

Alkaline phosphatase activity

To evaluate the degree of osteoblast differentiation following the osteogenic induction of hUC-MSCs, alkaline phosphatase activity was assessed. Cells were lysed in RIPA lysis buffer (R0010, Solarbio) on ice for 30 min. The lysates were centrifuged at 13,000× g for 10 min at 4 °C to remove cellular debris, and the supernatant was collected. ALP activity was determined using a commercial assay kit according to the manufacturer’s instructions (B8360, Solarbio). Finally, absorbance was measured at 405 nm using a spectrophotometer.

Oil red O staining

The adipogenic differentiation potential of hUC-MSCs was assessed via Oil Red O staining. HucMSCs were seeded at a density of 1 × 10⁴ cells per well in 6-well plates and cultured in adipogenic induction medium. Upon reaching confluence, the cells were fixed with 4% paraformaldehyde for 20 min. Subsequently, the cells were washed with PBS and stained with freshly prepared 1% Oil Red O working solution (PWL081, Meilunbio) for 10 min to label neutral lipids. Finally, intracellular lipid droplets were observed and imaged using an optical microscope (Model XD-202, Jiangnan).

Cell counting kit-8 (CCK-8) assay

Cell viability was assessed using the CCK-8 assay. HucMSCs were seeded at a density of 4 × 10³ cells per well in 96-well plates. At 0, 24, 48, and 72 h, 10 μl of CCK-8 reagent (CA1210, Solarbio) was added to each well. The cells were then incubated for 2 h at 37 °C. Finally, the optical density (OD) was measured at 450 nm using a microplate reader (Eppendorf, Biospectrometer basic).

5-ethynyl-2′-deoxyuridine (EdU) assay

Cell proliferation was evaluated using the EdU assay (C0078, Beyotime). HucMSCs were seeded in 96-well plates and incubated with 100 μL of medium containing EdU for 2 h at 37 °C. The cells were then fixed with 4% paraformaldehyde and permeabilized with 0.5% Triton X-100 (IR9073, Solarbio). Subsequently, the incorporated EdU was stained with Apollo solution for 30 min in the dark. Nuclei were counterstained with DAPI for 10 min. Finally, fluorescent images were captured using a fluorescence microscope (Ti2, Nikon).

Wound healing assay

HucMSCs were seeded into 6-well plates at a density of 3 × 10⁵ cells per well. Upon reaching confluence, a linear scratch was created on the monolayer using a sterile pipette tip. After washing with PBS to remove debris, the cells were incubated for 24 h. Images were captured at 0 h and 24 h using an inverted optical microscope (model XD-202, Jiangnan). Finally, wound closure was quantified using ImageJ software.

Transwell assay

The migratory ability of HucMSCs was evaluated using 24-well Transwell chambers with 8 μm pores (Cat# 353097, FALCON). Cells in the logarithmic growth phase were harvested, washed with PBS, and resuspended in serum-free medium at a density of 2 × 10⁵ cells/ml. To establish a chemotactic gradient, 600 µl of complete medium with serum was added to the lower chamber, and 200 µl of the cell suspension was added to the upper chamber. After incubation for 18 h, the cells that had migrated to the lower surface of the membrane were fixed with 4% paraformaldehyde (P1110, Solarbio) and stained with 1% crystal violet solution (G1064, Solarbio). Non-migrated cells remaining in the upper chamber were removed. Images of the migrated cells were acquired using a digital microscope (DM3000 LED, Leica).

Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay

Apoptosis was evaluated using the TUNEL assay (C1090, Beyotime). HucMSCs grown in 6-well plates were fixed with 4% paraformaldehyde for 45 min and permeabilized with 0.3% Triton X-100 for 20 min. Subsequently, the cells were incubated with TUNEL reaction mixture at 37 °C for 45 min. Nuclei were counterstained with DAPI for 5 min. Finally, images were acquired using a fluorescence microscope (Ti2, Nikon).

Real-time polymerase chain reaction (RT-PCR)

Total RNA was isolated using TRIzol reagent (R1100, Solarbio) and reverse transcribed into cDNA using the Revert Aid First-strand cDNA Synthesis Kit (KR118, TIANGEN). qPCR was performed with FastStart Universal SYBR Premix ExTaqTM II (FP205, TIANGEN) on an ABI PRISM^® 7900HT Real-Time PCR System (7900HT, ABI). GAPDH served as an internal control to normalize variations, and relative mRNA expression levels were calculated using the 2-ΔΔCT method. The primer sequences are listed in Table 1.

Table 1.

Sequences of Primers used in RT‒qPCR

Gene	Forward primer (5′–3′)	Reversed primer (5′–3′)
FGFR1	CCCGTAGCTCCATATTGGACA	TTTGCCATTTTTCAACCAGCG
osteocalcin	CACTCCTCGCCCTATTGGC	CCCTCCTGCTTGGACACAAAG
ALP	ATAGCAGCCACAAACATTCGC	ACACACATGCCGAAGGTATTG
RunX2	CGGAATGCCTCTGCTGTTAT	TTCCCGAGGTCCATCTACTG
OPG	TGGCACCAAAGTAAACGCAGAG	CTCGAAGGTGAGGTTAGCATGTC
BMP2	ACTACCAGAAACGAGTGGGAA	GCATCTGTTCTCGGAAAACCT
LPL	TCATTCCCGGAGTAGCAGAGT	GGCCACAAGTTTTGGCACC
PPARγ	GGAAGACCACTCGCATTCCTT	GTAATCAGCAACCATTGGGTCA
GAPDH	GACGTGCCGCCTGGAGA	GAAGAGTGGGAGTTGCTGTTGAA

Western blot

Total cellular proteins were extracted using RIPA lysis buffer supplemented with protease and phosphatase inhibitor cocktails (P0013B, Beyotime). Proteins were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred onto polyvinylidene difluoride membranes (IPVH00010, Millipore). The membranes were blocked with 5% nonfat milk (D8340, Solarbio) and incubated with primary antibodies against FGFR1 (ab76464, Abcam); osteocalcin (ab133612, Abcam); ALP (ab229126, Abcam); RUNX2 (ab236639, Abcam); OPG (ab73400, Abcam); BMP2 (66383-1-Ig, Proteintech); LPL (ab172953, Abcam); PPARγ (ab178860, Abcam); ERK (ab184699, Abcam) and phospho-ERK (ab278538, Abcam); P38 (8690, CST) and phospho-P38 (9216, CST); phospho-β-catenin (9561, CST); GSK3-β (ab93926, Abcam) and phospho-GSK3-β (ab75814, Abcam); total β-catenin (ab32572, Abcam); and Frizzled7 (16974-1-AP, Proteintech). Following incubation with HRP-conjugated secondary antibodies, protein signals were visualized using an enhanced chemiluminescence assay kit (MA0186, Meilunbio).

Coimmunoprecipitation (CoIP)

Co-IP was performed using the Dynabeads™ Protein G Immunoprecipitation Kit (sc-2003, Santa Cruz) according to the manufacturer’s instructions. Antibodies against FGFR1 (ab76464, Abcam) and the Flag epitope (F1804, Sigma) were used for immunoprecipitation. The immunoprecipitates were washed to remove non-specific binding and denatured by boiling in Laemmli sample buffer for 10 min. Proteins were separated by SDS-PAGE and stained with Coomassie Brilliant Blue (P1305, Solarbio). Bands of interest were excised and analyzed by liquid chromatography-tandem mass spectrometry to identify potential FGFR1-interacting proteins. Finally, these interactions were confirmed by Western blot analysis.

Statistical analysis

All experiments were conducted in triplicate. Statistical analyses were performed using GraphPad 8.0 software. Data are presented as mean ± standard deviation (SD). Comparisons between two groups were performed using Student’s t-test, while differences among multiple groups were analyzed by one-way analysis of variance (ANOVA) or two-way Analysis of Variance followed by Tukey’s post hoc test. P value < 0.05 considered statistically significant.

Results

HucMSCs possess osteogenic and adipogenic differentiation potential

Peripheral blood samples were collected from 56 patients for comprehensive genetic lineage analysis. Notably, in four families, offspring showing clinical features of hemifacial microsomia were identified despite the phenotypically normal status of the parents. The segregation of the condition showed no sex bias. For detailed information on these subjects, please refer to the Table 2. Exonic results identified a novel frameshift mutation in exon 1 of FGFR1 (c.225dupA, p. I76fs), which is predicted to abrogate binding affinity for FGF1.Detailed sequencing parameters are provided in Additional file 3: Table S3. Sequence analysis confirmed that the A insertion at chr8:38285567 (Fig. 1A). Given their accessibility, minimal ethical concerns, and broad applicability, HUCB-MSCs were selected as a cellular model to investigate the pathogenesis of CFM.

Table 2.

Clinical and genetic characteristics of patients with craniofacial microsomia and their family members

Pedigree ID	Sample submission name	Sample name in final report	Original sample ID	Sex	Relationship	Affected status (please select from dropdown: P: affected; N: normal)	Library type	Contracted data volume	Skeletal malformation
F_2	Changming huang	H_2	FKDO210263634-1A	M	Son	P	WES	10G	S
F_2	Changming huang Mother	H_3	FKDO210263635-1A	F	Mother	N	WES	10G	N
F_2	Changming huang Father	H_4	FKDO210263636-1A	M	Father	N	WES	10G	N
F_3	Haoran Hong	H_5	FKDO210263637-1A	M	Son	P	WES	10G	N
F_3	Haoran Hong Mother	H_6	FKDO210263638-1A	F	Mother	N	WES	10G	N
F_3	Haoran Hong Father	H_7	FKDO210263639-1A	M	Father	N	WES	10G	N
F_5	Zi Meng Cui	H_10	FKDO210263642-1A	F	Daughter	P	WES	10G	S
F_5	Zi Meng Cui Mother	H_11	FKDO210263643-1A	F	Mother	N	WES	10G	N
F_7	Yang KXuan Lyu	H_14	FKDO210263646-1A	F	Daughter	P	WES	10G	S
F_7	Yang KXuan Lyu Mother	H_15	FKDO210263647-1A	F	Mother	N	WES	10G	N
F_8	Yansen Hao	H_16	FKDO210263648-1A	M	Son	P	WES	10G	N
F_8	Yansen Hao Mother	H_17	FKDO210263649-1A	F	Mother	N	WES	10G	N
F_8	Yansen Hao Father	H_18	FKDO210263650-1A	M	Father	N	WES	10G	N
F_9	Qingwen Li	H_19	FKDO210263651-1A	F	患者本人	P	WES	10G	S
F_11	Lincheng Wang	H_20	FKDO210263652-1A	M	Son	P	WES	10G	N
F_11	Lincheng Wang Mother	H_21	FKDO210263653-1A	F	Mother	N	WES	10G	N
F_11	Lincheng Wang Father	H_22	FKDO210263654-1A	M	Father	N	WES	10G	N
F_12	Qihan Wang	H_24	FKDO210263656-1A	M	Son	P	WES	10G	S
F_12	Qihan Wang Mother	H_25	FKDO210263657-1A	F	Mother	N	WES	10G	N
F_13	Muke Shi	H_26	FKDO210263658-1A	F	Daughter	P	WES	10G	N
F_13	Muke Shi Mother	H_27	FKDO210263659-1A	F	Mother	N	WES	10G	N
F_13	Muke Shi Father	H_28	FKDO210263660-1A	M	Father	N	WES	10G	N
F_16	Shenglei Yuan	H_33	FKDO210263665-1A	M	患者本人	P	WES	10G	S
F_17	Yike Wang	H_34	FKDO210263666-1A	F	Daughter	P	WES	10G	S
F_17	Yike Wang Father	H_36	FKDO210263668-1A	M	Father	N	WES	10G	N
F_18	Linxin Yan	H_37	FKDO210263669-1A	F	Daughter	P	WES	10G	M
F_18	Linxin Yan Mother	H_38	FKDO210263670-1A	F	Mother	N	WES	10G	N
F_18	Linxin Yan Father	H_39	FKDO210263671-1A	M	Father	N	WES	10G	N
F_19	Siyan Liu	H_40	FKDO210263672-1A	F	Daughter	P	WES	10G	N
F_19	Siyan Liu Mother	H_41	FKDO210263673-1A	F	Mother	N	WES	10G	N
F_19	Siyan Liu Father	H_42	FKDO210263674-1A	M	Father	N	WES	10G	N
F_20	Xiaomi Zhu	H_43	FKDO210263675-1A	F	Daughter	P	WES	10G	S
F_20	Xiaomi Zhu Mother	H_44	FKDO210263676-1A	F	Mother	N	WES	10G	N
F_21	Congjie Wang	H_45	FKDO210263677-1A	F	Daughter	P	WES	10G	S
F_21	Congjie Wang Mother	H_46	FKDO210263678-1A	F	Mother	N	WES	10G	N
F_21	Congjie Wang Father	H_47	FKDO210263679-1A	M	Father	N	WES	10G	N
F_23	Bo Rui Tan	H_49	FKDO210263681-1A	F	Daughter	P	WES	10G	M
F_23	Bo Rui Tan Mother	H_50	FKDO210263682-1A	F	Mother	N	WES	10G	N
F_23	Bo Rui Tan Father	H_51	FKDO210263683-1A	M	Father	N	WES	10G	N
F_24	Fengqing Liu	H_52	FKDO210263684-1A	M	Son	P	WES	10G	M
F_24	Fengqing Liu Mother	H_53	FKDO210263685-1A	F	Mother	N	WES	10G	N
F_24	Fengqing Liu Father	H_54	FKDO210263686-1A	M	Father	N	WES	10G	N
F_26	Huanhuan Wei	H_57	FKDO210263689-1A	F	Daughter	P	WES	10G	M
F_26	Huanhuan Wei Mother	H_58	FKDO210263690-1A	F	Mother	N	WES	10G	N
F_26	Huanhuan Wei Father	H_59	FKDO210263691-1A	M	Father	N	WES	10G	N
F_27	Zhibin Wang	H_60	FKDO210263692-1A	F	Daughter	P	WES	10G	M
F_27	Zhibin Wang Mother	H_61	FKDO210263693-1A	F	Mother	N	WES	10G	N
F_27	Zhibin Wang Father	H_62	FKDO210263694-1A	M	Father	N	WES	10G	N
F_28	Qiyun Du	H_63	FKDO210263695-1A	F	Daughter	P	WES	10G	S
F_28	Qiyun Du Mother	H_64	FKDO210263696-1A	F	Mother	N	WES	10G	N
F_29	Zibo Fan	H_65	FKDO210263697-1A	M	Son	P	WES	10G	S
F_29	Zibo Fan Mother	H_66	FKDO210263698-1A	F	Mother	N	WES	10G	N
F_31	Shibo Qian	H_69	FKDO210263701-1A	M	Son	P	WES	10G	N
F_31	Shibo Qian Mother	H_70	FKDO210263702-1A	F	Mother	N	WES	10G	N
F_31	Shibo Qian Father	H_71	FKDO210263703-1A	M	Father	N	WES	10G	N
F_33	Ziming Huang	H_74	FKDO210263706-1A	M	heself	P	WES	10G	S

Fig. 1.

Characterization of HucMSCs. A Partial Pedigrees of the Patients’ Families; exome sequencing revealed the (c.225dupA, p.I76fs) mutation in the exons of the FGFR1 gene; B, C Alizarin Red staining revealed HucMSC activity after 7, 14, and 21 days of treatment with both nonosteogenic and osteogenic induction media; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using two-way ANOVA, followed by Tukey’s multiple comparisons test. D ALP assay revealed ALP activity within HucMSC cells after 7, 14, and 21 days of treatment with nonosteogenic and osteogenic induction media; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using two-way ANOVA, followed by Tukey’s multiple comparisons test. E ALP assay measured ALP activity in the supernatant of HucMSC cells after 7, 14, and 21 days of treatment with nonosteogenic and osteogenic induction media; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using two-way ANOVA, followed by Tukey’s multiple comparisons test. F–J RT-PCR was used to assess the mRNA expression levels of osteocalcin, ALP, RunX2, OPG, and BMP2 in HucMSC cells after 7, 14, and 21 days of exposure to nonosteogenic and osteogenic induction media; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using two-way ANOVA, followed by Tukey’s multiple comparisons test. K–P Western blot analysis was used to determine the protein expression levels of osteocalcin, ALP, RunX2, OPG, and BMP2 in HucMSC cells after 7, 14, and 21 days of culture in nonosteogenic and osteogenic induction media; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using two-way ANOVA, followed by Tukey’s multiple comparisons test. Q–R Oil Red O staining was used to evaluate the lipid content in HucMSC. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using two-way ANOVA, followed by Tukey’s multiple comparisons test. S, T RT-PCR was used to assess the mRNA expression levels of LPL and PPARγ in HucMSC cells after 7 and 14 days; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using two-way ANOVA, followed by Tukey’s multiple comparisons test. U–W Western blot analysis was used to determine the protein expression levels of LPL and PPARγ in HucMSC cells after 7 and 14 days; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using two-way ANOVA, followed by Tukey’s multiple comparisons test

To induce osteogenic differentiation, HUCB-MSCs were cultured in osteogenic induction medium for 7, 14, or 21 days. Quantitative analysis showed a time-dependent increase in calcium deposition (Fig. 1B, C) as well as ALP activity in both cell lysates and culture supernatants (Fig. 1D, E). RT-qPCR revealed progressive upregulation of key osteogenic markers, including osteocalcin, RUNX2, and BMP2. Specifically, RUNX2 mRNA expression peaked at day 14 (Fig. 1F–J). Consistent with these transcriptional changes, Western blotting confirmed elevated protein levels of these markers; notably, no statistically significant difference in RUNX2 protein expression was observed between days 14 and 21 (Fig. 1K–P).

For adipogenic differentiation, HUCB-MSCs were cultured in induction medium for 7 or 14 days, which resulted in substantial accumulation of intracytoplasmic lipid droplets (Fig. 1Q–R). Concurrently, the expression levels of PPARγ and LPL were significantly upregulated at both the mRNA and protein levels (Fig. 1S–W). These findings confirm the multilineage differentiation potential of HUCB-MSCs.

FGFR1 mutation inhibits the osteogenic differentiation of HucMSCs

To investigate the functional consequences of the FGFR1 mutation, HUCB-MSCs were transduced with lentiviral vectors encoding either wild-type FGFR1 (FGFR1-WT) or the mutant FGFR1 variant. Compared with the empty vector control, FGFR1 mRNA levels were significantly increased in both transduction groups (p < 0.01); however, the transcriptional abundance in the FGFR1-mut group was approximately 40% lower than that in the FGFR1-WT group (p < 0.05). RT-qPCR analysis on day 7 of osteogenic induction revealed that while FGFR1-WT enhanced the expression of osteogenic markers such as osteocalcin, RUNX2, and BMP2, the FGFR1-mut exerted a suppressive effect, reducing levels by 2.1- to 3.8-fold compared to the FGFR1-WT group (Fig. 2A–F).

Fig. 2.

The FGFR1 mutation suppressed the osteogenic differentiation of HucMSCs. A–F RT-PCR analysis of FGFR1, osteocalcin, ALP, Runx2, OPG, and BMP2 mRNA expression levels in the empty vector group, FGFR1-WT group, and FGFR1-mutant group; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using One-way ANOVA followed by Tukey’s multiple comparisons test. G, H Alizarin Red staining to assess the density of calcium deposits in the empty vector group, FGFR1-WT group, and FGFR1-mutant group; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using One-way ANOVA followed by Tukey's multiple comparisons test. I, J ALP assay to measure the ALP content in induced differentiated cells and cell supernatants from the empty vector group, FGFR1-WT group, and FGFR1-mutant group; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using One-way ANOVA followed by Tukey’s multiple comparisons test. K–P Western blot analysis of osteocalcin, ALP, Runx2, OPG, and BMP2 protein expression levels in the empty vector group, FGFR1-WT group, and FGFR1-mutant group. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using One-way ANOVA followed by Tukey’s multiple comparisons test. Full-length blots/gels are presented in Supplementary File

Alizarin Red S staining quantification indicated that mineralized nodule formation in the FGFR1-mut group was 62% lower than in the FGFR1-WT group. Alizarin Red S staining quantification indicated that mineralized nodule formation in the FGFR1-mut group was 62% lower than in the FGFR1-WT group (Fig. 2G, H). Furthermore, while ALP activity was significantly elevated in both FGFR1-WT and FGFR1-mut groups relative to the vector control, the enzymatic activity in the FGFR1-mut group was significantly reduced compared to that in the FGFR1-WT group (Fig. 2I, J). Western blot analysis further confirmed decreased protein levels of osteogenic markers in the FGFR1-mut group (Fig. 2K–P), suggesting that this specific FGFR1 insertion mutation impairs the osteogenic differentiation process.

Knockdown of FGFR1 suppresses the osteogenic differentiation of HucMSCs

To investigate the role of FGFR1 in the osteogenic commitment of HucMSCs, we constructed three lentiviral vectors encoding short hairpin RNAs (shRNAs) targeting FGFR1 and transduced HucMSCs. Following screening, shFGFR1-1 and shFGFR1-2 demonstrated superior silencing efficacy and were selected for subsequent functional assays. Upon osteogenic induction, RT-qPCR analysis revealed that transcript levels of key osteogenic markers, including osteocalcin, ALP, RUNX2, OPG, and BMP2, were significantly diminished in both shFGFR1-1 and shFGFR1 -2 groups compared to the negative control group (Fig. 3A–F). This transcriptional downregulation indicates that FGFR1depletion exerts a significant suppressive effect on the osteogenic differentiation program.

Fig. 3.

FGFR1 knockdown inhibited the osteogenic differentiation of HucMSCs. A–F RT-PCR analysis of FGFR1, osteocalcin, ALP, Runx2, OPG, and BMP2 mRNA expression levels in the NC group, ShFGFR1-1 group, and ShFGFR1-2 group; *p < 0.05; * p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using One-way ANOVA followed by Tukey's multiple comparisons test. G, H Alizarin Red staining to assess the calcium deposit density in the NC group, ShFGFR1-1 group, and ShFGFR1-2 group; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using One-way ANOVA followed by Tukey’s multiple comparisons test. I, J ALP assay to measure the ALP content in induced differentiated cells and cell supernatants from the NC group, ShFGFR1-1 group, and ShFGFR1-2 group; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using One-way ANOVA followed by Tukey’s multiple comparisons test. K–Q Western blot analysis of osteocalcin, ALP, Runx2, OPG, and BMP2 protein expression levels in the NC group, ShFGFR1-1 group, and ShFGFR1-2 group. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using One-way ANOVA followed by Tukey’s multiple comparisons test. Full-length blots/gels are presented in Supplementary File

Consistent with the molecular data, Alizarin Red S staining after 21 days of induction revealed a substantial reduction in mineralized nodule density in the shFGFR1 groups compared to the NC cohort, indicating compromised extracellular matrix calcification (Fig. 3G, H). Furthermore, spectrophotometric analysis showed that ALP enzymatic activity was significantly lower in both cell lysates and culture supernatants from the shFGFR1-1 and shFGFR1-2 groups than in the shRNA-NC control (Fig. 3I, J). Concordantly, Western blot analysis confirmed a marked reduction in the protein levels of osteocalcin, RUNX2, OPG, and BMP2 in the FGFR1-knockdown groups (Fig. 3K–Q). Collectively, these findings underscore the pivotal role of FGFR1 signaling in promoting osteogenic differentiation of HucMSCs.

Mutation or knockdown of FGFR1 inhibits the proliferation and migration of HucMSCs while promoting apoptosis

To elucidate the comprehensive functional consequences of FGFR1 modulation on cellular behavior, we systematically evaluated the phenotypic effects of FGFR1 overexpression (wild-type versus mutant) and knockdown on HUCB-MSC proliferation, migratory dynamics, and apoptotic susceptibility. CCK-8 assays revealed that, relative to the empty vector control, the FGFR1-WT group exhibited significantly elevated OD values at 450 nm at 24, 48, and 72 h, indicative of a hyper-proliferative state. In marked contrast, the FGFR1-MUT group demonstrated markedly diminished OD values, suggesting a significant impairment in proliferative capacity (Fig. 4A). These observations were further corroborated by EdU incorporation assays (Fig. 4B, C), wherein the FGFR1-WT cohort displayed a significantly higher fraction of EdU-positive nuclei, whereas the FGFR1-MUT group exhibited a dramatic decline in the proliferative index.

Fig. 4.

FGFR1 mutation modulated the proliferation, migration, and apoptosis of HucMSCs. A CCK8 assay was used to determine the OD values at 450 nm in the empty vector group, FGFR1-WT group, and FGFR1-mutant group; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using two-way ANOVA followed by Tukey’s multiple comparisons test. B, C EDU staining was used to evaluate the cell proliferation rate in the empty vector group, FGFR1-WT group, and FGFR1-mutant group; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using One-way ANOVA followed by Tukey's multiple comparisons test. D, E Transwell staining was used to assess the cell migration rate in the empty vector group, FGFR1-WT group, and FGFR1-mutant group; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using One-way ANOVA followed by Tukey’s multiple comparisons test. F, G scratch assay was used to evaluate the cell migration rate in the empty vector group, FGFR1-WT group, and FGFR1-mutant group; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using One-way ANOVA followed by Tukey’s multiple comparisons test. H, I TUNEL assay was used to analyze the apoptosis rate in the empty vector group, FGFR1-WT group, and FGFR1-mutant group; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using One-way ANOVA followed by Tukey’s multiple comparisons test. J–T HucMSCs infected with FGFR1-WT or FGFR1-Mutant were treated with osteogenesis induction media for 21 days. The protein expression levels of p-ERK, p-P38, p-β-catenin, p-GSK3-β, Frizzled7, BMP2, p-SMAD4, PI3K and p-AKT were determined via western blotting. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using One-way ANOVA followed by Tukey’s multiple comparisons test. Full-length blots/gels are presented in Supplementary File

Transwell migration assays demonstrated that ectopic expression of FGFR1-WT significantly augmented the number of cells traversing the porous membrane within 24 h. Conversely, the FGFR1-MUT group showed a pronounced deficit in migratory potential (Fig. 4D, E). Similarly, wound healing assays revealed that FGFR1-WT expression accelerated gap closure, whereas the FGFR1-MUT group presented a substantial delay in this process (Fig. 4F, G). Concurrently, TUNEL assays indicated that FGFR1-WT exerted an anti-apoptotic effect compared to the vector control, whereas the FGFR1-MUT group displayed a marked elevation in the percentage of apoptotic cells (Fig. 4H, I).

To delineate the molecular mechanisms underpinning FGFR1-mediated osteogenic regulation, we performed a comprehensive analysis of key signaling cascades associated with osteogenesis. Western blot analysis revealed that overexpression of FGFR1-WT significantly increased the phosphorylation ratios (p-ERK/ERK, p-β-catenin/β-catenin, p-P38/P38, p-GSK-3β/GSK-3β, p-SMAD4/SMAD4, and p-AKT/AKT) and total protein levels of pivotal signaling mediators, including PI3K, Frizzled7, and BMP2, compared to the control. In marked contrast, the FGFR1-MUT group exhibited a widespread downregulation of these critical signaling molecules, suggesting that the frameshift mutation disrupts the activation of multiple pro-osteogenic pathways (Fig. 4J–T).

FGFR1 mutation impedes the activation of osteogenesis-associated pathways in HucMSCs

To provide further validation for the phenotypic consequences of FGFR1 modulation, loss-of-function experiments were conducted by silencing endogenous FGFR1 using specific short hairpin RNA constructs (designated shFGFR1-1 and shFGFR1-2). Both knockdown cohorts exhibited significantly attenuated OD values at 24, 48, and 72 h relative to the negative control group, thereby validating the suppression of cellular proliferative kinetics (Fig. 5A). Concordantly, EdU incorporation staining delineated a marked reduction in the fraction of EdU-positive cells within the FGFR1-knockdown groups (Fig. 5B, C). Furthermore, Transwell migration and wound healing assays collectively demonstrated that the ablation of FGFR1 significantly compromised the HUCB-MSC migratory potential (Fig. 5D–G). In parallel, TUNEL assays corroborated that FGFR1 knockdown elicited a pronounced increase in apoptotic rate (Fig. 5H, I). Synthesizing these data, we conclude that both genetic mutation and depletion of FGFR1 consistently disrupt HUCB-MSC proliferative and migratory capabilities while concomitantly potentiating apoptotic pathways.

Fig. 5.

FGFR1 mutation regulated osteogenesis-related signaling pathways in HucMSCs. A CCK8 assay was used to determine the OD values at 450 nm in the NC group, ShFGFR1-1 group, and ShFGFR1-2 group; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using Two-way ANOVA followed by Tukey’s multiple comparisons test. B, C EDU staining was used to evaluate the cell proliferation rate in the NC group, ShFGFR1-1 group, and ShFGFR1-2 group; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using One-way ANOVA followed by Tukey's multiple comparisons test. D, E Transwell staining was used to assess the cell migration rate in the NC group, ShFGFR1-1 group, and ShFGFR1-2 group; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using One-way ANOVA followed by Tukey’s multiple comparisons test. F, G Wound healing assay was performed to assess HucMSCs migration ratios in the NC group, ShFGFR1-1 group, and ShFGFR1-2 group; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using One-way ANOVA followed by Tukey’s multiple comparisons test. H, I TUNEL assay was performed to assess HucMSCs apoptosis ratios in the NC group, ShFGFR1-1 group, and ShFGFR1-2 group; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using One-way ANOVA followed by Tukey’s multiple comparisons test. J–T Western blot analysis of p-ERK, p-P38, p-β-catenin, p-GSK-3β, Frizzled7, BMP2, p-SMAD4, PI3K and p-AKT expression in the NC group, ShFGFR1-1 group, and ShFGFR1-2 group; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using One-way ANOVA followed by Tukey’s multiple comparisons test. Full-length blots/gels are presented in Supplementary File

At the molecular level, comparative analysis indicated that FGFR1 knockdown (shFGFR1-1 and shFGFR1-2) significantly suppressed the activation/levels of the aforementioned critical signaling components compared to the NC group (Fig. 5J–T). These mechanistic insights suggest that FGFR1 mutations or knockdown exert a negative regulatory effect on osteogenic differentiation, likely via the inhibition of key anabolic signaling cascades, specifically involving the ERK/P38/β-catenin/GSK-3β and SMAD4/BMP2/AKT/PI3K axes.

FGFR1 mutations suppress osteogenic differentiation in HucMSCs by downregulating vimentin and collagen type I alpha 1 chain (COL1A1)

To elucidate the downstream molecular interactome of FGFR1, we performed immunoprecipitation coupled with mass spectrometry to systematically profile protein–protein interaction networks, contrasting the effects between the FGFR1-WT and FGFR1-MUT groups. Through this comprehensive screening, Vimentin and COL1A1 were identified as pivotal binding partners for FGFR1 and were selected for in-depth validation. In comparison to the FGFR1-WT group, the FGFR1-MUTANT group exhibited significant alterations in the expression levels of various proteins. Notably, Heat shock protein 90 beta family member 1, L-lactate dehydrogenase A chain, Fructose-bisphosphate aldolase, and Immunoglobulin kappa constant demonstrated pronounced upregulation in the FGFR1-MUTANT cohort relative to the FGFR1-WT cohort. In contrast, Cytoskeleton-associated protein 4 and Elongation factor 1-gamma showed substantial downregulation within the FGFR1-MUTANT group when juxtaposed with the FGFR1-WT group (Fig. 6A, B). Follow-up immunoprecipitation assays confirmed the physical association between FGFR1-WT and Vimentin, COL1A1, and FGF1, whereas the presence of FGFR1 mutations was shown to disrupt these critical protein–protein interactions (Fig. 6C, D). Consistent with this, knockdown of FGFR1 significantly attenuated the protein levels of Vimentin, COL1A1, and FGF1 (Fig. 6E–I). Collectively, these findings underscore a mechanistic paradigm wherein FGFR1 mutations impede the osteogenic differentiation trajectory of HUCB-MSCs by dampening the expression of key mediators, including Vimentin, COL1A1, and FGF1, thereby disrupting associated osteogenic signaling.

Fig. 6.

FGFR1 mutation negatively regulates HucMSC osteogenesis via vimentin and COL1A1. A, B Bioinformatics analysis from Co-IP-MS assay. C, D HucMSCs were infected with FGFR1-WT or FGFR1-Mutant before osteogenic induction for 21 days. Vimentin and COL1A1 protein levels were measured via western blotting. The associations of FGFR1 with vimentin, COL1A1, or FGF1 were confirmed by CO-IP. E–I HucMSCs infected with ShFGFR1-1 or ShFGFR1-2 were incubated in osteogenic induction media for 21 days. Western blotting was used to measure FGFR1, vimentin, COL1A1, and FGF1 protein expression. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. n = 3 independent biological replicates. Data are presented as mean ± SD. Statistical analysis was performed using One-way ANOVA followed by Tukey's multiple comparisons test. Full-length blots/gels are presented in Supplementary File

Discussion

Craniofacial Microsomia represents a complex congenital craniofacial anomaly with a genetic etiology that remains incompletely understood [20]. Utilizing high-precision exome sequencing, this study identified a pathogenic frameshift mutation (c.225dupA, p. I76fs) in the FGFR1 exon. Functional assays demonstrated that this variant markedly suppresses FGFR1 expression, concomitantly attenuating osteogenic differentiation, proliferation, and migration of human umbilical cord-derived mesenchymal stem cells while activating pro-apoptotic pathways. Mechanistically, the mutation perturbs a sophisticated network of osteogenic signaling cascades, encompassing the ERK/P38, Wnt/β-catenin, BMP/SMAD, and PI3K/AKT axes. Moreover, significant downregulation of key structural proteins, Vimentin and COL1A1, further compromises the osteogenic commitment of HUCB-MSCs.

The FGF signaling axis is critical for craniofacial morphogenesis and skeletogenesis. FGFR family proteins exhibit a conserved architecture comprising an extracellular ligand-binding region with three immunoglobulin-like (Ig) domains, a single transmembrane helix, and an intracellular split tyrosine kinase domain. High-affinity FGF binding is localized to the D2 and D3 domains and the intervening linker. The D1–D2 linker contains an acidic motif (“acid box”), while a conserved positively charged site within D2 facilitates heparan sulfate proteoglycan binding. The D3 domain, encoded by exon IIIa and alternatively spliced exon IIIb or IIIc, dictates receptor isoform-specific ligand affinity [21]. Pathogenic FGFR1 variants are associated with diverse pathologies, including dysembryoplastic neuroepithelial tumors, and are validated via fluorescence in situ hybridization, copy number variation analysis, and Sanger sequencing [22]. While missense mutations contribute to craniofacial dysostosis [23], the mechanistic role of FGFR1 in CFM pathogenesis remains elusive. Herein, we identified a frameshift mutation (c.225dupA, p. I76fs) via exome sequencing, providing novel evidence for the genetic etiology of CFM. Consequently, we advocate for the integration of FGFR1 screening into standard diagnostic protocols for hemifacial microsomia to elucidate the mutational spectrum and optimize clinical diagnostic precision.

FGF18-induced FGFR1 signaling facilitates MSC osteogenic differentiation via the ERK1/2-MAPK and PI3K transduction axes [24]. Recognized as a central regulatory node, FGFR1 governs the osteogenic commitment of periodontal ligament mesenchymal stem cells [25]. Concordantly, our data indicate that the identified FGFR1 variant precipitates a significant downregulation of endogenous FGFR1, thereby inhibiting HucMSC osteogenesis. This functional deficit is evidenced by diminished calcium deposition, attenuated alkaline phosphatase activity, and suppressed transcription of pivotal osteogenic markers, including osteocalcin, RUNX2, OPG, and BMP2. Furthermore, the variant compromised HucMSC proliferative and migratory capacities while potentiating apoptotic pathways. Lentiviral shRNA-mediated FGFR1 silencing recapitulated these pathological phenotypes, confirming the deleterious impact of the mutation. Collectively, these findings demonstrate that the FGFR1 variant attenuates receptor levels and profoundly disrupts fundamental MSC biological behaviors, including osteogenic commitment, proliferation, migration, and apoptosis regulation.

A substantial body of evidence highlights the critical role of complex signaling networks in regulating mesenchymal stem cell osteogenic differentiation. Inhibition of the P38/ERK MAPK axis abrogates osteogenesis in dental pulp-derived MSCs [26]; while Isorhynchophylline enhances bone marrow-derived MSC osteogenesis via P38/ERK and autophagy pathways [27]. Moreover, Lysosome-associated membrane protein type 2 functions as a critical positive modulator of BMSC osteogenic differentiation, exerting its influence via the Wnt/β-catenin/GSK3β signaling axis [28], wherein GSK-3β and β-catenin serve as pivotal stimuli  [29, 30]. Conversely, the RNA-binding protein QKI suppresses osteogenesis by downregulating Wnt5b, Frizzled7, and β-catenin [31]. Additionally, the BMP2/SMAD4 cascade facilitates osteogenic induction in adipose-derived MSCs [32, 33]. Similarly, Exendin-4 and eldecalcitol synergistically enhance BMSC osteogenesis via PI3K/AKT activation [34], and Glial cell line-derived neurotrophic factor amplifies osteogenic proliferation and differentiation in jaw bone marrow-derived MSCs through PI3K/AKT signaling [35]. Within this regulatory context, the FGFR1 variant significantly downregulates p-ERK, p-P38, p-β-catenin, and Frizzled7, while upregulating pGSK-3β. Concurrently, it reduces levels of BMP2, p-SMAD4, p-AKT, and PI3K. Notably, targeted FGFR1 knockdown phenocopied these mutational effects, implicating the coordinated dysregulation of the P38/ERK, Wnt/β-catenin, BMP2/SMAD4, and PI3K/AKT pathways as the fundamental mechanism underlying FGFR1-mediated inhibition of MSC osteogenic differentiation.

Vimentin, a type III intermediate filament protein constitutively expressed in mesenchymal populations, serves as an integral cytoskeletal scaffold, conferring the mechanical resilience essential for withstanding tensile and shear stress. Concurrently, actin stress fibers facilitate mechanical signal transduction, enhancing cellular rigidity and promoting osteogenic commitment [36, 37]. While fibroblast growth factor 9 and FGFR1 are implicated in neoplastic progression [38]. The present investigation reveals that the molecular interplay between FGFR1 and its partners, vimentin, COL1A1, and FGF1, is significantly compromised by the FGFR1 mutation. Quantitative analyses demonstrate substantial downregulation of vimentin and COL1A1 following the introduction of the FGFR1 variant or targeted FGFR1 knockdown. We postulate that the concomitant depletion of vimentin and COL1A1, acting synergistically with the suppression of osteogenic differentiation, constitutes a pivotal pathogenic determinant in craniofacial malformations. The FGFR1 c.225dupA frameshift mutation, located within the extracellular domain, introduces a structural aberration that perturbs the canonical FGF1 binding interface, as predicted by in silico modeling. Physiologically, FGFR1 engages FGF1 to initiate receptor dimerization, a prerequisite for activating Ras/MAPK and PI3K/AKT transduction. Conversely, the mutant FGFR1 fails to efficiently assemble a stable ternary complex with FGF1 and essential co-factors, including vimentin and COL1A1. This attenuation of FGFR1 signaling remodels the osteogenic microenvironment; given that COL1A1 and vimentin function as critical auxiliary elements potentiating FGF/FGFR1 signal transduction, their diminished expression further compromises cellular sensitivity to FGF1 and downstream signaling efficiency. Collectively, these data indicate that the mutation undermines osteogenic signal propagation through structural perturbation and functional blockade.

In the present investigation, experimental inquiries were predominantly executed utilizing in vitro cultured HucMSCs. Nevertheless, it is acknowledged that the in vitro microenvironment is intrinsically limited in its capacity to fully recapitulate the intricate and multifaceted in vivo physiological milieu characteristic of patients afflicted with CFM. To bridge this translational gap, prospective research endeavors are dedicated to the establishment of transgenic murine models harboring the specific FGFR1 c.225dupA mutation, or alternatively, the deployment of CRISPR/Cas9-mediated genome editing to engineer animal models harboring precise point mutations. Through the execution of comprehensive in vivo investigations, we intend to rigorously evaluate the specific repercussions of this genetic aberration on craniofacial skeletal morphogenesis. Furthermore, these studies will validate the in vivo modulation of key signaling cascades, including the ERK/P38/β-catenin/GSK-3β and SMAD4/BMP2/AKT/PI3K pathways, whose altered activities were identified in the current study. These concerted efforts are designed to elucidate the complex pathogenesis of CFM with significantly enhanced precision and resolution.

Beyond genetic validation, the present study demonstrates that the FGFR1 mutation significantly disrupts the molecular interaction between FGFR1 and its core binding partners: vimentin, COL1A1, and FGF1. The present study still has certain limitations. Future work will prioritize resolving the high-resolution three-dimensional structure of this protein complex via X-ray crystallography or cryo-electron microscopy, to delineate at the atomic level how the FGFR1 frameshift mutation alters receptor conformation and disrupts the specific architecture mediating protein–protein interactions. While clinical translation requires extensive further validation, this work establishes a theoretical foundation centered on the haploinsufficiency pathogenic mechanism of the FGFR1 mutation. Based on this framework, we hypothesize that residual signaling from functional FGFR1 receptors could be augmented via targeted administration of exogenous FGF ligands or pharmacologically optimized small-molecule agonists. Alternatively, during the critical window of early embryonic development, the CRISPR/Cas9 gene editing system may enable precise excision of the duplicated adenine residue to restore the canonical open reading frame. In current clinical practice, preimplantation genetic testing for monogenic disorders integrated with in vitro fertilization remains the only mature, clinically validated modality to prevent vertical transmission of this pathogenic mutation.

In summary, our study identified a frameshift mutation of FGFR1 in CFM patients and demonstrated that the variant FGFR1 was involved in mediating the dysfunction of MSCs. This study suggested that the frameshift mutation in the FGFR1 gene might be a cause of CFM.

Abbreviations

CFM

Craniofacial microsomia

FGFR1

Fibroblast growth factor receptor 1

MSCs

Mesenchymal stem cells

HucMSCs

Human umbilical cord-derived mesenchymal stem cells

FGF

Fibroblast growth factor

ERK

Extracellular signal-regulated kinases

GSK3-β

Glycogen synthase kinase 3 beta

LPL

Lipoprotein lipase

PPARγ

Peroxisome proliferator-activated receptor gamma

LAMP2A

Lysosomal-associated membrane protein 2A

BMP2/SMAD4

Bone morphogenetic protein 2/SMAD family member 4

PI3K/AKT

Phosphatidylinositol 3-kinase/protein kinase B

CCK-8

Cell counting kit-8

EdU

5-Ethynyl-2′-deoxyuridine

TUNEL

Terminal deoxynucleotidyl transferase dUTP nick end labeling

RT-qPCR

Reverse transcription quantitative polymerase chain reaction

Co-IP

Coimmunoprecipitation

IP-MS

Immunoprecipitation mass spectrometry

GAPDH

Glyceraldehyde 3-phosphate dehydrogenase

Author contributions

S.B.G.Z.: conceptualization, formal analysis, writing—original draft; P.L. and L.L.: data curation, methodology; C.K. and Z.F.L.: investigation, validation; B.Y.L.: resources; L.K.M.: visualization; T.Y.Z.: project administration; H.W.L.: methodology; X.X.: methodology; T.H.: methodology; B.C.W. and S.B.G.Z.: writing—review and editing; X.J.T. and S.B.G.Z.: conceptualization, writing-review and editing.

Funding

The present investigation was underwritten by the CAMS Initiative for Innovative Medicine (CAMS-I2M) (2017-I2M-3-006). The funding entity remained uninvolved in the conceptualization of the study, the collation of data, the conduct of analysis and interpretation, the drafting of the report, or the determination to forward the manuscript for publication.

Data availability

All the original data images are included in the Additional files 1 and 2, the images underlying this article will be shared on reasonable request to the corresponding author.

Declarations

Ethics approval and consent to participate

The ethical title of this article is ‘FGFR1 Frameshift Mutation Mediates Dysfunction of Mesenchymal Stem Cells in Craniofacial Microsomia.” This blood samples and the umbilical cord were approved by the Medical Ethics Review Board of the Plastic Surgery Hospital, Chinese Academy of Medical Sciences (Serial number of auditing: 2024 (341), Auditing time: Oct/24/2024). All participants patients provided informed consent, and their relatives also signed the consent form in accordance with the hospital’s requirements prior to participation. Written informed consent was obtained from all participants (patients) or their guardians/legally authorized representatives prior to the collection and use of umbilical cord samples, in accordance with the ethical principles and requirements of the hospital. All procedures were conducted in compliance with the approved protocol and ethical guidelines. Informed consent was obtained from all individual participants included in the study. Statement about AI: The authors declare that they have not use AI-generated work in this manuscript.

Competing interests

The authors declare no competing interests.