An orthogonal CRISPR/Cpf1 platform for precise spatiotemporal gene regulation and osteoporotic fracture repair

Summary

CRISPR-Cas systems enable powerful gene editing and regulation, yet single-modality control often fails to achieve orthogonal, spatiotemporally precise regulation of multiple endogenous genes. We engineered OREC, an orthogonal platform integrating chemogenetic and optogenetic modalities for precise, reversible, multiplex gene control. OREC comprises two components: OREC^C regulated by doxycycline (Dox) and OREC^o controlled by light. By assembling catalytically dead Cpf1 (dCpf1), gene regulatory elements, and crRNA arrays on single transcripts, OREC enables robust simultaneous manipulation of multiple genes. We demonstrated OREC’s therapeutic potential in vitro for osteoblast function modulation and in vivo for osteoporotic fracture repair. OREC effectively activated Bmp2 while inhibiting Dkk1, significantly enhancing bone formation and fracture healing in mouse models. These results establish OREC as a versatile platform for precise multiplex gene regulation, offering significant advancement for CRISPR-based gene therapy applications in complex tissues where coordinated control of multiple therapeutic targets is essential.

Graphical abstract

Highlights

• •Orthogonal CRISPR/Cpf1 platform integrates chemogenetic and optogenetic gene control
• •OREC enables multiplex spatiotemporal gene regulation with high precision
• •OREC enhances osteoporotic fracture repair through dual regulation of Bmp2 and Dkk1

Motivation

Complex biological processes and therapeutic applications require precise spatiotemporal control of multiple genes simultaneously. However, existing CRISPR systems predominantly rely on single-modality regulation, limiting their capacity for orthogonal, multiplexed gene manipulation. This constraint is particularly evident in treating osteoporotic fractures, where optimal therapeutic outcomes demand coordinated activation of osteogenic factors while suppressing inhibitors. To address this challenge, we developed OREC, an orthogonal dual-modality platform integrating chemogenetic and optogenetic control. This system enables independent, reversible regulation of multiple endogenous genes with spatiotemporal precision, providing a versatile tool for CRISPR-based gene therapy applications.

Zhao et al. developed OREC, an orthogonal CRISPR/Cpf1 platform combining chemical and light control for simultaneous regulation of multiple genes with spatiotemporal precision. The system successfully enhanced osteoporotic fracture repair in mice by activating bone formation while suppressing inhibitory signals, demonstrating therapeutic potential for complex gene therapy applications.

Introduction

CRISPR-associated proteins are natural genome editing tools used by prokaryotic adaptive immune systems.^1,2 These programmable RNA-guided DNA endonucleases have become fundamental in various genetic research applications, including gene knockout, knockin, transcriptional regulation, epigenetic programming, chromatin imaging, and DNA-protein interaction studies.^3,4,5With ongoing research, more CRISPR nucleases are being identified.⁶ Notably, the CRISPR/Cpf1 system has broadened the scope of genome editing.^7,8,9,10 Cpf1 exhibits both DNA cleavage endonuclease activity and RNase activity, enabling it to process pre-crRNA into mature crRNA independently. This allows for multiple genome edits using a single, compact construct, facilitating efficient multiplex gene regulation.⁷ Despite its potential, clinical applications of CRISPR/Cpf1 face challenges in delivery, specificity, and control of editor exposure in vivo. A key safety consideration is uncontrolled or prolonged nuclease activity after administration, which can increase off-target risk and immunogenicity.¹¹ Therefore, it is crucial to develop CRISPR platforms that provide tightly inducible control with spatial and temporal precision, preventing uncontrolled exposure and enabling rigorous interrogation of complex biology.

Researchers have developed various inducible systems for precise genome editing, among which chemogenetic and optogenetic methods are most widely used.^{12,13,14,15,16,17} Chemogenetic models, such as the Tet-On/Off system derived from the bacterial Tn10 operon, provide temporally specific gene activation or repression.^18,19 Optogenetic approaches offer spatially precise control by incorporating light-responsive elements into biomolecules, enabling rapid and localized regulation of CRISPR activity.^13,20,21 However, neither method alone sufficiently addresses the complexity of precise multiplex gene regulation. By combining temporal and spatial control modalities, this dual-input strategy overcomes the limitations of single-modality systems while avoiding the pharmacokinetic interactions and systemic exposure challenges associated with dual small-molecule approaches. This integrated approach enables advanced spatiotemporal regulation of multiple genomic loci through CRISPR-based systems for in-depth study of complex biological processes.²²

Orthogonal CRISPR-based gene regulation systems hold promise for therapeutic applications, especially in regenerative medicine. Facilitating fracture repair has long been a challenging task in clinical practice.^{23,24,25,26,27} While recombinant bone morphogenetic protein 2 (BMP2) is clinically approved for specific fracture healing applications, its use is limited by safety concerns and restriction to select indications. Current pharmacological approaches remain inadequate for the broad spectrum of fracture types, particularly osteoporotic fractures.^28,29 However, CRISPR-based orthogonal gene regulation technology sheds new light on enhancing fracture repair. It has been demonstrated that BMP2 acts as a powerful osteogenic factor essential for bone formation, whereas DKK1 inhibits osteogenesis by negatively regulating osteoblast differentiation through the Wnt signaling pathway.^30,31,32 By integrating chemogenetic and optogenetic methods into a unified, orthogonal CRISPR system, precise multiplex regulation of these critical osteogenic genes can be achieved. This integrated strategy allows simultaneous activation of osteogenic factors like BMP2 and inhibition of suppressors such as DKK1, significantly advancing our ability to therapeutically address complex osteoporotic fractures.

In this study, we developed an orthogonal CRISPR/Cpf1 platform integrating chemogenetic and optogenetic strategies to achieve precise, multiplexed genome editing. By combining chemical induction and light-responsive modules, our system enables highly specific spatiotemporal control of genome modifications. Through rational design, we assembled a compact transcript encoding dCpf1, regulatory elements, a crRNA array, and Triplex tertiary structural motifs. Triplex refers to 110-nt structural elements derived from the 3′ end of the mouse non-coding RNA Malat1 that stabilize transcripts lacking poly(A) tails through tertiary structure formation, thereby enabling simultaneous expression of protein and crRNA from the same transcript.³³ This design minimizes the genetic payload while maintaining functional integrity, facilitating efficient packaging and delivery. This orthogonal system achieved precise regulation of critical endogenous genes, including BMP2 and DKK1, both of which are essential modulators of skeletal development and bone remodeling, thus demonstrating robust control over osteogenic pathways and osteoporotic fracture healing. Our findings highlight the potential of this integrated CRISPR/Cpf1 approach to precisely manipulate complex gene networks, significantly advancing therapeutic strategies in regenerative medicine and beyond.

Results

Engineering of an orthogonally regulatable CRISPR/Cpf1 system

We have developed an OREC system integrating chemogenetic (OREC^c) and optogenetic (OREC^o) methods (Figures 1A and 1B). In this system, the expression of CRISPR-Cas effectors is tightly controlled and inducible by Dox or light stimulation, allowing precise spatiotemporal gene editing and regulation. To validate this inducible system, HEK293T cells were transfected with luciferase reporter constructs under OREC-responsive promoters. Upon Dox treatment, luciferase expression significantly increased (Figure 1C), driven by rTetR binding to TRE elements, facilitating activator recruitment to the miniCMV promoter. Similarly, luciferase activity robustly increased upon exposure to light (Figure 1D), mediated by the blue-light-induced interaction between CIBN and CRY2,³⁴ which recruits transcriptional activators to the promoter region. Importantly, immunofluorescence analysis confirmed that CRISPR-Cas effectors can be effectively induced by either Dox or light (Figures S1A–S1D), demonstrating the versatility and precise control afforded by the OREC system.

Figure 1.

Engineering and validation of the orthogonally regulatable CRISPR/Cpf1 (OREC) system

(A and B) Schematic overview illustrating the design and regulatory mechanisms of the OREC system, incorporating chemogenetic (OREC^c) and optogenetic (OREC^o) components. DR, direct repeat sequence; TRE, tetracycline response elements.

(C and D) Luciferase reporter assays validating inducible gene expression via chemical (C) or light stimulation (D) in HEK293T cells. Data represent mean ± SEM (n = 3).

(E) Representative fluorescence microscopy images confirming successful gene knockin of GFP and mCherry via HDR following OREC activation by doxycycline or light. Scale bars, 100 μm.

(F) Quantitative analysis of knockin efficiencies for GFP and mCherry genes mediated by OREC systems. Data shown as mean ± SEM (n = 3) (∗∗∗∗p < 0.0001).

(G and H) Sequencing analysis confirming precise CRISPR-mediated knockout of the endogenous CCR5 locus. Representative sequences demonstrating targeted insertion and deletions.

See also Figure S1.

Next, we investigated the gene editing capability of the OREC system by performing knockin experiments of GFP and mCherry reporter genes into the endogenous GAPDH locus in HEK293T cells via homology-directed repair (HDR) (Figures S1E–S1G). Specifically, HEK293T cells were co-transfected with the OREC system constructs along with a plasmid dsDNA donor carrying 5′/3′ homology arms flanking a recoded GAPDH exon 9 linked in-frame to P2A-GFP or P2A-mCherry. Following an essential-gene targeting strategy, the nuclease cleaves within the GAPDH exon and precise HDR both reconstitutes GAPDH function and integrates the reporter cargo, while cells with non-productive indels are naturally depleted, enriching for successful knockin events.³⁵ The results showed significant expression of GFP and mCherry proteins after adding Dox or light in both OREC^c and OREC^o systems (Figure 1E). Quantification indicated knockin efficiencies of approximately 16.27% for GFP and 13.73% for mCherry, demonstrating effective gene knockin by OREC systems (Figure 1F). Furthermore, we assessed the gene knockout capability of the OREC system by targeting the endogenous gene CCR5 (Figures S1H–S1J). After transfection of HEK293T cells with the OREC system, monoclonal cells were isolated and analyzed by Sanger sequencing, confirming precise knockout or insertion mutations at the targeted CCR5 locus (Figures 1G and 1H). To quantify editing efficiency at the population level, we performed TIDE (Tracking of Indels by Decomposition) analysis on bulk cell cultures, revealing CCR5 editing frequencies of approximately 30.79% for the OREC^c and 27.26% for the OREC^o (Figures S1K and S1L). To assess potential off-target effects, we used CRISPR to predict the top candidate off-target sites for the CCR5 guide and quantified indels at CCR2, CCDC15, and TRPM1 loci by TIDE analysis in bulk populations under both doxycycline and light induction. Importantly, no significant indels were detected at any of these predicted off-target sites (Figures S1M and S1N), indicating high specificity of the OREC system. These results collectively demonstrate the robust and precise gene editing capability of the orthogonally inducible OREC system.

Multiplex transcriptional activation and repression using the OREC system

To develop methods for controlling multiple endogenous gene transcription, we optimized OREC systems. First, we used catalytically dead AsCpf1 (dAsCpf1), which lacks DNA cleavage activity but retains pre-crRNA processing activity. The transcription activation domain sp65p₃-HSF1 (Active) was then fused to the C-terminus of dAsCpf1 to enable gene transcription activation.³³ Additionally, we constructed a CRISPR array targeting the ASCL1, IL1RN, and IL1B genes, which were selected from loci previously validated for Cpf1-based regulation with strong dynamic ranges, to enable multiplex gene regulation.^33,36 To co-express dAsCpf1 and crRNAs from a single transcript, a triplex structural motif was inserted between the dAsCpf1 coding sequence and the CRISPR array (Figures 2A and 2B). To evaluate the multiplex transcriptional activation capability of the optimized OREC systems under Dox or light conditions, we transfected them into HEK293T cells. Upon induction with Dox or light, an approximate 20- to 30-fold increase in the mRNA levels of ASCL1, IL1RN, and IL1B was observed (Figures 2C and 2D). To confirm target specificity and rule out cross-regulatory effects, we tested single-target constructs expressing individual crRNAs targeting ASCL1. The results demonstrated that ASCL1 activation occurred without affecting IL1RN or IL1B expression levels (Figure S2A), confirming direct regulation without cross-interference. This demonstrates that OREC system can effectively and simultaneously activate the expression of multiple endogenous genes with high specificity. We then explored the potential of the OREC systems for multi-transcriptional repression. We selected dLbCpf1 instead of dAsCpf1 because Cpf1 proteins from different species recognize distinct crRNA scaffolds, thereby preventing mutual interference and allowing for orthogonal regulation of targeted genes. A transcriptional repression domain KRAB₃ (Rep) was fused to the C-terminus of dLbCpf1.³³ Finally, a CRISPR array targeting the RAB7A, RAB9A, and FZD1 genes, which were selected based on previously validated Cpf1 repression targets,^33,36 was constructed to assess the system’s multi-transcriptional repression capability (Figures 2E and 2F). Our results showed that, upon induction with Dox or light, the mRNA levels of RAB7A, RAB9A, and FZD1 significantly decreased by 70%–85% (Figures 2G and 2H). Similarly, single-target validation using individual crRNAs targeting RAB9A confirmed specific repression without affecting RAB7A or FZD1 expression (Figure S2B), demonstrating target specificity in the repression system.

Figure 2.

Multiplex transcriptional activation and repression by the OREC system

(A and B) Schematic illustration of OREC constructs enabling multiplex transcriptional activation of endogenous genes (ASCL1, IL1RN, and IL1B) via dAsCpf1 fused to activation domains sp65p₃-HSF1, referred to as “Active.”

(C and D) RT-qPCR analysis demonstrating simultaneous activation of ASCL1, IL1RN, and IL1B gene transcription at 72 h following induction by Dox (C) or light (D). Data shown as mean ± SEM (n = 3) (∗∗∗∗p < 0.0001).

(E and F) Diagram depicting multiplex transcriptional repression mediated by dLbCpf1 fused with transcriptional repression domains KRAB₃, referred to as “Rep,” targeting RAB7A, RAB9A, and FZD1 genes.

(G and H) RT-qPCR analysis showing robust and simultaneous transcriptional repression of RAB7A, RAB9A, and FZD1 at 72 h after induction with Dox (G) or light (H). Data represent mean ± SEM (n = 3) (∗∗∗∗p < 0.0001).

See also Figure S2.

To contextualize these regulatory effects, we analyzed the expression levels of target genes relative to GAPDH. The genes targeted for activation (ASCL1, IL1RN, and IL1B) showed basal expression levels approximately 1000-fold lower than GAPDH. Upon activation, these genes were substantially upregulated while still remaining well below housekeeping gene levels (Figure S2C). In contrast, the genes targeted for repression (RAB7A, RAB9A, and FZD1) exhibited basal expression levels ranging from 2% to 8% of GAPDH expression, which were effectively suppressed upon Dox-induced repression (Figure S2D). These analyses demonstrate that our orthogonal CRISPR/Cpf1 platform can effectively regulate genes across a broad range of basal expression levels. Collectively, these findings confirm that the OREC systems enable significant multiplex transcriptional activation or repression, thereby providing a robust platform for orthogonal gene regulation.

Orthogonal transcriptional control and gene editing

To evaluate the efficiency of orthogonal transcriptional control, the OREC system was used for sequential activation of ASCL1 and repression of RAB7A, both previously validated CRISPR targets that enable demonstration of independent regulatory pathways without cross-interference (Figure 3A).^33,36 HEK293T cells transfected with the OREC system were subjected to three conditions: Dox alone from 0 to 24 h, both Dox and light from 24 to 36 h, and light alone from 36 to 48 h. To evaluate post-termination effects and assess the reversibility of gene regulation, we continued data collection through 72 h. As shown in the time course analysis (Figure 3B), during the first 24 h, RAB7A expression decreased in a time-dependent manner, while ASCL1 remained constant. This indicated that RAB7A is repressed by Dox-induced OREC^c system, whereas ASCL1 is unaffected. Between 24 and 36 h, RAB7A expression continued to decline, while ASCL1 mRNA levels began to rise. This demonstrated that RAB7A remains inhibited by OREC^c, while ASCL1 activation is initiated by the OREC^o system, with no cross-interference between the two pathways. Following Dox removal at 36 h, RAB7A expression showed partial recovery by 48 h and returned to baseline levels by 60–72 h, with levels statistically indistinguishable from baseline (0 h). Similarly, after light removal at 48 h, ASCL1 expression declined and returned to baseline levels by 72 h. These results demonstrate that the OREC system enables reversible transcriptional regulation, with gene expression returning to baseline levels upon inducer removal. To further validate the independence and stability of orthogonal regulation, we performed an additional experiment in which cells were simultaneously treated with Dox and blue light from the beginning of the experiment through 48 h (Figure S2E). We observed a sustained decrease in RAB7A expression and a concurrent increase in ASCL1 expression throughout the 48-h period. These results demonstrate that gene expression can be specifically and orthogonally manipulated using the OREC system, allowing for precise control of complex biological activities.

Figure 3.

Multiplex orthogonal gene regulation and programmable editing by OREC with therapeutic application in osteoblasts

(A) Experimental illustration of orthogonal activation and repression of ASCL1 and RAB7A gene expression in HEK293T cells using distinct chemical and light inductions.

(B) RT-qPCR time course analysis demonstrating orthogonal and reversible control of ASCL1 activation and RAB7A repression in response to alternating Dox and light stimuli. Data represent mean ± SEM (n = 3). vs. RAB7A baseline (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001); vs. RAB7A 36 h (^$p < 0.05, ^$$$p < 0.001); vs. ASCL1 baseline (^&&p < 0.01, ^&&&&p < 0.0001); vs. ASCL1 48 h (^##p < 0.01, ^####p < 0.0001).

(C) Schematic diagram illustrating crRNA guide-length-dependent transcriptional repression and HDR-based gene editing using OREC^c system.

(D) Transcriptional repression of RAB7A and FZD1 genes following 72 h doxycycline treatment using 15 bp crRNA guides. Data represent mean ± SEM (n = 3) (∗∗∗∗p < 0.0001).

(E) Luciferase activity analysis following HDR-mediated luciferase knockin using 20 bp crRNA guides. Data represent mean ± SEM (n = 3) (∗∗∗∗p < 0.0001).

(F) Flow cytometry analysis showing percentage of GFP-positive cells following HDR-mediated GFP knockin. Data represent mean ± SEM (n = 3) (∗∗∗∗p < 0.0001).

(G) Schematic representation illustrating simultaneous orthogonal activation of Bmp2 and repression of Dkk1 expression in MC3T3-E1 cells.

(H) RT-qPCR analysis demonstrating significantly enhanced Bmp2 activation and Dkk1 repression under simultaneous orthogonal induction compared with single-gene regulatory conditions. Data are presented as mean ± SEM (n = 3). ∗p < 0.05, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

(I) Representative images of alizarin red S (ARS) and alkaline phosphatase (ALP) staining showing significantly enhanced osteogenic differentiation following orthogonal OREC induction compared to single-gene controls. Color code for experimental conditions: white: empty vector control; gray: OREC^o construct only; crosshatched: OREC^c construct only; red: OREC^o + Light (optogenetic activation); green: OREC^c + Dox (chemogenetic activation) (∗∗∗p < 0.001, ∗∗∗∗p < 0.0001).

(J) RT-qPCR analysis demonstrating synergistic upregulation of osteogenic marker genes (Alp, Bglap, and Sp7) upon simultaneous Bmp2 activation and Dkk1 repression. Data represent mean ± SEM (n = 3) (∗∗∗p < 0.001, ∗∗∗∗p < 0.0001).

(K) The CCK-8 assay quantifies viable cells via dehydrogenase-mediated formazan formation measured at 450 nm, showing that OREC treatments do not affect cell viability in MC3T3-E1 cells. Data shown as mean ± SEM (n = 3).

See also Figures S2 and S3.

To further explore the potential of the OREC system in genome engineering, we investigated whether a single CRISPR array could simultaneously control transcriptional regulation and gene editing through differential crRNA guide design. Previous studies have demonstrated that Cpf1 nucleases can perform distinct functions depending on crRNA guide sequence length, with shorter guides primarily mediating transcriptional regulation through DNA binding without cleavage, while longer guides enable robust nuclease activity for gene editing.^33,37 Building on this principle, we designed an experimental system using the OREC^c platform to test orthogonal transcriptional control and gene editing (Figures 3C and S2F). We constructed AsCpf1-Rep with a CRISPR array containing two short crRNA guide sequences (15 bp) targeting the RAB7A and FZD1 genes for transcriptional repression and one long crRNA guide sequence (20 bp) targeting the GAPDH locus for luciferase gene integration via homology-directed repair (Figure S2G). As predicted by the guide length-dependent model, AsCpf1-Rep effectively inhibited the expression of RAB7A and FZD1 when guided by the 15 bp crRNAs (Figure 3D), while the 20 bp crRNA successfully facilitated gene editing at the GAPDH locus, resulting in significant luciferase activity (Figure 3E). To quantify editing efficiency at the population level, we performed a parallel GFP knockin experiment using identical HDR conditions and observed that approximately 14.48% of cells became GFP-positive (Figure 3F). These findings demonstrate that the OREC system can be rationally designed to achieve orthogonal gene knockin and transcriptional repression. Further exploration of different orthogonal gene editing and control methods will enhance our understanding of fine gene regulation and provide powerful tools for analyzing complex genetic patterns and advancing gene therapy.

Orthogonally regulated osteogenic effect in vitro

To assess the potential of our orthogonal gene editing system as a gene therapy tool for promoting fracture healing, we first investigated its ability to enhance osteogenesis in MC3T3-E1 cells, a well-established pre-osteoblast cell line commonly used to study osteogenic differentiation and bone formation in vitro.³⁸ Given the critical roles of BMP2 and DKK1 in osteogenesis,^39,40 we targeted these genes for precise orthogonal regulation. MC3T3-E1 cells were transduced with lentiviral OREC system, employing OREC^o to activate Bmp2 expression and OREC^c to repress Dkk1 expression (Figures 3G and S3A). Notably, simultaneous orthogonal regulation demonstrated synergistic effects, further enhancing Bmp2 upregulation and Dkk1 downregulation compared to single-gene interventions (Figure 3H). To contextualize these fold changes, we examined the expression levels relative to Gapdh. Bmp2 showed low basal expression that was substantially elevated upon light-induced activation, yet remained well below housekeeping gene levels. Dkk1, which had higher basal expression, was effectively suppressed to minimal levels upon Dox-induced repression (Figures S3B and S3C). Additionally, we observed reciprocal changes in gene expression: Dox-mediated Dkk1 repression secondarily elevated Bmp2 levels, while light-induced Bmp2 activation secondarily reduced Dkk1 expression. To determine whether this reflects system leakiness or endogenous pathway crosstalk, we performed independent validation experiments. Direct Bmp2 overexpression via lentiviral transduction increased miR-29 levels and reduced Dkk1 expression (Figures S3D–S3F), while siRNA-mediated Dkk1 knockdown reciprocally increased Bmp2 (Figures S3G and S3H). These results suggest that Bmp2 and Dkk1 exist within interconnected Wnt/β-catenin-miR-29 regulatory circuits in osteoblasts (Figure S3I), which is consistent with previous studies.^41,42,43 These findings confirm that the reciprocal effects result from physiological pathway crosstalk rather than unintended activation of our inducible modules. Furthermore, to assess potential genotoxicity, we quantified DNA double-strand breaks via γH2AX staining in MC3T3-E1 cells. Importantly, γH2AX foci did not increase relative to controls (Figure S3J), indicating that OREC did not measurably induce DNA damage under our experimental conditions and supporting a favorable initial safety profile.

Alizarin red staining and ALP staining were then employed, which evaluate early and late osteoblast differentiation, respectively. The results demonstrated significant enhancement of calcium deposition and ALP activity following orthogonal regulation (Figure 3I). Importantly, simultaneous Bmp2 activation and Dkk1 repression with OREC system exerted a synergistic effect, substantially amplifying osteogenic outcomes. This synergy was further confirmed by significantly increased expression levels of osteogenic markers Alp, Bglap, and Sp7 compared to single-gene regulation conditions (Figure 3J). Additionally, cell viability assays confirmed that OREC system did not significantly affect cell viability (Figure 3K). Collectively, these findings validate that the OREC system enables precise and efficient orthogonal control of critical osteogenic genes, significantly enhancing osteogenesis and presenting promising therapeutic potential for promoting fracture healing.

Orthogonally regulated osteogenic effect and fracture healing in vivo

To investigate the in vivo effects of OREC-mediated regulation on fracture healing, we used C57BL/6 mice with osteoporotic tibial fractures and locally administered separate AAV2/9 encoding OREC^o (for light-controlled Bmp2 activation) and OREC^c (for Dox-controlled Dkk1 repression) (Figures 4A–4D andS4A). This multi-vector approach enabled simultaneous delivery while maintaining each construct within adeno-associated virus (AAV) packaging constraints. In vivo bioluminescence imaging demonstrated that OREC expression was specifically localized to the tibial site under both light and Dox induction, with minimal signal detected in distant tissues (Figures 4E–4H). These findings confirm the spatial precision of our delivery approach and demonstrate that systemic Dox exposure does not activate OREC constructs outside the targeted injection site. Beginning one day post-fracture, the mice were either treated with oral Dox or exposed to light. A control group received direct recombinant BMP2 protein treatment at the fracture site for comparison. On days 7 and 21 post-fracture, we assessed Bmp2 and Dkk1 expression levels using RT-qPCR. Compared to controls, orthogonally regulated mice exhibited the highest Bmp2 levels and lowest Dkk1 levels (Figure 4I). To provide context for these fold changes, we analyzed the expression levels relative to Gapdh at day 7 post-fracture. In the fracture site tissue, Bmp2 showed low basal expression that was significantly elevated upon light-induced activation. Dkk1, which exhibited relatively higher basal expression in the fracture, was effectively suppressed upon Dox administration (Figure 4J). Meanwhile, inducer-only negative controls showed that neither Dox nor light exposure alone significantly altered Bmp2 or Dkk1 expression in animals without OREC transduction (Figures 4K and 4L). Further analysis of absolute expression levels in these control groups confirmed that both Bmp2 and Dkk1 maintained stable basal expression relative to Gapdh regardless of light or Dox treatment in the absence of OREC vectors (Figure S4B). This confirms that the observed gene expression changes require the presence of OREC components and demonstrates that the OREC system effectively activates Bmp2 expression while inhibiting Dkk1 expression in vivo.

Figure 4.

AAV-delivered OREC system design and in vivo orthogonal gene regulation in fracture healing

(A) OREC^o system comprises two vectors encoding dAsCpf1-Activer (upper) and TetR-CIBN-P2A-CRY2-VP16₃ with Bmp2-targeting crRNA (lower).

(B) OREC^c system consists of vectors encoding dLbCpf1-Repr (upper) and another vector expressing rtTA-Advanced with Dkk1-targeting crRNA (lower). See also Table S5. All constructs designed within AAV packaging constraints.

(C) Experimental setup showing AAV intratibial injection procedure with custom LED patch device and power supply system for localized light delivery.

(D) Experimental timeline: AAV intratibial injection 14 days before fracture (day 14), tibial fracture induction (day 0), and blue light treatment (470 nm, twice daily for 30 min) beginning one day post-fracture (day 1) through endpoint (day 42).

(E and F) AAV2/9 vector constructs for bioluminescence analysis of OREC activity in vivo.

(G) Experimental timeline showing AAV intratibial injection (day 0) followed by 14-day treatment with light stimulation (470 nm, 30 min twice daily) or oral doxycycline (30 mg/kg daily) prior to bioluminescence imaging.

(H) Representative bioluminescence images demonstrating inducible luciferase expression at tibial sites by OREC with no detectable signal in distant tissues.

(I) RT-qPCR analyses of Bmp2 and Dkk1 expression levels in orthogonally treated fracture sites at days 7 and 21 post-fracture. Data represent mean ± SEM (n = 6 mice/group) (∗∗p < 0.01, ∗∗∗∗p < 0.0001).

(J) Analysis of Bmp2 and Dkk1 expression relative to Gapdh using the 2^−ΔCt method in fracture callus tissue 7 days after light or doxycycline induction in OREC-transduced mice. Data represent mean ± SEM (n = 6 mice per group).

(K and L) RT-qPCR analysis of Bmp2 (K) and Dkk1 (L) expression in fracture callus tissue from mice exposed to inducers without OREC constructs. Data represent mean ± SEM (n = 3 mice per group). See also Figure S4 and Table S5.

Subsequent X-ray and micro-computed tomography (micro-CT) analyses were performed. X-ray scans confirmed that the fracture fixation was well maintained across all groups. While there were no significant differences in fracture healing among the groups on day 7, by day 21, more pronounced callus formation was observed in the groups where OREC activated Bmp2 or inhibited Dkk1 expression. The most substantial callus formation and mineralization occurred in the group receiving orthogonal treatment (Figure 5A), which demonstrated superior healing compared to direct recombinant BMP2 protein treatment. Micro-CT images taken on the day 21 post-fracture revealed that bone defects were still visible in the control group, whereas these defects had completely resolved in the orthogonal treatment group. Notably, the bone cortex in the orthogonal group appeared continuous and smooth, indicating that the fracture had nearly healed (Figure 5A). Quantitative analysis further demonstrated that the orthogonal treatment group showed a significantly greater increase in bone volume compared to the groups receiving either Bmp2 activation alone or Dkk1 inhibition alone and notably exceeded the therapeutic efficacy of direct BMP2 protein treatment, suggesting enhanced osteogenic activity and more effective fracture healing (Figure 5B). Additionally, the total callus volume in the orthogonal treatment group was slightly reduced, likely due to the resorption of excess callus during the healing process. As a result, the bone volume-to-total volume (BV/TV) ratio was most significantly elevated in the orthogonal treatment group (Figure 5B). These findings indicate that the fracture healing process was more rapid in the orthogonal treatment group, characterized by a strong bone tissue repair response that led to the resorption and remodeling of the callus tissue into normal bone.

Figure 5.

OREC-mediated orthogonal gene regulation enhances fracture healing outcomes in vivo

(A) Representative X-ray and micro-CT images demonstrating fracture healing progression in orthogonally regulated mice vs. control groups at days 7 and 21 post-fracture.

(B) Quantitative micro-CT analysis of total callus volume (TV), bone volume (BV), and BV/TV at day 21 post-fracture indicate superior callus formation, mineralization, and accelerated healing in orthogonally regulated groups compared to single interventions. Data represent mean ± SEM (n = 6 mice/group). (∗∗∗p < 0.001, ∗∗∗∗p < 0.0001).

(C) Alcian blue and hematoxylin and eosin (H&E) staining of fracture callus tissue at day 21 post-fracture. Scale bars, 500 μm.

(D) Mechanical testing results showing significantly improved biomechanical properties (ultimate load, stiffness, and energy to failure) in orthogonally treated groups compared to single interventions and controls at day 21 post-fracture. Data represent mean ± SEM (n = 6 mice/group) (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001).

See also Figure S5.

To further characterize the mechanisms underlying enhanced fracture repair, we performed histological analysis using Alcian blue and hematoxylin and eosin (H&E) staining (Figure 5C). At day 21 post-fracture, OREC-treated groups exhibited accelerated callus remodeling, with markedly reduced cartilaginous tissue within the fracture callus compared to controls. The orthogonal treatment group demonstrated the most pronounced reduction in Alcian-blue-positive areas, exceeding recombinant BMP2 protein treatment (Figure 5C). These findings indicate superior promotion of callus maturation and endochondral bone formation through OREC-mediated gene regulation. To assess potential genotoxicity associated with CRISPR-mediated interventions, we performed γH2AX immunostaining to quantify DNA double-strand breaks at day 21. Notably, no significant γH2AX-positive signals were detected across any treatment groups (Figure S4C), indicating that our OREC-mediated gene regulation approach did not induce detectable DNA damage in the healing tissue, supporting the safety profile of this therapeutic strategy.

Serum bone turnover analysis revealed dynamic changes in bone metabolism during fracture healing, as shown in Figures S4D and S4E. On day 7 post-fracture, the orthogonal group exhibited elevated levels of osteocalcin (OCN), indicating active osteoblast-mediated bone formation during the reparative phase. Meanwhile, CTX-1 levels declined, suggesting that osteoclast activity was transiently suppressed to prioritize new bone deposition (Figure S4D). By day 21, OCN levels remained high, reflecting ongoing osteoblastic participation in callus remodeling. In contrast, CTX-1 levels increased compared to day 7, indicating reactivation of osteoclasts involved in resorption of excess callus and structural optimization of the newly formed bone (Figure S4E). These time-dependent changes in bone turnover markers are consistent with the characteristic stages of murine fracture healing and further support that orthogonal regulation accelerates and coordinates bone regeneration through precise temporal modulation of osteogenic and osteoclastic activity. Additionally, we assessed the mechanical properties of bone tissue in mice 21 days post-fracture. The results indicated significant increases in ultimate load, stiffness, and energy to failure in the orthogonal group (Figure 5D), which demonstrated superior biomechanical recovery compared to direct BMP2 protein treatment. These improved mechanical properties suggest that the bones in the orthogonal treatment group healed more rapidly and were better equipped to restore physiological function.

To assess long-term stability and potential immunogenicity of AAV-delivered OREC, we extended our in vivo analysis through day 56 post-injection. CIBN was selected as a representative OREC component for stability analysis. CIBN expression in tibial tissue, as measured by RT-qPCR, increased progressively from day 7 through day 28 and remained stable through day 56, demonstrating sustained transgene expression without evidence of silencing (Figure S4F). Concurrently, we monitored adaptive immune responses to AAV by quantifying serum anti-AAV9 capsid antibodies via ELISA. Anti-AAV9 immunoglobulin G (IgG) antibodies emerged by days 14–21, peaked around day 21, and then gradually declined through day 56 (Figure S4G). Anti-AAV9 IgM antibodies were detectable by day 7 and subsequently declined, reflecting the expected early humoral response (Figure S4G). Importantly, the presence of anti-AAV9 antibodies did not correlate with reduced OREC expression levels, indicating that the systemic immune response did not compromise local transgene function over the 8-week observation period.

To comprehensively evaluate the biosafety of the OREC system, we conducted multiple assessments throughout the treatment period. To assess potential off-target effects, we computationally predicted candidate off-target sites for Bmp2 and Dkk1 crRNAs and quantified mRNA expression of corresponding genes in callus tissues at days 21 post-fracture. None of the predicted off-target genes showed significant expression changes compared to controls (Figure S4H), indicating high specificity of the OREC system for its intended targets. Long-term radiological analysis at day 42 revealed successful fracture healing with no evidence of tumor formation, ectopic bone growth, or other pathological abnormalities at the treatment site (Figure S4I). Additionally, histological examination of major organs including heart, liver, spleen, lung, and kidney showed no observable pathological changes across all treatment groups (Figure S4J). To assess system controllability, we evaluated OREC activity after inducer withdrawal. Bmp2 and Dkk1 expression levels at days 3 and 7 post-cessation showed no significant differences between treatment groups, confirming OREC inactivation and tight regulatory control (Figures S5A–S5D). We further examined potential systemic effects by analyzing distant skeletal sites. Bone mineral density (BMD) and biomechanical testing of lumbar spine and contralateral femur revealed no significant alterations in the OREC treatment group compared to controls (Figures S5E–S5L). In contrast, recombinant BMP2 protein treatment produced measurable increases in spinal BMD and ultimate load, indicating that OREC-mediated regulation remained spatially confined to the target fracture site while conventional BMP2 therapy exhibited systemic skeletal effects (Figures S5E–S5L). These results provide strong evidence supporting the biocompatibility and safety of the OREC system for potential clinical translation.

In summary, our results demonstrated that the engineered orthogonal gene regulation system can significantly accelerate fracture healing and enhance the recovery of bone function. This advancement offers a promising solution to the long-standing clinical challenge of prolonged fracture healing, providing a novel tool and platform for treating orthopedic conditions such as osteoporotic fractures.

Discussion

In this study, we developed an orthogonally regulatable CRISPR/Cpf1 system that integrates chemogenetic and optogenetic approaches to achieve precise, spatiotemporal, and multiplexed gene editing. By leveraging orthogonal regulation, our system enables independent control of gene expression through chemical or light stimulation. This capability extends existing CRISPR systems by enabling orthogonal, spatiotemporal perturbations that allow distinct genes to be modulated simultaneously within the same cell.^17,20,22 We demonstrated that OREC effectively regulates gene editing and regulation in vitro and enhances fracture healing in vivo by simultaneously activating Bmp2 and inhibiting Dkk1. These findings underscore the potential of OREC as a versatile gene therapy platform for regenerative medicine and beyond.

CRISPR-based gene regulation technologies face limitations in spatiotemporal control and multiplexed gene regulation. Traditional inducible CRISPR systems rely primarily on either chemical induction or optogenetic activation but fail to provide independent dual control within the same system. Our OREC platform overcomes these limitations by integrating both approaches, ensuring high specificity and tunability. The independent activation of two distinct gene-editing pathways allows for synergistic effects, as demonstrated by the coordinated upregulation of Bmp2 and suppression of Dkk1, which enhances osteogenic differentiation and accelerates fracture healing. This finding highlights the potential of our system to improve treatment strategies for osteoporotic fractures, a condition with limited effective pharmacological options. Additionally, the programmable and tunable properties of OREC make it adaptable for different therapeutic applications by substituting different crRNA arrays or regulatory elements, expanding its applicability beyond bone repair to other regenerative medicine fields, such as cartilage regeneration, neurodegeneration, and immune modulation.

Beyond its orthogonal regulatory capacity, our system integrates Triplex structures, enabling the assembly of dCpf1, gene regulatory elements, and a crRNA array into a single Pol II-derived transcript. This design minimizes genetic payload while maintaining synchronized expression of all components within a compact and efficiently delivered system. By optimizing the OREC system, we successfully reduced the vector size to comply with the stringent 5.0 kb (including ITRs) packaging limit of AAV, thereby enhancing its suitability for in vivo applications. This structural optimization facilitates efficient gene delivery and precise therapeutic control. Furthermore, the incorporation of Triplex structures improves crRNA processing and transcript stability, further enhancing the precision, efficiency, and robustness of gene regulation.

While our system demonstrates robust gene regulation capabilities, clinical translation faces two primary challenges. First, the Dox-dependent chemogenetic component encounters regulatory barriers, as tetracycline-based systems raise concerns regarding antibiotic resistance and off-target effects. Second, optogenetic control requires custom light delivery devices, increasing implementation complexity. The modular architecture of OREC, however, enables systematic adaptation to address these limitations. For chemogenetic components, recent advances have validated inducible systems responsive to clinically approved compounds (rapamycin, tamoxifen) and natural dietary molecules (vanillic acid, resveratrol) with established safety profiles and regulatory acceptance.^{15,44,45,46,47} The Dox module can be replaced with these validated inducers while preserving orthogonal functionality. For optogenetic components, emerging wireless and miniaturized LED technologies promise to significantly reduce implementation barriers, making optical control more clinically feasible.⁴⁸ Beyond addressing current limitations, expanding inducible modalities to include mechanical or metabolic signals could enhance OREC’s versatility across diverse biomedical applications.^49,50

In summary, the OREC system represents a significant advancement in CRISPR-based gene regulation, offering precise, multiplexed, and spatiotemporal control of gene expression. By enabling independent regulation of multiple genes, OREC holds great promise for therapeutic applications, particularly in regenerative medicine and gene therapy. The ability to fine-tune gene expression with high precision opens up new possibilities for treating complex diseases, including osteoporotic fractures, musculoskeletal disorders, and metabolic syndromes. As we continue to refine and expand this technology, it has the potential to transform the landscape of precision medicine, providing novel strategies for treating complex diseases and improving patient outcomes.

Limitations of the study

Although this study demonstrates the effectiveness of the OREC platform for precise multiplex gene regulation in osteoporotic fracture models, several limitations should be noted. First, translating findings from mouse models to clinical applications requires validation in larger animal models with bone healing characteristics more comparable to humans. Second, our study focuses specifically on fracture healing in an osteoporotic context; broader therapeutic applications of OREC remain to be explored in future work.

Resource availability

Lead contact

Additional information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Jie Zhao (zhaojie@tmu.edu.cn).

Materials availability

Plasmids generated in this study are available from the lead contact without restriction.

Data and code availability

• •All data reported in this paper will be shared by the lead contact upon request.
• •This paper does not report original code.
• •Any additional information required to re-analyze the data reported in this study is available from the lead contact upon request.

Acknowledgments

This work was supported by the National Key Research and Development Plan (2022YFC3601900 and 2022YFC3601904), National Natural Science Foundation of China (no. 82572860, 82302347, and 82102639), Tianjin Health Industry High-Level Talent Program (TJSQNYXXR-D2-133), Young Elite Scientists Sponsorship Program by CAST (2023QNRC001), Natural Science Foundation of Tianjin (25JCYBJC01510, 22JCQNJC00850, 22JCYBJC00780, and 24JCYBJC01450), Tianjin Health Science and Technology Project (TJWJ2021QN047, TJWJ2023MS020, and TJWJ2024ZD005), and National Key Clinical Specialty Construction Project (Digital Orthopedic Diagnosis and Treatment Center).

Author contributions

Conceptualization and methodology, J.Z., J.J., and X.M.; investigation and in vitro experiments, J.Z., Z.W., L.L., Y.Z., G.B., and Y.G.; in vivo model establishment and analysis, J.Z., L.L., Z.M., and G.B.; histological analysis and imaging, L.L., J.M., G.B., and Z.Y.; data interpretation, J.Z., J.J., and X.M.; writing—original draft, J.Z.; writing—review & editing, J.J. and X.M.; figure generation, J.Z.; supervision, J.J. and X.M.; J.Z., J.J., and X.M. had unrestricted access to all data. All authors read and approved the final draft, agree to submit the manuscript, and take full responsibility for its content.

Declaration of interests

The authors declare no competing interests.

Declaration of generative AI and AI-assisted technologies in the writing process

During the preparation of this work, the authors used ChatGPT (OpenAI) in order to improve the readability, clarity, and linguistic quality of the manuscript. After using this service, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Anti-AsCpf1	Cell Signaling Technology	Cat# 93300
Anti-γH2AX	Cell Signaling Technology	Cat# 2577
Goat Anti-Mouse IgG H&L (Alexa Fluor® 488)	Abcam	Cat# ab150113; RRID: AB_2576208
Goat Anti-Rabbit IgG H&L (Alexa Fluor® 488)	Abcam	Cat# ab150077; RRID: AB_2630356
Peroxidase-conjugated goat anti-mouse IgG	Jackson ImmunoResearch	Cat# 115-035-003
Peroxidase-conjugated goat anti-mouse IgM	Jackson ImmunoResearch	Cat# 115-035-075
Bacterial and virus strains
E. coli DH5a	Thermo Fisher	Cat# 18258-012
pLV-ORECo	This paper	N/A
pLV-ORECC	This paper	N/A
pAAV-ORECo	This paper	N/A
pAAV-ORECo	This paper	N/A
Chemicals, peptides, and recombinant proteins
Doxycycline	Sigma-Aldrich	Cat# D9891
Alizarin Red S	Sigma-Aldrich	Cat# A5533
Cetylpyridinium chloride	Sigma-Aldrich	Cat# C0732
Recombinant BMP2 protein	MedChemExpress	Cat# HY-P7006
T4 DNA ligase	NEB	Cat# M0202S
Dulbecco’s Modified Eagle Medium (DMEM)	Gibco	Cat# 11965092
Minimum Essential Medium-alpha (α-MEM)	Gibco	Cat# 12561056
Fetal bovine serum (FBS)	Lonsera	Cat# S711-001
Penicillin-streptomycin (P/S)	Gibco	Cat# 15140122
Entranster-H4000	Engreen Biosystem	Cat# 4000-4
Critical commercial assays
Alkaline phosphatase activity assay kit	BioVision	Cat# K412-500
PINP assay kit	Cloud-Clone Corp	Cat# SEA957Mu
CTX-1 assay kit	Cloud-Clone Corp	Cat# CEA665Mu
Dual-Luciferase Reporter Assay System	Promega	Cat# E1910
ClonExpress Ultra One Step Cloning Kit	Vazyme	Cat# C115-02
miRNA Unimodal SYBR qPCR Master Mix	Vazyme	Cat# MQ102
ChamQ Universal SYBR qPCR Master Mix	Vazyme	Cat# Q711
Experimental models: Cell lines
Human: HEK293T cells	ATCC	Cat# CRL-3216
Mouse: MC3T3-E1 cells	Cell Bank of Chinese Academy of Sciences	Cat# GNM15
Experimental models: Organisms/strains
C57BL/6 mice (female)	BEIJING HFK BIOSCIENCE CO.,LTD	N/A
Oligonucleotides
Guide RNA sequences, see: Table S1	This paper	N/A
siRNA sequences, see: Table S2	This paper	N/A
qPCR primers, see: Table S3	This paper	N/A
TIDE analysis primers, See: Table S4	This paper	N/A
Recombinant DNA
OREC Effector sequences, see: Table S5	This paper	N/A
pAAV-MCS2	Addgene	Cat# 46954
pAAV2/9	Addgene	Cat# 112865
pHelper plasmid	Addgene	Cat# 112867
pcDNA3.1+N-DYK	GenScript	NA
pUC57-Kan	Addgene	121344
pCDH-CMV	Addgene	Cat# 72265
pCDH-CMV-MCS-EF1-copGFP	System Biosciences	Cat# CD511B-1
pMDLg/pRRE	Addgene	Cat# 12251
pRSV-Rev	Addgene	Cat# 12253
pMD2.G	Addgene	Cat# 12259
Software andalgorithms
GraphPad Prism 9	GraphPad Software	https://www.graphpad.com
NanoDrop 2000	Thermo Fisher Scientific	https://www.thermofisher.com
Skyscan Micro-CT software	Bruker	https://www.bruker.com
ImageJ software	NIH	https://imagej.nih.gov/ij/

Experimental model and study participant details

Ethics statement

All experiments were performed with the approval of the Institutional Animal Care and Use Committees of Tianjin Hospital.

Animal models

Ovariectomy (OVX) in rodents has been used extensively as a model of postmenopausal osteoporosis. Bilateral ovariectomized (OVX) surgery was performed on skeletally mature female C57BL/6 mice. OVX mice were kept under the same standard conditions (room temperature: 20°C–24°C; relative humidity: 30%–40%). Tibial osteoporosis fracture were established 12 weeks post-OVX surgery as previous studies.^32,51 Shortly, a small blade was used to create a defect in the midshaft of the tibia, followed by a gentle bending force to induce the fracture. The fracture site was carefully manipulated to ensure a complete, transverse break. A 0.4-mm diameter stainless steel pin was then inserted into the intramedullary canal, extending the full length of the tibia. Mice were then assigned into six groups of 6 mice each randomly: (1) OVX fracture + empty vector, (2) OVX fracture + OREC construct only, (3) OVX fracture + OREC^o + light, (4), OVX fracture + OREC^c + Dox, (5), OVX fracture + OREC + orthogonal treatment, (6) OVX fracture + recombinant BMP2 protein treatment. For the recombinant BMP2 protein control group, a collagen sponge (RCM6 Resorbable Collagen Membrane) soaked with 10 μL of saline containing BMP2 (5 μg) was sutured around the fracture site.⁵²

Method details

AAV vector preparation and in vivo transduction

AAV2/9 vectors encoding the OREC constructs were utilized due to their excellent tropism for bone tissue and efficient transduction capability in vivo.^53,54 To accommodate AAV packaging capacity limitations, each OREC system was divided into two separate vector components using pAAV-MCS2 backbone vectors (Figures 4A and 4B). Each component was individually produced, where HEK293T cells were co-transfected with the respective pAAV-OREC plasmid, helper plasmid (pAdDeltaF6), and rep/cap plasmid encoding AAV2/9 (pAAV2/9), using Entranster-H4000 according to the manufacturer’s protocol. The cells were cultured for 72 h post-transfection, after which rAAV particles were harvested from cell lysates and culture media by centrifugation and purified using an iodixanol gradient ultracentrifugation. The purified viral vectors were concentrated and dialyzed into phosphate-buffered saline (PBS), aliquoted, and stored at −80°C until use.

For in vivo transduction, AAV vectors were locally injected two weeks prior to the establishment of osteoporotic tibial fractures in C57BL/6 mice.^55,56 Specifically, a total volume of 10 μL AAV (approximately 1 × 10¹¹ genome copies) was administered at the intended fracture site using a 30-gauge needle. Beginning one day post-fracture, mice were administered either doxycycline (30 mg/kg body weight) via daily oral gavage or received localized blue light stimulation through custom-designed LED patches positioned directly over the fracture site. Blue light treatment (470 nm, 0.02 mW mm⁻²) was administered twice daily for 30 min (3s light, 60s dark cycles) throughout the treatment period (Figures 4C and 4D).

X-ray and micro-CT analysis

X-ray radiographic analysis was performed in mice 7, 21 and 42 days after fracture. Then, the mice were euthanized, the tibia and serum were collected for subsequent experiments. Tibia were fixed in 10% neutral buffered formalin solution for 48 h. The Micro-CT scanning (Skyscan 1276, Bruker microCT, Kontich, Belgium) was performed to obtain high-quality images and quantative analysis.⁵⁷

Bone turnover analysis

The serum was isolated, then PINP and CTX-1 levels were quantified by a chemiluminescence-based assay kit as manufacturer’s instructions.

Plasmid constructs

Genetic elements and constructs were obtained through gene synthesis (Genscript Biotech). The synthesized elements were then cloned into pcDNA3.1+N-DYK backbone vectors for HEK293T cell experiments, pCDH-CMV lentiviral vectors for MC3T3-E1 cell transduction, or pAAV-MCS2 backbone vectors for in vivo AAV delivery, using Gibson assembly (ClonExpress Ultra One Step Cloning Kit V2, Vazyme). HDR donor templates containing reporter sequences were cloned into pUC57-Kan vectors. Comprehensive design and construction details for all expression vectors are listed in Table S5. All constructions have been confirmed by sequencing (Sangon Biotech). The guide RNA sequences used in this study is provided in Table S1.

Cell lines and transfection

HEK293T obtained from ATCC were maintained in Dulbecco’s modified Eagle’s medium (GIBCO) supplemented with 10% fetal bovine serum (FBS, Lonsera) and 1× penicillin/streptomycin (P/S, GIBCO). MC3T3-E1 obtained from Cell Bank of Chinese Academy of Sciences were maintained in Minimum Essential Medium-α (GIBCO) supplemented with 10% FBS and 1× P/S.

Plasmid transfections were performed using Entranster-H4000 (Engreen Biosystem Co., Ltd.) according to manufacturer’s protocol. Cells were plated 16 h prior to transfection to ensure 70–80% confluency. After 6 h, culture medium was replaced with fresh medium optionally supplemented with doxycycline or subjected to light stimulation as previously reported.²¹ For siRNA experiments, cells were transfected with siRNA targeting Dkk1 (siDkk1) or negative control siRNA (siNC) at 50 nM using Lipofectamine RNAiMAX (Thermo Fisher Scientific). siRNA sequences are provided in Table S2.

Lentivirus production and transduction

The OREC system components were cloned into two separate lentiviral vectors, with detailed constructs provided in Figure S3A. For overexpression of Bmp2, the Bmp2 coding sequence was directly cloned into pCDH-CMV-MCS-EF1-copGFP vector. Lentiviral particles were produced by co-transfecting HEK293T cells with the respective transfer vectors and packaging plasmids pMDLg/pRRE, pRSV-Rev, and pMD2.G using Entranster TM-H4000.⁵⁷ After 6 h, culture medium was replaced with Freestyle 293 Expression Medium supplemented with 50 U/mL penicillin/streptomycin, 2 mM L-glutamine, 1 mM sodium pyruvate, and 5 mM sodium butyrate. Viral supernatants were collected at 48 and 72 h post-transfection, filtered through 0.45 μm filters, and concentrated using Lenti-X Concentrator (Takara Bio 631232) according to manufacturer’s protocol.

For MC3T3-E1 cell transduction, cells were seeded at 70% confluency and co-infected with the lentiviral constructs in the presence of 8 μg/mL polybrene. After 24 h, medium was replaced with fresh α-MEM containing lentiviral constructs for a second round of transduction. Cells expressing multiple fluorescent reporters indicating successful transduction of all OREC components were isolated by fluorescence-activated cell sorting 72 h post-transduction and expanded for functional analyses. For Bmp2 overexpression experiments, successfully transduced cells were isolated and expanded for subsequent analyses.

Dual luciferase reporter assay

Luciferase readouts of Fluc and Rluc were sequentially measured using the DualGlo Luciferase Assay System (Promega) on a GLOMAX Multi Detection System (Promega) according to the manufacturer’s protocol. Fluc luminescence levels were normalized to the corresponding Rluc luminescence levels to generate the normalized change in protein levels.

GFP, mCherry and luciferase knock-in assays

For GFP gene knock-in with OREC^c system, HEK293T cells were co-transfected with OREC^c and GFP template (Table S5). After 6 h, the culture medium was replaced by fresh cell culture medium supplemented with doxycycline (1 μg/mL), After 72 h of doxycycline intervention, GFP-positive cells were analyzed by flow cytometry and knock-in efficiency was calculated as the percentage of GFP-expressing cells.

For mCherry gene knock-in with OREC^o system, HEK293T cells were co-transfected with OREC^o system and mCherry template (Table S5). After 6 h, the culture medium was replaced by fresh cell culture medium supplemented with blue light (0.02 mW mm⁻², cycles of 3s light and 60 s dark), After 72 h of light intervention, mCherry-positive cells were analyzed by flow cytometry.

For luciferase gene knock-in demonstrating orthogonal transcriptional control and gene editing, the OREC^c system was designed with a multiplexed crRNA array containing both short guides (15 bp) for transcriptional repression and a long guide (20 bp) for HDR-mediated integration. Cells were co-transfected with the OREC^c construct and luciferase reporter template (Table S5), then treated with doxycycline (1 μg/mL) as described above. Luciferase activity was measured 72 h post-treatment using the Dual-Luciferase Reporter Assay System (Promega). To validate knock-in efficiency, parallel experiments were performed using GFP reporter template under identical conditions, with GFP-positive cells quantified by flow cytometry to determine the percentage of successful integration events.

RT-qPCR assay

Total RNA were isolated using Trizol (Invitrogen) according to manufacturer’s instructions. The RNA concentration was measured with a NanoDrop 2000 (Thermo Fisher Scientific). RNA was reverse-transcribed into complementary DNA (cDNA) using HiScript III 1st strand cDNA synthesis kit (Vazyme). The qPCR was performed using ChamQ qPCR SYBR Green Master Mix. For miR-29 quantification, miRNA Unimodal SYBR qPCR Master Mix was used. For cell-based experiments, RT-qPCR was performed 72 h after Dox or light induction unless otherwise indicated; for in vivo studies, RT-qPCR was conducted at predetermined tissue collection timepoints as detailed in figure legends. Data are shown as fold change relative to the control group and normalized to GAPDH or U6 using a 2^−ΔΔCt method. The primers are listed in Table S3.

Immunofluorescence analysis

Cells transfected with OREC constructs were treated with doxycycline (1 μg/mL) or blue light for 72 h, then fixed with 4% paraformaldehyde and permeabilized with 0.1% Triton X-100. Following blocking with 5% normal goat serum, cells were incubated overnight with anti-AsCpf1 primary antibody (1:500), γH2AX antibody (1:500) at 4°C, followed by Alexa Fluor 488-conjugated secondary antibody (1:1000) for 1 h at room temperature. Nuclei were counterstained with DAPI and images acquired by confocal microscopy.

For tissue immunofluorescence, paraffin sections were deparaffinized in xylene and rehydrated through graded alcohols. Antigen retrieval was performed using citrate buffer (pH 6.0) in a pressure cooker for 15 min. After cooling and blocking with 5% normal goat serum for 1 h, sections were incubated overnight at 4°C with primary antibodies against γH2AX (1:500). Following washing, sections were incubated with Alexa Fluor 488-conjugated secondary antibody (1:1000) for 1 h at room temperature. Nuclei were counterstained with DAPI, sections were mounted with anti-fade mounting medium, and images were acquired by confocal microscopy.

Alizarin red staining assay

For alizarin red staining (ARS) assay, After osteoblast MC3T3-E1 were transfected with OREC system, the cells were fixed with 4% polyformaldehyde for 15 min and stained with 1% Alizarin red S (pH 4.2, Sigma-Aldrich) for 5 min. Subsequently, the mineralized matrix of osteoblast was stained with alizarin red. After that, the stained mineralized matrix was destained using 10% cetylpyridinium chloride. Finally, the absorbance of the sample was measured at 562 nm using a UV spectrophotometer and the calcium content was calculated.

ALP staining assay

For ALP staining assay, After osteoblast MC3T3-E1 were transfected with OREC system, the ALP activity of cells was measured with an ALP colorimetric assay kit according to the manufacturer’s protocol (BioVision, USA). In addition, ALP staining was also performed on the cells according to the manufacturer’s instructions.

Cell viability assay

HEK293T cells seeded in 96-well flat-bottomed plates were exposed to doxycycline (1 μg/mL) or blue light (0.02 mW mm⁻², cycles of 3s light and 60 s dark) for different duration of time. After that, the medium was changed and suplemented with 10 μL of CCK-8 reagent (Dojindo Molecular Technologies, Inc.), then incubated at 37°C for 1 h. The absorbance at 450 nm was measured using a microplate reader.

TIDE analysis

Genomic DNA was extracted using the DNeasy Blood & Tissue Kit (Qiagen) according to manufacturer’s instructions. Target loci were amplified by PCR using primers flanking the CRISPR cut sites (Table S4). PCR amplification was performed using Q5 High-Fidelity DNA Polymerase (NEB) with the following conditions: initial denaturation at 98°C for 30 s, followed by 35 cycles of 98°C for 10 s, 60°C for 20 s, and 72°C for 30 s, with a final extension at 72°C for 2 min. PCR products were purified using the QIAquick PCR Purification Kit (Qiagen) and submitted for Sanger sequencing (Sangon Biotech). TIDE analysis was then conducted using the online TIDE tool (https://tide.nki.nl/).

In vivo bioluminescence and imaging

On day 14 post-AAV injection, following light stimulation and doxycycline treatment, each mouse was intraperitoneally injected with 15 mg/mL D-Luciferin solution (150 mg/kg) and anesthetized with 2% isoflurane dissolved in oxygen using an economical animal anesthesia machine (HSIV-S, Raymain). Ten minutes after Luciferin injection, bioluminescence images of the mice were taken using the IVIS Lumina II in vivo imaging system (PerkinElmer).

Serum AAV antibody analysis

Serum samples were collected from mice at days 7, 14, 21, 28, 42 and 56 post-AAV injection. Anti-AAV9 IgG and IgM concentrations in mouse serum were measured using ELISA as previously described.⁵⁸ Briefly, 96-well plates were coated with AAV9 capsid protein (1 μg/mL) in carbonate buffer overnight at 4°C. After blocking with 5% non-fat dry milk in PBS-Tween 20 (0.05%) for 2 h at room temperature, plates were incubated with mouse serum samples (1:100 dilution for IgG, 1:50 for IgM) for 1 h at 37°C. Following washing, plates were incubated with horseradish peroxidase-conjugated goat anti-mouse IgG (1:5000) or IgM (1:2000) secondary antibodies for 1 h at room temperature. Color development was performed using TMB substrate (Thermo Fisher Scientific) and stopped with 2M sulfuric acid. Optical density was measured at 450 nm using a microplate reader (BioTek).

Bone mineral density and biomechanical testing

Bone mineral density (BMD) was measured using dual-energy X-ray absorptiometry (DXA) on a PIXImus densitometer (GE Lunar) following manufacturer protocols.⁵⁹ Lumbar vertebrae (L1-L6) and contralateral femur were scanned using small animal software with automated region of interest (ROI) analysis. BMD values were expressed as g/cm².

For biomechanical testing, tibiae, L6 vertebrae and femora were harvested. Specimens were tested using a universal testing machine (Bio-electroforce 3230). Tibiae and femora were subjected to three-point bending with supports spaced 10 mm apart and loading applied to the mid-diaphysis at 0.5 mm/min until failure. The L6 vertebrae were subjected to a vertebral compression test on the Bio-electroforce 3230 with a ramp of 0.5 mm/min.⁵⁹ The analysis was run using Bluehill2 software. Ultimate load, stiffness and energy to failure were derived from the load/displacement curves obtained during the tests.

Histological analysis

Tibiae tissues were fixed in 10% formalin, decalcified in 10% EDTA, embedded in paraffin and sectioned at 5 μm. Sections were stained with Alcian blue/H&E to assess general morphology and cartilaginous callus formation during endochondral ossification. In addition, H&E staining was also performed on major organs, including the heart, liver, spleen, lung, and kidney, to evaluate potential systemic toxicity of OREC treatments. Images were acquired using a light microscope (Leica Microsystems) under identical settings across groups.

Quantification and statistical analysis

All in vitro data represent mean ± SEM of three independent experiments (n = 3). For mouse experiments, each treatment group was composed of six mice. For comparisons between two groups, unpaired two-tailed Student’s t tests were used. For multiple group comparisons, one-way or two-way ANOVA followed by Tukey’s post hoc test was applied. Differences were considered statistically significant at p < 0.05. All analyses were performed using GraphPad Prism software (version 10.3.1, GraphPad Software Inc.).

Published: February 11, 2026