Engineering metal-organic frameworks as precision nanomedicines for myocardial infarction therapy

Abstract

Myocardial infarction (MI) remains a major cause of mortality worldwide. However, current therapeutic paradigms are limited in their ability to modulate the post-infarction pathological microenvironment with spatiotemporal precision. Metal–organic frameworks (MOFs), an emerging class of crystalline porous coordination polymers, offer exceptional structural programmability through their ultra-high specific surface areas, tunable pore architectures, and customizable surface chemistries. These properties have positioned MOFs as promising platforms for targeted MI therapy. In this review, the molecular design principles for constructing MOFs as precise nanomedicines for myocardial infarction are examined. Structural strategies—including stimulus-responsive drug delivery, biomimetic nanoplatforms, theranostic systems, and biohybrid constructs—are systematically analyzed with respect to their capacity to modulate the post-infarction microenvironment. The mechanisms underlying MOF-mediated cardiac repair are critically evaluated, with emphasis on oxidative stress regulation, inflammatory microenvironment reprogramming, and angiogenesis promotion. Key translational barriers, including biosafety assessment, scalable manufacturing, delivery-route optimization, and validation in large-animal models, are also discussed. Finally, future directions involving multi-omics-guided architectural design, organ-on-chip technology, 3D-printed cardiac patches, regulatory science, and a testable research agenda are proposed. Although MOF-based MI therapies remain at the preclinical stage, these convergent efforts may accelerate the bench-to-bedside translation of MOF-based precision nanomedicines for cardiovascular disease.

Introduction

Pathological features and unmet clinical needs

Acute coronary artery obstruction precipitates myocardial infarction (MI), a catastrophic cardiovascular event characterized by sustained myocardial ischemia and hypoxia that ultimately result in irreversible cardiomyocyte necrosis. The pathological course of MI is temporally dynamic (Fig. 1). Cardiomyocyte death can begin within 20–40 min after coronary blood flow is interrupted. Although ischemia–reperfusion (I/R) therapy rescues jeopardized myocardium, reperfusion simultaneously induces severe oxidative stress through a burst of reactive oxygen species (ROS) that overwhelms endogenous antioxidant defenses [1]. The subsequent inflammatory cascade—characterized by neutrophil infiltration, imbalanced macrophage polarization, and a cytokine storm—further aggravates tissue injury and ultimately leads to cardiomyocyte apoptosis and replacement by fibrotic scar tissue [2, 3]. The resulting ventricular remodeling can progress to heart failure, malignant arrhythmias, and cardiogenic sudden death. Globally, ischemic heart disease remains the leading cause of death and has been estimated to account for more than 9 million deaths annually, or approximately 16% of all deaths [4, 5]. Although acute-phase mortality has been reduced by reperfusion strategies, heart failure develops within 5 years in approximately 30% of patients with MI, imposing a substantial burden on healthcare systems [6, 7]. MI has also been associated with dysfunction in distant organs, including vascular dementia and other neurological sequelae, underscoring its systemic consequences [8]. Effective interruption of the post-infarction pathological cascade therefore remains a major unmet need in cardiovascular medicine.

Fig. 1.

Temporal dynamics of the post-infarction pathological cascade and associated unmet clinical needs (Schematic diagram). The timeline depicts pathological changes after MI, with time shown on the horizontal axis (0–72 h and the chronic phase) and relative levels of ROS, inflammatory-cell infiltration (neutrophils and macrophages), cardiomyocyte viability, and fibrosis shown on the vertical axes. Four critical therapeutic windows are indicated: the reperfusion window (0–2 h), antioxidant window (2–6 h), anti-inflammatory window (6–48 h), and anti-fibrotic window (48–72 h). The lower panel summarizes the structural limitations of current therapeutic strategies, including reperfusion injury (an extremely short therapeutic window, immediate damage induced by the ROS burst, and limited cardioprotection beyond minutes to hours), drug-delivery barriers (poor penetration into injured myocardium, rapid clearance and off-target distribution, and inadequate timing and local retention), and regenerative bottlenecks (repair impaired by persistent inflammation, cell engraftment limited by extracellular-matrix remodeling, and insufficient endogenous regeneration)

Limitations of current therapeutic strategies

Current standard-of-care treatments for MI—including coronary artery bypass grafting (CABG), percutaneous coronary intervention (PCI), thrombolytic therapy, antiplatelet and anticoagulant drugs, β-blockers, ACEI/ARB agents, and statins—are limited by fundamental structural constraints that prevent precise modulation of the pathological microenvironment.

The double-edged nature of reperfusion therapy remains the primary challenge. Although coronary blood flow is rapidly restored by PCI, reperfusion is accompanied by electron leakage from the mitochondrial respiratory chain, which causes ROS production to increase by 10–100-fold relative to the ischemic period and thereby overwhelms endogenous antioxidant capacity [9]. Calcium overload simultaneously activates calcium-dependent proteases, promotes opening of the mitochondrial permeability transition pore (mPTP), and induces cytochrome c release, thereby initiating the apoptotic cascade. Secondary injury caused by reperfusion has been estimated to account for approximately 50–75% of the final infarct size, making the reconciliation of revascularization with myocardial protection a persistent clinical paradox [9, 10].

Conventional pharmacotherapy is similarly limited in its capacity to promote tissue repair. Although prognosis is improved through antiplatelet, lipid-regulating, and ventricular-remodeling-inhibitory mechanisms, necrotic myocardium cannot be regenerated, nor can the complex inflammatory microenvironment be precisely regulated. The clinical translation of antioxidants (e.g., vitamins C and E) and natural enzymes (e.g., superoxide dismutase, SOD) has repeatedly failed. Intravenously administered SOD has a half-life of only 6 min, making sustained therapeutic blood concentrations difficult to maintain; small-molecule antioxidants have bioavailability below 5% and penetrate mitochondria poorly, despite mitochondria being the principal site of ROS generation. In addition, nonselective systemic antioxidant administration may disrupt physiological ROS signaling, including HIF-1α stabilization and the regulation of cell proliferation, thereby causing systemic adverse effects [11, 12]. More fundamentally, the post-MI microenvironment is highly heterogeneous: pH differences of 0.5–1.0 units can occur between the ischemic core and border zone, ROS concentrations span 2–3 orders of magnitude, and inflammatory-cell infiltration changes dynamically over time. Therapeutic strategies capable of sensing and responding to these changes—rather than relying on conventional one-size-fits-all dosing paradigms—are therefore required.

Regenerative-medicine approaches have encountered similar translational barriers. Although mesenchymal stem cells (MSCs) possess paracrine and immunomodulatory activity, their clinical application has been limited by several bottlenecks: more than 90% of transplanted cells die within 24 h because of hypoxia, inflammation, and mechanical shear; homing efficiency remains below 1%, with most cells being retained in peripheral organs such as the lungs and liver; batch heterogeneity complicates standardized quality control; and potential tumorigenicity hinders long-term safety assessment [13]. It is worth emphasizing that the clinical evidence for cell therapy in myocardial infarction remains controversial and of limited efficacy. As one of the largest randomized controlled studies in this field, the BAMI trial included patients with acute myocardial infarction with left ventricular ejection fraction (LVEF) ≤ 45%, and randomly assigned them to standard treatment alone or coronary artery infusion of autologous bone marrow mononuclear cells (BMCs). This study used all-cause mortality as the primary endpoint and was prematurely terminated. Ultimately, it was confirmed that cell therapy could not significantly reduce the mortality rate of patients [14, 15]. Similarly, the phase II CONCERT-HF trial conducted a double-blind, placebo-controlled study on patients with chronic ischemic heart failure, randomly assigning them to the autologous bone marrow MSCs treatment group, the c-kit positive cardiac progenitor cells (CPCs) treatment group, the combined treatment group, and the placebo group [16]. Although this study confirmed the overall safety of cell therapy, no significant improvement was observed in core efficacy endpoints such as left ventricular ejection fraction and left ventricular end-diastolic volume at 12 months. Only the combined treatment showed a slight advantage in some exploratory subjective prognostic indicators [16]. In summary, current high-quality clinical trials consistently indicate that the clinical efficacy of current cell regenerative therapies for cardiac repair after myocardial infarction is still very limited and unstable. Cell-free therapies, including small extracellular vesicles (sEVs) such as M2 macrophage-derived sEVs (M2-sEVs), produced promising results in animal models by promoting M1-to-M2 macrophage polarization, reducing infarct size, and improving cardiac function [17]. Their clinical translation, however, has been restricted by extremely low yields (10⁹–10¹⁰ particles per milliliter of culture medium), pronounced heterogeneity in size, cargo, and surface markers, poor in vivo stability with clearance occurring within hours after intravenous injection, and unresolved concerns regarding immunogenicity [17].

Taken together, these limitations define four critical unmet technical needs: (1) targeted delivery across biological barriers, including the vascular endothelium and cardiomyocyte and mitochondrial membranes; (2) controlled release that is temporally matched to pathological events, such as peaks in oxidative stress and inflammation and the onset of fibrosis; (3) integrated, multimodal synergy involving antioxidant, anti-inflammatory, pro-angiogenic, and anti-apoptotic effects; and (4) theranostic integration of diagnostic imaging, therapeutic intervention, and efficacy assessment.

The rise of nanomedicine and the MOF paradigm

The emergence of precision nanomedicine has provided a potentially transformative approach to overcoming these bottlenecks. Metal–organic frameworks (MOFs) are crystalline porous materials that are self-assembled from metal nodes and organic ligands [18]. Their high specific surface areas, high porosity, and tunable pore sizes allow the framework to be tailored to the physicochemical properties of a target molecule [19]. For example, a drug-loading capacity of 57.93% was achieved with IRMOF-1 when its pore size was matched to the oridonin (ORI) molecule, and approximately 90% of the drug was released within 24 h [20]. γ-CD-MOFs also displayed exceptional encapsulation efficiency for hydrophobic drugs with molecular weights of approximately 300 Da, reaching an experimental loading of 143.0 mg drug/g MOF; halogen atoms and benzene-ring structures were identified as major contributors to the enhanced encapsulation [21]. The structural programmability of MOFs can be further expanded by introducing different metal centers (Zn, Zr, Fe, or Cu) or functionalized organic ligands, through which specific catalytic, imaging, or therapeutic functions can be conferred [22]. Zeolitic imidazolate frameworks (ZIFs) constructed using zinc nucleotide gel templates were used to encapsulate enzymes in situ, which significantly improved their stability and catalytic performance in acidic environments [23]. This build-to-order feature makes MOFs particularly suitable for the construction of multifunctional biomedical platforms. The structural programmability of MOFs and their diverse biomedical applications are shown in Fig. 2.

Fig. 2.

Structural programmability of MOFs and the cross-disease transfer of technology (Schematic diagram). MOFs are assembled from metal nodes (Zn, Zr, Fe, and Cu) and organic ligands into crystalline porous frameworks. Their principal structural attributes include tunable pore sizes, ultra-high specific surface areas, and programmable surface chemistries. These features have supported diverse biomedical applications, including cancer therapy (drug delivery, chemodynamic therapy [CDT], and photodynamic therapy [PDT]), radiation protection (ROS scavenging), antibacterial treatment (ZIF-8 hydrogel), inflammatory bowel disease, and bone disease. The lower arrow indicates the transfer of MOF-based nanomedicine technologies to cardiovascular disease, specifically myocardial infarction (MI) therapy

Although the application of MOFs to MI therapy remains at an early stage, their use in cancer and other diseases has provided important technical precedents. In oncology, MOFs have been widely used for intelligent drug delivery, combination therapy, and theranostic applications. Doxorubicin (DOX) was loaded onto Cu-TCPP nanosheets through π–π stacking, and a drug-loading rate of 33% was achieved while glutathione (GSH)-scavenging activity, antitumor efficacy, and biocompatibility were retained [24]. Carboplatin was delivered by PEGylated iron-doped MOF-808 nanocomposites that also exhibited magnetic resonance imaging (MRI) potential, illustrating the promise of multimetal-doped MOFs for multimodal therapy [25]. MOFs have also been developed as carrier-free nanodrugs, in which metal ions such as Fe or Cu catalyze Fenton reactions for chemodynamic therapy (CDT), or porphyrin-based ligands mediate photodynamic or photothermal therapy [22]. Beyond oncology, MOFs have been investigated for radiation protection, antibacterial treatment, inflammatory bowel disease, and bone disease. ROS was efficiently scavenged by PEGylated MIL-101(Cr) loaded with WR-1065, and the 30-day survival rate of irradiated mice was increased to 90% [26]. Potent antibacterial activity was also demonstrated by nZIF-8@PAM/starch hydrogel composites, with IC₅₀ values as low as 0.51 ± 0.05 µg/mL [27]. Collectively, these cross-disease studies support the multifunctionality of MOFs and provide a technical foundation for their application to cardiovascular disease [22, 28].

It is instructive to position MOFs relative to more established nanocarriers. Liposomes and polymeric micelles are clinically mature with favorable biocompatibility and controllable sustained or triggered release, but their drug-loading capacity is generally limited (typically 5–20 wt%), they lack intrinsic catalytic or imaging activity, and they are prone to opsonization and rapid reticuloendothelial clearance after intravenous administration [29, 30]. Inorganic nanoparticles (e.g., gold, iron oxide, silica) provide robust imaging or magnetic properties but often suffer from slow, incomplete biodegradation, potential long-term hepatic and splenic accumulation, and comparatively modest payload loading [31]. Against this backdrop, MOFs combine several attributes that are particularly valuable for MI therapy: ultra-high and tunable drug loading (frequently exceeding 20–60 wt%), intrinsic metal-node-derived catalytic activity (e.g., SOD- and catalase-mimetic ROS scavenging), stimulus-responsive framework degradation that enables spatiotemporally controlled release, and the capacity to integrate imaging and therapy within a single theranostic platform. These properties map directly onto the four unmet needs identified above. Nevertheless, MOFs also carry important disadvantages—including limited physiological stability in some frameworks, potential toxicity of metal ions and organic ligands, and current difficulties in scalable, reproducible manufacturing—which are discussed in Sect. 4. Table 1 summarizes the comparative positioning of MOFs against liposomes, polymeric micelles, dendrimers, and inorganic nanoparticles across drug-loading capacity, stimulus-responsiveness, biocompatibility and degradability, theranostic integration, scalability, and cost.

Table 1.

Comparison of major performance characteristics between MOFs and classic nanocarriers for myocardial infarction therapy

Nanocarrier	Drug-loading capacity	Stimulus responsiveness	Biocompatibility and degradability	Theranostic potential	Scalable manufacturing	Cost	Core advantages	Main limitations
MOFs	High, 20–60 wt%, tunable	Strong; responsive to pH, ROS, enzymes and other microenvironmental cues	Tunable; partial frameworks are degradable; some frameworks suffer from insufficient physiological stability	Excellent; intrinsic metal-node-based catalytic/imaging functions, integrated theranostics in one platform	Moderate; high challenges in batch reproducibility and scale-up synthesis	Relatively high	Ultrahigh drug loading; intrinsic enzyme-mimetic catalytic ROS scavenging; spatiotemporally controlled release; modular multifunctional design	Poor physiological stability of some frameworks; potential toxicity of metal ions/organic ligands; difficulties in scalable production
Liposomes	Low–moderate, 5–20 wt%	Good; pH/enzyme/temperature-responsive systems well-established	Favorable, clinically approved	Limited; extra contrast agents required	Fully mature, industrialized production	Moderate	Excellent biocompatibility; sufficient in vivo safety data; mature drug encapsulation technology	Low loading capacity; no intrinsic catalytic activity; prone to RES clearance after intravenous injection
Polymeric micelles	Low–moderate, 5–15 wt%	Good; multiple stimulus-responsive polymers have been developed	Favorable; biodegradable polymers are available	Limited; additional conjugation of imaging probes required	Good, scalable	Moderate	Long-circulation potential; facile surface modification; low cytotoxicity	Low maximum loading; no intrinsic catalytic activity; disassembly upon dilution
Dendrimers	Moderate, 10–25 wt%	Good; end-group modification enables stimulus response	Moderate; high generations induce obvious cytotoxicity	Acceptable; drugs and imaging moieties can be conjugated simultaneously	Poor, tedious synthesis, hard to scale up	High	High structural homogeneity; precise multivalent modification	Complex synthesis; high cytotoxicity at high generations; risk of in vivo accumulation
Inorganic nanoparticles (Au, iron oxide, silica)	Low–moderate, 5–20 wt%	Weak; only a few pH-responsive systems available	Poor; slow and incomplete biodegradation	Excellent, inherent MRI/CT/photoacoustic imaging capacity	Good	Moderate	Superior imaging performance; photothermal/magnetothermal functions	Long-term hepatic and splenic accumulation; poor biodegradability; limited drug loading

Scope and unique contributions of this review

Several reviews have discussed MOFs in biomedicine or nanomedicine in general [19, 32–34]. The present review differs from these predecessors in two respects. First, to our knowledge, it is the first review to organize MOF design strategies—stimulus-responsive delivery, biomimetic nanoplatforms, theranostic systems, and biohybrid constructs—specifically around the post-infarction pathological microenvironment and its four therapeutic windows (Fig. 1), rather than around MOF chemistry alone. Second, beyond summarizing existing studies, the review proposes a forward-looking, testable research agenda with recommended experimental designs (Sect. 5.6). These distinguishing features position this work not merely as a literature synopsis but as a practical roadmap to guide future investigations in this field.

Functional design strategies of MOFs for MI treatment

To address the unmet needs identified above, MOFs can be engineered through multiple structural strategies—stimulus-responsive drug delivery, biomimetic nanoplatforms, theranostic systems, and biohybrid constructs—that are systematically analyzed in Fig. 3 with respect to their capacity to modulate the post-infarction microenvironment. Representative systems are compared side by side in Table 2 (composition, cargo/function, targeting or responsive mechanism, administration route, disease model, outcome, and key limitation).

Fig. 3.

Functional design strategies of MOFs for MI therapy. (a) Stimulus-responsive MOFs (Schematic diagram). Controlled drug release from MOF carriers is triggered by pathological features of the post-infarction microenvironment, including elevated ROS, reduced pH in the acidic MI zone, and external stimuli such as light or ultrasound. (b) Biomimetic MOFs. MOFs are coated with cell membranes to inherit source-cell surface properties and evade immune clearance; mannan-decorated MOFs are used to target macrophage mannose receptors in inflamed myocardium; and MOFs are embedded in extracellular-matrix (ECM) fibrous matrices to mimic the native myocardial microenvironment. (c) Theranostic integration. Imaging (e.g., MRI), therapeutic delivery, and real-time biosensing or monitoring are integrated into a single MOF platform. (d) Biohybrid systems. Synergistic cardiac repair is achieved through multicomponent constructs in which ZIF-8 hydrogel scaffolds, stem-cell encapsulation, and microneedle patches are combined for localized and sustained delivery to infarcted myocardium

Table 2.

Summary comparison of representative MOF systems discussed in Sects.  2 and 3

MOF / material composition	Therapeutic cargo or function	Targeting / responsive mechanism	Administration route	Disease / animal model	Major therapeutic outcome	Key limitation
ZIF-8 zinc nanozyme injectable hydrogel[35]	Zn²⁺ release; SOD/CAT-mimetic ROS scavenging	ROS-responsive; sustained Zn²⁺ release	Local (injectable hydrogel)	Rat MI	Reduced ROS; M2 polarization; promoted angiogenesis; improved cardiac function	Hydrogel injectability and scale-up; long-term Zn toxicity unassessed
Que@MOF/Man[36]	Quercetin; antioxidant and anti-inflammatory	Mannan modification → mannose-receptor (macrophage) targeting	Intramyocardial / intrapericardial	Rat acute MI	Reduced oxidative stress; M2 polarization; attenuated remodeling	Targeting efficiency; manufacturing scale-up
SOD-ZrMOF[47]	SOD cargo; antioxidant	Zr-MOF surface conjugation	Systemic / local	Mouse MI	ROS scavenged; mitochondrial protection; reduced infarct; improved function	In vivo enzyme stability; long-term Zr clearance
Cu-Zn-MOF (CZM) [39]	Cu/Zn nodes; ROS scavenging + iron chelation; ferroptosis inhibition	Bimetallic catalytic activity	Systemic	DOX-induced cardiotoxicity (non-MI)	Antioxidant; ferroptosis inhibition; cardioprotection exceeding dexrazoxane	Not yet evaluated in MI models
VB@MOF/TA[49]	VB (verbascoside) cargo; mitochondria-targeted antioxidant/anti-inflammatory	Mitochondria-targeting ligand; tannic acid	Systemic	MI/I-R (in vitro; in vivo)	2.59-fold reduced apoptosis; M2 polarization; 2.50-fold vascular density	Mitochondrial targeting efficiency; in vivo kinetics
Cu-TCPP-Mn bimetallic nanozyme[48]	SOD/CAT-mimetic ROS scavenging	Catalytic activity of Cu/Mn nodes	Systemic	MI and I/R injury	Reduced inflammation; myocardial protection	Long-term Mn toxicity
Zn-MOF (mitochondria-targeting peptide) [50]	Zn²⁺; PI3K/Akt/eNOS activation	Mitochondria-targeting peptide; ultralow dose	Systemic	Hindlimb ischemia (non-MI)	Improved perfusion; 100-fold dose reduction	Hindlimb-ischemia model, not MI
ZIF-8 exoskeleton-coated hucMSCs[45]	Cell protection; preserved paracrine activity	Cell-in-shell partial coating	Local transplantation	Rat MI	Enhanced retention; improved LVEF; reduced infarct	Long-term cell viability in vivo
ZIF-8/PAM-starch hydrogel[27]	Antibacterial; pH-responsive delivery	pH-responsive degradation	Local hydrogel	Non-cardiac (anti-infective proof-of-concept)	Antibacterial activity	Not evaluated in MI; proof-of-concept
Fe-MIL-88[42]	miRNA biosensing (miR-21/499/133a)	Nonenzymatic cascade amplification	In vitro diagnostic	Acute-MI-related miRNA detection	Ultrasensitive detection (13–40 pM)	Diagnostic proof-of-concept; not therapeutic

Stimulus-responsive MOFs

The high porosity, tunable pore sizes, and stimulus-responsive properties of MOFs have been widely exploited to construct intelligent drug-delivery systems capable of on-demand release for MI therapy. The local post-MI microenvironment is characterized by marked oxidative stress, and abnormally elevated ROS levels provide endogenous triggers for ROS-responsive MOFs. ZIF-8-based zinc nanozyme injectable hydrogels were reported to scavenge excess ROS in the infarct zone by mimicking superoxide dismutase and catalase activities and to permit the controlled release of drugs or metal ions in response to ROS, thereby modulating inflammation and promoting angiogenesis [35]. In Que@MOF/Man nanomedicines, quercetin was selectively delivered to inflammatory myocardial sites through mannan modification; intracellular ROS in macrophages was neutralized, oxidative stress was alleviated, and cardiomyocyte survival was improved [36]. Local acidosis provides another relevant trigger because acid-sensitive MOFs such as ZIF-8 dissociate at low pH, allowing site-specific drug release to be achieved [23]. Although light-, ultrasound-, and enzyme-responsive MOFs have been extensively investigated in cancer therapy, their use in MI has remained preliminary. Nevertheless, their spatiotemporally controlled release capabilities provide a promising basis for future precision intervention [22, 37].

Biomimetic MOFs

To improve in vivo biocompatibility and targeting, biomimetic strategies have been incorporated into the functional design of MOFs. Through cell-membrane camouflage, the surface properties of source cells can be inherited by MOFs, allowing immune clearance to be reduced and lesion accumulation to be enhanced [38]. Recognition of injured myocardium may also be conferred by myocardium-specific membrane coatings. In the Que@MOF/Man system [36], targeted delivery to macrophages with highly expressed mannose receptors in inflamed myocardium was achieved through mannan modification. Here, mannan—a polysaccharide ligand that is specifically recognized by the mannose receptor (CD206) highly expressed on macrophages within inflamed myocardium—serves as the targeting moiety, illustrating a precise delivery strategy mediated by biomimetic ligands. MOFs have also been combined with extracellular-matrix (ECM) components to mimic the native myocardial microenvironment. As a transferable design concept developed outside the MI setting—in an antibacterial photodynamic therapy study—the nZIF-8@PAM/starch material was formed by incorporating ZIF-8 into polyacrylamide/starch hydrogels, and its self-adhesion and viscoelasticity reproduced key physicochemical features of the ECM while providing structural support and bioactive cues for tissue repair [27]. The biomimetic Cu-Zn-MOF (CZM) exhibited both ROS-scavenging and iron-chelating activities and protected myocardium by maintaining mitochondrial homeostasis [39]. It should be noted that this system was evaluated in a doxorubicin-induced cardiotoxicity model rather than in an MI model, and its relevance to MI is therefore conceptual.

Theranostic systems

The structural tunability of MOFs enables the integration of imaging and therapeutic functionalities to construct versatile theranostic platforms. Benefiting from their ultrahigh specific surface areas and abundant open metal sites, MOFs can efficiently load various contrast agents; additionally, the MOF framework itself can endow the system with inherent imaging capabilities. MOF materials doped with manganese, iron, or copper have been extensively exploited for biomarker detection, molecular imaging, and combinatorial drug delivery [22, 40]. For instance, MIL-88B-NH/Pt nanocomposites have realized colorimetric and fluorescent detection of specific enzymes, though this work only provides an extracardiac proof-of-concept for biosensing applications [41]. Similarly, Fe-MIL-88 platforms have achieved the ultrasensitive in vitro detection of microRNAs (miR-21, miR-499, and miR-133a) closely associated with acute myocardial infarction (AMI) [42]. Notably, the strategy of integrating such biosensing modules with therapeutic units to achieve dynamic, real-time monitoring of inflammation, oxidative stress, and angiogenesis progression in infarcted myocardium remains a conceptual framework rather than a maturely validated technology, which requires further verification in cardiac disease models. The combination of imaging functions with gas or small-molecule drug delivery systems allows synchronous therapeutic intervention and efficacy evaluation, which greatly promotes the advancement of precision cardiovascular nanomedicine [43].

Biohybrid systems

Multicomponent biohybrid systems can be constructed by combining MOFs with biomaterials or cell-based therapies, thereby enhancing myocardial repair through complementary mechanisms. Hydrogels provide three-dimensional mechanical support and sustained release of bioactive components, whereas MOFs confer catalytic, drug-loading, and ion-releasing functions. Injectable multifunctional ZIF-8 hydrogels were reported not only to scavenge ROS and regulate macrophage polarization but also to release zinc ions through gradual degradation, which continuously promoted angiogenesis and improved cardiac function [35]. As noted in Sect. 2.2, ZIF-8 nanocrystals were incorporated into polyacrylamide/starch hydrogels to form composite platforms with pH responsiveness and antibacterial activity [27]; although this platform was developed for localized anti-infective treatment, the underlying design concept could be adapted to myocardial repair.

The combination of MOFs with stem-cell therapy is also promising. Although MSCs have been shown to improve post-MI cardiac function and promote angiogenesis [44], their therapeutic efficacy is limited by poor survival and low retention after transplantation. To address these limitations, an exoskeleton partial-coating strategy was developed by He et al., in which human umbilical cord mesenchymal stem cells (hucMSCs) were partially encapsulated with ZIF-8 nanoparticles [45]. In this cell-in-shell structure, the cells were protected from mechanical injury and oxidative stress, while nutrient exchange and paracrine-factor secretion were preserved by the partial coating. In a rat MI model, cardiac retention was enhanced, left ventricular ejection fraction was improved, infarct size was reduced, and neovascularization was increased at 28 days after transplantation [45]. Stem-cell survival and paracrine activity may also be enhanced by MOF-mediated modulation of the local microenvironment, including the alleviation of oxidative stress and suppression of excessive inflammation.

Microneedle patches incorporating nanoparticles provide an innovative strategy for localized delivery in MI therapy. A nanoparticle–patch system was constructed by Chen et al. for the localized, effective, and sustained administration of miRNAs to infarcted myocardium [46]. Through sustained release and local action, therapeutic miRNAs were precisely delivered to the ischemic region, and myocardial ischemia–reperfusion injury was alleviated. Such biohybrid patch systems may represent an emerging paradigm in which MOF or nanoparticle functions are combined with structural support to enhance cardiac repair.

Spatiotemporal matching of MOF delivery to the therapeutic windows

The four therapeutic windows defined in Fig. 1 provide a rational framework for aligning MOF administration timing and release kinetics with the evolving pathology of MI. Because the administration route determines the spatiotemporal distribution of a nanoplatform in vivo—and hence the dose and duration of exposure actually achieved at the infarct site—window matching must be considered along two dimensions: when a given MOF system is delivered and how it is delivered.

For the reperfusion window (0–2 h) and the antioxidant window (2–6 h), the requirement is immediate and quantitative ROS interception. Systems such as SOD-ZrMOF [47], the Cu-TCPP-Mn nanozyme [48], the ZIF-8 nanozyme hydrogel [35], and Que@MOF/Man [36] are therefore most rationally administered before or during reperfusion (e.g., as an intravenous bolus at the time of PCI, or as a pre-reperfusion intramyocardial injection in surgical models), so that circulating nanoparticles are present in the coronary bed when the ROS burst occurs. Mitochondria-targeted systems such as VB@MOF/TA [49] and mitochondria-targeting-peptide-modified Zn-MOFs [50] must additionally cross the cardiomyocyte and mitochondrial membranes within this narrow period, which imposes constraints on particle size, surface charge, and targeting-ligand density. In this hyperacute phase, intravenous administration is the only route compatible with the clinical reperfusion workflow; yet after conventional intravenous injection, MOFs are readily cleared by the mononuclear phagocyte system of the liver and spleen, and cardiac accumulation is limited [29, 30]. Even when a ROS-scavenging MOF is administered intravenously at reperfusion, the fraction that actually reaches the infarcted myocardium may therefore be too small to achieve effective concentrations during the brief ROS burst. Surface modifications such as PEGylation or cell-membrane camouflage can prolong circulation and alter organ distribution [38, 51], but the magnitude of their improvement in cardiac uptake has not been quantified in MI models. Matching a system to these early windows is thus constrained not only by its intrinsic catalytic activity but also by a delivery gap that current pharmacokinetic data cannot yet close.

For the anti-inflammatory window (6–48 h), rapid ROS clearance alone is insufficient; sustained modulation of macrophage phenotype is required. Here the degradation kinetics of the framework become critical: the gradual Zn²⁺ release and framework erosion of ZIF-8-based hydrogels [35] and the sustained availability of mannan-targeted quercetin [36] are well matched to this multi-day window, whereas a burst-release system would be poorly suited. For the anti-fibrotic window (48–72 h) and the chronic phase, pro-angiogenic systems such as Zn-MOFs acting through the PI3K/Akt/eNOS pathway [50] and the vascular-density enhancement observed with VB@MOF/TA [49] are most relevant. For these later windows, local depot formulations (hydrogels, patches) may outperform systemic administration in maintaining exposure over days; however, whether the release kinetics of such depots are actually aligned with the windows has not been systematically validated. Key unresolved questions include whether Zn²⁺ release from a ZIF-8-based hydrogel can cover the 48–72 h anti-fibrotic window rather than being exhausted earlier, and to what extent the release profiles of MOF–hydrogel composites are modulated by the infarct microenvironment (e.g., local pH, enzyme activity, and mechanical stress). Because these later windows tolerate a longer interval between administration and effect, they also permit delivery routes that are impractical in the hyperacute phase, including intramyocardial and intrapericardial injection; the procedural feasibility and translational aspects of these routes are discussed in Sect. 4.3.

We emphasize that this kind of matching is still largely theoretical in nature. Quantitative pharmacokinetic/pharmacodynamic (PK/PD) data on the release kinetics, cardiac uptake, and infarct-to-remote distribution of MOFs in MI models are currently lacking, and obtaining such data should be regarded as a prerequisite for rational window matching rather than a detail to be settled after efficacy is established. Addressing this gap is directly reflected in the research agenda proposed in Sect.  5.6 (Hypotheses H1, H3, and H4).

Mechanisms of MOF-Mediated cardiac repair

The protective effects of MOFs against post-infarction injury converge on three principal pathways—oxidative stress regulation, inflammatory microenvironment reprogramming, and angiogenesis promotion—that are depicted in Fig. 4 and critically evaluated below. Throughout this section we link each observed biological effect to the specific structural feature (metal node, ligand, pore architecture, surface modification, degradation behavior, or loaded cargo) that is responsible, and we distinguish direct evidence obtained in MI or ischemia–reperfusion (I/R) models from transferable concepts established in other cardiac or non-cardiac settings. For ease of comparison, Table 2 summarizes the key attributes of the representative MOF systems discussed in Sects.  2 and 3.

Fig. 4.

Multimechanistic synergy of MOF-mediated cardiac repair from the ischemic or infarct core to functional tissue (Schematic diagram). The pathological microenvironment of the ischemic or infarct core is shown in the left panel and is characterized by ROS accumulation, damaged mitochondria, apoptotic cardiomyocytes, and M1 macrophage infiltration. Three convergent pathways are targeted by MOF nanoparticles: (1) oxidative-stress regulation, in which ROS is scavenged and SOD/CAT-mimetic activity is used to protect mitochondrial-membrane integrity; (2) inflammatory-microenvironment reprogramming, in which drug or ion release drives M1-to-M2 macrophage polarization and reduces pro-inflammatory cytokines (IL-1β and TNF-α); and (3) angiogenesis promotion, in which activation of the Zn²⁺/PI3K-Akt-eNOS pathway induces endothelial-cell tube formation and neovascularization. The repaired tissue shown in the right panel contains newly formed vessels, viable cardiomyocytes, and restored myocardial architecture. Key cellular and molecular components are identified in the lower legend

Oxidative stress regulation and mitochondrial protection

After MI, ischemia–reperfusion causes massive ROS accumulation, which aggravates oxidative stress, damages mitochondrial structure and function, and triggers cardiomyocyte apoptosis. Owing to their high specific surface areas and functionalizable structures, MOFs are well suited for oxidative-stress regulation and mitochondrial protection. In animal models, ROS was efficiently scavenged by zirconium-based MOFs loaded with superoxide dismutase (SOD-ZrMOF); oxidative stress was suppressed, mitochondrial function was preserved, cell death and infarct size were reduced, and cardiac function was improved [47]. Here the antioxidant effect is conferred by the SOD cargo anchored to the Zr-MOF surface, illustrating a cargo-conjugation design.

A copper–zinc bimetallic MOF (Cu-Zn-MOF) displayed strong antioxidant activity in a doxorubicin-induced cardiotoxicity model (rather than an MI model). ROS was eliminated, iron ions were efficiently chelated by the bimetallic framework, the expression of mitochondrial respiratory complexes was upregulated, and ferroptosis was alleviated through inhibition of the AMPK/mTOR signaling pathway; the resulting cardioprotection exceeded that provided by dexrazoxane [39]. The catalytic and iron-chelating functions of this system arise directly from its Cu/Zn bimetallic nodes. Because the Cu-Zn-MOF was evaluated in a DOX-induced cardiotoxicity model rather than an MI model, it constitutes transferable evidence of mitochondrial protection that remains to be confirmed in MI or I/R models.

In the VB@MOF/TA nanosystem, tannic acid (TA) exhibits mitochondrial targeting, and its antioxidant capacity is markedly augmented, achieving effective suppression of cardiomyocyte apoptosis in vitro [49]. Such mitochondrial enrichment and amplified antioxidant performance stem from mitochondria-targeted surface functionalization rather than the intrinsic properties of the MOF framework. At ultralow dosages, peptide-modified Zn-MOFs with mitochondrial targeting capability activate the PI3K/Akt/eNOS pathway [50]. This intervention ameliorates mitochondrial function within ischemic lesions, facilitates angiogenesis, improves bioavailability, and mitigates toxic risks. Consistently, mitochondria-targeted Zn-MOFs markedly restore blood perfusion in ischemic tissue in a hindlimb ischemia model at a 100-fold lower dose. Zn²⁺ liberated from the metal nodes of Zn-MOFs serves as the effector molecule triggering PI3K/Akt/eNOS signaling, while the mitochondria-targeting peptide enables this 100-fold dose reduction. Notably, this investigation was implemented in a hindlimb ischemia model, a noncardiac ischemic setting. Although its outcomes are deemed potentially translatable to myocardial ischemia, validation in myocardial infarction (MI) models remains pending. Taken together, these findings demonstrate that MOFs can synergistically alleviate oxidative stress and mitochondrial damage via multifaceted mechanisms, favoring myocardial repair.

Inflammatory microenvironment reprogramming

Appropriate initiation and timely resolution of the inflammatory response after MI are essential for tissue repair. Myocardial injury is aggravated by excessive or persistent inflammation, whereas an anti-inflammatory and reparative microenvironment is promoted by M2 macrophage polarization. Considerable potential for regulating macrophage phenotype has been demonstrated by MOFs. In the Que@MOF/Man system [36], mannan-mediated recognition targets quercetin to macrophages within inflamed lesions. This nanoplatform scavenges intracellular ROS, promotes reparative M2 macrophage polarization, remodels the inflammatory microenvironment, preserves cardiomyocyte viability, and mitigates ventricular remodeling. Surface-conjugated mannan enables specific binding to mannose-receptor-expressing macrophages, whereas the encapsulated quercetin cargo exerts antioxidant activity to trigger M2 polarization—illustrating a canonical design–function paradigm of ligand–cargo engineering.

With the VB@MOF/TA system, cardiomyocyte apoptosis was reduced by 2.59-fold and macrophage polarization was markedly modulated, producing synergistic anti-inflammatory and reparative effects [49]. The dual action here stems from the mitochondria-targeted delivery of the VB cargo together with tannic-acid-based anti-inflammatory surface chemistry. A ZIF-8-based zinc nanozyme hydrogel mimicked superoxide dismutase (SOD) and catalase (CAT) activities. Excess ROS was removed from the infarct area, the ROS-driven inflammatory cascade was inhibited, immunoregulatory activity was enhanced through sustained zinc-ion release, macrophage polarization was shifted toward the M2 phenotype, and inflammatory-factor secretion was reduced [35]. Here the SOD/CAT-mimetic activity derives from the Zn-based framework nodes, whereas the sustained immunomodulatory effect is attributable to gradual zinc-ion release during hydrogel degradation—a degradation-controlled design. SOD-like and CAT-like activities were also exhibited by the Cu-TCPP-Mn bimetallic nanozyme, and inflammation was significantly alleviated while myocardial tissue was protected in MI and ischemia–reperfusion injury models [48]. The catalytic activity of this bimetallic system is a direct property of its Cu and Mn nodes, and—in contrast to the Cu-Zn-MOF discussed above—its efficacy has been confirmed in direct MI and I/R models. These findings showed that macrophage polarization and the inflammatory microenvironment could be precisely modulated by MOFs through ROS scavenging, anti-inflammatory drug delivery, or functional metal-ion release.

Angiogenesis promotion and cardiomyocyte regeneration

Neovascularization after MI is essential for restoring local oxygen supply, sustaining viable myocardium, and supporting tissue regeneration. Angiogenesis can be promoted by MOFs through the delivery of pro-angiogenic factors or activation of relevant signaling pathways. Zn-MOFs were shown to induce vascular endothelial-cell morphogenesis through activation of the PI3K/Akt/eNOS pathway; after mitochondria-targeting modification, blood perfusion in ischemic regions was significantly improved even when the dose was reduced by 100-fold [50]. The pro-angiogenic signal is driven by Zn²⁺ released from the metal nodes, with the mitochondria-targeting peptide enhancing efficiency at ultralow doses; as noted above, this evidence derives from a hindlimb-ischemia model and should be confirmed in MI models. The SOD-ZrMOF system constructed by Guo et al. not only attenuated oxidative stress but also markedly promoted angiogenesis and inhibited pathological myocardial remodeling in a mouse MI model [47]. Here the antioxidant and pro-angiogenic effects both trace back to the SOD cargo conjugated onto the Zr-MOF framework, and they have been validated directly in an MI model. In vivo, a 2.50-fold increase in vascular density was observed in the VB@MOF/TA treatment group, indicating strong pro-angiogenic activity [49]. This effect is attributable to the mitochondrial-targeted antioxidant/anti-inflammatory cargo that stabilizes the vascular niche. Although direct cardiomyocyte regeneration induced by MOFs has not yet been established, a supportive environment for cardiomyocyte survival and potential regeneration may be created indirectly through the alleviation of oxidative injury, modulation of inflammation, and promotion of vascularization. Cardiac function was significantly improved and fibrosis was reduced by Que@MOF/Man treatment, suggesting a potential role in myocardial structural recovery [36].

Clinical translation challenges

Before MOF-based precision nanomedicines can reach clinical practice, formidable barriers in biosafety evaluation, scalable manufacturing, delivery-route optimization, and large-animal validation must be overcome (Fig. 5).

Fig. 5.

Critical challenges and evaluation frameworks for the laboratory-to-clinic translation of MOF-based nanomedicines (Schematic diagram). Progress from laboratory development to clinical application depends on four interconnected domains. Biosafety evaluation includes assessment of metal-ion and organic-ligand toxicity, clarification of in vivo metabolic pathways, profiling of long-term immunogenicity, and monitoring of organ accumulation. Scalable manufacturing requires batch consistency, standardized synthesis, physicochemical quality control (particle size, surface charge, and porosity), and stability validation in physiological media. Delivery-route optimization requires intravenous and intramyocardial injection to be compared, targeted surface modifications to be evaluated, therapeutic time windows to be defined, and transendothelial transport mechanisms to be elucidated. Large-animal validation requires standardized porcine or ovine MI models, functional assessment at multiple time points, integrated histological and imaging analyses, and cardiac electrophysiological evaluation

Biosafety assessment

Biological safety remains the primary barrier to the clinical translation of MOFs for MI therapy. Although favorable biocompatibility has been reported for some MOFs, toxic effects may be induced in vivo by their constituent metal ions and organic ligands. Cyclodextrin-based MOFs can achieve loading efficiencies as high as 97.80%, but their pharmaceutical use is limited by the potential toxicity of synthetic reagents and framework components [52]. Angiogenesis can be promoted by zinc-based MOFs (Zn-MOFs) through activation of the PI3K/Akt/eNOS pathway, but dose-dependent toxicity remains a concern [50]. The degradation of MOFs under physiological conditions must also be considered because rapid or uncontrolled degradation may reduce bioactivity and generate harmful byproducts [53]. In phosphate-containing media, the physiological stability of some nano-MOFs has been found to be poor, further increasing safety concerns [51]. A systematic framework for long-term toxicity assessment must therefore be established, and the in vivo metabolic pathways, organ accumulation, and immunogenicity of MOFs must be clarified before clinical translation can proceed [19].

Importantly, biosafety profiles are not uniform across MOF systems and must be assessed on a case-by-case basis according to metal identity, ligand chemistry, particle size, surface modification, and degradation rate. For example, iron- and zinc-based MOFs draw on essential micronutrients and are generally regarded as more readily tolerated, whereas cobalt-, nickel-, or chromium-containing frameworks (e.g., MIL-101(Cr)) raise greater concerns regarding chronic metal accumulation and potential carcinogenicity [19, 26]. Ligand chemistry is equally decisive: biologically derived or endogenous ligands (e.g., nucleotides, cyclodextrins, or amino acids) tend to degrade into nontoxic metabolites, whereas some synthetic imidazolate or carboxylate linkers may release cytotoxic fragments [52]. Particle size and surface modification influence both acute toxicity and pharmacokinetics—smaller, uncoated particles are cleared and accumulate in the liver and spleen more rapidly, whereas PEGylation or cell-membrane camouflage can prolong circulation and alter organ distribution; these modifications may also reduce immunogenicity [38, 51]. Faster-degrading frameworks may avoid long-term accumulation but can produce abrupt, high-concentration release of metal ions and ligands, whereas more stable frameworks carry a greater risk of sustained tissue retention. Such composition-dependent differences must be characterized systematically for each candidate before clinical translation.

Scalable manufacturing and quality control

Clinical translation of MOFs depends not only on biological performance but also on reproducible and stable large-scale manufacturing. At present, MOF synthesis is largely confined to small laboratory batches, and stringent clinical-grade requirements for batch consistency, purity, and physicochemical attributes—including particle size, morphology, porosity, and surface charge—cannot yet be reliably met [54]. Structural heterogeneity can readily be introduced by variations in temperature, solvent, or ligand ratio, thereby altering drug-loading efficiency and release behavior. Marked differences in drug loading and release were observed among UiO-66 and its functionalized derivatives (UiO-66-NH₂, UiO-66-NO₂, and UiO-66-OH), indicating that material performance is strongly influenced by functional groups [55]. Although mechanical stability and corrosion resistance can be improved by combining MOFs with materials such as bioactive glasses (BGs) [53], standardized fabrication protocols and quality-control criteria for such composites have not yet been established. Full-process quality standards covering raw-material control, synthesis, final-product characterization, and stability testing are urgently required to ensure the safety and efficacy of MOF formulations.

Administration route optimization

Therapeutic efficacy depends critically on whether MOFs can be delivered to and retained within infarcted myocardium. After conventional intravenous injection, MOFs are readily cleared by the liver and spleen, and cardiac accumulation is therefore limited. Several local and targeted delivery approaches have been investigated. In a rat model of acute MI, retention in the infarct region was prolonged by intramyocardial or intrapericardial injection [36]. Targeting efficiency was also improved through biomimetic modification: macrophages in inflamed infarcted myocardium were specifically recognized by mannose-modified Que@MOF/Man nanomedicine, allowing quercetin to be delivered precisely and oxidative stress and inflammation to be effectively regulated [36]. Mitochondria-targeting peptide modification reduced the effective dose of Zn-MOFs by 100-fold to 4.4 µg/kg, while vascularization in ischemic regions was significantly improved without muscular injury [50]. However, the distribution kinetics, transendothelial transport mechanisms, and optimal administration windows of MOFs in the cardiovascular system have not been systematically characterized. In large-animal models, the procedural feasibility, safety, and drug-diffusion range of pericardial injection must be further evaluated.

Large animal model validation

Although the therapeutic potential of MOFs for myocardial repair has been demonstrated in numerous rodent studies, efficacy and safety must be rigorously confirmed in large-animal models before clinical translation. Differences between small animals and humans in cardiac anatomy, hemodynamics, immune responses, and drug metabolism limit the predictive value of rodent models. In rat models, engineered mesenchymal stem-cell sheets combined with superparamagnetic iron oxide nanoparticles achieved cell retention for more than 28 days and enabled MRI tracking [56]; however, feasibility in porcine or non-human primate models has not been reported. The long-term biocompatibility of MOFs, clearance pathways of their degradation products, and effects on cardiac electrophysiology must all be systematically evaluated in models that more closely reproduce human physiology. Large-animal studies of MOFs for MI therapy remain extremely scarce. Standardized porcine or ovine MI models should therefore be established, and comprehensive assessments at multiple time points—including functional, histological, and imaging analyses—should be conducted to provide reliable evidence for subsequent clinical trials.

Future perspectives

To accelerate the bench-to-bedside translation of MOF-based precision nanomedicines, convergent directions involving conductive composites, multi-omics-guided design, organ-on-chip evaluation, 3D-bioprinted cardiac patches, and regulatory science are proposed (Fig. 6).

Fig. 6.

Integrated technology roadmap for MOF-based precision cardiovascular nanomedicine (Schematic diagram). Five convergent directions are presented for accelerating bench-to-bedside translation. In conductive composites, MOFs are integrated with MXene or conductive polymers to restore electrophysiological function. Multi-omics-guided design uses genomic, proteomic, and metabolomic data to inform rational MOF engineering. In organ-on-chip evaluation, cardiac microphysiological systems are used for real-time monitoring of the biological effects of MOFs, thereby reducing reliance on animal experiments. In 3D-bioprinted cardiac patches, MOFs are combined with biomimetic scaffolds to provide sustained therapeutic release and functional myocardial reconstruction. Regulatory science is used to establish quality-control standards, define toxicological endpoints, and delineate clinical translation pathways. The lower banner emphasizes that industry–university–research–clinical collaboration will be required to achieve these goals

Electrical signal conduction and multifunctional platforms

Scar formation after MI disrupts the continuity of myocardial electrical conduction and predisposes patients to arrhythmias. Although MOFs are not intrinsically conductive, recovery of electrical function may be facilitated indirectly when they are combined with conductive components or used to improve the myocardial microenvironment. In a rat MI model, a DPC/MXene composite hydrogel increased electrical conductivity, promoted the survival and proliferation of H9c2 cardiomyocytes, improved cardiac function, and increased ventricular-wall thickness, thereby demonstrating the value of conductive materials in myocardial repair [57]. PLA/DP/MXene electrospun patches likewise promoted cardiomyogenesis and angiogenesis by providing a conductive microenvironment [58]. No study has yet demonstrated that electrical conduction is directly improved by MOFs. Nevertheless, myocardial electrophysiological integrity may be maintained or restored indirectly through the alleviation of oxidative stress, inhibition of fibrosis, promotion of angiogenesis, and support of cardiomyocyte survival. Left ventricular ejection fraction was increased by 7.72% ± 3.71% after VB@MOF/TA treatment [49], and cardiac function was significantly improved by SOD-ZrMOF [47]; these functional benefits may have been accompanied by greater electrical stability. In the future, multifunctional platforms in which MOFs are integrated with conductive polymers or two-dimensional materials such as MXene could combine catalytic, drug-loading, and conductive properties and may provide a new strategy for restoring myocardial electrical conduction.

Multi-Omics-Guided precision design

With advances in high-throughput sequencing and multi-omics technologies, including genomics, transcriptomics, proteomics, and metabolomics, the molecular mechanisms of MI have increasingly been resolved, providing new guidance for precise MOF design. MOF carriers could be customized according to patient-specific biomarker profiles so that targeted recognition and individualized therapy can be achieved. Using an Fe-MIL-88 MOF platform, ultrasensitive detection of acute-MI-related microRNAs (miR-21, miR-499, and miR-133a) was achieved with detection limits of 13, 25, and 40 pM, respectively, demonstrating the diagnostic potential of MOFs [42]. In future studies, disease-microenvironment maps could be integrated with multi-omics data to guide ligand selection, pore-size regulation, and the design of responsive mechanisms, thereby improving targeting and therapeutic efficacy in myocardial repair.

Organ-on-chip technology for accelerated evaluation

Organ-on-chip systems reproduce key features of the human microphysiological environment and provide highly biomimetic platforms for the in vitro evaluation of novel biomaterials. Compared with conventional two-dimensional cell culture and animal models, cardiac chips can more faithfully reproduce the mechanical stimulation, electrophysiological activity, and inflammatory microenvironment of myocardial tissue, allowing the biological effects and biosafety of MOFs in MI therapy to be evaluated systematically. MOFs have been extensively investigated in preclinical studies of drug delivery, anti-inflammatory therapy, and tissue repair [32]; however, a dynamic evaluation system tailored to the cardiovascular system has not yet been established. By integrating MOFs into myocardial-chip models, their effects on cardiomyocyte survival, angiogenesis, and electrical conduction could be monitored in real time. The research-and-development cycle could thereby be shortened, reliance on animal experiments could be reduced, and preclinical screening could be made more efficient and reliable.

3D Bioprinting of myocardial patches

The development of 3D bioprinting has enabled complex tissue-engineering scaffolds to be fabricated with high precision. By exploiting the high drug-loading capacity and tunable structure of MOFs, myocardial patches that provide sustained drug release, mechanical support, and biological activity could be produced. The nZIF-8@PAM/starch composite hydrogel was developed as a pH-responsive drug-delivery platform with favorable mechanical strength and self-adhesion [27]. Such materials could be 3D-printed into biomimetic patches that conform to the infarct region and continuously release therapeutic factors. The mechanical stability and bioactivity of the scaffold may be further enhanced by combining MOFs with bioactive glass or other materials [53]. Future incorporation of conductive components or cell-loading strategies into 3D-printed MOF-based myocardial patches is expected to promote both structural reconstruction and functional recovery and represents an important direction for post-infarction tissue engineering.

Industrial translation pathways and regulatory science

Despite their promise for MI therapy, the clinical translation of MOFs is hindered by limitations in large-scale production, quality control, and regulatory guidance. Although diverse synthesis methods have been developed, batch-to-batch consistency, long-term stability, and in vivo metabolic pathways remain insufficiently defined [33]. A full-chain quality-management system should therefore be established to cover raw-material control, process validation, physicochemical characterization, and biological-safety assessment. MOF-specific regulatory guidance should be developed by adapting the review frameworks for nanomedicines issued by international drug-regulatory agencies, and toxicological endpoints, degradation-product monitoring, and administration-route specifications should be clearly defined [34]. Although local approaches such as pericardial injection can improve targeting, their procedural safety and clinical feasibility must be comprehensively evaluated in large-animal models. Ultimately, clinical implementation will require coordinated industry–university–research–clinical collaboration and standardized pathways for translation from the laboratory to the clinic.

Priority research agenda: testable hypotheses and proposed experimental designs

To move the field beyond summarization, we propose five high-priority, testable hypotheses, each with a recommended experimental design. These are intended as a practical roadmap for subsequent studies.

Hypothesis 1

(H1): MOF nanozymes administered at reperfusion can achieve on-demand, quantitative ROS scavenging within the 0–2 h window, thereby reducing infarct size beyond what reperfusion alone achieves. Proposed design: rat or porcine I/R model with intravenous bolus administration immediately before reperfusion; quantitative ROS measurement by electron paramagnetic resonance or spin-trapping in myocardial biopsies at 0.5, 1, 2, and 6 h; primary endpoints of infarct size (TTC/Evans blue) at 24 h and cardiac function (echocardiography) at 28 days; controls include empty MOF, free drug, and untargeted MOF.

Hypothesis 2

(H2): Among Zr-, Zn-, Fe-, and Cu-based MOFs, which framework provides the best balance between long-term in vivo safety and therapeutic efficacy in MI? Proposed design: 90-day rat study with repeated dosing at the therapeutic regimen, endpoints including organ accumulation by ICP-MS, hematology, serum biochemistry, histopathology, and anti-drug antibody/immunogenicity profiling; efficacy arm in a rat MI model comparing the same four frameworks head-to-head with identical cargo; controls include vehicle and clinically used antioxidant.

Hypothesis 3

(H3): Mitochondria-targeted MOFs can achieve a quantifiable upper limit of delivery efficiency to infarcted cardiomyocytes that substantially exceeds untargeted MOFs. Proposed design: MI model with fluorescence/PET dual-labeled mitochondria-targeted versus untargeted MOF; biodistribution quantification in heart, infarct border zone, and major organs at 1, 6, 24, and 72 h; mitochondrial colocalization by super-resolution microscopy or organelle fractionation; dose-escalation to define the delivery-efficiency plateau; controls include free peptide and non-targeted MOF.

Hypothesis 4

(H4): MOF-hydrogel systems with programmable, sequential cargo release can match the anti-inflammatory (6–48 h) and anti-fibrotic (48–72 h) windows more effectively than bolus delivery. Proposed design: porcine MI model, randomized to bolus versus sequential-release MOF-hydrogel; serial cardiac MRI and serum biomarker assessment; histological endpoints of macrophage phenotype, fibrosis (Masson staining), and angiogenesis at 7 and 28 days; controls include hydrogel alone and conventionally timed dosing.

Hypothesis 5

(H5): MOF-integrated conductive scaffolds can restore electrophysiological conduction across infarct scars and reduce arrhythmia susceptibility. Proposed design: rat MI model implanted with conductive MOF/MXene or MOF/conductive-polymer scaffold; endpoint assessment by epicardial mapping, optical mapping of conduction velocity, and programmed electrical stimulation for arrhythmia inducibility at 28 days; histological correlation of scaffold integration with conduction recovery; controls include non-conductive MOF scaffold and conductive scaffold without MOF.

Conclusion

Myocardial infarction remains a global healthcare challenge with high mortality and morbidity, and the limitations of current therapeutic strategies in precise regulation of the post-infarction pathological microenvironment highlight an urgent need for architecturally programmable therapeutic platforms. Metal-organic frameworks, as crystalline porous materials with molecularly tunable structures and versatile functionalities, have emerged as transformative candidates for precision nanomedicine in MI therapy. This review has systematically examined how the architectural design of MOFs—including stimulus-responsive drug delivery, biomimetic nanoplatforms, theranostic systems, and biohybrid constructs—enables targeted therapeutic agent delivery, microenvironment-responsive release, and synergistic reprogramming of the post-infarction niche.

The mechanistic underpinnings of MOF-mediated cardiac repair have been critically evaluated, encompassing oxidative stress regulation, inflammatory microenvironment reprogramming, and angiogenesis promotion. Multiple studies have confirmed that MOFs exert cardiac protective effects through convergent pathways: ROS scavenging, M2 macrophage polarization, pro-angiogenic signaling activation, and mitochondrial homeostasis maintenance. Nevertheless, clinical translation still faces formidable challenges, including biosafety evaluation, scalable manufacturing, administration route optimization, and large animal model validation—barriers intrinsically linked to MOF metal/ligand toxicity, synthesis heterogeneity, and suboptimal targeting efficiency.

Looking forward, the development of MOF-based precision nanomedicines will be advanced through multi-omics-guided architectural design, organ-on-chip accelerated evaluation, 3D-printed cardiac patches, and regulatory science development. Integration of materials science, molecular biology, and clinical medicine through industry-university-research-medical collaboration is expected to overcome current barriers and accelerate bench-to-bedside translation. We therefore conclude that MOFs represent an evolving and promising—but still preclinical—class of candidate nanomedicines for cardiovascular disease, and that their clinical value will be determined by rigorous biosafety evaluation, scalable manufacturing, and validation in large-animal models and, ultimately, clinical trials.

Author contributions

K. Liu: Conceptualization, Writing – Original Draft, Visualization, Supervision, Project Administration, Funding Acquisition. Y. Liu: Writing – Original Draft, Investigation, Data Curation, Visualization. A. Zhang: Conceptualization, Writing – Review & Editing, Supervision, Project Administration, Funding Acquisition. X. Han: Conceptualization, Visualization, Supervision. All authors read and approved the final version of the manuscript.

Funding

This project was supported by National Natural Science Foundation of China (82600821, China); Shandong Provincial Natural Science Foundation (ZR2026QC1699); Qingdao Natural Science Foundation (26-1-1-27-zyyd-jch); Hebei Natural Science Foundation (H2024405017).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.

Ethics

Ethics, Consent to participate, and Consent to publish declarations: not applicable.

Clinical trial number: Not applicable.