Modular design for drug delivery across hierarchical gastrointestinal barriers: From materials to systems Authors: Yaxin Chen, Qingyun Yang, Jian Yang, Jiquan Zhang Published: 2026 ABSTRACT AbstractDrug delivery to the gastrointestinal (GI) tract necessitates overcoming multi-level physiological barriers from organ to cellular levels, which exhibit significant heterogeneity under pathological conditions. Conventional delivery strategies, constrained by single-function capabilities, struggle to achieve cross-scale coordination, severely limiting therapeutic efficacy. Modular design (MD) offers an innovative approach to synergistically breaching multiple barriers by decoupling complex functions into standardized units and enabling intelligent assembly. This review systematically analyzes the characteristics and delivery challenges of each barrier level within the GI tract, establishing a materials library that encompasses functional modules for organ-level regulation, mucosal penetration, cellular delivery, and environmental responsiveness. Utilizing GI inflammation and infection as exemplars, the text goes on to highlight the mechanisms and applications of modular assembly strategies in achieving breakthroughs in organ-mucosa, organ-cell, and full-layer barrier technologies. Additionally, it assesses the challenges and prospects for clinical translation. This paper aims to provide design strategies for developing next-generation intelligent oral delivery systems capable of spatiotemporal programming and dynamic response to the multi-level barriers and pathological microenvironments of the digestive tract. This should drive a paradigm shift towards precise drug delivery within complex physiological settings.Keywords: Modular design, Physiological barriers, Drug delivery, GI tract, Pathological response, Targeting AbbreviationsAbbreviationFull nameBBRBerberineCDTChemodynamic therapyCPCalprotectinCSChitosanDAIDisease activity indexDATSDiallyl trisulfideDDSDrug delivery systemsDMYDihydromyricetinDNPsDNA nanopatchesEcNEscherichia coli Nissle 1917EWDPEgg white-derived peptidesGIGastrointestinalGSHGlutathioneHAHyaluronic acidHMWHigh-molecular-weightICGIndocyanine greenLMWLow-molecular-weightLNPsLipid nanoparticlesMAGMagnololMDModular designMMP-9Matrix metalloproteinase-9MMTMontmorilloniteMnCOManganese carbonylMOFMetal-organic frameworkMPNMetal-polyphenol networkNCsNanocrystalsNPsNanoparticlesNTRNitroreductaseOcurOxidized chitosanPDAPolydopaminePEGPolyethylene glycolPGPolyglycerolROSReactive oxygen speciesSASodium alginateSACDSuccinic acid-modified γ-cyclodextrinSCFSimulated colonic fluidSDTSonodynamic therapySGFSimulated gastric fluidSIFSimulated intestinal fluidSNRsSupramolecular nanoribbonsSPSpirulina platensisTJsTight junctionsZWZwitterionic Abbreviations Introduction Efficient drug delivery in the digestive system must overcome a series of hierarchical anatomical barriers, spanning from macroscopic organs to tiny organelles. As shown in Fig. 1, this hierarchical defense system begins with gastric acid degradation and GI motility at the organ level, then progresses to mucus filtration and tight junction restrictions at the mucosal level, membrane permeability barriers at the cellular level, and the acidic environment of lysosomes at the subcellular level. Simultaneously, precise localization to the target location is vital since specific accumulation plays an important role in efficacy and safety. Encouragingly, since Leserman introduced the inaugural antibody-modified liposome (Lip) for targeted therapy in 1980 [1], active targeting and responsive design within pathological microenvironments have consistently captivated researchers in pharmacology, materials science, and translational biomedicine [[2], [3], [4], [5], [6], [7], [8]]. This approach offers dual advantages: enhancing drug accumulation at the lesion site while reducing systemic toxicity. For instance, CD44 receptor-mediated active targeting achieved a threefold higher drug enrichment intensity in inflammatory lesions compared to healthy colon tissue [9]. This substantial enhancement of the efficacy-to-toxicity ratio positions targeting technologies as the primary driver for novel drug delivery systems (DDS). Consequently, GI nanomedicine delivery requires engineered solutions that simultaneously address the penetration of anatomical barriers and the precise recognition of pathological microenvironments. Multistage physiological barriers and challenges for GI drug delivery. Created with BioRender.com. Recent research has yielded significant advancements to overcome these limitations, notably in surface modification [[10], [11], [12], [13]], material refinement [[14], [15], [16]], and size/morphology control [[17], [18], [19], [20], [21]]. Nevertheless, a considerable number of strategies concentrate on optimizing a single barrier, frequently disregarding the functional synergies that exist across multiple scales. This may lead to functional imbalances at other levels following the resolution of one barrier. Hydrophilic polyethylene glycol coatings have been demonstrated to reduce mucin adsorption via hydration [[22], [23], [24], [25]]; however, they may also decrease the uptake of nanodrugs by the mononuclear phagocyte system [26,27]. Micron-sized particles designed to enhance gastric retention may experience reduced penetration in gastric mucus because their size exceeds the dimensions of mucus network pores [28,29], which impedes drug access to deeper mucosal layers. These cases illustrate that DDS face a common issue of "local optimization resulting in overall imbalance" when addressing the complex barrier system of the GI tract. Addressing cross-level functional adaptation is a pressing challenge. The concept of MD has been identified as a promising solution to this challenge. At its core lies the creation of diverse, complex systems through the flexible combination of standardized units, an idea traceable to ancient Roman architecture [30,31]. In recent years, driven by breakthroughs in DNA nanotechnology and targeted therapies, MD has shown immense potential in synthetic biology and programmable materials [32,33]. In the context of gastrointestinal (GI) drug delivery, MD is regarded as a design concept wherein intricate delivery requirements are disassembled into relatively independent components, including protection against harsh gastrointestinal conditions, mucosal penetration, lesion targeting, cellular internalization, and controlled release, and subsequently rationally integrated into coordinated systems [[34], [35], [36]]. The fundamental principle of MD resides not in the mere integration of multiple functionalities within a singular carrier but rather in the flexible organization, recombination, and adaptive deployment of these capabilities to address heterogeneous pathological requirements. While MD conceptually aligns with established strategies such as multifunctional nanomedicine and hierarchical delivery systems, it is distinctly characterized by its emphasis on modular flexibility, system generality, and structural reconfigurability. In this review, MD is therefore defined pragmatically: rather than strictly requiring fully standardized coupling interfaces, it encompasses a modular logic wherein separable components can be independently optimized and dynamically assembled to overcome the multi-level physiological barriers of the gastrointestinal tract. The crucial element enhancing MD's efficacy in overcoming multi-level delivery barriers is its ability to leverage advancements in diverse material coupling strategies. For instance, biorthogonal chemistry enables “click-like” coupling of functional modules, providing orthogonal connection tools for precise targeting and barrier penetration [37]. Self-assembly strategies serve as modular platforms that facilitate the sequential integration of various functional molecules (e.g., penetration peptides, phosphatase-responsive units) to programmatically construct intelligent DDS capable of surmounting multiple obstacles [38]. Furthermore, the modular assembly process often results in the formation of ordered architectures, such as core-shell structures, thereby ensuring orthogonal compatibility and synergistic interactions among functional modules [39]. These characteristics collectively establish the foundation for MD strategies in developing intelligent delivery systems. Fig. 2 outlines the pivotal developmental stages and breakthroughs in MD concepts through a timeline format. An outline presenting a timeline of pivotal developments and breakthroughs in MD concepts. 2024: Multiple shapes can be assembled from the same monomer pool by adding different unique sets of ssDNA connector strands (Reproduced with permission [32]. Copyright 2024, The American Association for the Advancement of Science.). The review aims to systematically explore how MD strategies can address the challenges of multi-scale delivery within the digestive tract. The multi-level barrier characteristics of the digestive tract are first examined from organ to cellular levels, alongside its pathological microenvironments. A modular library is then established, comprising functional units for barrier penetration, pathological response, external field regulation, and inherently multifunctional materials. Utilizing GI inflammation and infection as a model, we elucidate the mechanism and advantages of the MD strategy in overcoming multi-level physiological barriers through the intelligent coupling of functional modules. The review concludes with a critical discussion of the challenges in assembly complexity, in vivo controllability, and clinical translation, while outlining future directions. We hypothesize that spatiotemporally programmed drug systems, founded on MD principles, will precipitate a paradigm shift in precision drug delivery by establishing effective pathways through the body's diverse multilayer physiological barriers. Physiological barriers and pathological characteristics across digestive tract layers Following an organ–mucosal–cellular hierarchy, the physiological barriers of the digestive tract are elucidated at three levels. At the organ level, macro-dynamics (peristalsis, pH, enzymes) form the first line of defense; at the mucosal level, the bilayered mucus–epithelium constitutes the frontline barrier; at the cellular level, membranes, efflux pumps, and organelles pose intracellular obstacles. Pathological conditions alter these barriers, adding further complexity to drug delivery. Characteristics of anatomical barriers and delivery challenges Organ level: macro-dynamics and physicochemical barriers The oral cavity serves as the entry point for drugs into the digestive tract, where its keratinized epithelium, salivary secretion, and tongue movements collectively form a physical and mechanical barrier. Research has demonstrated that under dry conditions, the friction coefficient at the tongue-palate interface can reach as high as 1.12, and the tongue's complex movements can induce localized stress concentration. These factors may result in drug carrier rupture, particularly when wetting properties are inadequate or mechanical strength is deficient [40,41]. The esophagus, through rapid peristalsis synergizing with gravity, limits drug retention (residence time of only 5-6 s, extending to 24 s in the supine position), making effective absorption challenging [42]. To extend the duration of drug action in the esophagus, studies have proposed the use of adhesive hydrogels or expandable carriers to increase contact time with the esophageal wall. For instance, chitosan (CS)-based hydrogels exhibit the property of spontaneous gelation at body temperature, resulting in robust adhesion to esophageal wound surfaces. This property is further characterized by the sustained release of drugs over a period exceeding 10 days, as demonstrated by a 10 mg dose [43]. The primary challenge in the field of gastric drug delivery pertains to the necessity of overcoming the interference from a dynamic physiological environment characterized by high acidity (pH 1.88 in the fasting state to 4.98 in the fed state) [44], the action of digestive enzymes, vigorous peristalsis, and rapid gastric emptying. These factors result in a reduction of drug residence time to 0.5-3 h [45,46], while daily gastric secretion of 2-3 L further dilutes drug concentrations. It is imperative to prolong residence time to achieve controlled gastric release. Strategies such as gastric flotation, high-density formulations, and adhesion systems have been employed to anchor drugs within the stomach, thereby enabling sustained local drug release [47]. Following brief retention in the stomach, the drugs migrate to the intestines. Being the longest and most distal organ in the digestive tract, the intestines present the most challenging battlefield for drug delivery. In the small intestine, the continuous degradation of proteins, lipids, and various drugs is catalyzed by pancreatic enzymes, operating within a pH gradient that ranges from 6.30 (proximal) to 7.42 (distal) [44]. This environment also provides a foundation for enzyme-responsive delivery systems. For instance, the BAI [48] and BIXENMANN [49] teams designed lipase-responsive lignin-loaded nanoparticle systems and dual pH/lipase-responsive amphiphilic block copolymers using pancreatic lipase, respectively, achieving targeted drug delivery in the small intestine. Moreover, while a 5-6 m structure of the small intestine allows for an absorption period of approximately 4 h [46,50], its high-frequency peristalsis often leads to uneven drug distribution and premature migration. While bile salts in the small intestine are known to facilitate lipid absorption, the presence of insoluble macromolecules within the intestinal tract indicates that these macromolecules are not subject to degradation and thus enter the colon via mucus [51]. Analogous to the small intestine, colonic pH displays a proximal-to-distal gradient, ranging from 6.28 in the ascending colon to 7.10 in the descending colon [44]. Concurrently, the structure of the pouch, in conjunction with low-frequency peristalsis and high viscosity of the contents, has been shown to prolong the retention time of the drug to between one and three days [50,52]. This, in turn, is thought to increase the risk of exposure to the gut microbiota, which may have a role in the metabolism of the drug. The prevailing consensus among researchers posits that the gut microbiota primarily modulates the efficacy of drugs through two mechanisms. Firstly, microbial enzymes react directly with drugs or drug carriers, thereby altering their bioactivity or degrading the carriers. Secondly, the accumulation of drugs within microbiota cells directly reduces effective concentrations [53]. Consequently, the cascade barrier mechanism of the stomach-small intestine-colon poses significant challenges for intestinal targeted DDS. Nevertheless, the limited absorption of poorly soluble macromolecules in the small intestine, in combination with the unique microbial enzyme systems and pH gradient environment of the colon, provides a physiological basis for colon-targeted spatiotemporal controlled-release delivery systems. Mucosal layer: a bilayered defense system While the multilayered mucosal barrier of the digestive tract—comprising the mucus layer and tightly packed epithelium—serves as a vital defender against pathogens, it also presents a major obstacle to drug delivery [54]. The mucus layer is composed of water, mucins, carbohydrates, lipids, inorganic salts, other proteins, and cellular debris [[55], [56], [57]]. Within the GI environment, it forms a three-dimensional cross-linked gel-like network through covalent and non-covalent interactions of mucins [58]. The lubrication and protection of the digestive tract epithelium from gastric acid damage and pathogen invasion is attributable to its unique negative charge, viscoelasticity [59], and selective permeability [60,61]. Nevertheless, these properties can also impede the penetration of the mucus layer by inducing physicochemical interactions with drug carriers. For instance, experiments with porcine gastric mucin showed that under acidic (pH 5.2), high-viscosity mucus conditions, amino- and carboxyl-modified nanoparticles (NPs) were sequestered by mucus via electrostatic and hydrogen bonding interactions, respectively [10]. It was found that only high-density polyethylene glycol (PEG) achieved efficient penetration by forming a "brush-like conformation" that shielded particle-mucin interactions. Notably, under conditions of neutral pH (pH 7.0) and weakly alkaline pH (pH 7.7), due to reduced mucus viscoelasticity, nanoparticle aggregation was observed to dissipate, and diffusion capacity underwent a substantial enhancement [10]. Additionally, mucus undergoes dynamic renewal through continuous secretion and shedding [62], thereby helping to maintain the thickness and function of the mucus layer while limiting the retention time of drugs on the mucosal surface. It is important to note that there are notable differences between gastric and intestinal mucus in terms of their primary mucoprotein types, structural organization, glycosylation, and functional roles. Gastric mucus is predominantly composed of MUC5AC, while intestinal mucus is chiefly constituted of MUC2. The small intestine possesses a single-layered, loose, and more permeable mucus layer [63,64], whereas the colon exhibits a double-layered structure: a loose, bacteria-colonized outer layer and an inner layer tightly adhering to the epithelial cell surface. The latter functions as a "sieve" due to its dense pore size of 100-500 nm [65,66]. In order to overcome mucus layer barriers, drug carriers can be engineered with modified shapes, sizes, and stiffness, thus evading mucus entrapment. More specifically, given the negative charge of the mucus layer, the use of near-neutral charge designs has been shown to reduce mucus interactions and enhance drug penetration [62]. However, for drugs that do not require reaching the submucosal layer, prolongation of drug action time in target organs can be achieved through mucosal adhesion by leveraging electrostatic/hydrogen bond/covalent interactions between formulations and mucins [67]. The epithelial layer of the digestive tract undergoes continuous evolution from the oral cavity to the anus, comprising a diverse array of epithelial cells and other specialized cell types. It serves both as a protective barrier, maintaining mucosal integrity, and as a key permeation barrier for mucosal drug delivery [68]. The epithelium of the oral cavity and esophagus consists of squamous epithelial cells, whereas that of the stomach and intestines is primarily a single layer of columnar cells. The epithelium from the esophagus to the colon forms robust tight junctions (TJs) through a series of intercellular structures, including occludins, claudins, junctional adhesion molecules, and desmoglein [68,69]. This structure is capable of defending against harmful substances such as gastric acid while concomitantly increasing the difficulty of drug penetration. Beyond this physical defense, the apical membrane of intestinal epithelial cells has been found to express P-glycoprotein (P-gp) highly. Upon entering epithelial cells, orally administered drugs are promptly recognized by P-gp and pumped back into the intestinal lumen as substrates. This prevents the medications from successfully traversing the cells to enter the bloodstream or deeper mucosal layers [70]. Current strategies frequently employ nanoscale advantages to traverse intercellular gaps and utilize receptor-mediated endocytosis mechanisms to enhance drug transport efficiency through epithelial cells. Cellular/intracellular level: microscopic defense system Following the successful overcoming of numerous barriers at the organ and tissue levels, DDS must also contend with the "microscopic-level" defense system formed by cells and intracellular structures. This is regarded as the ultimate critical impediment to delivery efficiency and therapeutic efficacy. Firstly, the cell membrane, the fundamental boundary structure between cells and the external environment, is composed of a lipid bilayer. It exhibits extremely low permeability to hydrophilic drugs, while lipophilic drugs can enter cells to some extent via passive diffusion [71]. This structural selectivity is the primary factor limiting the delivery efficiency of most hydrophilic drugs. To surmount this limitation, common strategies involve the utilization of phagocytic mechanisms mediated by lipid carriers [72], ultrasound [73], or receptors [74,75] to introduce drugs into target cells. However, receptor-mediated phagocytosis exhibits a saturation effect, which complicates the fulfillment of high-dose delivery demands. In such cases, the high drug-loading capacity of nano delivery systems becomes particularly crucial. Upon entering cells, drugs face a new set of challenges posed by complex subcellular compartmentalization structures, particularly issues such as lysosomal degradation, efflux pump-mediated drug expulsion, and target localization. Lysosomes are the core intracellular compartments for processing and degrading exogenous substances. They are characterized by an acidic environment (pH approximately 4.5) that is rich in various hydrolases. If drugs are rapidly transported to lysosomes following cellular entry, they are highly likely to be completely degraded before exerting any therapeutic effects. Concurrently, intracellular therapeutic targets are frequently localized within specific subcellular structures (e.g., nucleus, mitochondria), which possess independent membrane systems. The membrane selectivity of these cells has been shown to impose further restrictions on the targeting and release of drugs, thereby forming multiple coupled barriers. In recent years, research on the intracellular fate of delivery systems has yielded new insights, particularly regarding how the physical properties of nanocarriers influence cellular uptake mechanisms and intracellular transport pathways. For instance, studies indicate that the geometric shape of nanocarriers exerts a significant influence on their phagocytic pathways within cells. Compared with spherical structures, tubular nanorods are more likely to activate the macrophage mechanism, thus avoiding the fate of entering the lysosomal pathway after traditional endocytosis. When delivering mangiferin via this mechanism, its intracellular accumulation levels increased by 2.1-fold relative to spherical carriers, significantly enhancing intracellular drug retention [18]. This finding suggests that when designing DDS, multiple factors must be comprehensively considered for their impact on cellular uptake and organelle targeting, including material properties, size, shape, and charge. This comprehensive strategy facilitates multi-level regulation of microscopic barriers, thereby enhancing drug release activity and efficacy. Specificity of pathological conditions in the digestive tract In pathological states of the GI tract, various parameters of the digestive tract microenvironment, notably transit time and mucosal characteristics, may be significantly altered [76]. These changes introduce a greater degree of complexity to the precise design of DDS. The onset of GI diseases is frequently accompanied by disturbances in the retention time of contents [[77], [78], [79]]. For example, most patients with Crohn's disease exhibit an accelerated colonic transit speed [76], which may result in the premature expulsion of drug carriers that utilize retention time for controlled release, thereby compromising therapeutic efficacy. Colonic transit time in ulcerative colitis patients demonstrates considerable inter-individual variability [80,81]. Also, delayed gastric emptying in conditions such as Crohn's disease and coeliac disease [76,82] may prolong drug residence in the upper GI tract, delaying peak plasma concentrations. At the mucosal level, inflammation has been shown to increase mucosal permeability [83] and thin the mucus layer [84] through a variety of mechanisms, thereby compromising mucosal barrier function. While this facilitates drug penetration through the mucus layer to a certain extent, it also non-selectively creates a rapid pathway for pathogens and foreign substances. In addition, as the mucosal layer at the site of inflammation is disrupted, positively charged proteins accumulate in situ within the mucosal exudate, including bactericidal/permeability-increasing proteins, antimicrobial peptides, and transferrin [85]. Beyond these macro-level alterations, pathological sites often exhibit particular physicochemical characteristics, offering potential, precise triggering and recognition signals for DDS. Abnormal pH levels are a common feature across many pathological sites. For instance, Helicobacter pylori (H. pylori) infection may cause either hypochlorhydria or hyperchlorhydria [86], while gastroesophageal reflux disease has been shown to lead to esophageal acidification [87]. Furthermore, intestinal inflammatory sites have been observed to exhibit decreased pH due to local metabolic hyperactivity and immune cell infiltration [88]. The existence of an acid-base gradient between diseased and healthy tissues offers targeting opportunities for pH-responsive DDS [89]. Redox imbalance is also a hallmark of inflammatory and infectious microenvironments. Reactive oxygen species (ROS) are physiological byproducts of cellular respiration and aerobic metabolism, and their levels are known to be significantly elevated in inflammatory diseases. ROS can also be generated exogenously upon exposure to agents such as chemotherapy, radiotherapy, or ultraviolet radiation [90]. Concurrently, glutathione (GSH)—the primary intracellular reducing agent—is often upregulated by inflammatory cells to protect against ROS damage [91]. This abnormal coexistence of elevated ROS and GSH provides triggering conditions for constructing ROS-responsive, GSH-responsive, or dual-responsive DDS, enabling carrier degradation, drug release, or structural reconfiguration at the lesion site [[92], [93], [94]]. Additionally, GI lesion tissues frequently overexpress specific proteases, such as matrix metalloproteinase-9 (MMP-9), hyaluronidase, and nitroreductase (NTR). The expression profiles of these enzymes have been shown to reflect the stages of disease progression, while simultaneously providing a molecular recognition basis for the design of carrier shells, protective groups, or linkers that can be cleaved by specific enzymes [95,96]. It is noteworthy that inflammatory sites may also exhibit substantial expression of calcium-binding proteins (e.g., calmodulin) [97]. Simultaneously, there is evidence to suggest that positively charged inflammatory mucosal proteins and changes in the membrane potential of inflammatory cells may interact selectively with negatively charged drugs or carriers [98,99]. Furthermore, in the context of inflammatory bowel disease, P-gp expression is abnormally elevated in peripheral blood lymphocytes and mucosal cells, leading to the rapid efflux of drug molecules from the cell. This process reduces intracellular drug concentrations, a phenomenon particularly pronounced in drug-resistant patients [100]. Therefore, effective drug delivery strategies for pathological GI conditions must account for macro-level changes in transit kinetics and mucosal barriers, while also leveraging disease-specific physicochemical and molecular signatures. To achieve precise spatiotemporal control, a promising approach involves integrating microenvironment-responsive mechanisms with molecular recognition elements. This dual-targeting strategy enables carriers to first accumulate at diseased sites through ligand-receptor interactions, followed by triggered drug release in response to local biochemical signals, thereby maximizing therapeutic efficacy while minimizing off-target effects. It is anticipated that this integrated targeting design approach will become a key direction for advancing personalized treatment of GI diseases and optimizing targeted DDS in the future. Module library: core functional units for GI drug delivery To achieve efficient drug delivery in the complex GI environment, sequential penetration capability across multi-scale barriers and spatiotemporal regulation mechanisms are essential. By reviewing the latest advances in GI delivery systems, this section constructs a component library covering barrier penetration modules, pathological environment-responsive/active targeting modules, and intrinsic multifunctional modules. It further analyzes their design principles and application scenarios, laying a foundation for the development of programmable delivery systems. Organ-level delivery regulation To address organ-level barriers, drug delivery strategies aim to maintain drug stability and prolong GI retention. The GI environment, with its high concentrations of gastric acid and enzymes, often induces drug degradation and premature release. To enhance stability, various micro- and nano-carriers have been developed to shield drugs from gastric acid by leveraging their distinctive physicochemical properties. Targeted release platforms are a pivotal strategy in this regard, using their targeting capabilities to ensure precise spatiotemporal drug release. Among these, pH-dependent and enzyme-responsive DDS are two core types. The former utilizes the pH gradient of the digestive tract for controlled release and is a mainstream design due to its broad adaptability and technological maturity. The latter primarily relies on enzymatic metabolism by the gut microbiota and is widely used for colon-targeted delivery. For overcoming kinetic barriers, time-dependent DDS is the prevailing approach. It prolongs the residence time of the formulation in specific GI regions, thereby increasing local drug concentration. A major challenge, however, is its sensitivity to individual variations in GI kinetics [101]. Thus, caution is warranted in patients with conditions such as delayed gastric emptying, slow intestinal transit, diarrhea, or heightened sensitivity to systemic side effects [102]. The advanced strategies outlined above are exemplified by a range of materials employed for GI drug delivery. Among these, synthetic polymer-mediated and polysaccharide-based systems represent two classical approaches. A comparative summary of these two systems, categorized by the primary barriers they address, is provided in Table 1, Table 2. Table 1Advanced representative of synthetic polymer-mediated delivery systems for GI administration.Table 1ObjectiveMechanismDDSCompositionOutcomeRefBoth target organ release and drug stabilityLow pH: protonated, insoluble. High pH: deprotonated, soluble. A higher carboxyl group ratio lowers the dissolution pH (e.g., L100-55)rDNPs@P-LEudragit L100-55, DNA origami nanosheets, phage capture strandsThe coated formulation showed a 1000-fold higher survival rate after 20 min in gastric juice, with the coating dissolving at pH > 5.5.[103]pH and H2O2-Responsive Enteric CapsulesEudragit L100-55, Ag/Ag2S JNPs, DCNPs, DexCapsules maintain structural integrity in simulated gastric fluid (SGF) (pH 1-3) and gradually disintegrate in simulated intestinal fluid (SIF, pH 5.5-7.5).[104]CuS/DSF/EL/PVPCuS, Disulfiram, Eudragit L100-55, PVPIn SGF, only 10% Cu2+ was released at 24 h, and TEM confirmed structural integrity; whereas in SIF >60% was released.[105]EFA-AC-IFXFolic acid-modified aminoclay, Eudragit S100, and infliximabpH 1.2 (gastric): Drug release <10%;pH 7.2 (small intestinal): 48% release after coating dissolution;pH 5.5 (inflamed colon): Electrostatic repulsion-driven release of 40%.[106]MCNMesoporous Cerium Oxide, TJ-5, Dextran Sulfate Sodium, Eudragit S100Only 5.05% release rate after 2 h in SGF; Colon-specific release at pH > 7.0[107]CeO2@S100CeO2 NPs, Eudragit S100Dimensionally stable in gastric acid, dissociates in the intestine, and accumulates up to 3 times more than uncoated CeO2 at sites of inflammation[108]Physical barrier coating and freeze-drying dehydration to extend shelf lifeLGG-M@PLGG, iMXene, mPEG-d-PEI, plasmid DNA, Eudragit®FS30DProtected from gastric acid degradation; colonic disintegration rate >95%.[109]Target organ retentionRapid formation of physically cross-linked hydrogelsPEI/PAA powderPolyethyleneimine, polyacrylic acidStrong adhesion to diverse wet tissues (e.g., chicken skin, porcine heart, GI mucosa); adhesion stress is significantly higher than that of pre-formed PEI/PAA hydrogel, dried PEI-cPAA gel powder, or fibrin glue.[110]Synergistic effect of thermosensitive gelation and mechanical strength of hydrogels5-FU + DDP/PLEL hydrogelPoly (D,L-lactic acid)-polyethylene glycol-poly (D,L-lactic acid) copolymer, 5-Fluorouracil, dichloro diamine Platinum20 wt% hydrogel retained in the peritoneal cavity for >168 h (disappeared at 24 h in the free group).[111]Both target organ retention and drug stabilityEnhanced adhesion on wet tissue by water drainage and subsequent water uptake. Its swelling behavior is controlled by physical interactions and hydrophobic phase separation. PAA carboxyl groups protonate to enhance stability.Adhesive hydrogelPolyacrylic acid, Gelatin (Gel), Hyperbranched polymer-modified catechol, CS oligosaccharideAvoiding the problem of reduced adhesion properties due to swelling of conventional hydrogels. Ability to tolerate extreme acidic environments in gastric perforation repair.[112] Advanced representative of synthetic polymer-mediated delivery systems for GI administration. Table 2Advanced representative of polysaccharide-based delivery systems for GI administration.Table 2ObjectiveMechanismDDSCompositionOutcomeRefEnsure drug release or retention at target organsProtonation in acidic conditionsCAMOHyaluronic acid (HA), Ag+, Mesoporous organosilica NPs,ClarithromycinReleases ∼50% at pH 1.0; releases <10% at pH 7.4.[113]Schiff base cross-linkingTCP-25 peptide hydrogelsCarboxymethyl CS,TCP-25 PeptideOptimized drug release performance; 92% hemostasis rate within 9 s.[114]Starch content affects crosslinking strength. Low-density oil achieves buoyancy.S-CS-PCCS, Starch, Corn oilProanthocyanidinsEnhanced sustained-release performance; 3 h gastric suspension.[115]Binds to the intestinal mucus via electrostatic interactionsLMWC-BRNPsLow-molecular-weight (LMW) CS, BilirubinEnrichment in the colon was 3.2-fold greater for 10K HA-BRNPs.[116]Binds to the intestinal mucus via electrostatic interactionsCOS-GL NanoparticlesLMW CS oligosaccharide, GlycyrrhizinHigher retention in the colon at all time points compared to free Glycyrrhizin.[117]Enhance stabilityElectrostatic self-assembly forms a protective coatingLBL-CO@MPDACS, Alginate, Fe3(CO)12, Mesoporous polydopamine (PDA) NPsMaintain integrity in the gastric acid and small intestinal environments.[118]Electrostatic interactions and hydrogen bondingEcN@TACS, Aromatic Thioacetal, Escherichia coli Nissle 1917 (EcN)A 5.1-fold better survival rate in SGF; a 7.7-fold better survival rate with ApAm.[119]CS-Sodium alginate (SA) self-assemblyDNPs@P-LMaltodextrin, CS, EcN, SA, rDONsA 17-fold higher survival rate after SGF/SIF exposure for 2h.[120]Encapsulating glycerol CS and SA layer-by-layerSe@EcN-C2/A2Glycerol CS, Selenium NPs, SA, EcNThe coating enhanced SGF survival by 100-fold (>106 CFU).[121]Both target organ release and drug stabilitySA exhibits pH sensitivity.CS@FL-D@lzd/MSSA, CS, Gel, Dendritic silicon framework, FluidosomeSuppressed in SGF (swelling <30%, release <20%); a 9-fold swelling increase and 85% release in SIF.[122]Schiff base bonds & electrostatic interactions mediated by amino protonationEcN-Apt@HGOxidized HA, Polyamide-amine dendritic polymer, IL-6 aptamer, EcNIntestinal live bacteria increased tenfold; tolerance to bile salts and H2O2 enhanced fivefold; intestinal transit time prolonged.[123]Covalent network formation via Schiff base; interfacial mediation by Epigallocatechin gallate.ECA@EcNEpigallocatechin gallate, CS, AuNPs, EcNEnhanced survival in gastric acid (>1400 CFU count within 2 h); sustained intestinal fluorescence for 48 h; and a 4.3-fold higher accumulation in colitis tissue[124]Lip structure ensures oral stability; Colonic microbiota enzyme sensitivity of prebiotic xylan derivative Sxy.FLSKCholesterol, 1,2- (DPPE), Sxy, Hydrogenated soybean phosphatidylcholine, 5-FUStable in the stomach; slow release in the small intestine; rapid, ∼100% 5-FU release in the caecum within 24 h.[125]CBS-specific colonic germination; enzyme-responsive degradation of CSCBS-CS@Lipo/Oxp/MTZClostridium butyricum spores, CS,Release <20% in SGF; ∼30% in SIF; >80% in simulated colonic fluid (SCF) with β-glucosidase.[126]Gastric acid protection is provided by the PDA layer; colonic microbiota enzymes can degrade the Gel microspheres.P@G@NGOxidized HA, L-Arginine, PDA, Gel microspheresMaintains integrity in SGF but surface-disrupts in SIF and ruptures in SCF, enabling colonic targeting.[127]Both target organ retention and drug stabilityViscoelastic modulus of inulin gel (G' = 2000-3000) ensures intestinal retention; colonic inulinase-responsive release.SAM NCs@B. subtB. subtilis, SAM NCs, InulinStable under harsh (gastric acid/SGF/SIF) and neutral (pH 6.5) conditions; Accumulates in the colon >48 h after oral administration.[128]High bioadhesion of inulin gel; macroporous structure of Cu2(Olsa)/gel compositeCu2(Olsa)/GelCu2(Olsa) nanopins, inulin gelMinimal release (<1%) in SGF within 2 h; burst release (71.7%) in SIF at 4 h. Drug retention in the colon is prolonged.[129]Inulin prolongs mucosal adhesion and colonic inulinase-responsive release.UIRNInulin, Ursodeoxycholic acid, RegorafenibMaintains a high concentration of >40 μg g−1 in the mucus layer for 12 h; achieves microbiota-triggered release.[130] Advanced representative of polysaccharide-based delivery systems for GI administration. Inert material-mediated Passivated nanocarriers use inert materials to physically shield against gastric acid erosion and reduce instability driven by interface energy, thereby mitigating the risks of drug degradation, carrier aggregation, and protein coating formation. For example, Ta2C nanosheets maintained their morphology, particle size, and surface charge during simulated digestion, with no reduction in radical scavenging capacity, providing a promising platform for sustained oral delivery [131]. However, this passive protection strategy limits precise drug release. To overcome this limitation, Qin et al. developed tungsten-coated zinc NPs (W@ZnNPs) synthesized in one step [132]. These NPs exhibit remarkable stability in gastric acid (with a Zn2+ release rate of only 7.9% after 10 h) and enhance the targeted release of tungsten ions in inflamed intestines, improving local therapeutic efficacy and offering new insights for oral nanotherapeutics targeting colonic diseases. Synthetic polymer material-mediated Synthetic polymers offer significant advantages for GI drug delivery, primarily due to their programmable nature, which permits precise molecular design to control material behavior across diverse physiological conditions. For instance, Feng et al. pioneered the use of Eudragit L100-55—which dissolves at pH > 5.5—to achieve precise intestinal delivery of EcN [133]. This strategy enhanced bacterial survival in gastric acid by 14.9-fold, establishing a viable pathway for oral delivery of live biologics. To address the challenge of encapsulating liquid drugs, a co-precipitated droplet-S100 system was developed [134]. This system utilizes liquid-liquid phase separation to spontaneously concentrate the drug, and when combined with the colon-targeting S100 coating, it ensures stability in gastric acid and subsequent rapid release of emodin in the colon, offering a novel strategy for delivering poorly water-soluble small molecules. Plant-derived vesicles, which are rich in bioactive macromolecules such as proteins, lipids, and nucleic acids, are highly susceptible to degradation in the GI tract. Relying on a single protective material is often insufficient to preserve their structural integrity. To this end, the Liu research team [135] integrated an S100 coating with freeze-drying technology, which markedly improved the stability of plant vesicles in the harsh gastric environment. Likewise, to protect macromolecular drugs, Huang et al. developed the RNACap capsule, a system featuring a triple-layer structure composed of an enteric coating, a sealing membrane, and an inner coating [136]. It employs a unique dual "pH-mechanical force" response mechanism. By leveraging intestinal peristaltic pressure, the capsule achieves rapid release of 50% of its mRNA nanoparticle payload within 13 min, effectively overcoming the limitations of traditional sustained-release systems for nucleic acid therapeutics. Synthetic polymers also demonstrate potent capabilities in managing dynamic mechanical stress in the stomach. For example, the PEI/PAA powder system developed by Peng et al. adheres to wet gastric tissues within 2 s via rapid water absorption and gelation, providing immediate protection against acute injury [110]. The material exhibited excellent stability, showing no significant degradation after 30 days of implantation, which suggests its potential as a long-term physical barrier. In another advanced design, LIANG et al. created a Janus structure [112]. On one side, a cationic polymer forms a dense layer through hydrogen bonding and electrostatic interactions. The opposite hydrophobic side integrates PAA-Gel and hyperbranched catechol polymers. This spatially partitioned design simultaneously addresses anti-swelling, robust wet adhesion, and postoperative anti-adhesion, successfully resolving the functional conflicts common in traditional adhesive materials within complex physiological environments and representing a major advance in smart adhesion technology. These studies demonstrate that synthetic polymer materials for GI delivery have evolved from simple pH-responsive systems to those with multiple responsive mechanisms, and from basic adhesion to intelligent environmental adaptation. This technological progression provides a solid foundation for clinical translation. Nonetheless, the biodegradability of these synthetic, chemical coupling systems remains a critical challenge that requires ongoing investigation. Polysaccharide-mediated materials Natural polysaccharide materials exhibit unique advantages in GI DDS due to their excellent biocompatibility, degradability, and abundant chemical modification sites. A substantial proportion of polysaccharide molecules possess ionizable functional groups that undergo protonation under low pH conditions. While this characteristic may compromise acid resistance, it provides a unique responsive mechanism for gastric-targeted drug delivery. HA and CS are representative examples of such polysaccharides. Leveraging this principle, FAN et al. employed HA to load Ag+ and the antibiotic CLR concurrently, significantly suppressing antibiotic resistance issues in H. pylori infections [113]. Results showed that under physiological conditions, the release rates of both drugs were <10%, whereas in gastric acid environments, Ag+ and CLR release rates reached 49.7% and 56.6%, respectively, thereby achieving synergistic effects through the dual-drug mechanism of antibiotic action and urease inhibition. However, the highly acidic gastric environment poses significant challenges for the controlled release of drugs, particularly in preventing premature drug release. To address this issue, researchers have developed various multi-unit delivery systems. Wen's team utilized a TCP-25/carboxymethyl CS hydrogel constructed via Schiff base cross-linking, enabling simultaneous dissolution of CS and contraction of the gel network at low pH [114]. This dual-response behaviour facilitated sustained drug release under conditions of gastric acidity, concomitant with a high haemostasis rate of 92% within 9 s. More ingeniously, Chen et al. employed emulsification-extrusion technology to engineer pH-responsive CS-starch composite gel beads [115]. By adjusting starch content to optimise crosslinking strength, the sustained-release performance of the delivery system was significantly enhanced. Combined with low-density corn oil encapsulation, this achieved stable gastric flotation throughout the 3-h digestion process. The swelling and charge properties of polysaccharide materials were fully exploited. Beyond physical buoyancy, leveraging polysaccharide swelling and charge characteristics to achieve mucosal adhesion represents another crucial strategy for overcoming GI emptying kinetics. The chitin-guar gum graft copolymer carrier developed by Fetouh's team exhibited a swelling rate of up to 980%, significantly enhancing physical bonding with the mucus layer [137]. As a highly deacetylated derivative of chitin, CS generates far stronger adhesion than CS through electrostatic attraction with negatively charged mucus. A systematic comparison by Rahman et al. revealed that LMW-CS-modified NPs exhibited optimal mucus adhesion properties compared to high-molecular-weight (HMW) CS and HA-modified NPs [116]. The intensity of the colonic fluorescence reached 3.2 times that of the HA-modified system 6 h post-oral administration, demonstrating superior efficacy in treating colitis. Continued innovation in cross-linking strategies has further improved the performance of polysaccharide materials. Protective properties have been significantly enhanced through various approaches, ranging from physical interactions (such as CS-HA electrostatic composite coatings [138] and CS/SA ‘nano-armour’ [118]) to covalent cross-linking systems. Moreover, combining interaction forces with different characteristics enables the construction of more robust protective layers. For instance, the Pan group combined electrostatic interactions with amide bond formation to fabricate multilayer protective structures, which increased the survival rate of probiotics in gastric fluid by 247-fold compared to unmodified controls [139]. Advancing this trend, the acid-tolerant DNA microcapsules developed by Zhao et al. exemplify a more sophisticated intelligent design [140]. This system uses HA methacrylate as the structural matrix and incorporates oligonucleotide sequences that form dense A-/i-motif structures under acidic conditions. This design enhances probiotic protection via a dual mechanism involving chemical cross-linking and biomolecular structural transformation. As reported, the contraction of these microcapsules under acidic conditions strengthens protective efficacy, enabling a probiotic survival rate exceeding 87% after 5 h. Meanwhile, reversible expansion at neutral pH facilitates colon-targeted release. Alginate rapidly forms a hydrogel network in the highly acidic gastric environment due to its unique G-block structure, while deprotonating to release drugs under neutral colonic conditions. This property renders it an ideal material for colonic targeted delivery. For example, Peng et al. used SA to co-deliver the probiotic EcN and the anti-inflammatory drug 5-ASA [141]. The system achieved synchronous release in the colon, thus demonstrating the feasibility of probiotic-drug synergistic therapy. To further improve performance, researchers commonly cross-link alginate with Ca2+ to form an “egg-crate” structure [[142], [143], [144], [145]]. In one study, Bhagat et al. reported that calcium alginate microcapsules remained intact for 24 h in pH 2.0-6.5 buffers, with rapid degradation occurring only at pH ≥ 7.0 [146]. Moreover, recent studies have identified new polysaccharide materials with high acid resistance. For instance, Lei's team developed a novel pH-responsive microcapsule based on calcium-crosslinked drip-shaped Leonurus polysaccharides, which protected Lactobacillus plantarum with a gastric acid survival rate of 80.75% [147]. Refining structural design represents another key research direction. Huang's team utilized microfluidic technology to fabricate SA-Ca2+/GelMA multi-leaf microparticles, which leveraged their distinctive morphology to significantly improve intestinal adhesion and retention [148]. Similarly, Wu et al. developed a Janus-structured droplet system using microfluidic double-emulsion templating [149]. This system featured PDA in one compartment and a complex of probiotics with montmorillonite (MMT), in the other. Following this, Ca2+ induced cross-linking and solidification of the Gel-SA network. The resulting system not only preserves the gastric acid protection and intestinal sustained-release properties of calcium alginate but also incorporates PDA's adhesive functionality and MMT's ability to promote probiotic growth. Collectively, these attributes effectively enhance the colonization efficiency of orally administered probiotics in the gut. Enzyme-responsive polysaccharide systems offer unique advantages for colon-targeted drug delivery. The XXXG-Dex system directly employs xyloglucan heptasaccharide as a carrier, which is hydrolyzed by the colonic microbiota-specific GH5_4 endo-xylanase to release the drug. This approach reduced the required anti-inflammatory dose of dexamethasone in the colon to one-third to one-tenth that of the free drug [150]. Inulin is regarded as one of the most promising enzyme-responsive materials for colonic delivery due to its inherent acid resistance, microbial degradability, and prebiotic properties. Delivery systems based on inulin, developed by Zhang [151] and Xie's team [152], exhibited drug release rates below 7% in SGF; their release is triggered by inulinase secreted by colonic bifidobacteria. Additionally, gelation-enhanced retention constitutes another core value of inulin. The ternary inulin hydrogel reported by Cao's group [153] significantly enhanced therapeutic efficacy by combining prolonged intestinal retention with targeted intestinal degradation. In summary, the development of polysaccharide-based GI DDS shows a clear trend toward functional integration and intelligent design. These materials have evolved from providing basic protection to enabling multifunctional platforms that feature smart responses and therapeutic synergy. This advancement is propelled by innovative cross-linking strategies, refined structural designs, and the adoption of bio-inspired intelligent mechanisms. Metal-organic framework (MOF) -mediated MOFs are widely used for encapsulating sensitive active ingredients due to their high specific surface area and tunable pore structures. Studies have shown that structural advantages of MOFs enable the design of acid-resistant carriers. For instance, Li et al.immobilized thermostable superoxide dismutase (HtSOD) in ZIF-8 via biomimetic mineralization [154]. Although ZIF-8 degrades gradually in acid, its slow decomposition helps preserve enzyme activity, allowing HtSOD to retain about 70% activity at pH 2.0-4.0 and recover to 97% in the intestine. Further improvement was achieved by introducing covalent crosslinking, which enhanced acid resistance. In one case, TCOF in TCD was covalently connected with a thioacetal crosslinker via γ-cyclodextrin MOF, forming stable hydrophobic supramolecular NPs [155]. This structure limited water ingress and framework hydrolysis in gastric acid, resulting in less than 20% release in SGF within 2 h. Thus, the acid-resistant barrier strategy in MOFs has evolved from delayed degradation to active resistance. In the realm of MOFs, the metal-polyphenol network (MPN) stands out as an up-and-coming candidate for intestinal targeted delivery, leveraging dual metal chelation and polyphenol-mediated adhesion. Gao et al. systematically compared iron-coordinating networks of small-molecule polyphenols (gallic acid, epigallocatechin, and epigallocatechin gallate), revealing that the intestinal adhesion of MPN@L complexes increased with polyphenol molecular weight [156]. Epigallocatechin gallate showed 2.3-fold higher adhesion work than gallic acid. However, in SIF I, gallic acid- and epigallocatechin-based MPN@L exhibited superior bile salt resistance. Considering adhesion, stability, and microbial modulation, epigallocatechin-based MPN@L demonstrated optimal performance. Based on structure-activity insights of macromolecular polyphenols, Xie et al. developed an EcN@Fe-TA@mGN probiotic system using tannic acid [157]. The Fe-TA network provided protection, while the carboxymethylated β-glucan coating enhanced mucin binding via hydrogen bonding and hydrophobic interactions. This system increased the survival rate in SGF by 1720-fold and achieved 47.54% retention in mouse intestines after 16 h. Wang et al. further integrated adhesion and response-triggered release functions [158]. A TA-Ce3+ network enhanced colonic mucus retention via catechol-mediated hydrogen bonding, while tannic acid hydrolase from colon microbiota triggered network dissociation for precise release of bee venom peptides and antioxidant enzymes, collectively modulating gut-brain axis pathways. Natural plant-based carrier In contrast to isolated polysaccharides, intact plant structures and their biomimetic analogs provide unique solutions for GI delivery, leveraging their natural architecture and bioactivity. For instance, Mao's team designed a biomimetic carrier that replicates the plant cell structure: a cellulose nanofiber/pectin shell encapsulating a lipid bilayer with an aqueous core [159]. This cell wall-like structure protects drugs from gastric acid. In the small intestine, bile salts trigger lipid bilayer contraction, releasing ∼59% of small-molecule drugs. Subsequently, colonic enzymes degrade both cellulose nanofiber and pectin, enabling complete macromolecular drug release. This system overcomes the instability of Lips in the small intestine and the leakage issues of shell-less carriers, offering a robust platform for targeted colon therapy. Similarly, the microalgae Spirulina platensis (SP) has garnered attention due to its spontaneous buoyancy, gastric acid stability, high polysaccharide content, and probiotic effects. Zhong et al. constructed SP@Curcumin via a simple one-step process [160]. With a high drug loading capacity (>80%), this formulation suppressed colon tumors and reduced pro-inflammatory factors, achieving minimal gastric release (8%) and sustained intestinal release (66%). Zhang's team developed an SP@Amifostine system using a freeze-drying-rehydration strategy, increasing drug loading to 0.48 mg mL−1 [161]. SP provided sustained intestinal release (50% retention after 12 h) and promoted uniform drug distribution, while its inherent probiotic effect enriched beneficial gut bacteria. For gastric retention, SP's instant buoyancy makes it an effective platform. To enhance this, researchers have combined buoyancy with adhesion strategies. Inspired by algal blooms, the SP-RSV-Bi system leverages SP's buoyancy and bismuth pectin's adhesiveness, maintaining stable adhesion-buoyancy for 24 h in SGF [162]. In vivo, this system extended gastric retention to 48 h and reduced acute alcoholic gastric injury area by ∼50%. This synergistic design overcomes the limitations of single-flotation methods, yielding a low-cost, high-bioactivity multifunctional strategy. However, batch-to-batch size variability of SP remains a significant challenge for large-scale production. In summary, the evolution from synthetic polymers to natural microalgae signifies a deepening shift toward safer, multifunctional approaches in GI-targeted delivery. Drug self-administration system Self-assembled, carrier-free DDSs, where the active pharmaceutical ingredient itself forms the delivery vehicle, represent a promising strategy to overcome the limitations of traditional excipient-based carriers, primarily by maximizing drug loading and intrinsic biocompatibility. In these systems, drugs are not passive cargo but active building blocks that can be engineered to perform delivery functions [163], such as resisting digestive degradation and enabling sustained release. A prime example is supramolecular nanoribbons (SNRs) formed through the hydrogen bonding and π-π stacking of single-molecule quercetin. These SNRs exhibit exceptional acid stability, maintaining their integrity and antioxidant activity in SGF and SIF for over 4 h. The release rates of quercetin from SNRs were markedly low (0.67% in SGF and 1.69% in SIF), significantly delaying drug release compared to free quercetin and thereby optimizing GI delivery [164]. Inspired by the classical "Huanglian-Houpo Decoction," Xu's team [165] co-assembled berberine (BBR) and magnolol (MAG) into functionalized BM NPs via electrostatic attraction and π-π stacking. The BM NPs provided sustained release in SGF for 20 h, far exceeding the 4-h complete release of free BBR, due to effective protection against gastric acid. This system also demonstrated spatiotemporally controlled delivery, with a threefold higher colon accumulation than the control, likely triggered by dissociation under alkaline conditions in the intestine. Beyond pure molecular self-assembly, click chemistry offers an emerging method for constructing carrier-free DDSs. Multiple probiotics possess the inherent property of naturally binding to intestinal mucus [156,166]. Cao et al. utilized the natural adhesive characteristics of probiotics such as Bifidobacterium longum to anchor an artificial enzyme B-SA onto the bacterial surface via a boronic acid-diol click reaction [166]. This coating conferred stability against digestive fluids while preserving ROS-scavenging capacity. The probiotics' adhesion enabled prolonged retention at inflamed colon sites, and the synergistic action of probiotics and artificial enzymes effectively promoted colonic healing in a canine colitis model. Similarly, Ma et al. conjugated dextran with an aspirin prodrug via click chemistry to form self-assembling nanomicelles (P3C-Asp) [167]. These nanomicelles provided exceptional protection, releasing less than 20% of salicylic acid in SGF over 48 h, a rate significantly slower than that of free aspirin. These strategies exemplify the potential of click chemistry in programming complex, carrier-free systems for targeted GI therapy. External device control system Conventional protective and responsive drug delivery strategies often depend on the natural physiological pathways, making them susceptible to individual variability and multiple biological barriers. To achieve precise targeting, BUSS et al. developed the CLAMP technique, which amplifies targeting with an external magnetic field [168]. CLAMP uses micron-sized magnetic particles to generate a gradient over 1.0× 104 T·m-1 in the mouse GI tract. This tripled the capture of magnetically tagged bacteria, with 17% retained in the small intestine after 9 h—30 times more than without a magnet. For non-magnetic strains like EcN, Mg2-mediated electrostatic binding increased capture from 4.7% to 22%, eliminating the need for magnetic labeling and its associated problems. CLAMP allows precise targeting 10 mm from the source, surpassing the range of previous magnetic systems. Studies indicate that 9-h of magnetic guidance helps probiotics establish a foothold, enabling colonization to persist for 19 days after the magnet is removed. This active control method overcomes the random distribution of standard oral doses, offering a paradigm for probiotic delivery. Collectively, the diverse strategies discussed above represent distinct approaches to overcoming GI organ-level barriers, each offering unique advantages while exhibiting inherent limitations that ultimately define their therapeutic scope. Notably, optimizing GI delivery requires balancing functional performance with translational feasibility, as no single platform simultaneously maximizes stability, permeability, and manufacturability. To provide a clearer perspective on these trade-offs, Table 3 systematically compares representative platforms across key dimensions, including targeted barriers, design advantages, inherent limitations, optimal cargo types, and translational considerations, thereby facilitating the rational selection of delivery strategies for specific scenarios. Table 3Comparative analysis of current strategies for organ-level GI drug delivery.Table 3PlatformPrimary GI organ level barriers addressedAdvantagesKey limitationsOptimal cargo typesTranslational considerationsRefInert nanomaterialsGastric acid erosionSuperior stability during digestionPassive protection mainly; Limited versatilityAntioxidants; Metal ionsBiocompatibility needs evaluation[131,132]Synthetic polymersInsufficient organ targeting; Gastric acid degradation; Rapid GI transit; Dynamic mechanical stressMultiple responsive mechanisms; Site-specific delivery capabilityBiodegradability concerns; Uncertain long-term safety; Complex synthesis processLive biologics (probiotics); Macromolecular drugs; Liquid drugsPotential long-term toxicity & accumulation[110,112,[133], [134], [135], [136]]PolysaccharidesInsufficient organ targeting; Poor mucosal adhesion; Gastric acid degradationExcellent biocompatibility; Natural degradability; Abundant chemical modification sites; Inherent mucoadhesive properties; Enzyme-responsive releaseCompromised acid resistance without cross-linking; Batch variability from natural sourcesAntibiotics; Probiotics; Anti-inflammatory drugs; Hemostatic agentsStandardization needed[[113], [114], [115], [116]][[146], [147], [148], [149]]MOFsGastric acid degradation; Poor intestinal retentionHigh specific surface area; Tunable pore structures; High encapsulation efficiency; Dual adhesion (MPN)Uncertain long-term safety; Complex preparation and characterization processesEnzymes; Polyphenols; Probiotics; Hydrophobic drugsComprehensive toxicological evaluation; Regulatory pathway needs establishment[[154], [155], [156], [157], [158],169]Plant-based carriersGastric acid instability; Rapid gastric emptying; Poor colonic targetingExcellent biosafety; Cost-effective and sustainable sourcing; High drug loading capacityHigh batch-to-batch variability; Unclear composition; Challenges in standardization & scale-upSmall-molecule drugsNeed for standardization protocols and quality control measures[[159], [160], [161], [162]]Drug self-assembly systemsGastric acid degradation; Poor bioavailability of poorly soluble drugsMaximized drug loading; Biocompatibility & low cost without excipient; Sustained release capabilityPoor structural stability; Limited to specific drug classes; High batch-to-batch variabilityFlavonoids; Alkaloids; Phenolic compoundsPoor batch-to-batch reproducibility[[164], [165], [166], [167]]External device controlIndividual physiological variability; Poor targeted retentionActive, precise targeting (physiology-independent); Overcomes individual variability in GI transitPatient compliance; Technical complexity; High cost; Limited clinical feasibilityProbioticsUser-friendly design required; Safety and reliability need to be proven[168,170] Comparative analysis of current strategies for organ-level GI drug delivery. Strategies for overcoming mucosal barriers For oral drugs targeting systemic absorption or action in the deeper layers of the mucosa, it is a prerequisite that the mucosal barrier be traversed. This process involves two sequential major challenges: first, the penetration of the mucus layer to establish direct contact with the epithelial cells, and subsequently, the crossing of the epithelial barrier itself to achieve effective absorption or local action. The ensuing sections delineate the strategies devised to surmount these distinct yet interconnected impediments. Strategies for overcoming the mucus barrier Strategies to overcome the mucus barrier have evolved from early mucus dissolution or adhesion towards approaches focused on enhanced penetration. Initial methods like mucus disruptors were abandoned due to irreversible barrier damage, and pure adhesion risks rapid clearance; recent efforts aim to balance adhesion with penetration. For instance, Liu et al. developed a double-network hydrogel inspired by mushroom deformation [171]. This system uses silicon NPs for strong acid-resistant adhesion, achieving an adhesion force of 81 ± 5 kPa at pH = 1. Simultaneously, the dual-network DMA/SA hydrogel crosslinked with CBA/Ca2+ actively adapts to gastric motility through compression-expansion behavior, penetrating the mucus layer to adhere to the perforation site. Another strategy leverages molecular chirality to precisely tune this balance. PEI-L-TA@MON with L-chiral ligands exhibits stereoselective adsorption to mucosal glycans, achieving robust adhesion (66.7% retention post-flushing). Concurrently, the L-form structure minimizes steric hindrance, significantly enhancing penetration depth in inflammatory mucus to 143.7 μm, far exceeding the 91.7 μm of its D-form counterpart. This provides a novel paradigm for synergistically enhancing adhesion and penetration [172]. However, compared to balancing these properties, there is increasing interest in strategies that directly improve mucus penetration. Recent research has systematically identified key factors—including size, charge, shape, surface properties, and responses to external fields—that influence the penetration efficiency of DDS through mucus, uncovering complex underlying mechanisms. The ability of particles to traverse the mucus layer is primarily governed by their size, due to the sieve-like nature of the mucus mesh. Research consistently shows that smaller carriers are less likely to be retained by mucus fibers. For instance, Henriques et al. found that CS microspheres with reduced size and lower mucoadhesion exhibited the most pronounced penetration through gastric mucus [173]. The critical size for efficient diffusion was identified by the Erstling team to be approximately 20 nm [20]. Silica NPs below this threshold displayed diffusion coefficients (1-33 μm2 s−1) in simulated intestinal mucus sufficient to reach the epithelium before mucus clearance, whereas larger particles were significantly hindered. Beyond size, the surface charge of particles is another critical parameter determining their penetration rate, given the negative charge of the mucus matrix. It is well-documented that neutral or negatively charged particles typically penetrate more easily by minimizing electrostatic binding to mucins [174]. This is evidenced by a fucoidan coating that shifted nanoparticle zeta potential to −50 mV and nearly doubled penetration through porcine gastric mucus [175], and by PEGylated, near-neutral NPs that showed excellent penetration across various pH conditions [10]. However, the influence of charge is not absolute, as positively charged particles (e.g., CS-modified) can also achieve efficient penetration under specific conditions [[176], [177], [178]]. Zhang et al. proposed a mechanism of reversible electrostatic adsorption, allowing for gradual permeation through the dynamically renewing mucus layer [176]. This anomaly highlights that the effect of charge is modulated by the complex and dynamic properties of the mucus environment. Finally, the geometric configuration of NPs constitutes a pivotal factor influencing penetration efficiency. Anisotropic nanostructures, such as nanorods, leverage a higher aspect ratio to enhance mobility within the mucus network. Their rotational and translational diffusion mechanisms facilitate deeper penetration, increasing epithelial contact opportunities [179]. Experimentally, nanorods exhibit significantly higher GI permeability and prolonged retention (2-6 h) compared to nanospheres. This principle is exemplified by Salmonella-mimicking gold nanorods (SM-AuNRs), which demonstrated more unrestricted trajectories and diffusion coefficients 3.7 times greater than spherical Lips [180]. Consistently, Liang's team observed that rod-shaped exosomes had mean square displacement values 9.2 times higher than their spherical counterparts, alongside significantly increased penetration depth in 3D mucus models [181]. In light of the established mechanisms by which size, charge, and shape influence mucus penetration, a series of sophisticated surface engineering strategies has been developed by researchers to regulate the interaction between nanocarriers and the mucus network. A common strategy to improve mucosal permeability is surface hydrophilic modification, which leverages a steric hindrance effect from a surface-bound hydration layer to reduce mucus adsorption. Although PEG coatings were historically the mainstream choice for this purpose, issues such as immunogenicity have prompted a shift toward zwitterionic materials. The Veide group directly compared lipid nanocrystals (NCs) modified with zwitterionic (ZW), polyglycerol (PG), or PEG ligands for their ability to penetrate mucus and cross the intestinal epithelial barrier [11]. Their results showed that ZW/PG-NCs achieved mucus permeability comparable to PEG-NCs, while ZW-NCs exhibited a 4.5-fold higher cellular internalization rate in a Caco-2/HT29-MTX co-culture model. Thus, ZW and PG modifiers represent highly promising alternatives to PEG for enhancing the efficacy of oral lipid NCs. Effective surface chemical regulation requires coordinated control of hydrophilicity and charge. For instance, the PDA-modified silica nanoparticle designed by Ma et al. displayed peak hydrophobicity at pH 3.0, strengthening hydrophobic binding with mucin and severely impeding penetration [12]. At the isoelectric point (pH 5.6), however, the particles achieved both electroneutrality and minimal hydrophobicity, reducing the mucin binding constant by 7.7-fold (to 2.93 × 107 m−1) and increasing the diffusion rate threefold. Natural lipid coatings can function through multi-mechanistic synergistic effects. For example, tea tree leaf lipids disrupt the mucus barrier via a triple synergistic mechanism [182]: phospholipid hydrophilic heads form a hydration layer; charge neutralization reduces adhesion; and hydrophobic sugar chains insert into the mucin network, inducing temporary depolymerization. Externally applied devices have opened novel avenues for active penetration. Under near-infrared irradiation, for instance, gold nanocapsules generate thermophoretic forces that propel particles through the mucus barrier at 13.9 μm s−1—a tenfold increase in speed [183]. Notably, protein corona regulation is emerging as a novel strategy for breaching the mucus barrier. Modifying NPs with small circular FcRn-binding peptides enhances their effective diffusion coefficient in intestinal mucus by 10.6-fold. This improvement stems from a reduced mucin corona thickness, which significantly optimizes penetration depth. Strategies for overcoming the epithelial barrier Transepithelial transport is known to occur via two pathways: the paracellular route through TJs or the transcellular route via endocytosis. The paracellular route is a well-established penetration strategy, especially in intestinal DDS. Erstling et al. reported that C′ dots smaller than 10 nm passively diffuse through Caco-2 TJs with an apparent permeability coefficient >1 × 10−6 cm s−1, whereas particles larger than 20 nm shift to active transport [20]. Li's group found that highly rigid particles improve penetration by reversibly reorganizing TJs via JNK signaling inhibition, disrupting ZO-1 localization. Unfused nanotubes (20 nm diameter, 200 nm length) can reversibly open TJs, but this effect weakens when the length increases to 1 μm. After glutaraldehyde cross-linking, short tubes stiffened ∼3-fold, optimizing TJ modulation. A sharp TEER drop within 2 h and reduced ZO-1 expression (to 0.22) confirmed that smaller size and greater stiffness enhance TJ regulation [18]. Transcellular pathways are governed by receptor-mediated or physicochemical mechanisms. The receptor-mediated pathway achieves highly efficient transport through diverse mechanisms. A prime example is FcRn-mediated endocytosis. Milk-derived exosomes exploit this pathway to attain at least a 2.5-fold higher transport efficiency than polyethylene glycol-modified Lips [184]. Similarly, FcBP-peptide-modified NPs designed to target FcRn induce clathrin-independent endocytosis, yielding an 8.2-fold increase in transport within 6 h [185]. Other receptor-mediated strategies in ordinary epithelial cells include: lectin-1-mediated endocytosis, which significantly boosts the permeability coefficient of coenzyme Q10 via the lactoferrin pathway [186]; and nutrient transporter-directed approaches, such as glucose-modified NPs that bind to the SGLT1 receptor for site-specific absorption in the proximal small intestine [176]. Beyond these, M cells, which are pivotal in intestinal immune surveillance, actively phagocytose exogenous substances for delivery to the lymphatic system. Targeting M cells has thus become a major research focus for enhancing intestinal transport. For instance, yeast β-glucan-modified carriers enhance absorption in Peyer's patches by activating phagocytosis through the Dectin-1 receptor [187]. Oxidized CS (Ocur), a modified form of CS, also enables the selective targeting of M cells within microparticles, significantly enhancing transcellular efficiency and promoting drug accumulation in mesenteric lymph nodes [188]. Furthermore, Li et al. found that engineering outer membrane vesicles to display the Co1 peptide improves their recognition and transport by M cells [189]. Non-receptor pathways depend largely on the physicochemical characteristics of the carrier. While NPs between 25 and 500 nm are internalized by Caco-2 cells [17], their transport efficiency and mechanisms across the epithelium are strongly influenced by particle size [28]. For instance, Guo et al. observed that polystyrene particles of 100-150 nm were most efficiently internalized by epithelial cells via the clathrin-dependent pathway, whereas 300 nm particles were predominantly taken up by caveolae-mediated endocytosis [190]. Similarly, He et al. found that 50 nm polycaprolactone NPs exhibited significantly greater intestinal penetration and subsequent hepatic transport compared to 2000 nm particles [191]. Beyond size, particle shape and surface charge also considerably affect cellular uptake. Non-spherical particles, for example, exhibit prolonged gut retention and enhanced uptake efficiency [179,192,193]. In a study by Wu et al., the hydrolysis of phosphoserine by intestinal alkaline phosphatase shifted the nanoparticle surface charge from −1.87 mV to +7.37 mV, exposing the cationic cell-penetrating peptide octa-arginine [38]. This modification increased Caco-2 uptake threefold and improved trans-epithelial transport efficiency. Zhang et al. employed Polyampholytic P (OEDMA-co-CL) as a carrier material, which promoted transport across intestinal epithelium by resisting protein adsorption and forming weak interactions with the cell membrane [194]. Metabolite-mediated uptake offers another promising strategy. Li et al. designed butyrate-functionalized Lips that leverage butyrate's role as an energy source for colonic epithelial cells [13]. These Lips showed a ∼4-fold increase in cellular uptake in NCM460 and IEC-6 cell lines over unmodified ones. Consistent with this, in vivo results demonstrated high epithelial targeting specificity, with a 7.3-fold greater uptake in intestinal epithelial cells (EpCAM+) than in hematopoietic cells (CD45+), and predominant Lip accumulation within the epithelial layer. Overcoming epithelial barriers also requires efflux pump inhibition. Self-assembled BBR-MAG NPs raised BBR's Cmax and AUC by 3.8- and 3.2-fold, combining nano-size with MAG-induced P-gp inhibition [165]. In a related approach, Han et al. developed hybrid vesicles derived from milk exosomes and Lips, which retain a positively charged surface and membrane proteins [195]. These vesicles efficiently traverse multiple intestinal barriers—apical, intracellular, and basal—in ileal epithelial models. Mucus penetration and epithelial penetration represent sequential yet interrelated stages in oral drug delivery, exhibiting synergistic potential when carrier properties are optimally matched. Synergy is demonstrated by small particles, which diffuse readily through the mucus mesh and enter epithelial cells via TJs or transcellular pathways. Similarly, zwitterionic surface modifications reduce mucin adhesion and can enhance transport by promoting cell membrane fusion. Nanorod morphologies also offer a dual advantage by improving mucus penetration and potentially increasing the contact area with cells to facilitate uptake. These examples illustrate how rational engineering of carrier physicochemical properties can achieve concurrent benefits across sequential barriers. Strategies for overcoming the bacterial/cellular barrier After crossing the epithelial barrier, drug carriers must further infiltrate submucosal immune or stromal cells to exert therapeutic effects. Efficient drug uptake can be achieved by targeting receptors highly expressed on specific cells. While this approach shares mechanistic similarities with trans-epithelial transport, it often involves distinct receptors and intracellular fates. For example, LMW-heparin targets integrin αM, which is highly expressed on inflammatory macrophages but minimal on epithelial cells. In a colitis model, this specificity resulted in a 12-fold higher accumulation in lesions than in healthy tissue, with deep penetration into the αM+ immune cell-rich submucosa [196]. Besides, glycan ligands can be employed for the specific targeting of macrophage receptors. Mannose-modified peptide dendrimers achieved 98% uptake in activated macrophages, substantially higher than the 65% uptake of unmodified controls [197]. Chondroitin sulfate modification similarly enhanced nanozyme uptake by macrophages at inflammatory sites by 2.3-fold via scavenger receptor SR-A [107]. A more refined strategy targets receptor activation states. These are unique molecular surfaces exclusively presented on the activated form of a receptor, often resulting from a change in its three-dimensional structure. As demonstrated by Dammes et al., RG7-decorated lipid NPs (LNPs) selectively bind the high-affinity conformation of α4β7 integrin on inflammatory leukocytes [198]. These particles exhibited eightfold stronger binding to activated versus resting integrins, accumulated in inflamed colon regions in correlation with disease severity, and restored colon length to ∼8 cm in mice treated with siIFNγ-loaded LNPs, compared to ∼6.5 cm in non-targeted controls, without inducing systemic immunosuppression. Some single ligands inherently enable targeting across multiple cell types. Lactoferrin, for instance, not only promotes epithelial transcytosis but also mediates LRP1-dependent endocytosis of Lactoferrin-Lips by colonic M1 macrophages, enhancing uptake by 2.5-fold [199]. Similarly, HA promotes carrier internalization by binding to the CD44 receptor, which is highly expressed on both epithelial cells and macrophages. This allows HA-modified carriers to penetrate both barriers, a strategy reported to enhance uptake by 3- to 85-fold [142,152,[200], [201], [202]]. Alternatively, delivery efficiency can be enhanced by engineering dual-targeting strategies that employ multiple ligands to overcome several biological barriers simultaneously. A prime example is SST/MOS@EUP-SeNP, which simultaneously targets SSTR2/SSTR5 receptors on epithelial cells and CD206 receptors on macrophages. This design increased overall cellular uptake by 82%, elevating fluorescence signal intensity by 4.3-fold in epithelial cells and 5.1-fold in macrophages compared to non-targeted controls [145]. Beyond chemical ligands, biomimetic membrane modifications leverage natural membrane proteins to achieve cell-specific targeting. For example, Han et al. developed Effero-RLP, a biomimetic Lip employing phosphatidylserine from apoptotic erythrocyte membranes to target the macrophage Tim-4/MerTK receptor [203]. These Lips showed fivefold greater macrophage internalization than control cells and eightfold higher accumulation in inflamed colons compared to conventional Lips, while also inducing efferocytosis. Similarly, macrophage-coated NPs retaining adhesion molecules and cytokine receptors exhibit enhanced chemotaxis toward inflammatory sites and improved immune recognition through homophilic interactions [204,205]. Gut microbiota-derived vesicles, such as defatted polysaccharide outer membrane vesicles from EcN, also possess intrinsic affinity for intestinal mucosa and immune cells, increasing drug-encapsulated cell uptake threefold [144]. The physical properties of drug carriers—such as size, shape, stiffness, and surface charge—critically govern their cellular internalization and interactions with cell membranes [206]. Size plays a particularly decisive role: smaller particles are less efficiently phagocytosed by macrophages than larger particles [[207], [208], [209]]. Beyond uptake, particle size significantly influences downstream immune responses, including macrophage polarization and lysosomal damage potential [210]. The effect of nanoparticle shape on cellular uptake has been shown to be different in various cells. Spherical particles typically benefit from lower membrane bending energy, facilitating endocytosis [[211], [212], [213]]. In contrast, high-aspect-ratio nanostructures like nanorods and carbon nanotubes can achieve superior uptake in specific cells (e.g., macrophages) via increased contact area or direct penetration [129,[214], [215], [216], [217], [218]]. This principle extends to nanoneedles, which locally stress the double membranes of H. pylori to induce invagination [219], and is consistent with the high uptake efficiency of rigid particles [220,221], as observed in intestinal epithelia. Surface charge conventionally favors positive values for membrane adsorption and endocytosis. However, recent studies reveal more complex behavior. For example, Ag9 nanoclusters with a strong negative charge (−42.5 mV) show higher uptake and ROS scavenging in inflammatory cells than Ag NPs with weaker electronegativity [222]. This may stem from specific protein corona formation or electrostatic interactions within the inflammatory microenvironment, rather than simple charge attraction. Such findings caution against predicting charge effects solely based on zeta potential. Notably, the protonatable nature of ionizable lipids in LNPs may outweigh apparent surface charge: negatively charged LNPs can protonate within acidic endosomes, significantly enhancing mRNA delivery and expression [223]. Nanoparticle composition also modulates cellular internalization behavior through tailored chemical properties. For mammalian cells, both lipid-based and synthetic polymeric carriers are effective. LNPs with optimized DMPE-PEG (5%) promote macropinocytosis and endosomal escape, doubling transfection in intestinal epithelia [136]. Natural lipid exosomes, such as those from milk or ginger, exploit their PE/PC-rich layer—entering macrophages via clathrin-mediated endocytosis [195,224]. Synthetic chemical carriers, tyramine-oxidized aggregates, enhance uptake by boosting endocytosis [225]. For bacterial targeting, representative strategies include: polyphenol-mediated membrane adhesion to shorten the distance between antimicrobial agents and bacteria [226]; maltodextrin-modified DNA origami structures that exploit bacteria-specific transporters for efficient targeting [120]; and phage-PSg chains achieving over 90% targeting efficiency against Sg [103]. Therapeutic efficacy depends not only on total cellular uptake but also on the ability to deliver drugs to specific organelles and achieve efficient release. Lysosomal escape is a critical step for enhancing intracellular bioavailability, as drugs often require active escape from lysosomes after entering immune cells to exert their functions. The following strategies have been proposed: the utilization of high aspect ratio structures to evade lysosomes via macropinocytosis; the design of "proton sponge" effects [109,126,227,228]; and generating intracellular CO2 to rupture lysosomal membranes [229]. Near-infrared-triggered mild hyperthermia has also been suggested to facilitate escape [183]. In some cases, however, restoring—not escaping—lysosomal function is required, such as in pathogen clearance. This is exemplified by calcitriol, which promotes H. pylori elimination by reinstating lysosomal acidification in host cells [174]. Organelle targeting can be accomplished using specific ligands or physical design. For example, triphenylphosphine modification enables mitochondrial targeting via membrane potential, yielding colocalization coefficients of 0.76-0.85 [151,230]. Nanorods developed by Liang et al. preferentially target the endoplasmic reticulum, showing significantly higher colocalization than spherical particles [181]. Pathological environmental response enrichment/release/activation Targeted enrichment strategies for GI inflammation and infection Distinct physicochemical features of the pathological site drive smart drug delivery systems that sense pathological cues for improved site-specific accumulation and controlled release. The inflamed GI microenvironment, characterized by disrupted epithelial barriers and immune cell infiltration, facilitates targeted drug accumulation. For instance, Lu et al. demonstrated that GSH-modified gold nanoclusters (∼1.95 nm) accumulate in inflamed colon tissue by leveraging epithelial barrier disruption, enhanced phagocytosis, and vascular permeability, with surface GSH further improving colon affinity [231]. Furthermore, a widely adopted strategy for inflammatory targeting involves imparting a sustained negative surface charge to delivery systems, promoting their electrostatic adhesion and retention at inflamed mucosal sites. This principle is demonstrated across material classes: Natural clays such as MMT, with its intrinsic negative charge, enhance adhesion as shown by CeO2@MMT (−29.7 mV, 9.45-fold adhesion increase) [232], MMT-engineered EcN (22.6-fold colonization enhancement) [233], and DPs@MMT for prolonged retention [234]. Polymer-based systems achieve targeting through diverse mechanisms. Electrostatic contributions are evident in HMW-HA gels (4-fold signal increase) [9], alginate-CS coatings for bacterial adhesion [121], and CS/alginate layer-by-layer assemblies [118]. Synthetic polymer coatings like PDA (shifting charge to −2.5 mV for 3-fold accumulation) [235], Polyacrylic acid [236], and ethyl methacrylate copolymers (−25 mV) [105] further exemplify this approach. Small-molecule modifiers provide another versatile route. Tannic acid-modified nanoplatforms exhibited a zeta potential of −37 mV, with further charge enhancement in GI environments and significantly improved accumulation at inflammatory sites [237]. The innovative use of L-tartaric acid introduces chiral recognition, yielding 1.3-fold greater mucosal adhesion and >24-h inflammatory retention—far exceeding the 6-h retention in healthy tissues [172]. These studies collectively establish that imparting a negative surface charge represents a simple, versatile, and effective strategy for inflammatory targeting. To further enhance targeting precision, active targeting mechanisms have been integrated with passive strategies. This synergy is exemplified by LMW-HA, which overcomes the steric hindrance of HMW-HA to enable concurrent CD44 receptor-mediated endocytosis and electrostatic interactions. For instance, LMW-HA-modified micelles exhibited 2.9-fold higher inflamed colon accumulation than free drug [201], while Spore@CH spores utilized HA-mediated targeting and charge effects to increase colonic retention 1.4-fold [138]. Similarly, SHSe microspheres combined HA-mediated active targeting with a negatively charged alginate shell (−39.7 mV), tripling colonic selenium levels and prolonging retention to 24 h [143]. This dual-targeting logic also applies to chondroitin sulfate-based systems, which leverage CD44 binding and negative charge to enhance both targeting and residence time beyond 72 h [131]. Building on the molecular synergy of HA, the gel system further leverages its targeting capability to achieve robust adhesion and prolonged retention. A decellularized porcine intestinal submucosa hydrogel integrated HA targeting with epigallocatechin gallate-derived negative charge [238], whereas the thermosensitive metabolic nanoregulator (TMNR) hydrogel combined α-cyclodextrin/HA dual adhesion with rapid gelation (<15 s) for eightfold inflammatory site retention [239]. Similarly, oxidized HA nanogels promoted M1 macrophage uptake and extended fluorescence retention in colitis models [127]. Furthermore, superior enrichment efficacy can be achieved by integrating multiple targeting motifs or even whole cells. The EcN-Apt@HG system combined electrostatic adsorption, CD44 targeting, and IL-6 aptamer recognition to enhance bacterial colonization by 197.2-fold over bare bacteria [123]. Beyond molecular design, cell-mediated targeting offers a bio-hybrid strategy; macrophage-bacterial complexes (GM-EcN) exploited innate macrophage homing to the inflamed gut, achieving selective adhesion and 72-h intestinal retention [240]. Stimulus-response strategies for GI inflammation and infection Based on realizing efficient enrichment, precise control over drug release within specific microenvironments is pivotal to enhancing therapeutic efficacy. Among various strategies, pH responsiveness remains a classic approach. For instance, RHL/Cl-Ch-cal vesicles maintain stability in gastric acid but undergo structural relaxation and accelerated drug release at the weakly acidic sites of infection, due to the partial deprotonation of rhamnolipid carboxyl groups [174]. Similarly, the core-shell structured W@ZnNPs leverage the high gastric stability of their tungsten shell to protect the zinc core, enabling controlled zinc and tungsten release in the weakly acidic inflamed colon [132]. Redox-responsive nanostrategies exploit the elevated levels of ROS and GSH in inflammatory lesions to achieve intelligent drug release, local retention, and even diagnostic capabilities. For example, ratio-fluorescence sensing systems [104] and superoxide anion-specific probes [142] have been developed for quantifying H2O2 and O2−•. ROS-responsive systems based on dynamic sulfur bonds follow two main design strategies: one focuses on precise inflammatory-site delivery through bond cleavage. Lin et al. constructed Cas9/sgRNA nanoclusters using disulfide bonds, achieving >20% release within 2 h under 10/100 mm H2O2 and enabling conditional gene editing in inflammatory cells [144]. Kuang et al. used an aromatic thioacetal linker to synchronize drug and probiotic EcN release in colitis, enhancing local enrichment [119]. Xiong et al. developed a thio-ketone crosslinked system that shows minimal release in SGF/SIF but rapid release in 1 mm H2O2, confirming inflammatory-triggered drug deployment [155]. Another strategy aims for simultaneous drug release and ROS scavenging. Xu et al. used a tetrasulfur bond to confer anti-oxidant and inflammation-responsive properties to mesoporous organosilica NPs [172]. ROS-triggered release of gaseous signalling molecules offers another avenue for anti-inflammatory therapy. L-Arg-based nanogels release ∼3 μm NO upon H2O2 stimulation, significantly reducing intracellular ROS [127]. In contrast, the CO release system exhibits a unique immunomodulatory function. Fe3(CO)12 decomposes under oxidative stress to release CO, which promotes M2 macrophage polarization via the Notch/Hes1/Stat3 pathway, suppressing TNF-α and IL-6 while elevating IL-10 [118]. CORM-401-based nanomedicine rapidly releases CO in high-ROS environments, exerting antioxidant, anti-inflammatory, and antibacterial effects to remodel the gut immune-microbiome interface [197]. A microfluidic approach produced probiotic EcN and manganese carbonyl (MnCO) co-encapsulated microcapsules, where MnCO reacts with ROS to generate CO-propelled migration and NO-sensitive linkers enable targeted bacterial release in inflamed regions [140]. These systems leverage ROS-responsive mechanisms to release gas signaling molecules with distinct anti-inflammatory actions: NO directly scavenges ROS, while CO modulates immune cell function for broader immunoregulation. ROS-responsive strategies also enhance local drug retention. Nie et al. described tyramine-modified NPs that undergo ROS/MPO-mediated oxidative cross-linking, forming aggregates that anchor in inflamed tissues [225]. Huang et al. developed Lac@HBD probiotics where phenylboronic acid groups cleave under high ROS, exposing catechol groups for strong mucosal adhesion and prolonged retention [241]. The former operates via “crosslinking-aggregation,” the latter via “cleavage-exposure,” showcasing versatile adhesion control. ε-Polylysine-based NPs containing phenylboronic ester groups further demonstrate dual ROS responsiveness and scavenging, modulating intestinal redox balance [196]. Similar to ROS-responsive systems, GSH-responsive ones also operate through dynamic covalent bonds (e.g., disulfide, diselenide) to control drug release at target inflammatory sites. Zhang et al. achieved 54.31% release of cannabidiol in a 10 mm GSH environment simulating the inflammatory cell cytoplasm by exploiting disulfide bonds in α-lipoic acid [151]. Fan et al. showed complete disintegration of disulfide-containing MON carriers in 3 mg mL−1 GSH, mimicking H. pylori infection [113]. In another study, the PSBA-63@Mag system exhibited a 40% increase in drug release in 10 mm GSH due to disulfide cleavage [242]. The resulting thiol groups enhanced colonic fluorescence retention threefold in colitis mice—a “cleavage-exposure” mechanism akin to Huang et al. [241]. GSH-triggered gas therapy has been advanced using diallyl trisulfide (DATS) as an H2S prodrug. Zhang et al. reported a DATS prodrug that releases 3.58 μm H2S within 2 h in 2 mm GSH, polarizing macrophages toward an anti-inflammatory phenotype [243]. Similarly, Jin et al. constructed DATS@PDNs nanogels for GSH-triggered H2S release, yielding antioxidant, anti-inflammatory, and antibacterial outcomes [234]. Collectively, redox-responsive strategies—whether ROS- or GSH-driven—exploit shared principles of the inflammatory microenvironment. Future development of multi-responsive systems capable of sensing combined redox signals could allow more precise discrimination and modulation of inflammatory states. Beyond redox-responsive strategies, enzyme-responsive DDS leverage proteases that are overexpressed in inflamed or infected tissues. MMP-responsive designs often incorporate specific cleavage sequences (e.g., GPLGVRG) or natural substrates like Gel. For example, Zhang et al. developed porous Gel microgels encapsulating AKK bacteria, enabling precise probiotic release and sustained anti-inflammatory therapy in the colon via Gel degradation by MMPs [15]. Similarly, Chen et al. used oral Gel microspheres to achieve targeted drug release at infection sites by exploiting MMP-9 overexpression [122]. To precisely control intestinal retention, Xie et al. designed a series of hydrazone linkage-crosslinked Gel-PEG hydrogels (GP-2, GP-4, GP-8) with tunable degradation rates. The more densely crosslinked GP-8 hydrogel degraded more slowly under high MMP-9 conditions, supporting extended drug release and highlighting the value of degradability-matched design [244]. The degradation kinetics of such systems often depend on enzyme concentration. Lai et al. reported an ascorbyl palmitate hydrogel that degraded over 90% within 6 h at MMP concentrations ≥6.25 × 10−2 mg mL−1—significantly faster than in PBS (25%)—demonstrating enzyme-driven release at inflammatory sites [178]. Other enzyme-responsive mechanisms also show promise. NTR, abundant in inflamed colons, can trigger material disassembly. Chen et al. developed an NTR-sensitive hydrogel containing a 4-nitrobenzyl cage; upon NTR activation, the gel becomes hydrophilic and dissociates, releasing over 90% of probiotics within 6 h SCF [245]. Calprotectin (CP), another inflammation-associated protein, has been harnessed for ion-responsive therapy. Yang et al. designed calcium tungstate NPs and a sodium tungstate-alginate gel, both of which release tungsten ions in simulated inflammatory environments. The released tungsten selectively inhibits Enterobacteriaceae with minimal impact on probiotics, thereby preserving microbial balance. Tungsten release from gel increased nearly 13-fold at CP concentrations of 500 μg mL−1 [246,247]. Together, these systems—responsive to MMPs, NTR, or CP—illustrate a general design paradigm: evince a precision design approach predicated on the enzymatic profile characteristics of the lesion microenvironment. The expansion from classical proteases to broader enzyme targets underscores the versatility and adaptability of enzyme-responsive strategies for GI inflammation and infection. Synergistic enrichment-response strategies for GI inflammation and infection Recent research has increasingly focused on synergistic strategies that combine targeted drug enrichment at pathological sites with local stimulus-responsive release. By achieving site-specific accumulation followed by controlled drug release, this approach enhances therapeutic precision while minimizing systemic side effects. Current strategies based on enrichment mechanisms mainly fall into two categories, both often integrated with ROS-responsive release. One strategy leverages electrostatic enrichment via the positive surface charge of inflamed tissues. For instance, Chi et al. constructed hesperidin-arginine NPs through Mannich reaction and π–π stacking [248]. These NPs carry a negative zeta potential (−22 mV) in the GI environment, promoting specific accumulation in positively charged regions of the inflamed colon. Similarly, Sun et al. developed self-assembled quercetin nanoribbons that exhibit enhanced enrichment in the inflamed colon due to their negative charge (−18.5 mV) and large size (∼13.9 μm) [164]. Both systems exert anti-inflammatory and antioxidant effects by scavenging free radicals and triggering NO release in high-ROS microenvironments. Another strategy employs receptor-ligand interactions for active targeting, with HA-based nanosystems serving as a prominent example. Representative systems illustrate how HA can be engineered to achieve diverse therapeutic functions. For instance, Xu et al. developed a diselenide-bridged HA nanogel that concurrently provides redox-responsive ROS scavenging and active targeting to CD44-overexpressing inflammatory cells [200]. Uthaman et al. similarly incorporated a thioacetal linker into an HA-based carrier to enable ROS-triggered drug release [202]. Separately, Jin et al. adopted an alternative strategy by constructing a nanogel from a hyperbranched polymer (HBPAK) and HA-methacrylate (HA-MA) [142]. This system leverages the ROS-scavenging capability of HBPAK while retaining the CD44-targeting function of HA-MA, emphasizing a dual-functional design beyond linker chemistry. In a related approach, Lee et al. developed HABN, in which HA serves as a CD44-targeting ligand for inflammatory site accumulation, while bilirubin (BR) functions as both a hydrophobic core and a ROS-responsive unit that oxidizes to initiate drug release [249]. In contrast, Chen et al. designed mRNA/EPHB particles by grafting BR onto HA to form an antioxidative shell that protects mRNA from degradation [228]. In this system, HA preserves CD44 targeting and improves colloidal stability. These distinct designs result in divergent ROS-responsive behaviors: the former prioritizes controlled drug release, whereas the latter emphasizes structural stability for effective gene delivery. Overall, the “enrichment-response” cascade offers an effective framework for precision therapy in inflammatory bowel disease. Design strategies have accordingly evolved from passive accumulation to multi-level responsiveness and from electrostatic adsorption to specific biological recognition. Release/activation of external field force response Compared to strategies relying on endogenous microenvironmental changes, externally controlled systems—especially photodynamic and sonodynamic therapy (SDT) provide superior spatiotemporal precision and treatment reproducibility through external ultrasound or light modulation. These approaches are particularly advantageous for deep-tissue sterilization and immunomodulation. For example, Wang et al. developed an antibody-conjugated Lip (Hp Ab-LiP-ICG) that targets H. pylori surface antigens [250]. Under ultrasound excitation, it activates indocyanine green (ICG) to generate singlet oxygen (1O2), eradicating bacteria without inducing resistance while enabling photoacoustic imaging for therapy monitoring. Similarly, Yin's team constructed ICG@FCS nanosystems that, upon ultrasound activation, enhance 1O2 production and promote autophagy, exerting dose-dependent bactericidal effects against both planktonic bacteria and biofilms, with optimal efficacy after 10 min of irradiation [177]. Sonodynamic strategies can also synergize with chemodynamic therapy (CDT) to overcome hypoxia-limited SDT efficacy. Yu et al. developed Fe-HMME@DHA@MPN, integrating an iron-based sonosensitizer with dihydroartemisinin to supply exogenous H2O2 and enhance peroxidase-like activity [226]. Under ultrasound, this system generated 2.1-fold more 1O2 than pure HMME, increasing the bactericidal rate from 40% (CDT alone) to 100%, demonstrating strong SDT-CDT synergy. Acoustically responsive systems can further modulate the immune microenvironment. Jin's group engineered a probiotic system (IL10-EcN@DNPs) using a temperature-sensitive promoter and focused ultrasound to achieve on-demand IL-10 expression [120]. Ultrasound stimulation (30 kHz, 5.66 W cm−2, 30 min) induced IL-10 secretion up to 1500 pg mL−1, whereas no expression occurred without stimulation. The system remained responsive over five on-off cycles and promoted macrophage polarization from M1 to M2, demonstrating reversible and reproducible immunomodulation. Modern externally responsive DDS have thus evolved from simple antimicrobials to multifunctional platforms integrating immune regulation and real-time imaging. By optimizing photosensitizers, designing cascade reactions, and engineering genetic circuits, these systems increasingly overcome limitations such as microenvironmental constraints, poor targeting, and limited control precision, offering powerful tools for precise intervention in GI inflammation. Intrinsic multifunctional materials Developing oral formulations that can withstand the harsh GI environment and precisely target intestinal inflammation remains a significant challenge. While conventional approaches often rely on complex multi-component assemblies, recent studies show that certain single materials can intrinsically integrate multiple functions—such as digestive resistance and inflammatory targeting—thereby simplifying system design and improving therapeutic outcomes. These materials can be classified into natural/biomimetic, synthetic, and high-performance inorganic nanomaterials. Natural biomaterials are widely used to validate this simplified strategy due to their inherent biocompatibility and structural resilience. A key example is plant-derived vesicles, which inherently integrate stability and targeting functions. Gao et al. isolated turmeric-derived nanovesicles via ultracentrifugation; these lipid- and protein-rich nanostructures resist gastric degradation without further modification and promote macrophage uptake and polarization [251]. Expanding on this, Zhang et al. loaded curcumin into ginger-derived vesicles, retaining the natural gastric protection and targeting properties while enhancing immunomodulatory efficacy [252]. A broad range of plant-derived nanovesicles—from grapes, grapefruit, broccoli, sweet orange, honeysuckle, ginseng, and tea—have similarly shown promise in IBD models, owing to their distinctive native compositions [253]. Beyond plant-based systems, milk-derived exosomes offer a highly efficient platform for nucleic acid delivery. Han et al. developed M-Exo/siR, an oral system that protects TNF-α siRNA from degradation and improves intestinal uptake and colonic accumulation via natural lipid components [195]. This dual functionality reduced TNF-α expression by 67.2%, significantly surpassing free siRNA or cell-derived exosomes. Similarly, certain protein-based nanostructures exhibit intrinsic multifunctionality. NPs derived from Bacillus thermophilus spore coat protein, for instance, maintain structural integrity under GI conditions and target inflammatory sites through negative surface charge, facilitating epithelial repair [254]. Other natural systems employ distinct functional mechanisms. Escherichia coli membrane-coated NPs leverage their membrane lipid structure for gastric acid resistance and use surface proteins to actively target the colonic mucus [255]. Similarly, yeast β-glucan shells provide both acid resistance and M cell targeting via Dectin-1 binding [243]. Fucans enhance transport by reducing mucoprotein adsorption, achieving an apparent permeability coefficient of 2.84 × 10−5 cm s−1, while also facilitating uptake in gastric epithelial cells and macrophages through scavenger receptor mediation [175]. Natural minerals such as CaCO3-based NPs offer yet another multifunctional platform, neutralizing gastric acid to protect payloads while promoting intestinal cellular delivery via dissolution-precipitation processes [256]. Inspired by natural systems, biomimetic materials further expand functional integration through rational design. The engineered probiotic EcN-Mel, for example, synthesizes melanin in situ via tyrosinase expression. This pigment protects the bacteria from GI degradation and, due to its negative charge, enhances retention and colonization in the inflamed colon [257]. In a related biomimetic strategy, Park et al. developed self-degrading NPs that mimic ROS scavenging and self-protection mechanisms in pyomelanin-producing microbes [258]. Their pyomelanin NPs, formed from homogenistic acid-γ-lactone polymerization, remain stable in gastric acid without mucosal adhesion. Upon intestinal entry, carboxyl group deprotonation triggers disassembly into sub-5 nm particles, which utilize small size and a near-neutral charge for efficient mucus penetration. Collectively, these systems demonstrate that intelligent material selection and bio-inspired design can achieve efficient, multifunctional oral delivery without excessive engineering. Synthetic materials offer superior structural and functional design flexibility compared to their natural counterparts. Representative examples include PDA and melanin-like materials, which exhibit tunable surface chemistry and excellent environmental stability. As a protective coating, PDA significantly improves nanogel stability in SGF [127]. When used as a core component, such as in BPCD systems, it enables pH-responsive drug release via π–π stacking and hydrogen bonding, while its negative surface charge facilitates active targeting to inflamed mucosa [235]. Deng et al. demonstrated the multifunctionality of PDA: (1) as a physical barrier protecting probiotics from gastric degradation; (2) as a mesoporous reservoir for controlled W6+ release in the alkaline intestine; and (3) through catechol-mediated covalent binding to the mucosa, prolonging intestinal retention [259]. Similarly, Wang et al. developed melanin nanobottles via dopamine self-aggregation [260]. These hollow structures achieved a high drug loading capacity (70%), combined with inflammation targeting, extended retention (>200 h), and antioxidant activity, leveraging melanin's acid stability (only 0.49% drug release in SGF) and negative surface charge. These cases illustrate how PDA and melanin-based materials integrate multiple functions—protection, controlled release, targeting, and retention—into a single platform, thereby simplifying conventional multi-component DDS designs. Beyond PDA, other synthetic materials also achieve integrated functionality through rational molecular design. For instance, ascorbyl palmitate forms a physical barrier and responds to MMPs in inflammatory microenvironments via enzyme-sensitive bonds, enabling targeted drug release and subsequent hydrogel-mediated tissue repair [261]. The zwitterionic polymer POSR [262] resists GI erosion through strong hydration, while its embedded ultrasound-sensitive thioflavone units release therapeutic gases (CO/H2S) upon external stimulation, enabling on-demand therapy. Advances in computational and synthetic technologies are shifting research toward tunable, high-performance inorganic nanomaterials. SiH NPs, for example, use silicon–hydrogen bonds as active sites for ROS scavenging, while their inherent negative charge and stability confer both gastric acid resistance and inflammatory targeting [263]. Machine learning further accelerates the discovery of such materials. For instance, Ni3S4, identified via high-throughput screening, exhibits high acid stability (>85% enzyme activity retained) and an inherent negative zeta potential (−18.2 mV), enabling intrinsic inflammatory targeting [264]. Similarly, SrDy2O4 nanozymes integrate ROS scavenging, acid resistance, targeting, and in vivo imaging, demonstrating multifunctional potential [265]. These systems represent a paradigm shift from multi-material composites toward the rational functionalization of single materials based on their intrinsic physicochemical properties, highlighting the power of data-driven design in multifunctional materials development. Module design strategy The global burden of GI inflammation and infection is considerable and rising, posing a significant public health threat to human health [[266], [267], [268]]. These diseases are characterized by high incidence rates and chronic pathogenicity, and are frequently accompanied by organ dysfunction, compromised mucosal barriers, and cellular immune dysregulation, significantly complicating drug delivery and treatment. Therefore, overcoming the triple barriers from organ to mucosa to cell, while achieving responsiveness and targeting tailored to pathological microenvironment characteristics, is pivotal to enhancing therapeutic efficacy. The MD strategy, by integrating functional modules such as barrier penetration, pathological response, and targeted recognition, offers a solution for drug delivery in complex disease scenarios. The present chapter will focus on the therapeutic needs of GI inflammation and infection, exploring the application of the MD strategy within this disease context. Organ-mucosal BI-level barrier breakthroughs The inflammatory and infectious microenvironment of the digestive tract exhibits significant spatiotemporal heterogeneity, posing a dual challenge for oral drug delivery: first, to withstand the harsh acidic conditions of the upper GI tract to maintain drug integrity (“organ-level” stability), and second, to effectively penetrate the complex mucus barrier at lesion sites (“mucosal-level” penetration). This section explores the design of dual-functional systems that intelligently integrate organ-targeting and mucus-penetrating modules to sequentially overcome gastric degradation and intestinal mucus barriers, thereby promoting drug accumulation at inflammatory sites and enhancing local efficacy. In one biohybrid strategy, Xiao et al. developed a temporal system, mExos@DSPE-Hyd-PMPC (Fig. 3A) [269], by fusing membranes of milk exosomes and Lips, encapsulated in Eudragit L100-55 enteric capsules to prevent gastric degradation. The surface-grafted zwitterionic polymer PMPC enhanced mucus penetration 26.5-fold via a hydrated electrically neutral layer. Upon intestinal entry, pH-sensitive hydrazone bonds are cleaved to expose positive charges and exosomal proteins, substantially improving PepT1-mediated trans-epithelial transport and achieving 8.7% oral bioavailability—double that of natural exosomes. In contrast to this natural targeting approach, Wang et al. employed artificial peptide modification, conjugating cyclic FcBP peptides to mesoporous silica NPs and encapsulating them in CS-alginate microspheres (MSNs-FcBP) [185]. The microspheres limited gastric release to below 5%, while FcBP reduced the protein corona thickness to 4.0 ± 1.2 nm and increased the mucus diffusion coefficient by 11.6-fold. This promoted FcRn-mediated transcellular transport, yielding 11.3% oral bioavailability—2.3 times that of unmodified NPs. This multi-step functional assembly strategy is exemplified in other systems. Lai et al. constructed AP@CS@Lip@HKUST-1 via sequential layering, which minimized copper ion release in the stomach [178]. The positively charged surface, combined with phospholipids, enhanced mucus penetration and activated intracellular autophagy to exert antibacterial effects. Liu et al. alternately deposited bifidobacteria, antigen peptides, CS, and Ocur to shield probiotics and drugs from gastric degradation [188]. In the intestine, the Ocur outer layer promoted M cell-mediated transcytosis across the mucosal barrier. In another study, Zhang et al. developed positively charged glucosylated NPs (Fig. 3B) that prevented drug release in the GI tract for 12 h [176]. The positive charge facilitated mucus penetration, while glucose targeting of SGLT1 receptors promoted uptake in proximal intestinal cells, yielding higher blood drug concentration than monofunctional NPs. Notably, although such multi-step assembly systems can achieve superior bioavailability, their complex fabrication may challenge scalable production. Delivery system integrating organ targeting and mucosal permeation capabilities: (A) Milk exosome-Lip hybrid vesicles. Reproduced with permission [269]. Copyright 2024, American Chemical Society. (B) Glucosylated nanocarriers. Reproduced with permission [176]. Copyright 2022, Springer Nature. (C) DMY-1000 NAs. Reproduced with permission [270]. Copyright 2023, John Wiley and Sons. Moreover, material self-assembly is a prevalent method for constructing such systems. For instance, Zheng et al. developed ultra-minute nanoenzymes (UIPNs) smaller than 10 nm through the coordination of iron ions with natural polyphenols and hydrogen bonding with Poly(vinylpyrrolidone) [271]. The resulting MOF structure and polymer coating confer resistance to gastric acid, while the ultrasmall size enables rapid penetration of the gastric mucosa. Similarly, Tang et al. used a one-step hydrothermal method to assemble dihydromyricetin (DMY) with PEG 1000 into ∼5 nm NPs (denoted as DMY-1000 NAs, Fig. 3C) [270]. PEG disrupts DMY intermolecular hydrogen bonds, thereby enhancing solubility and gastric acid stability, as well as increasing mucus penetration capacity due to its compact size. In a more complex design, Yang et al. constructed a core-shell nanosystem via electrostatic interaction between quaternary ammonium CS (HTCC) and succinic acid-modified γ-cyclodextrin (SACD) [272]. Egg white-derived peptides (EWDP) were co-assembled on the surface through noncovalent interactions, while curcumin was encapsulated in SACD's hydrophobic cavities. This system remains stable in acidic gastric environments due to HTCC protonation, and upon reaching the neutral colon, enhances mucosal adhesion and penetration through SACD targeting and EWDP modification. The therapeutic advantage of integrating dual functionalities—gastric stability and mucosal permeability—is clearly demonstrated in comparative studies. Systems lacking either function show limited efficacy, whereas those combining both exhibit superior outcomes. For example, UIPNs outperformed the acid-suppressing drug omeprazole in treating gastritis [271]. Similarly, the multifunctional system AP@CS@Lip@HKUST-1 achieved a 98% antibacterial rate, exceeding both conventional antibiotics and simpler nanocarriers [178]. Moreover, although the HTCC-SACD-Cur formulation itself offers some acid stability and drug loading, omitting EWDP co-assembly markedly reduced its therapeutic effect. The complete system was significantly more effective in alleviating colon shortening, weight loss, histopathological damage, and inflammatory cytokines, as well as in restoring intestinal barrier integrity [272]. This finding highlights the advantages of the multi-component self-assembly strategy in synergistically integrating multiple functional modules. Organ-cell BI-level barrier breakthroughs Under inflammatory or infectious conditions, the permeability of the mucosal barrier increases. This heightened permeability justifies the development of advanced delivery strategies that minimize reliance on active mucus-penetration techniques. Instead, such strategies can integrate organ-level protective mechanisms with those for cellular targeting and internalization. In recent years, various smart coupling strategies have been employed to achieve such dual-barrier synergistic crossing through physical encapsulation, chemical conjugation, self-assembly, or bioaffinity interactions. Single-stage construction strategies, such as molecular self-assembly and chemical coupling, enable the direct fabrication of multifunctional delivery systems. For instance, Xu et al. developed the "BM" system through molecular self-assembly driven by electrostatic and π–π interactions between berberine and magnolol, a structure that remains intact in the stomach but dissociates in the intestine to enhance intracellular drug concentration via P-gp inhibition [165]. Gong et al. constructed an antimicrobial peptide hydrogel (C12G2) via supramolecular self-assembly, which disassembles in gastric acid into membrane-penetrating α-helical peptides that exert rapid bactericidal and immunomodulatory effects against H. pylori [219]. In the realm of chemical coupling, Uthaman et al. covalently conjugated HA to stearic acid via a thioacetal bond, creating the amphiphilic copolymer HTS [202]. This self-assembles into micelles that combine CD44 targeting with ROS-responsive drug release. Beyond classical covalent bonds, Yu et al. employed coordination-driven assembly between tannic acid and Fe (III) to form a metal-polyphenol network that dissociates in gastric acid, facilitating bacterial adhesion and ROS-mediated antibacterial activity [226]. Multi-stage modular integration involves the sequential assembly of pre-formed components to achieve complex functionalities, primarily falling into two categories: spatiotemporal control and stimuli-responsive activation. Strategies for spatiotemporal control focus on site-specific delivery and release. A notable example of sequential electrostatic assembly is the CCS system developed by Zhang et al., which was constructed stepwise: CS was first modified with triphenylphosphine, then combined with caseinate micelles, and finally incorporated with alginate to form a stable, cauliflower-like architecture capable of colon-targeted release and subcellular targeting [230]. This approach shares the MD principle with protective polymer coatings, such as the Eudragit S100 coating applied by Lee et al. to folic acid-modified clay, which provides both gastric protection and folate receptor-β targeting for infliximab delivery [106]. A more integrated example is provided by Yu et al., who combined iMXene/mPEG-d-PEI gene carriers (Fig. 4A) with probiotics and encapsulated them in enteric capsules, thereby incorporating acid protection, cellular targeting, and endosomal escape into a single system [109]. Liu et al. further demonstrated this approach by adsorbing drug-loaded ceria nanozymes onto carboxymethyl cellulose and applying an Eudragit coating, conferring acid resistance and dextran sulfate-mediated targeting of macrophage SR-A [107]. Natural polysaccharide-based microencapsulation also serves this purpose effectively; Zhang et al. encapsulated HSA NPs within calcium alginate-CS microcapsules (Fig. 4D) [34], which protect against gastric acid and enable colon-specific release, while the small size of HSA NPs (∼30 nm) promotes cellular uptake. Similarly, HA-Se NPs encapsulated in calcium alginate microbeads (Fig. 4C) exhibit pH-dependent release, with minimal Se release (<5%) in gastric acid and rapid intestinal liberation [143]. The delivered HA-Se showed enhanced uptake in CD44-high iBMDM cells and specific accumulation in inflamed colon tissue for over 24 h post-administration. Ye et al. encapsulated dual-targeted selenium NPs in alginate microspheres to combine acid protection with dual cellular targeting [145]. In a related approach, Fu et al. developed a multi-stage delivery system by first forming a nanogel from oxidized HA and L-arginine, leveraging HA's affinity for CD44-overexpressed macrophages on inflamed tissues to enable transepithelial transport and targeting [127]. This nanogel was then encapsulated in Gel microspheres and coated with PDA to form P@G@NG microcapsules (Fig. 4B), which provide pH- and enzyme-responsive controlled release, thereby protecting the nanogel from premature degradation in the GI tract. Another strategy involves synthetic matrix embedding, as shown by Zhang et al., who embedded Cu2(Olsa) nanoneedles in an inulin gel to achieve enzymatic-triggered release in the colon while leveraging the needles' penetrating capability for intracellular delivery [129]. Bio-hybrid strategies have also been employed: Jin et al. attached maltodextrin-modified DNA origami nanosheets to engineered bacteria via base pairing, creating DNA nanopatches (DNPs) that significantly improve gastric survival [120]. Zhao et al. extended this concept by similarly combining DNPs with phages for precise pathogen recognition, followed by enteric polymer encapsulation to finalize the delivery system [103]. Delivery system integrating organ targeting and cell targeting capabilities: (A) iMXene/mPEG-d-PEI gene carriers. Reproduced with permission [109]. Copyright 2024, John Wiley and Sons. (B) P@G@NG microcapsules. Reproduced with permission [127]. Copyright 2024, John Wiley and Sons. (C) SHSe hydrogel microbeads. Reproduced with permission [143]. Copyright 2023, American Chemical Society; (D) MTX@HSA/AlgC microcapsules. Reproduced with permission [34]. Copyright 2023, John Wiley and Sons. Meanwhile, stimuli-responsive activation strategies focus on systems that undergo functional changes in response to pathological cues. Huang et al. constructed a system comprising an outer layer of low molecular weight heparin for targeting macrophage integrin αM, a middle CS layer that gels in stomach acid to form a protective barrier, and a ROS-responsive core (OsEPL) for controlled release in inflamed tissue [196]. In a distinct approach, Nie et al. synthesized tyramine-functionalized PLGA NPs that undergo oxidative cross-linking in inflammatory microenvironments, forming aggregates with improved acid stability and cellular uptake [225]. Comparative studies confirm that integrating gastric acid stability with cellular bioavailability enhancement significantly improves the therapeutic performance of oral delivery systems. This multi-module strategy outperforms single-function approaches across multiple metrics, including disease activity index (DAI), inflammatory cytokines, tissue repair, and pathogen clearance. For example, a folate-modified amino clay system combining Eudragit coating and folate targeting markedly improved colitis outcomes (DAI = 3) compared to the non-targeted group (DAI = 8) [106]. Similarly, Liu et al. reported that an integrated formulation of nanoenzymes, DSS, and enteric coating restored colon length by 98% in a colitis model, surpassing uncoated or physically mixed systems [107]. In another study, the LGG-M@P system—comprising enteric capsules, iMXene, and PEI—reduced TNF-α levels by 50% and increased colon length recovery by 40% relative to iMXene alone [109]. These results indicate that merely stacking functional modules is insufficient to overcome GI barriers. Instead, effective drug enrichment and utilization in diseased cells require rational designs that systematically incorporate gastric protection, targeting, environmental responsiveness, and intracellular delivery. Organ-mucosal-cellular three-level barrier breakthrough Recent advances in nanomaterial assembly have led to intelligent delivery systems that function synergistically across the organ-mucosa-cellular continuum. These designs reflect a growing understanding of biological interfaces and the modular integration of functional components. Static functional integration strategies rely on intrinsic material properties to achieve multifunctionality. For example, Chen et al. developed RHL/Cl-Ch-cal lipid vesicles via thin-film hydration [174]. The bilayer structure protects against gastric acid, with only 20% drug leakage at pH 2.0 over 6 h. Rhamnosyl lipids confer a negative charge, enhancing mucus penetration and reducing mucin binding to 13.4%. Meanwhile, cholesterol-PEG and calcitriol synergistically promote membrane repair and lysosomal recovery, lowering intracellular bacterial survival to 14.8%. Beyond lipid-based systems, synthetic polymers, particularly those in the Pluronic series, as well as natural polysaccharides (e.g., inulin, HA, CS, SA), also demonstrate diverse applications in these systems. Xie et al. designed an INB/IND-NC system (Fig. 5A) [152] with inulin for colon-targeted release. Dual modification with Pluronic F127 and HA enhanced mucus penetration and CD44-mediated targeting, resulting in increased drug levels in the lamina propria and a nearly threefold reduction in DAI scores compared to HA-only controls. Zu et al. coated tea-lipid-curcumin NCs with F127, suppressing burst release and improving mucus transit distance threefold, while galactose moieties promoted macrophage uptake and polarization, yielding 3.5-fold higher fluorescence in colitis tissue [274]. Chen et al. developed CS@FL-D@lzd/MS microspheres (Fig. 5C) that respond to pH and MMP-9 [122]. CS enhances mucopenetration and retention, while a fluid lipid layer fuses with Gram-negative bacterial cell membrane, enabling “organ-tissue-bacteria” targeting and raising antibacterial activity more than 16-fold. Similarly, Liu et al. used CS in ZAN@CS MNDs for mucopenetration [275]. Thiol acetamide protonation in the stomach enhances gastric retention, and the resulting positive charge targets H. pylori, achieving efficacy comparable to clinical drugs without disrupting gut microbiota. This ingenious design demonstrates that surface charge regulation can synergistically integrate multiple functions—such as organ retention, mucus penetration, and bacterial targeting—thus offering new insights for simplifying the design of oral delivery systems. Zou et al. reported that a nanocrystalline core self-assembled from metformin hydrochloride derivatives and linoleic acid conferred structural stability [175]. Meanwhile, surface modification with fucosyl polysaccharides facilitated efficient transport across both mucus and cellular barriers, leading to over 99% eradication of H. pylori. This system demonstrated markedly superior efficacy compared to both physically mixed formulations and conventional triple therapy. Biomimetic designs are also emerging in oral delivery. Xu et al. constructed PEI-L-TA@MON with a chiral silanized structure for acid resistance [172]. L-Tartaric acid aids chiral recognition, improving retention and penetration in inflamed mucosa via combined chiral, hydrophobic, and electrostatic interactions, achieving 84.9% colon length recovery. Delivery system capable of simultaneously overcoming the triple barrier of organ-mucosa-cell: (A) INB/IND-NC system. Reproduced with permission [152]. Copyright 2023, Elsevier. (B) Nucleic acid drug-delivery system of MSNs. Reproduced with permission [273]. Copyright 2022, American Chemical Society. (C) CS@FL-D@lzd/MS. Reproduced with permission [122]. Copyright 2025, American Chemical Society. (D) P127-AuS@CURs. Reproduced with permission [183]. Copyright 2024, John Wiley and Sons. Beyond static design, numerous studies show that dynamic responsive systems have the capacity to endow DDS with the potential for environmental adaptation, thereby enabling more precise overcoming of multiple biological barriers. Cheng et al. designed the HADLA system with a size-switching strategy: 900 nm aggregates protect the drug in acidic conditions, while 200 nm particles form in the colon to enhance mucus penetration via electrostatic repulsion and CD44-mediated uptake, increasing colon accumulation by 190% [201]. The MSNs system (Fig. 5B) developed by Li et al. integrated dual enzyme and redox responsiveness to construct a more sophisticated multistage-responsive nanocomposite [273]. An alginate shell protects against gastric acid and degrades via bacterial enzymes in the colon; subsequent polymeric nanocapsules (∼30 nm) penetrate mucus and enter cells, where GSH cleaves disulfide bonds to release drugs. This system achieves a controlled release cascade responsive to gastric pH, colonic enzymes, and intracellular GSH, enabling nucleic acid delivery into the colonic muscle layer and resulting in full restoration of colon length. Subsequent studies have further advanced this dynamic response paradigm through the incorporation of intelligent photothermal strategies. Ma et al. encapsulated Pluronic F127-modified gold-shell-polymeric core NPs loading with curcumin in a CS/ALG hydrogel for colon targeting (Fig. 5D) [183]. Under NIR irradiation, the gold shell generates thermophoretic forces that accelerate mucus penetration (from 1.3 to 13.9 μm s−1) and promote lysosomal escape, illustrating the potential of combining biological and physical triggers. The extant literature demonstrates the diversity and innovation in oral delivery system design, ranging from Lips to NCs, and from photothermal-driven to chiral recognition approaches. Despite the heterogeneity of their methodologies, these approaches are unified by a shared objective: the attainment of a tripartite functionality, entailing gastric acid protection, mucus penetration, and cellular delivery. Multi-module systems consistently outperform single-function designs, underscoring the value of integrated strategies for future oral drug development. Design trade-offs across multilevel barriers Notably, the sequential penetration of multi-scale GI drug delivery barriers, including the organ-level barrier, the mucosal barrier and the cellular barrier, is often accompanied by inherent trade-offs in strategy design. That is, a strategy that facilitates penetration at one barrier may impede traversal of another. Therefore, beyond the successful strategies summarized above, an analysis of design failures and inherent trade-offs is essential for rational carrier design. The most prominent and systematic conflict lies between mucus barrier penetration and epithelial barrier absorption. A key example is surface charge: mucus penetration favors neutral or negative charges, while epithelial uptake often relies on the electrostatic adsorption of positive charges. Consequently, a single charge state cannot optimally address both barriers. Likewise, PEGylation enhances mucus penetration, but its stealth effect often diminishes interactions with epithelial membranes, reducing transport efficiency. The second major trade-off exists between organ-level retention and mucus penetration. Robust mucoadhesive materials (e.g., CS, PDA) and large-scale carriers (e.g., microscale hydrogels) confer gastric acid resistance and prolong gastrointestinal retention, yet impose kinetic barriers to mucus penetration. A representative example is the TMNR hydrogel system, which achieves a 24-h colonic residence time through rapid temperature-triggered gelation combined with robust mucoadhesion [239]; however, such adhesion anchors carriers to the mucus layer, resulting in a failure to reach the epithelial surface despite prolonged residence time [239]. Similarly, carriers exceeding 20 nm in diameter, such as multilobe microparticles with enhanced mucoadhesion owing to their lobed structure, are substantially hindered by the sieve-like architecture of the mucus mesh, resulting in compromised mucus penetration [148]. Additionally, conflicts also exist between mucus/epithelial penetration and bacterial/cellular internalization: ultra-small nanoparticles, such as the ∼1.95 nm GSH-modified gold nanoclusters that efficiently penetrate mucus and epithelial tight junctions [231], are less efficiently phagocytosed by macrophages and immune cells that prefer particles larger than 60 nm [[207], [208], [209]], thereby hindering targeted enrichment in lesion cells. A further critical challenge is to balance permeability enhancement with safety. For instance, NPs modified with mucolytic agents can disrupt the mucus layer to promote NPs uptake [190]. While this strategy improves penetration, it compromises the protective function of the mucus barrier, thereby increasing the risk of pathogen invasion and local inflammation. The absence of mucin leaves the underlying epithelium vulnerable to the harsh gastrointestinal environment, leading to potential damage [276]. Besides, externally driven strategies, such as ultrasound, which enhance mucosal permeability, may also cause epithelial damage. Similarly, cationic surface modifications, designed to improve epithelial adhesion, can cause membrane disruption and cytotoxicity at high concentrations [277]. These safety concerns often necessitate a trade-off between maximizing permeability and maintaining biocompatibility. These inherent strategy conflicts not only reflect the complexity of the multiscale GI microenvironment but also point to a central direction for future research in programmable GI drug delivery systems: the sequential modulation of multilayered barriers through the rational design of carrier properties (size, shape, charge, and surface modification) and the precise control of stimulus-responsive timing, thereby resolving trade-offs in barrier penetration across different levels and further improving the bioavailability and therapeutic precision of oral drugs. Discussion and outlook In recent years, MD has attracted increasing interest as a pragmatic design philosophy for addressing the intricate barriers posed by the GI tract. For GI inflammatory and infectious diseases, MD enables on-demand integration of key functional units—such as resistance to digestive fluids, mucus interaction and permeation, lesion-associated enrichment, and controlled release—thereby offering a flexible route to overcome multiple, sequential obstacles along the oral delivery pathway. Nevertheless, despite encouraging preclinical progress, the clinical translation of Modular DDS remains non-trivial. As discussed in previous reviews, the leap from bench to bedside critically involves stringent Chemistry, Manufacturing, and Controls (CMC) requirements, predictable biointerface interactions, and clinically meaningful spatiotemporal control over drug exposure [278]. Firstly, CMC readiness remains a substantial barrier to large-scale implementation. This is largely because most GI delivery systems discussed here are under a pragmatic “modular logic” rather than built on fully standardized coupling interfaces. In principle, true modularity should enable plug-and-play reconfiguration; however, current GI modular DDS systems are still dominated by case-specific modularity, where module exchange often necessitates reconsidering critical quality attributes/process parameters and revalidating release assays, increasing batch-to-batch variability risks under Good Manufacturing Practice expectations. One-pot in situ synthesis may reduce unit operations and improve loading in selected systems, but its applicability is typically chemistry-constrained and cannot be generalized across modules. Accordingly, a practical near-term direction is to drive GI modular DDS toward platformization. By establishing a universal conjugation interface and standardized production processes, diverse ligands can be interchanged without chemical modification [279]. This approach ensures modular exchange without compromising manufacturability—an important step toward an authentic “Lego-like” customization era and, ultimately, personalized precision medicine. Secondly, balancing GI stability with site-specific activation is a core translational challenge in oral drug delivery. Delivery modules must protect payloads from acid, enzymes, bile salts, and rapid transit, yet still ensure timely release at target lesions. However, the profound pathological heterogeneity of the GI tract frequently disrupts this “stability–release coupling”—a vulnerability highlighted by the clinical failures of advanced targeted systems. A prominent example is Mongersen (GED-0301), an oral antisense oligonucleotide for Crohn's disease that utilized a pH-dependent coating to trigger release in the terminal ileum and right colon. Despite promising early-phase data, it failed in Phase III trials (NCT02596893). Researchers speculate this failure stemmed from enrolling patients with broader lesion distributions (including the left colon), which likely resulted in insufficient local drug concentrations at the target tissues [280]. Furthermore, even clinically approved colon-targeted formulations are hindered by unpredictable in vivo performance. Imaging studies of enteric-coated mesalazine reveal that severe interindividual variability often overshadows the intended targeting benefits of distinct coating compositions, even when dual-trigger mechanisms are employed [52]. These clinical realities underscore a fundamental principle for GI DDS: while robust stabilization is essential to survive the harsh GI environment, over-stabilization or reliance on a single biological trigger can be just as detrimental as premature leakage if it prevents adequate drug exposure at the lesion. Consequently, future development must evolve beyond simplistic physiological dependencies. Next-generation platforms should integrate multi-modal or disease-specific triggers, alongside modular designs that allow for personalized titration of release kinetics to match individual lesion topographies. Thirdly, inflammation-driven heterogeneity in lesion accessibility emerges as a primary driver of the variable clinical benefit observed for modular GI DDS. While active targeting strategies have shown considerable promise in preclinical studies, their clinical translation has been extremely limited due to variable target expression, manufacturing and regulatory complexities, and loss of function in the harsh GI environment. Consequently, most clinically advanced GI platforms inherently rely on passive enrichment. Even BIND-014, one of the few ligand-targeted nanomedicines to reach Phase II (NCT01812746), did not advance further, underscoring the translational gap facing active targeting strategies. However, the mechanisms governing passive accumulation present a critical bottleneck. Theoretically, inflamed gut tissue offers permissive features for accumulation, such as epithelial barrier disruption and altered mucus properties. In practice, these characteristics vary considerably across patients, disease phenotypes, and even among different lesions within the same patient. This heterogeneity manifests in two critical ways. First, at the inter-patient and inter-phenotype levels, the relationship between inflammation and payload accumulation is context-dependent, influenced by factors such as the type of inflammation, tissue environment, and the nature of the payload itself [281,282]. Second, at the intra-patient level, lesion geography is frequently discontinuous. For example, in a cohort of 202 Crohn's disease patients, 24.7% exhibited small-bowel skip lesions, highlighting the patchy nature of mucosal involvement [283]. Overcoming this barrier, therefore, demands shifting focus from merely optimizing average carrier performance toward patient stratification, using identifiable endoscopic features or mucosal biomarkers to prospectively select patients likely to respond [[284], [285], [286]]. Fourthly, the limitations of animal models present a significant challenge for GI DDS. Disparities between species in digestive physicochemical properties, gut microbiota, and pathological microenvironments can bias predictions of both efficacy and safety [287,288]. For instance, while DSS-induced colitis remains the most widely used preclinical IBD model given its imitation of many human disease features, its inability to fully replicate patient clinical manifestations and immune responses [[289], [290], [291]]. Therefore, evaluating a wider range of animal species is crucial to identifying those with pathology that closely resembles that of humans. Additionally, future research should prioritize the advancement of organoid-on-a-chip and related microphysiological systems as more precise evaluation platforms that better recapitulate human physiology, enabling more supporting regulatory-facing decision-making. Finally, GI modular DDS warrants greater biosafety scrutiny than any other micro/nano. Its safety should be considered in microbial composition or function, at the level of the assembled formulation, rather than inferred from the drug or each material component alone [292]. Because their biological effects are shaped not only by composition but also by how different functional elements are integrated, transformed, and exposed in vivo [292,293]. In the GI, particles are continuously corona-coated by contact with mucus, bile salts, digestive enzymes, dietary components, and other luminal biomolecules, which can alter their bioidentity and, in turn, their interaction with the mucosa, epithelial uptake, immune sensing, and systemic translocation [292,294]. This environmental remodelling implies that design features introduced to improve delivery performance, such as stealth coatings, charged surfaces, uptake-promoting ligands, or stimuli-responsive modules, may also create new safety liabilities, including immunogenicity, complement or innate immune activation, membrane perturbation, and reduced tolerability upon repeated dosing [295,296]. These concerns may be further amplified in diseased tissue, where barrier disruption can magnify formulation-dependent differences in retention, penetration, and systemic exposure [292]. In addition, because such systems often rely on novel excipients and tightly controlled critical quality attributes, even modest batch-to-batch variation may shift their exposure–safety profile in clinically meaningful ways [293]. Repeated administration should also assess microbiome perturbation, as carriers with consequences for long-term tolerability and therapeutic consistency [297]. Conclusion Technological advancements, including multimodal AI and emerging fabrication platforms such as microfluidics, are beginning to reshape the design paradigm of GI modular DDS. By enabling more systematic selection and integration of functional modules, AI-assisted approaches may improve formulation optimization, adaptability, and disease-specific tailoring, moving these systems closer to a more programmable and potentially personalized design framework. At the same time, meaningful clinical translation will depend not only on advances in computational design, but also on scalable and reproducible manufacturing, human-relevant preclinical models, and early consideration of formulation safety and developability. Progress in these areas, together with clearer regulatory expectations, should help convert the conceptual versatility of modular DDS into clinically actionable platforms for GI disease treatment. Ethics approval and consent to participate Not applicable. Funding This work was supported by the Science and Technology Commission of Shanghai Municipality (14401901400). CRediT authorship contribution statement Yaxin Chen: Conceptualization, Data curation, Formal analysis, Investigation, Visualization, Writing – original draft, Writing – review & editing. Qingyun Yang: Supervision, Validation, Visualization, Writing – review & editing. Jian Yang: Conceptualization, Supervision, Writing – original draft. Jiquan Zhang: Conceptualization, Funding acquisition, Project administration, Writing – review & editing. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Data availability No data was used for the research described in the article. References References Associated Data Data Availability Statement No data was used for the research described in the article.