The Role of Cell-Cell Communication in Renal Damage and the Therapeutic Targeting of Diabetic Kidney Disease Free

Diabetic kidney disease (DKD) is one of the most prevalent and lethal complications of diabetes mellitus (DM). It is characterized by progressive proteinuria and a persistent decrease in glomerular filtration rate [1, 2]. The global number of adults with DM was estimated to be 536.6 million in 2021 and is projected to rise to 783.2 million by 2045 [3]. Approximately 30%–40% of individuals with DM develop DKD [4]. Thus, the prevalence of DKD is increasing worldwide owing to the rising incidence of DM. According to the Global Burden of Disease Study in 2019, the number of people with DKD worldwide has reached 135 million [5]. DKD has become the leading cause of end-stage renal disease, which poses a heavy burden on society [6]. Hence, it is imperative to explore the pathogenesis of DKD and identify new therapeutic targets.

Hyperglycemia is the major driving force behind the progression of DM to DKD. Various changes occur in kidney cells in response to a high-glucose environment, including the production of advanced glycation end products and the activation of protein kinase C and reactive oxygen species (ROS) [7]. However, the kidney is a multicellular organ consisting of more than 20 cell types, and these different cell types interact with each other, both in a normal setting and in response to hyperglycemia [8]. The crosstalk between intrinsic renal cells is believed to play a pivotal role in the pathogenesis of DKD.

Cell communication, also known as cellular crosstalk, is typically facilitated through direct cell-cell contact or transfer of secreted molecules and extracellular vesicles (EVs) [9]. Cellular communication is necessary for many physiological processes. However, aberrant cellular communication patterns may contribute to an increased risk of kidney disease. The role of intercellular communication in the pathogenesis of DKD has not been fully elucidated. This review examines the role of intercellular communication in the pathogenesis of DKD from the perspective of pathophysiological changes in four intrinsic renal cell types: podocytes, glomerular endothelial cells (GECs), mesangial cells, and renal tubular epithelial cells (RTECs). Furthermore, it summarizes novel therapeutic strategies targeting cellular crosstalk, including stem cell-based approaches and pharmacological interventions.

We searched PubMed and Web of Science for publications from 2000 to 2025 using keywords including “DKD/diabetic nephropathy”, “cell communication/crosstalk”, “podocyte”, “GEC”, “mesangial cell”, “RTEC” and “stem cell”. We included high-quality original research articles and reviews related to intercellular communication mechanisms and therapeutic strategies in DKD and also examined the reference lists of these articles. And we categorized and summarized the literature, with a particular focus on the molecular mechanisms of intercellular communication and their clinical translational potential.

GECs, which constitute the first layer of the glomerular filtration barrier, are characterized by fenestrations. Their luminal aspects are covered with a layer of negatively charged endothelial glycocalyx composed of glycoproteins, which helps block the penetration of albumin [10]. High-glucose exposure can induce GEC apoptosis and degradation of the endothelial glycocalyx, subsequently increasing cellular permeability, which contributes to the development of DKD [11, 12]. Abnormal angiogenesis is also a key feature of DKD. It may be attributed to the activation of vascular endothelial growth factor (VEGF) and other pro-angiogenic factors in the diabetic microenvironment [13].

Podocytes wrap around glomerular capillaries and play a pivotal role in the maintenance of the integrity of the glomerular filtration barrier [14, 15]. During the pathogenesis of DKD, the podocyte apoptosis, hypertrophy, and dedifferentiation work together to damage the filtration barrier, leading to proteinuria [16]. However, as terminally differentiated cells, podocytes cannot proliferate or maintain their cellular pool. An insufficient number of podocytes leads to glomerulosclerosis. To compensate for the depletion of the podocyte population, mammalian target of rapamycin signaling is activated and mediates podocyte hypertrophy in the early stages [17]. Moreover, under the influence of hyperglycemia, podocytes undergo dedifferentiation, resulting in structural and functional alterations that eventually disturb their normal role in the glomerular filtration barrier [18].

Mesangial cells are smooth muscle-like pericytes located in the central stalk of the glomerulus that hold and support multiple capillary loops. While the contractile properties of mesangial cells allow them to regulate intraglomerular capillary flow and the glomerular filtration rate [19], these cells are also characterized by abnormal expansion through active proliferation and increased matrix protein production in response to diabetic-induced injury [20]. Furthermore, mesangial cells not only proliferate excessively but also release inflammatory factors triggered by high glucose, leading to glomerular sclerosis [21, 22]. A recent study found that activated mesangial cells act as antigen-presenting cells to regulate T-cell participation in immune response [23].

The prevailing belief is that the most crucial pathological changes in DKD are related to glomeruli; however, renal tubules are also involved in the pathogenesis of DKD. Severe renal tubular lesions have been observed in some patients with type 2 diabetes, despite relatively mild glomerular lesions [24]. On the one hand, similar to glomeruli, polyol, transforming growth factor β1 (TGFβ1), angiotensin, VEGF, ROS, and other intracellular pathways directly activated by glucose and advanced glycation end products contribute to RTEC hypertrophy, apoptosis, and tubulointerstitial fibrosis [25]. On the other hand, albuminuria caused by glomerular filtration barrier injury also leads to RTEC damage by the activation of inflammatory mediators, ROS, nuclear factor κ-B, and other molecules or harmful events [26‒31].

The glomerulus, which contains podocytes, GECs, and mesangial cells, plays an essential role in the development and pathology of the kidney, owing to its unique filtration structure [32]. Because these cells are spatially adjacent, intercellular communication occurs frequently. Persistent hyperglycemia induces pathological changes through dysregulated cellular communication in injured glomerular cells. The communication between glomerular cells is summarized in Figure 1.

In the pathogenesis of DKD, the impairment of GEC homeostasis is a pivotal event, with dysregulated crosstalk between endothelial cells and podocytes playing a fundamental role in driving disease progression. The members of VEGF family, including VEGFA, VEGFB, VEGFD, VEGFE, and placental growth factor, are key regulators of endothelial cell function [33]. Among these, VEGFA, the most extensively studied angiogenic factor in kidney, is secreted by both podocytes and RTECs, regulating renal blood vessel formation and RTEC proliferation [34]. It can bind to two different tyrosine kinase receptors on vascular endothelial cells, VEGF receptor 1 (VEGFR1) and VEGFR2 [35]. Studies have shown that mice with a podocyte-specific deletion of VEGFA or a constitutive deletion of VEGFR2 exhibit barely visible glomerular capillaries. This suggests that podocytes regulate the function of GECs via a paracrine pathway involving VEGFA [36, 37].

The expression of VEGF in podocytes and VEGFR2 in GECs is elevated in DKD. Moreover, the VEGF mRNA expression is positively correlated with neovascularization in the early stages of DKD [38, 39]. Under physiological conditions, VEGF produced by podocytes can activate endothelial nitric oxide synthase by binding to either VEGFR1 or VEGFR2, promoting the generation of nitric oxide (NO) in GECs [40, 41]. However, high glucose or glycosylation products can reduce the production and bioavailability of NO in GECs, leading to decoupling between the VEGF axis and NO [41]. Moreover, elevated level of VEGF activates VEGFR2, promoting the pathological proliferation of GECs through the extracellular signal-regulated kinase pathway, leading to abnormal neovascularization [42].

Notably, VEGFA expression may exhibit dynamic changes during the different stages of DKD. Hohenstein et al. [43] showed that decreased local expression of VEGFA/VEGFR was associated with sparse glomeruli. As VEGF is actively produced by podocytes, this decrease in VEGFA expression may be associated with the loss of podocytes during the late stage of DKD [44]. In summary, the VEGFA/VEGFR pathway plays a crucial role in crosstalk between podocytes and GECs in DKD. Maintaining this communication at a moderate level is optimal because both high and low levels can disrupt GEC homeostasis and contribute to renal injury.

In addition, the angiopoietin (Angpt) family, which mainly consists of Angpt1-4 and its receptors tyrosine kinase with Ig and epidermal growth factor (EGF) homology domains 1 and 2 (Tie1 and Tie2), also plays an essential role in renal angiogenesis [45‒47]. Angpt1, primarily secreted by podocytes, binds to the Tie2 receptor on GECs to promote endothelial cell survival and vascular stability [48]. Jeansson et al. [49] demonstrated that Angpt1-deficient mice with DM had a considerably higher mortality and exhibited more severe pathological changes than those exhibited by mice with DM. However, restoration of podocyte-specific Angpt1 expression in mice with DM elevated Tie2 phosphorylation and ameliorated GEC proliferation [50]. In fact, Angpt1 derived from podocytes is reduced, impairing Angpt1-mediated communication between podocytes and GECs, as well as destabilizing the microvasculature under high-glucose conditions [51]. Unlike Angpt1, Angpt2 is an inhibitory ligand of Tie2. It is secreted by both podocytes and endothelial cells to inhibit the phosphorylation of the Tie2 receptor and disrupt vascular integrity [52‒54]. Overexpression of Angpt2 in podocytes can result in GEC apoptosis [55]. Interestingly, elevated levels of Angpt2 have been observed in animal models and patients with diabetes [56]. Podocytes secrete excess Angpt2 under high-glucose conditions, thereby inhibiting the phosphoinositide 3-kinase/protein kinase B (PI3K/AKT) signaling pathway and promoting GEC damage and vascular regression [57]. In addition, the downregulated PI3K/AKT pathway can also activate nuclear factor κ-B and promote endothelial cell inflammation [58]. Therefore, dysregulation of the Angpt1/2-Tie2 signaling pathway directly contributes to the disruption of endothelial cell homeostasis, representing a key mechanism in DKD progression. In summary, the disruption of GEC homeostasis caused by aberrant communication between podocytes and GECs plays a critical role in the pathogenesis of DKD.

Apoptosis is an important cause of podocyte loss in patients with DKD. Recent studies have demonstrated that GECs, which form a filter barrier with podocytes, are also involved in podocyte apoptosis. Retinoic acid receptor reactive protein 1 (RARRES1) is a tumor suppressor gene that has previously been implicated in the pathogenesis of various tumors [59]. RARRES1, a type 1 membrane protein, is predominantly expressed in podocytes during kidney disease. It is upregulated by tumor necrosis factor-α, promoting podocyte apoptosis and contributing to glomerular dysfunction [60, 61]. Nevertheless, another mechanism has been proposed in which increased RARRES1 in endothelial cells acts on podocytes through a paracrine pathway to promote podocyte apoptosis in glomeruli in DKD [62]. In addition, pathological changes in GECs can induce podocyte apoptosis. Casalena et al. [63] found that high glucose specifically induced mitochondrial dysfunction and mitochondrial superoxide release in GECs. They further demonstrated that dysfunctional GECs caused podocyte apoptosis and that this effect could be blocked by mitoTEMPO (a selective mitochondrial antioxidant) [63]. However, the specific messengers that mediate the communication between the two cell types still need to be explored.

Similar to other cells in the glomeruli, mesangial cells can affect podocyte apoptosis. Under high-glucose conditions, specific humoral factors secreted by mesangial cells inhibit the endoplasmic reticulum-associated degradation pathway of podocytes, consequently suppressing renin phosphorylation and causing podocyte apoptosis [64]. Similarly, exosomes have also been shown to participate in the communication between mesangial cells and podocytes. Exosomes released from mesangial cells induced by high-glucose levels lead to podocyte apoptosis. This mechanism may be linked to the transfer of TGFβ1 mRNA from mesangial cells to podocytes via exosomes [65].

In addition to apoptosis, many other mechanisms can cause podocyte injury, such as cytoskeletal changes and epithelial-mesenchymal transition (EMT). Hill et al. [66] demonstrated remarkable changes in the transcriptome of podocytes co-cultured with EVs from GECs treated with high glucose compared to those treated with normal glucose. Among these changes, miR-200c-3p levels are notably upregulated, resulting in decreased VEGF secretion and alterations in the podocyte cytoskeleton [66]. Moreover, another study found that a high expression of TGFβ1 mRNA was shown in exosomes derived from GECs treated with high glucose, and exosomes could deliver this mRNA to podocytes, inducing EMT in podocytes [67].

Similarly, Kruppel-like factor 2 (KLF2) is also involved in the communication between GECs and podocytes. It belongs to the KLF family, whose expression is induced by laminar shear stress generated by the blood within the endothelium to play a protective role [68]. Zhong et al. [69] observed decreased levels of KLF2 in both the glomeruli of rats with DM and the kidney tissue of patients with DKD. Additionally, they demonstrated that KLF2-specific knockout in the GECs of diabetic mice resulted in more severe kidney damage, particularly severe podocyte and GEC damage, suggesting that KLF2 derived from GECs might be a potentially important molecule in podocyte injury [69].

Mesangial cell proliferation and extracellular matrix (ECM) production are pathological manifestations of DKD. In recent years, intercellular communication has been found to play an important role in them.

Endothelin-1 (ET-1) is a vasoconstrictor, mainly secreted by endothelial cells. In the glomerulus, including mesangial cells, various cells can express endothelin receptor type A and endothelin receptor type B [70]. López-Ongil et al. [71] reported that the secretion of ET-1 by endothelial cells is influenced by mesangial cells. Notably, ET-1 is upregulated in the kidneys of patients with DKD [72]. Therefore, increased expression of ET-1 from GECs interacts with endothelin receptor type A in mesangial cells, promoting mesangial cell proliferation and ECM production through the RhoA/Rho-kinase pathway [73].

Platelet-derived growth factor B (PDGFB) is another effective stimulator of mesangial cell proliferation [74]. In the kidneys, PDGFB is predominantly expressed in endothelial cells, whereas its receptor, PDGFB-R, is expressed in mesangial cells [75, 76]. A previous study showed that the suboptimal secretion of PDGFB in endothelial cells leads to mesangial cell depletion, indicating that GECs secrete PDGFB in a paracrine manner to maintain the development and function of mesangial cells [77]. This paracrine pathway is enhanced by hypoxia, which is also a characteristic of DKD [78]. Moreover, evidence indicates a considerable increase in PDGFB and PDGFR expression in DKD [79, 80]. Therefore, the PDGF-mediated communication between GECs and mesangial cells is abnormally activated and plays an important role in mesangial cell proliferation and hypertrophy.

Additionally, exosomes derived from endothelial cells can serve as mediators of mesangial cell proliferation. Studies showed that high-glucose conditions increase TGFβ1 mRNA-enriched exosome secretion from GECs. These exosomes are transferred into mesangial cells, activating the TGFβ1/small mothers against decapentaplegic pathway to upregulate α-smooth muscle actin and promote ECM production [81]. Chen et al. [82] also discovered that exosomes containing decreased expression of circRNF169 and circSTRN3 in GECs cultured in high glucose stimulated mesangial cells to produce more α-smooth muscle actin.

Zhou et al. [83] described the cellular crosstalk between mesangial cells and podocytes. Elevated expression of C-X-C chemokine receptor type 4 (CXCR4) in damaged podocytes promotes injury and activation of mesangial cells. In reverse, the increased expression of stromal cell-derived factor 1α, a ligand of CXCR4, in mesangial cells leads to podocyte injury and oxidative stress. However, this mechanism needs to be further studied in DKD.

RTECs, the main component cells of the tubules, participate in communication with other cells and play an important role in renal development and pathology. For instance, Liang et al. [84] showed that RTECs subjected to hypoxia/reoxygenation stimulated mesangial cell proliferation. The communication between RTECs and glomerular cells is summarized in Figure 1.

EVs not only play a vital role in intercellular communication among glomerular cells but are also engaged in crosstalk between podocytes and RTECs. Previous studies have shown that the EVs produced by podocytes participate in RTEC damage. Podocyte-derived microparticles are considerably increased in both mice and patients with DM, indicating that microparticles can serve as biomarkers for the detection of early DKD [85, 86]. Subsequent studies revealed that microparticles produced by podocytes can induce fibrosis in RTECs [87]. A recent study also discovered that miRNA-221 in podocyte-derived exosomes activated the Wnt/β-catenin signaling pathway under high glucose, resulting in the dedifferentiation of RTECs [88].

Cold-shock protein Y-box-binding protein 1 (YBX1) is mainly involved in inflammation during kidney disease [89]. A recent study confirmed that after podocyte-specific deletion of YBX1, renal tubular injury was aggravated via the activation of toll-like receptor 4 signaling [90]. However, its role in DKD requires further investigation.

Nicotinamide mononucleotide (NMN) is a biosynthetic intermediate of nicotinamide adenine dinucleotide (NAD+) and plays a potential role in the treatment of various diseases [91]. During salvage synthesis of NAD+, nicotinamide is catalyzed by nicotinamide phosphoribosyl transferase to synthesize NMN, which subsequently synthesizes NAD+ [92]. Sirtuin 1 (SIRT1) has been identified as a mediator of NMN function and exerts protective effects against diabetes and kidney disease [93, 94]. In a DKD model, Hasegawa et al. [95] observed that a reduction in SIRT1 levels within RTECs under high-glucose conditions influenced SIRT1 levels in podocytes, and this decrease was ameliorated by supplementation with decreased NMN in RTECs. Reduced SIRT1 levels in podocytes upregulated claudin-1, contributing to proteinuria in DKD. Therefore, NMN may mediate the crosstalk between RTECs and podocytes.

Bim, a BH3-only member of the B-cell lymphoma 2 protein family, induces cell apoptosis [96]. High-glucose conditions upregulated Bim expression in RTECs in vitro, leading to apoptosis [97]. Xu et al. [98] showed that high-glucose-induced upregulation of Bim in RTECs promotes nuclear factor of activated T cells 2 activation and nuclear translocation. These changes further downregulated lncRNA NONHSAT179542.1 expression in podocytes, leading to cytoskeletal damage [98].

Stem cells are undifferentiated cells that exhibit self-renewal and differentiation capabilities [99]. In recent years, stem cell transplantation therapy has gained momentum in various fields, including DKD treatment. On the one hand, stem cells are able to undergo differentiation to replenish damaged cells in injured kidney, and on the other hand, they help re-establish adaptive intercellular communication. Stem cells secrete molecules and EVs that are taken up by renal cells for renal repair [100] (Table 1 and Fig. 2). In brief, stem cell transplantation to establish adaptive cell crosstalk is a promising therapy for DKD.

A previous study showed that umbilical cord mesenchymal stem cells (UC-MSCs) injected into the tail vein of rats with DM alleviated kidney damage. To explore this protective mechanism, the conditioned medium of UC-MSCs was co-cultured with RTECs stimulated with TGFβ1, and it was found to inhibit the cell injury caused by TGFβ1. Hence, one of the protective effects of MSCs on kidney cells is likely achieved through secretion of unknown substances [116]. Several studies have shown that MSCs establish adaptive crosstalk with renal cells by secreting factors such as bone morphogenetic protein 7 (BMP7), soluble EGF, and hepatocyte growth factor to facilitate kidney repair [105, 114, 115, 117].

As discussed earlier, EVs participate in the intercellular communication among kidney cells and contribute to the pathology of DKD. Recently, the protective effects of stem cells on the kidneys have also been studied. In general, stem cell-derived EVs ameliorate DKD-induced damage through intercellular communication with different kidney cells.

Stem cell-derived EVs can carry various miRNAs involved in the communication among kidney cells. This has been confirmed by several recent studies. Particularly, Jin et al. [101] found that exosomes secreted by adipose-derived mesenchymal stem cells (ADMSCs) replenished the reduced miRNA-486 expression in podocytes under diabetic conditions to alleviate podocyte damage by enhancing autophagy. Similarly, exosomes secreted by ADMSCs can also transfer miR-215-5p [102], miR-26a-5p [103], and miR-15b-5p [104] to podocytes to play a protective role. Exosomes secreted by human urinary-derived stem cells also have a protective effect on podocytes achieved by transferring miR-16-5p to inhibit the expression of VEGFA [112]. Similarly, UC-MSCs reduce podocyte inflammation by delivering exosomes containing miR-22-3p [107]. ADMSC-derived exosomal miR-125a alleviates mesangial expansion in an animal model of DKD [106]. Moreover, UC-MSCs reduce hyperglycemic-induced apoptosis and EMT in RTEC by transferring miR-424-5p in exosomes [110]. UC-MSCs also secrete exosomes containing miR-23a-3p to inhibit the signal transducer and activator of transcription 3 signaling pathway by targeting KLF3 in RTECs [111].

Stem cell-derived exosomes can also transfer other molecules into kidney cells. Exosomes secreted by urine-derived stem cells have been shown to carry growth factors, TGFβ1, Angpt, and BMP7, which reduce apoptosis of podocytes and tubules and promote endothelial cell proliferation in DKD [113]. In vitro studies also revealed that UC-MSCs derived exosomes contain a variety of growth factors, including hepatocyte growth factor, EGF, fibroblast growth factor, and VEGF. These exosomal factors can alleviate inflammation in kidney cells, particularly RTECs and GECs [108, 109]. In summary, stem cells can achieve therapeutic effects by communicating with renal cells.

Increasing evidence has shown that the therapeutic effect of stem cells on DKD is largely realized through a paracrine effect, indicating that stem cell-based therapy is a promising therapeutic approach. Both in vitro and in vivo studies have shown that the paracrine effects of stem cells can alleviate renal damage of DKD. However, only two clinical trials on stem cells have been conducted. The first trial involving the infusion of allogeneic bone marrow-derived mesenchymal precursor cells into patients with DKD was completed in 2016 [118]. The second study, named the Novel Stromal Cell Therapy for DKD, evaluated the safety and efficacy of bone marrow-derived anti-CD362-selected allogeneic mesenchymal stromal cells in adults with DKD [119]. Both clinical experiments provided evidence for the effectiveness of stem cell therapy. Further clinical trials are needed to explore the feasibility, effectiveness, and reproducibility of stem cell-based therapies for DKD.

Sodium-glucose cotransporter 2 inhibitors (SGLT2i) reduce the reabsorption of glucose and sodium and play a role in hypoglycemia and cardiorenal protection by inhibiting SGLT2 in the proximal renal tubules [120]. Although SGLT2i primarily exert its therapeutic effects through the action on renal tubules, emerging evidence demonstrates its significant association with glomerular function. A study showed that empagliflozin, an SGLT2i, protects GECs by reducing the paracrine effects of podocyte-derived VEGFA [121]. This suggests that SGLT2i can block some pathological communication in kidney cells; however, more mechanisms need to be explored. Furthermore, astaxanthin, an antioxidant, alleviates podocyte-induced autophagy deficiency in mesangial cells [122]. In addition, as previously discussed, NMN supplementation has been demonstrated to restore dysregulated intercellular communication between podocytes and RTECs, consequently mitigating the development of proteinuria in DKD [95]. The effects of these drugs on intercellular communication are illustrated in Figure 2.

In addition, a variety of drugs, including renin-angiotensin-aldosterone system inhibitors, glucagon-like peptide-1 receptor agonists, and mineralocorticoid receptor antagonists, have been proven to play therapeutic effects on DKD. The ability of these drugs to slow the progression of DKD may be related to their improvement of abnormal intercellular communication, although the specific mechanisms of action require further in-depth research [123].

Substantial evidence shows that cellular communication plays an important role in the pathogenesis of DKD. Under physiological conditions, intercellular communication is crucial for kidney growth and development. However, under high-glucose conditions, abnormal communication among kidney cells is involved in the development of DKD (Table 2). These insights may offer new therapeutic directions.

However, several aspects require further investigation. First, some existing molecular mechanisms involved in intercellular communication, such as VEGFA, remain controversial. The opposite results have been reported in other studies. While the observed phenomenon may potentially correlate with the degree of podocyte depletion during distinct stages of DKD, additional investigations are necessary to validate this association. Second, the current research is insufficient, and further exploration of the cellular communication mechanisms is necessary. Similarly, many experiments are limited to in vitro conditions, and more in vivo experiments need to be performed. Third, drugs targeting abnormal intercellular communication or the adaptive cellular communication re-established by stem cells, which will provide new directions for the treatment of DKD, need to be developed. Finally, translating intercellular communication into clinical practice remains a critical challenge. The paracrine effect of stem cells has been elaborated by many studies. However, the factors secreted by stem cells are broad-spectrum and lack specificity. Pretransplantation genetic editing of stem cells to enable targeted exosome secretion and enhanced bioactive factor production represents a promising therapeutic strategy [124, 125]. In addition, current pharmacotherapies targeting intercellular communication exhibit limited clinical applicability due to metabolic instability and multi-target limitations. Nanoparticles engineered with precise size and charge parameters, conjugated with targeting peptides, enhance the drug delivery specificity and stability [126]. Another factor limiting its clinical translation is the species differences in rodent models. Human kidney organoids and organs-on-chips with diabetic microenvironments have partially compensated for this shortcoming [127, 128]. In conclusion, research on cellular crosstalk not only provides insight into the detailed pathophysiological mechanisms of DKD but also helps identify promising potential molecular targets for the treatment of DKD.

The authors have no relevant financial or nonfinancial interests to disclose.

This work was supported by National Natural Science Foundation of China (No. U21A20348), Natural Science Foundation of Henan Province (No. 222300420090), and Hong Kong Scholars Program (XJ2021043/P0034973).

Conceptualization was performed by Zhangsuo Liu, Sijie Zhou, and Alex Kwok Kuen Cheung. Original draft writing was performed by Linxiao Lv and Sijie Zhou. Figures and tables preparation were performed by Linxiao Lv, Wen Zheng, and Mingyang Hu. Funding acquisition was performed by Zhangsuo Liu, Sijie Zhou, and Gladys Lai Ying Cheing. Reviewed, revised, and editing were performed by Alex Kwok Kuen Cheung. All authors commented on previous versions of the manuscript, read and approved the final manuscript, and contributed to the study conception and design.