The cornea is a structurally and functionally integrated avascular organ with three cellular layers interrupted by two acellular layers [10]. It is resistant to minor wounds; however, major ocular trauma can compromise its transparency [11]. A significant abrasion might disrupt the basement and Bowman’s membranes, leading to interactions between the epithelial and stromal layers of the cornea. This interaction initiates the complex wound-healing process [12,13]. Various growth factors, such as fibroblast growth factor (FGF), epidermal growth factor (EGF), and IGF-1, cross the disrupted Bowman’s membrane from the tear film to reach the stromal layer of the cornea during the initial wound-healing process [14,15]. In the epithelial layer, wounded epithelial cells undergo apoptosis, migration, and proliferation. The concentration of interleukin-1 (IL-1) increases in the epithelial layer upon injury. Therefore, IL-1 crosses the disrupted Bowman’s membrane into the stromal layer and becomes a major factor in killing keratocytes [16].

In the stromal layer, the corneal injury increases the expression of keratocyte membrane-bound αVβ/αβ integrins [17] and various metalloproteases, such as matrix metalloprotease 2 (MMP2) or MMP9 [18]. This cumulatively activates TGF-β from latency [19,20]. Neutrophils, monocytes, and macrophages infiltrate the wounded area during the healing process. They either become incorporated into fibrous DNA-like structures known as neutrophil extracellular traps of ECM or secrete TGF-β, increasing the concentration of TGF-β in the stromal layer [21,22,23,24]. Therefore, TGF-β takes over the control of the corneal wound-healing process. Quiescent keratocytes transdifferentiate into myofibroblasts, enlarged spindle-shaped cells characterized by the neo-expression of α-SMA in the presence of TGF-β. Incorporating α-SMA into the corneal ECM generates a contractile force that disrupts the orthogonal arrangement of its collagen fibrils, altering its normal curvature [25,26].

The active form of TGF-β first binds to the keratocyte plasma-membrane-bound TGF-β type I/II receptor (TBRI/II) and switches on the downstream SMAD-dependent signaling cascade [27]. This increases the expression of various fibrotic genes, such as MMPs and connective tissue growth factor (CTGF) [28]. In addition to TGF-β, other neuropeptides, such as calcitonin gene-related peptide, vasoactive intestinal peptide, and substance P (SP), are released from the axonal nerve endings of corneal epithelial layers. These neuropeptides cross the disrupted Bowman’s membrane to reach the stromal layer of the cornea. Substance P binds to the neurokinin 1 receptor (NK1R) of keratocytes and activates three different signaling pathways [29]. The SP/NK1R complex activates the RhoA/ROCK pathway to phosphorylate LIM kinase (LIMK). This inactivates cofilin (an actin-cleaving enzyme) [30], preventing actin degradation and promoting actin polymerization. Actin mostly remains bound to myocardin-related transcription factors A/B (MRTFA/B); however, the increased actin polymerization allows myocardin-related transcription factor A/B (MRTFA/B) to bind to serum response factor, activating profibrotic genes, such as MMPs [31,32]. The SP/NK-IR complex activates phospholipase C-beta and adenylyl cyclase, increasing the intracellular calcium level and cyclic adenosine monophosphate (cAMP) concentrations. This switches on protein kinase A and activates the extracellular signal-regulated kinase/mitogen-activated protein kinase (ERK/MAPK) pathway, increasing the proliferation of keratocytes and fibroblast cells [33,34]. The increased Ca²⁺ ions bind to calmodulin and cohere with the Kca3.1 ion channel of keratocytes to hyperpolarize the cells. This causes the cells to skip the G1 phase of the cell cycle, promoting fibrotic cell proliferation [35]. The increase in cellular stress increases the intracellular reactive oxygen species (ROS) levels, which share a positive feedback loop with the secretion of various growth factors, such as TGF-β, proinflammatory cytokines, etc. [36] Interleukin 10 (IL-10), a proinflammatory cytokine, binds to its cognate receptor and activates the nuclear factor kappa B (NF-κβ) pathway, increasing stress and other inflammatory genes’ expression [37]. TGF-β also increases the expression of bromodomain-containing protein 4 (BRD4), increasing the expression of kelch-like ECH-associated protein 1 (keap1). Keap1 might bind to the antioxidant gene nuclear factor erythroid 2–related factor 2 (Nrf-2) and undergo proteasomal degradation [38,39]. Cellular stress also increases the expression of the ubiquitin-specific peptidase-10 (USP10) gene. The USP10 gene stabilizes p53, increasing apoptosis of wounded keratocytes. However, it also deubiquitinates integrins, contributing to the activation of TGF-β and fibrosis [40]. The apoptotic cells also release TGF-β and other inflammatory cytokines, completing the loop of fibrosis [41,42] (Figure 2).

The complex signaling cascade signifies the corneal wound healing and scarring process. The corneal scarring process continues even after the wound has healed, making the condition worse. Keratocytes return to their quiescent state after wound healing, and the immune cells disappear. However, the disrupted Bowman’s membrane does not regenerate because of its low regenerative capacity. This causes extracellular vesicles and persisting fibroblast cells to continuously invade the stromal layer, accumulating TGF-β and repeating corneal ECM remodeling. Therefore, a scarred cornea is an immediate outcome of the multiplexed corneal wound-healing process [43].

Good vision requires a healthy corneal stromal ECM. The wound-healing process replaces the clear transparent stromal tissues with fibrotic tissues containing a disorganized collagen fibril arrangement and different proteoglycan content. There is an increase in the synthesis of chondroitin sulfate, hyaluronan, and biglycan and a decrease in the synthesis of keratin sulfate and lumican, which are required for maintaining corneal transparency [44,45]. This disordered, newly remodeled corneal ECM also lacks normal curvature and possesses low mechanical strength and elasticity. It refracts the incident light in all directions; therefore, the light is not focused on the retina, leading to vision impairment [46]. Externally stimulating the body’s response in attenuating the signaling cascade of corneal fibrosis might be a better approach to recovering a scarless cornea.

Restoring useful vision is the primary concern of patients with corneal scarring. This can be achieved by preventing the excessive migration of corneal ECM components during wound closure. TGF-β, the growth factor responsible for disrupting the corneal ECM, must remain inactive during the wound-healing process.

Exosomes are membrane-bound vesicles released by cells into their extracellular space. They have emerged as a novel therapeutic approach over cell-based therapy for treating corneal scarring [47]. Most stem-cell-based therapies face several challenges, such as immunogenicity, the high cost of production, ethical concerns, stability, and survival. However, exosomes have less stringent safety and regulatory requirements than other stem-cell-based therapies. These microvesicles act as biological vehicles that deliver parent cell cargo to the target cells. The exosomes are internalized by the target cells by either receptor-mediated endocytosis or simple membrane fusion. Upon reaching the cytoplasm of the target cell, the exosomes release their contents, activating various signaling cascades [48]. Exosomes can be isolated from stem cells using simple techniques, such as ultracentrifugation, ultrafiltration, size exclusion chromatography, and immunity capture. These vesicles can deliver antifibrotic proteins and miRNAs from stem cells to the ocular surface to modulate the therapeutic signaling cascade of the scarred cornea.

An increase in the concentration of chemokines and cytokines at the injury site is the hallmark of wound healing. Interleukin 8 (IL-8), a proinflammatory cytokine, binds to glycosaminoglycans (GAGs). In this GAG-bound form, IL-8 interacts with the C-X-C motif chemokine receptor 1 (CXCR1) and CXCR2 of neutrophils that infiltrate at the site of injury, leading to oligomerization and a haptotactic gradient. This haptotactic gradient directs the migration of neutrophils to the wounded area and seeds the process of corneal fibrosis [49]. Corneal stromal stem cells (CSSCs) express tumor necrosis factor (TNF)-stimulated gene 6 (TSG-6), a hyaluronan-binding protein, during inflammation [50]. This 35 kDa protein interacts with IL-8 via its link module domain and disrupts the binding of IL-8 and GAGs, preventing neutrophil migration [51,52]. Shojaati et al. [53] isolated extracellular vesicles (EVs) from CSSCs, mixed them with fibrin gel, and administered the EV–fibrin gel to the surface of an eye with corneal debridement. After 2 weeks of debridement, the EV-treated cornea showed minor corneal scarring compared to CSSC-treated and untreated controls. EVs were more effective in preventing neutrophil infiltration, reducing fibrotic marker expression, and restoring corneal morphology.

Cultured CSSCs treated with exosomes isolated from adipose-derived stem cells (ADSCs) showed optimal proliferation, reduced apoptosis, increased aldehyde dehydrogenase 1 (ALDH1), decreased MMP1, MMP3, and MMP9 expression, and increased collagen I, II, III, IV, and V expression compared to untreated CSC [54]. Therefore, ADSC-derived exosomes might revive the plasticity of transformed corneal keratocytes or stromal cells by increasing keratocyte marker expression and ECM production. The therapeutic effects of ADSCs and ADSC-derived exosomes could be attributed to microRNA-19a, which binds the 3′UTR of homeodomain-containing serine/threonine kinase 2 (HIPK2). MicroRNA-19a post-transcriptionally suppresses homeodomain-interacting protein kinase 2 (HIPK2), preventing the activation of the Jun N-terminal kinase (JNK) pathway and fibrosis [55]. Corneal keratocytes, when co-cultured with ADSC-derived exosomes, reduce HIPK2, phosphorylated SMAD3, and p53 expression, preventing the transformation of keratocytes into myofibroblasts and apoptosis of the nearby injured cells. This condition was reversed by the lentivirus-mediated overexpression of HIPK2 in keratocytes, confirming the role of miRNA-19a in corneal scar healing [56].

Exosome-based therapy is still in its nascency and requires preclinical research to be used for treating scars. Exosomes isolated from corneal stromal stem cells effectively prevent corneal scarring and require more preclinical safety and toxicity studies. Similarly, ADSC-derived exosomes reduce apoptosis and restore keratocyte marker expression in corneal stromal cells. Further characterization of these experiments is required to check their preclinical efficacy and gain better molecular insight to understand their therapeutic benefits. Exosomes from HCEs can hasten the corneal wound-healing process and prevent corneal scarring (Figure 3).The parent cells could be engineered to overexpress specific antifibrotic proteins. Exosomes consisting of this protein can be used in combination with tissue-derived microparticles to increase the effectiveness of exosome-based therapy for a scarred cornea.

The idea of directly targeting the upregulated genes of the corneal scarring signaling cascade to heal a scarred cornea has attracted researchers over the past few years. Semaphorin 3A (SEMA-3A), ubiquitin-specific protease-10 (USP 10), and calmodulin/calcium-activated K+ channel 3.1 (Kca3.1) are upregulated during the corneal wound-healing process. The increased expression of these genes contributes to corneal fibrosis. Therefore, targeting these genes using siRNA can mitigate the scarred corneal environment. The complete knockout of a targeted gene is undesirable because genetically modifying a mammalian cell is difficult, time-consuming, and expensive; moreover, knocking out a gene could have lethal consequences. Therefore, the in vivo delivery of small oligonucleotide sequences, designed to target certain mRNAs, using a suitable delivery vehicle to silence a target gene might be an effective way to prevent corneal fibrosis.

3.3. Protein Overexpression in Preventing Corneal Scarring