CD13/Aminopeptidase N Is a Potential Therapeutic Target for Inflammatory Disorders Free

CD13, also known as aminopeptidase N or membrane alanyl aminopeptidase, is a type II membrane, 150-kDa metalloprotease with an extracellular-oriented catalytic domain. It is a seahorse-shaped molecule and usually forms a head-to-head homodimer by means of hydrophobic interactions (1, 2). Each monomeric molecule of CD13 possesses a seven-domain organization, which is characteristic of M1 metallopeptidases (1, 3). CD13 has been termed a “moonlighting enzyme” because of its multiple functions (4). Increasing evidence points to crucial regulatory functions for CD13 during normal and pathologic immune responses. CD13 has been suggested to play a role in Ag processing (5, 6), cell trafficking (7–10) and processing of inflammatory mediators, which are important features of immune responses. In this study, we review evidence concerning the involvement of CD13, including a soluble form of the molecule in inflammation and its potential as a therapeutic target in inflammatory disorders.

CD13 was named aminopeptidase N because of its preference for binding neutral amino acids and because it removes N-terminal amino acids from unsubstituted oligopeptides, with the exception of peptides with proline in the penultimate position (11). By cleaving N termini, CD13 regulates activity of numerous hormones, cytokines, and chemokines that participate in inflammation. Moreover, CD13 is coexpressed by MHC class II (MHC II)–bearing APCs (12), and is involved in the trimming of MHC II–associated peptides on the surface of APCs (5, 13, 14) (Fig. 1).

Some CD13 functions in the immune system are independent of its enzymatic activity. These mechanisms include cross-linking–induced signal transduction (15), enhancement of FcγR-mediated phagocytosis (16), acting as a receptor (17, 18), and binding of soluble CD13 (sCD13) to G-protein–coupled receptors (GPCRs) (19) (Fig. 1).

Various cellular responses, including homotypic aggregation, cell–cell adhesion, and migration, have been observed after cross-linking CD13 with mAbs. Interestingly, CD13, which includes a 7-aa cytoplasmic tail that was previously assumed to be inert, is itself phosphorylated in a Src-dependent manner (20). mAb cross-linking of CD13 induces clustering and phosphorylation of tyrosine 6 in the cytoplasmic domain of CD13, activation of Src, FAK, ERK, and potentially other components of the Ras/MAPK (JNK and p38) and PI3K pathways. Initiation of a Ca²⁺ flux by mAb against CD13 results in increased adhesion of monocytes (MNs) to endothelial cells (ECs) or to a monolayer of CD13 and upregulation of cytokines including IL-8 (15, 20). Anti-CD13 mAbs also induce integrin-independent homotypic aggregation of monocytic U937 cells independent of their effects on CD13’s enzymatic activity (7). Such CD13-induced cell–cell adhesion is also dependent on the carbohydrate-binding protein galectin-3. A possible mechanism could be that Ab cross-linking of CD13 triggers a signaling cascade leading to release of galectin-3, which then binds to another receptor and triggers adhesion or, in oligomeric form, binds to glycoconjugates on neighboring cells (21). CD13 can be constitutively phosphorylated at Tyr6 (10, 20) to interact with IQ motif containing GTPase activating protein 1 (IQGAP1), a scaffold protein that interacts with over 90 other proteins to mediate biological functions (22). CD13 serves to properly position IQGAP1 to perform its functions; thus, it was expected that CD13 participates in numerous functions mediated by IQGAP1, including cell motility, endocytosis, filopodia formation, angiogenesis, and phagocytosis (10). A recent study demonstrated that CD13 can tether IQGAP1 with active adenosine diphosphate ribosylation factor 6 (ARF6) and β1 integrin at the plasma membrane to allow β1 integrin recycling and cell migration (10). However, mAbs binding to distinct epitopes of CD13 mediate different consequences for cell adhesion. For example, cross-linking CD13 by some mAbs inhibits cell adhesion, whereas mAb 452 induces it (23).

CD13 colocalizes with FcγRs, including FcγRII/CD32 and FcγRI/CD64 after Fc receptor ligation with specific Abs (24). During FcγR-mediated phagocytosis, CD13 redistributes to the phagocytic cup and is simultaneously internalized into phagosomes with FcγRs, rendering ingestion more efficient by increasing the level and duration of Syk phosphorylation induced by FcγRs themselves (16). There is a possibility that CD13 positively modulates FcγR-mediated signals and functions when CD13 and FcγR bind to their ligands, as CD13 is a well-known receptor for viral proteins and other pathogen-derived molecules. Moreover, CD13 was reported to be a cell-surface receptor for some cytokines, including 14-3-3ε and 14-3-3σ (also known as [a.k.a] stratifin) (17, 18). By binding to CD13 on cells such as fibroblasts and chondrocytes, 14-3-3ε regulates production of matrix-degrading enzymes such as matrix metalloproteinases (MMP)–3 and –13 through p38 MAPKs and JNK-signaling cascades, whereas 14-3-3σ induces expression of MMP-1 through p38 MAPK activation (17, 18). Phosphorylation of the ectodomains of CD13 augments the binding affinity of 14-3-3 proteins to CD13, regulating the downstream signaling cascade. It has been proposed that enzymatic activity and binding to extracellular matrix proteins through Asn–Gly–Arg motifs is mediated through the catalytic site involving domains IV–VII; cell–cell adhesion is mediated by the C-terminal part, involving half of domain VI and VII, whereas signal transduction depends on the transmembrane and cytoplasmic domains (I and II) (8, 23).

CD13 can also be shed from the cell membrane by MMP-14 (25), and the resultant sCD13 functions by engaging GPCRs on cytokine-activated lymphocytes (19), MNs, ECs, and fibroblasts (26). However, the identity of such GPCRs has not yet been reported. With binding to GPCRs, sCD13 induces phosphorylation of Src, ERK1/2, and, to some extent, NF-κB, activating cell migration, angiogenesis, and other responses (26). Pertussis toxin, a GPCR inhibitor, significantly reduced chemotactic responses of cytokine-activated T cells (19) or monocytic cells as well as endothelial capillary morphogenesis in response to sCD13 (26).

CD13 is widely expressed. Within the hematopoietic system, CD13 is predominantly expressed on myelomonocytic lineage cells, and it has long been used as a lineage marker in the characterization and typing of leukemia or lymphoma cells (27). However, CD13 is not just a cellular marker, but also a molecule modulating development and function of immune-related cells (Table I).

CD13 has been implicated in MN/macrophage activation and differentiation (28). It is expressed on the surface of blood MNs and is upregulated by IFN-γ, LPSs, C5a, IL-4, and TGF-β but downregulated by IL-10 (28–31). There is also an increase of CD13 expression during the differentiation from MNs to macrophages (28). Meanwhile, blocking CD13 by mAb or by binding of human CMV (HCMV) inhibits macrophage differentiation (32).

CD13 mediates MN activation either by cross-linking of cell membrane CD13 or by binding of sCD13 to a GPCR. Cross-linking membrane CD13 by mAbs activates signaling pathways involving MAPK and PI3K and a Ca²⁺ flux, resulting in homotypic aggregation, cytokine secretion, and increased adhesive capacity (7, 15, 20, 21). SCD13 binding to a GPCR increases phosphorylation of multiple downstream signaling molecules and activates MNs (26). Moreover, CD13 interacts with FcγRs on the monocytic cell membrane and regulates FcγR-mediated phagocytosis (16, 33). CD13 also regulates MN migration and contributes to inflammation. The expression of CD13 is enriched specifically on the proinflammatory subset of MNs, and blocking CD13 by Ab or knockdown of CD13 significantly decreases MN infiltration, suggesting that CD13 may regulate trafficking and function of these cells (8). In addition, CD13 expression by ECs is also essential for MN infiltration (8, 34). CD13 is upregulated on activated ECs (35) and mediates MN/EC adhesion by homotypic interactions that involve the C-terminal domain (8). In CD13^−/− mice, lack of monocytic CD13 ablates anti-CD13–dependent MN adhesion to ECs; yet, CD13 appears to be dispensable for phagocytosis, proliferation, and Ag presentation of macrophages (36). Meanwhile, assessment of four inflammatory disease models (thioglycollate-induced peritonitis, collagen Ab-induced arthritis, DSS-induced colitis, and croton oil-induced contact hypersensitivity) showed that lack of CD13 had little effect on disease onset or progression (36). These models are, however, relatively independent of autoimmune T cell responses.

CD13 is involved in the development of dendritic cells from CD34⁺ hematopoietic progenitor cells (37). CD13 is highly expressed on dendritic cells and participates in cell-surface Ag processing through trimming of MHC II– or MHC I–bound peptides on APCs (12, 13, 38). Furthermore, CD13 is highly and specifically expressed on the CD8⁺ splenic dendritic cells, which are responsible for cross-presentation (39). It negatively regulates receptor-mediated, dynamin-dependent endocytosis of Ags to control T cell activation in adaptive immunity (39). CD13 restricts TLR4 endocytic signal transduction, balancing the innate response by maintaining the inflammatory equilibrium critical to innate immune regulation (40). However, development, maturation, Ag processing, and presentation of dendritic cells are normal in CD13^−/− mice (39).

Lymphocytes, including T cells and B cells, are important in inflammation and autoimmunity. However, the expression and function of CD13 on lymphocytes remain controversial. Generally, it is accepted that CD13 is absent on resting lymphocytes but may be inducible by stimulation or contact with some CD13-positive cells. CD13 expression is observed on hematopoietic stem cells and at the earliest stages of B or T cell differentiation but is then lost upon maturation (41, 42). Consistent with this, some malignant B cells were CD13 positive (43, 44). Human T cell lines HuT78 and H9 contained both Alanine-β-naphthylamide–hydrolyzing activity and CD13 mRNA; however, CD13 protein was undetectable (45). Mature B and resting T cells lack CD13 expression detectable by standard flow cytometry (46). In contrast, other studies suggested that CD13 mRNA is consistently detected in resting T cells, and its surface expression is markedly upregulated in response to T cell activation (47–51). For instance, CD13 expression on T cells can be upregulated by Con A (50) or other mitogens such as PHA and PMA (49). However, induction of CD13 in lymphocytes was not dominantly dependent on de novo protein biosynthesis or glycosylation, as inhibitors of protein biosynthesis and glycosylation only partly prevented mitogen-induced CD13 expression (49). CD13⁺ T cells have also been found outside the peripheral blood, for instance, in the synovial fluid of patients with arthritis (52), in the pericardial fluid of patients undergoing cardiac surgery (53), or in tumor microenvironment (54). The upregulation of CD13 in lymphocytes was mediated by adhesion to CD13⁺ cells, including fibroblast-like synoviocytes (FLS), MNs/macrophages, and ECs (55), which was attributed to increased CD13 promoter activity in lymphocytes through cell–cell contact (56). However, the possibility that lymphocytes increase CD13 on the cell surface by trogocytosis from nonlymphoid cells has not been excluded.

The CD13 inhibitors, actinonin and probestin, suppress inflammation through anti-proliferation and reduced production of proinflammatory cytokines such as IL-1β and IL-2, accompanied by augmented release of anti-inflammatory cytokines including TGF-β1 and IL-1R antagonist (IL-1RA) (46, 57). Although actinonin has no effect on chemotaxis or unstimulated cell migration, it selectively suppresses lymphocyte functions, including proliferation and production of IFN-γ, IL-17, and TNF-α, and ameliorates autoimmunity in vivo (58). Additionally, CD13 inhibitors could preserve and promote immunosuppressive functions of CD4⁺CD25⁺ regulatory T (Treg) cells. Inhibition of CD13 on CD4⁺CD25⁺ Treg cells by phebestin-enhanced expression of Foxp3 and TGF-β1 and alleviated acute colitis in mice (46, 59). Simultaneous application of dipeptidyl peptidase IV (DP IV; a.k.a DPP-4 or CD26) and CD13 inhibitors significantly suppressed DNA synthesis and increased TGF-β1 production in mitogen- or anti-CD3–stimulated human T cells in vitro (60, 61). In response to CD13 inhibitors, there is a marked increase of expression and activity of p42/ERK2 (62); meanwhile, a reduction of expression and phosphorylation of glycogen synthase kinase–3β (GSK-3β) occurs (63). However, another study showed that the CD13⁺CD4⁺CD25^hi Treg subpopulation exhibits stronger suppressive function among CD4⁺CD25^hi Treg cells by expressing a higher level of Foxp3, CTLA-4, membrane-bound TGF-β1, and B7H1 (64). A CD13 cross-linking Ab, WM15, suppresses expression of Foxp3, CTLA-4, B7H1, the secretion of TGF-β1 and IL-10 by CD13⁺CD4⁺CD25^hi Treg cells, and their ability to suppress CD25 expression and proliferation of CD4⁺CD25⁻ T cells (64).

CD13 is present on neutrophils and can be upregulated by proinflammatory agonists including fMLF, IL-8, and TNF-α (65). Anti-CD13 mAbs, WM15 and MY7, impair IL-8–mediated neutrophil migration in type I collagen gels and induce significant homotypic aggregation of neutrophils, which is dependent on CD13 cross-linking (65). Neutrophil apoptosis is critical to resolution of acute inflammation and prevention of granulocyte-mediated tissue injury (66). TNF-α was found as a priming agonist to enhance the rate of neutrophil apoptosis, whereas CD13 could regulate TNF-α–induced apoptosis in human neutrophils (67). Inhibition of cell-surface CD13 using CD13 inhibitors such as actinonin or bestatin significantly enhanced TNF-α–induced apoptosis of neutrophils (67). A possible mechanism was that CD13 inhibition interfered with shedding of TNFR1, resulting in augmented TNF-α–induced apoptosis, cell polarization, and respiratory burst (67).

Evidence of CD13’s role in mast cells is limited. One study claimed that CD13 is a negative regulator of mast cell activation, because Ag stimulation of CD13-deficient bone marrow–derived mast cells increased degranulation and proinflammatory cytokine production (68). This study also identified a functional interaction between FcεRI and CD13 on mast cells as Ab-mediated cross-linking of CD13 caused IL-6 production in an FcεRI-dependent manner (68). However, the pathophysiological or physical connection between CD13 cross-linking and FcεRI activation remains uncertain.

Angiogenesis is a critical step for tumor growth as well as metastasis and an integral component of the pathologic inflammatory response in arthritis. CD13 was exclusively expressed by angiogenic or activated ECs, but not resting ECs (35, 69). Hypoxia, angiogenic growth factors, and signals mediating angiogenesis potently induce CD13 transcription in primary ECs, whereas CD13 inhibitors interfered with tube formation, but not EC proliferation, suggesting that CD13 is a crucial regulator of endothelial morphogenesis during angiogenesis (35, 70). Otherwise, addition of exogenous CD13 is sufficient to restore arrested endothelial migration and morphogenesis resulting from inhibition of Ras/MAPK or PI3K signaling (71). In vivo, CD13^−/− mice showed a severely impaired angiogenic response under pathological conditions (72). One possible mechanism is that CD13 mediates galectin-3–induced angiogenesis through carbohydrate-recognition interactions (73). Alternatively, CD13, through its peptidase activity, can regulate endothelial invasion and filopodia formation by facilitating bradykinin (BK)–BK receptor B2 (B2R) internalization and plasma membrane protein organization (74). Pharmacological inhibitors of CD13 and anti-CD13 mAbs blocked angiogenesis and endothelial invasion (74). Moreover, recent work demonstrated that sCD13 can induce angiogenesis through GPCRs (26).

Synovial fibroblasts (a.k.a FLS) reside in the synovial lining of joints and act as key contributors to arthritis (75). In rheumatoid arthritis (RA), activated FLS are a major source of inflammatory cytokines and catabolic enzymes that promote joint damage (75). CD13 was found to be highly expressed on RA FLS (19, 76) and upregulated by proinflammatory cytokines, including IL-17, IFN-γ, and TNF-α (25). CD13 is more abundant in synovial fluids of RA compared with osteoarthritis (OA) (19). Moreover, CD13 is present as both a soluble molecule and on extracellular vesicles in biological fluids including plasma, synovial fluid, and FLS culture supernatant (25). SCD13 arises from the shedding of membrane CD13 on the FLS surface by MMPs such as MMP-14 (25). CD13 is localized predominantly in caveolae or lipid rafts on RA FLS, which are hotspot regions for functional cell–cell interactions (77). Inhibition of CD13 by either inhibitors of enzymatic activity or mAb resulted in decreased proliferation and diminished migration of RA FLS (25). Similarly, there was elevated expression of CD13 in dermal fibroblasts at wound sites (78). Blocking CD13 with mAbs (WM15, 3D8, and H300) reduced the migration activity of dermal fibroblast in a dose-dependent manner without any antiproliferative or cytotoxic effect (78). Recent work has found that sCD13 increases expression of some proinflammatory cytokines, including IL-6 and MCP-1, in RA FLS (26). Moreover, sCD13 activated signaling pathways in RA FLS through GPCRs and pertussis toxin blocked this effect (26).

Biological activities of inflammatory mediators can change robustly following short deletions or mutations at the N terminus. By cleaving N-terminal amino acids, CD13 can regulate the activity of numerous hormones, cytokines, and chemokines that participate in inflammation (Table I). Kinins are naturally occurring vasoactive peptide hormones, influencing inflammation, nociception, and cardiovascular homeostasis. The kinin family contains BK, kallidin (Lys–BK), and their des-Arg derivatives des-Arg⁹–BK and Lys–des-Arg⁹–BK, respectively) (79), and both kallidin and its derivative Lys–des-Arg⁹–BK are CD13 substrates. N-terminal Lys residue of Lys–des-Arg⁹–BK can be cleaved by CD13, yielding the much less potent des-Arg⁹–BK (80). Des-Arg⁹–BK or BK is not as susceptible to CD13 degradation, because the peptide bond preceding a proline is resistant to this enzyme (27, 81). Nevertheless, CD13 controls EC invasion in response to BK by facilitating signal transduction at plasma membrane after BK binding, but prior to ligand–receptor internalization; inhibition of CD13 enzymatic activity abrogates B2R internalization, leading to the attenuation of downstream events such as BK-induced activation of Cdc42 and filopodia formation, thus affecting EC motility (74). Interestingly, inhibitors of two BK receptors directly inhibited CD13 as well (82, 83).

Tuftsin, a natural tetrapeptide located in the Fc domain of the IgG H chain, can enhance phagocytosis, immune responses, bactericidal, and antitumor activities. It is hydrolyzed by CD13, generating an antagonist that competes for receptor binding and regulates tuftsin functions (81).

CXCL11 selectively recruits immune cells at sites of inflammation and inhibits angiogenesis upon binding to its receptors CXCR3 and CXCR7 (84). In combination with DP IV, CD13 rapidly processed CXCL11 to generate truncated forms, which may bind and desensitize CXCR3 as well as CXCR7, leading to a reduction of lymphocyte infiltration and an increase of EC migration (85). In vitro, CD13 cleaves synthetic oligopeptides corresponding to the N terminus of human IL-1β, IL-2, TNF-β (lymphotoxin), and murine IL-6, although natural IL-1α, IL-1β, IL-2, and G-CSF are resistant to CD13 (86). This suggests that proteolysis of cytokines by cellular exopeptidases may require one or more preceding steps. CD13 degraded IL-8 and inactivated its chemotactic activity (87), whereas other groups found that IL-8 was resistant to degradation by CD13 (85, 88). More studies are needed to clarify knowledge about inflammatory mediators processed by CD13.

RA is a chronic inflammatory disease characterized by persistent synovitis, systemic inflammation, and autoantibodies, causing cartilage and bone damage as well as disability (89). CD13 is strongly expressed in RA synovial tissue (28, 90), specifically, it is highly expressed by RA FLS, synovial MNs, and T cells and is also present in synovial fluid (19, 52, 76). In addition, CD13 is also found in serum and FLS culture supernatant, which contains sCD13 cleaved from FLS by MMP-14 or secreted in extracellular vesicles (25). However, measurement of sCD13 in RA sera did not show higher levels than in normal sera (19). cDNA microarray analysis showed that the gene encoding laeverin (an enzyme with sequence homology to CD13) was found to be the most significantly overexpressed gene in RA twins’ lymphoblastoid B cell lines compared with those from healthy twins (91).

CD13 facilitates cell trafficking and cytokine secretion of immune-related cells that play an important role in the pathogenesis of RA. SCD13 induced chemotaxis of lymphocytes, especially cytokine-activated T cells, a T cell population similar to that found in RA synovium (19, 76). This chemotactic activity was mediated through GPCRs which can be inhibited by pertussis toxin (19). Immunodepleting CD13 from RA synovial fluid removed a substantial fraction of the total cytokine-activated T cell (19) and MN (26) chemotactic activity from that fluid. Inhibitors of CD13 enzymatic activity and anti-CD13 Abs also decreased growth and migration of FLS (25). Neutralizing Abs against CD13 inhibited secretion of proinflammatory cytokines such as IL-6, IL-8, and MCP-1 from FLS (92) and RA synovial tissue ex vivo (26). Remarkably, sCD13 also increased MN migration, angiogenesis, and cytokine expression by RA FLS, which are key processes for joint inflammation (26). Injection of sCD13 into mouse knees caused severe joint inflammation in 24 h while upregulating cytokines such as IL-1β, IL-6, and MCP-1 and augmenting MN/macrophage infiltration (26).

Psoriasis is an immune-mediated disease manifesting in skin or joints or both (93). The skin plaques are characterized by epidermal hyperplasia with hyperproliferation, epidermal differentiation impairment, angiogenesis, and a brisk infiltration of MNs/macrophages, dendritic cells, and activated T cells in the dermis and the epidermis (94). CD13 has been identified as an activation marker in immune cells (27, 57) and keratinocytes (95, 96). Moreover, it is overexpressed in psoriatic skin and skin fibroblasts compared with skin of healthy control (97). Calcitonin gene–related peptide and IL-4 were implicated in the regulation of CD13 expression and activity (97). Inhibiting CD13 with actinonin, bestatin, or PAC-22 (a cytosol alanyl aminopeptidase–specific inhibitor) led to dose-dependent suppression of DNA synthesis in keratinocytes, correlating well with the simultaneous decrease in enzyme activity (96). In a mouse tail model of psoriasis, actinonin dose-dependently restored the stratum granulosum and ameliorated the impaired keratinocyte differentiation (96). In clinical situations, epidermal CD13 was especially upregulated in psoriatic plaques, and successful treatments were associated with a reduction of CD13 expression (98).

Multiple sclerosis (MS) is the most prevalent chronic inflammatory disease of the CNS, and it is currently incurable (99). A complex immune process, involving T cells, B cells, Abs, and cells of the innate immune system, such as bloodborne macrophages and microglia, is found in lesions of MS (99). Increased expression of CD13 on lymphocytes and PBMC of MS patients was found during acute exacerbation and chronic progression, compared with MS remission or other neurologic diseases, suggesting that CD13 can be a cell activation and migration marker in MS (100–102). Compared with use of a single ectopeptidase inhibitor, combined application of DP IV and CD13 inhibitors increased suppression of DNA synthesis as well as increased TGF-β1 production in human PBMC and isolated T cells (61). In experimental autoimmune encephalomyelitis (EAE), an animal model of MS, combined DP IV and CD13 inhibition, dramatically alleviated clinical severity of EAE compared with the response to each of the inhibitors alone (61). Therefore, a dual inhibitor of DP IV and CD13, PETIR-001, exhibits a therapeutic effect in EAE (103). Although inhibitors of DP IV and CD13 have no effect on migration of pathogenic effector T cells into target tissues, they induce release of TGF-β by T cells at the site of inflammation and suppress lymphocyte proliferation and production of IFN-γ, IL-17, and TNF-α (58). The amount of IL-17 and the number of CD13⁺IL-17R⁺ myeloid cells in peripheral blood were steadily decreased in MS patients with effective treatment (104).

Inflammatory bowel diseases (IBD), comprising Crohn disease and ulcerative colitis, result from an inappropriate inflammatory response to intestinal microbes (105). In a mouse model of colitis, simultaneously inhibiting DP IV and CD13 reduced colitis activity significantly (106). Endogenous opioid peptides such as Met- and Leu-enkephalin have beneficial impact on the function of the digestive system and exert anti-inflammatory effects through direct influence on immune cells. However, those peptides are rapidly degraded by endogenous enkephalinases, including neutral endopeptidase (NEP) and CD13. Blocking NEP and CD13 with Pal-KKQHNPR (an analogue of natural enkephalin) or sialorphin (a natural blocker) attenuated colitis in mice (107, 108).

It has been noticed that HCMV infection induces production of CD13-specific autoantibodies, which may promote inflammation and tissue damage (109–111). Evidence suggests an association between HCMV infection and IBD, but the mechanism is unclear. However, cytotoxic CD13–specific autoantibodies were identified in 66% of the sera obtained from HCMV-IgG–positive patients with ulcerative colitis and in 58% of the sera obtained from HCMV-IgG–positive patients with Crohn disease, but not in control individuals (112). These cytotoxic autoantibodies may interfere with cell functions and could thereby contribute to chronic inflammation in patients with IBD.

OA, once viewed as mechanical cartilage degradation, is now known to be a complex condition affecting the entire joint, in which matrix proteases, synovium, and systemic inflammation have key roles (113). OA shares some features with RA, and CD13 has also been detected in OA synovial tissue (19, 90). CD13 is expressed on articular chondrocytes, and a proinflammatory cytokine 14-3-3ε might directly bind to CD13, which transmits a signal in chondrocytes to induce a catabolic phenotype similar to that observed in OA (17). Therefore, 14-3-3ε–CD13 interaction could be a new therapeutic target in OA.

Significantly higher aminopeptidase activity was detected in bronchoalveolar lavage fluid from patients with interstitial lung diseases (ILD) because of RA, polymyositis/dermatomyositis, systemic sclerosis, and Sjogren syndrome than from normal volunteers or control patients who were free of ILD, and increased aminopeptidase activity and increased expression of CD13 were found in alveolar macrophages from patients with ILD (114). CD13 may also be a target for acne. DP IV and CD13 inhibitors upregulate IL-1RA in sebocytes and keratinocyte cell line, suppress Propionibacterium acnes–stimulated T cell proliferation and IL-2 production, and enhance TGF-β1 expression (115). CD13⁺CD33⁺ B cells were more numerous in Behcet disease and sepsis compared with healthy controls and RA or systemic lupus erythematosus patients (116), but the authors failed to further explore functions of this subset of B cells. Recently, another research group identified a distinct CD13⁺CD33⁺ population of leukemic cells contributing to a proinflammatory microenvironment that may be detrimental to long-term normal hematopoiesis within acute lymphoblastic leukemia bone marrow (117). With the capacity to secrete many proinflammatory cytokines, such as IL-1α, IL-1β, TNF-α, and GM-CSF, the CD13⁺ B cells may contribute to autoimmune diseases such as Behcet disease. Activated immune cells such as MNs are important factors in the systemic inflammatory response syndrome following trauma or sepsis. Increased number of tissue factor⁺CD13⁺ microparticles were detected in peripheral blood from patients in the acute phase of trauma and severe sepsis, which correlated significantly with severity, suggesting that tissue factor⁺CD13⁺ microparticles are important in the pathogenesis of early systemic inflammatory response syndrome following trauma or sepsis (118). A role for CD13 has been suggested in systemic lupus erythematosus, but the data are still preliminary (114, 119–121). CD13 is also a receptor for a strain of human coronavirus. A broad spectrum of variations in the CD13 domain were found critical for coronavirus binding (122). Some studies pointed out that single-site polymorphisms and alternative splicing of human CD13 are characteristic for acute myeloid leukemia, implying that CD13 could be a disease marker and target for therapy (123–125). However, to our knowledge, there are no data concerning CD13 polymorphisms in autoimmune diseases/inflammatory diseases.

Almost 100 CD13 inhibitors have been reported and tested (126, 127). Mina-Osorio (4) divided them into seven categories: natural inhibitors (betatin and amastatin), synthetic peptidomimetic inhibitors, synthetic nonpeptide inhibitors, tumor-homing peptides Asn–Gly–Arg, cholesterol-lowering drugs (ezetimibe), rAbs (single-chain, fragment-variable Ab fragments), and mAbs (WM15 and mAb 452). Most of the clinical studies with CD13 enzymatic inhibitors, such as bestatin and tosedostat, were performed in cancers including leukemia, lymphomas, and carcinomas of lung, bladder, esophagus, and stomach. Inhibition of aminopeptidase yielded a higher efficacy in some cancer types (128). Several other compounds inhibiting CD13 enzymatic activity are under preclinical development, for instance, LYP, a bestatin dimethylaminoethyl ester (129). Curcumin, a phenolic natural product, has been described as an irreversible inhibitor of CD13 and is currently being investigated for its effects on patients with cancer (130) as well as patients with inflammation (available from ClinicalTrials.gov: https://clinicaltrials.gov/ct2/show/NCT04032132?term=Curcumin&draw=2&rank=14; https://clinicaltrials.gov/ct2/show/NCT03122613?term=Curcumin&draw=2&rank=12). However, there are no available data concerning the clinical use of CD13 mAbs. To our knowledge, no clinical trial has been completed to test CD13 inhibitors in inflammatory diseases, except one ongoing phase II trial on psoriasis studying IP10.C8, a combined DP IV and CD13 inhibitor. The possible reason for this could be the complexity and extensiveness of CD13 functions. CD13 is essential for proper trafficking of inflammatory cells, which is necessary to prime and sustain the reparative responses (9). Therefore, the role of CD13 as well as the potential use of CD13 inhibitors in inflammatory diseases requires further investigation.

In summary, CD13 functions through enzyme-dependent mechanisms and enzyme-independent mechanisms. It not only can modulate development and activities of immune-related cells but also can regulate functions of inflammatory mediators. Furthermore, CD13 plays an important role in various inflammatory disorders, including RA, psoriasis, MS, and IBD. Many inhibitors against CD13 have been found and tested; yet, none of them has been used for the clinical treatment of human inflammatory disorders. The contribution of CD13 to various diseases and its significance as a therapeutic target need further study.

We thank Jonatan Hervoso and Mikel Gurrea from the University of Michigan for generous help in reviewing the manuscript. Special thanks are due to the University of Michigan librarians for assistance with access to literature.

The authors have no financial conflicts of interest.