Kallikrein-Kinin System: An Emerging Competitor or Collaborator for VEGF in Diabetic Macular Edema?

An increasing number of patients with diabetes suffer from vision-threatening diabetic retinopathy (DR), i.e., proliferative diabetic retinopathy (PDR) and diabetic macular edema (DME), worldwide (1). Vascular endothelial growth factor (VEGF) plays a key role in the angiogenic responses in PDR, and the development of anti-VEGF therapy has reduced the burdens of PDR patients (2,3). Retinal vascular permeability leads to morphological and functional damages in the neuroglial components in the retinas and concomitant visual disturbance in DME (4–6). Although anti-VEGF agents have been effective for many patients with DME, their effects are often slow and partial and the benefits are limited, suggesting that there are other cellular and molecular mechanisms in addition to VEGF (7).

Diabetes-induced disruption of the blood-retinal barrier (BRB) manifests as clinical findings (i.e., retinal edema, hemorrhages, hard exudates on fundus examinations, and dye leakage on fluorescein angiography). In physiological retinas, the vascular endothelium functions as an essential barrier, which is supported by neuroglial cells. In diabetic retinas, increased paracellular flux, transcellular transport, and cell death in vascular endothelial cells promote vascular hyperpermeability (5). Among many growth factors and cytokines, VEGF and tumor necrosis factor-α were shown to disrupt the BRB (8,9). In addition to autocrine/paracrine signaling, a classic pathway of vasodilation and inflammation, plasma kallikrein (PKal)-kinin cascade, has recently been revealed to exert potent effects on retinal vascular hyperpermeability (10). Proteomics approaches have discovered a new effect of carbonic anhydrase (CA)-I in the vitreous humor in PDR patients that increases pH and concomitantly activates the kallikrein-kinin pathway (11).

In this issue of Diabetes, Kita et al. (12) demonstrate that the PKal-kinin-nitric oxide synthase (NOS) pathway promotes vascular hyperpermeability in DME, independent of the biological effects of VEGF. Biochemical analyses of clinical specimens from DME patients revealed that both PKal and VEGF were increased in the intraocular fluids, although they were not related to each other. The authors conducted interesting translational research to test the hypothesis that the PKal-kinin pathway had VEGF-independent effects on vascular permeability in DME. The vitreous specimens from DME patients containing higher amounts of PKal induced vascular permeability at a longer time point, mediated via bradykinin (BK) B1 or B2 receptors (B1Rs or B2Rs) but not via VEGF signaling. In contrast, the vitreous specimens with high concentrations of VEGF promoted dye leakage at a shorter time point, which did not depend on B1Rs or B2Rs. Further animal experiments confirmed that BK or des-Arg⁹-BK (DABK) increased vascular permeability in diabetic rodents, although antagonists for B1R or B2R did not influence VEGF-induced permeability in vivo. The authors also investigated the functions in NOS as the downstream effectors and showed that endothelial NOS (eNOS) and inducible NOS (iNOS) stimulation by BK and DABK, respectively, promoted retinal vascular permeability.

In vivo experiments combined with clinical specimens, to some extent, answered the question raised in the clinic: what are the VEGF-independent molecular mechanisms in DME (12)? As the efficacy of anti-VEGF therapy is significant but limited in DME (7), the authors proposed a novel therapeutic target for vascular permeability in DME (12). Further in vivo pathway analyses using both inhibitors and gene ablation confirmed definitely the detailed molecular mechanisms in this cascade and suggested the candidates for additional molecular targets and possible adverse effects. Intriguingly, several publications reported that B1R and B2R increase the expression of VEGF and its main receptor, VEGFR2, and that B2R has the potential to transactivate VEGFR2 (13,14). BK and VEGF share eNOS activation and Src-mediated VE-cadherin phosphorylation (15,16). Molecules in the PKal-kinin cascade derive from the serum, which allows researchers to hypothesize that VEGF initiates BRB disruption and then PKal-kinin cascade promotes vascular permeability. These two major systems, PKal-kinin cascade and VEGF signaling, might promote the pathological effects reciprocally and concomitantly contribute to sustained vascular hyperpermeability in chronic pathogenesis of DME (Fig. 1).

In humans, tissue kallikreins, a subfamily of serine proteases, and PKal share substrates and downstream effectors and concomitant biological effects including angiogenic responses (17). In contrast, tissue kallikrein inhibits VEGF signaling via the cleavage of VEGF165 isoform, which might reduce vascular permeability in DME (18). The pathway analyses in the article by Kita et al. (12) showed that the downstream effectors of kallikreins (i.e., kinins, their receptors, and NOS) were redundant, which suggested that the bottleneck of this cascade was PKal and might be supportive of the ongoing phase I clinical trials studying the effects of PKal inhibitor on DME.

Systemic hypertension is a twin risk factor for DR progression, and the renin-angiotensin system, one of the major regulators of blood pressure, contributes to VEGF/VEGFR expression and retinal angiogenesis (19). As BK is metabolized by ACE, ACE inhibitors reduce the production of effector peptides and angiotensin II and the degradation of kinins, leading to additive effects on vascular tonus and reduced blood pressure (20). Clinicians might be concerned with the question about whether ACE inhibitors cause accumulation of kinins with concomitant vascular permeability in DME and whether inhibitors of PKal would reduce the accumulation of BK and DABK, thus canceling the possible adverse effects of ACE inhibitors and concomitantly becoming a better candidate for therapeutic targets.

The article by Kita et al. (12) provides definitive data regarding a PKal-kinin-NOS pathway independent of VEGF at least in part, which would elucidate a novel molecular mechanism as well as propose a therapeutic target in retinal vascular permeability in DME. Many questions in the molecular world are raised. What is the association of this pathway with secreted proteins other than VEGF, intracellular signaling, and diabetes-induced biochemical pathways? How does PKal stimulation promote paracellular flux, transcellular transports, or cell death in retinal vascular endothelial cells? How does the kallikrein-kinin cascade influence other retinal cells? Because of the multifaceted mechanisms in DME, would the combination therapy be planned after a convincing study of PKal inhibitors is completed? Or, if good biomarkers indicate the pathological patterns in DME, would customized medicine improve the visual prognosis? Novel aspects of the classic kallikrein-kinin system revealed by translational research are improving the basic and clinical research in DME.