β1 integrin monoclonal antibody treatment ameliorates cerebral cavernous malformations

Abstract

β1 integrins are important in blood vessel formation and function, finely tuning the adhesion of endothelial cells to each other and to the extracellular matrix. The role of integrins in the vascular disease, cerebral cavernous malformation (CCM) has yet to be explored in vivo. Endothelial loss of the gene KRIT1 leads to brain microvascular defects, resulting in debilitating and often fatal consequences. We tested administration of a monoclonal antibody that enforces the active β1 integrin conformation, (clone 9EG7), on a murine neonatal CCM mouse model, Krit1^flox/flox ;Pdgfb-iCreERT2 (Krit1^ECKO ), and on KRIT1-silenced human umbilical vein endothelial cells (HUVECs). In addition, endothelial deletion of the master regulator of integrin activation, Talin 1 (Tln1), in Krit1^ECKO mice was performed to assess the effect of completely blocking endothelial integrin activation on CCM. Treatment with 9EG7 reduced lesion burden in the Krit1^ECKO model and was accompanied by a strong reduction in the phosphorylation of the ROCK substrate, myosin light chain (pMLC), in both retina and brain endothelial cells. Treatment of KRIT1-silenced HUVECs with 9EG7 in vitro stabilized cell-cell junctions. Overnight treatment of HUVECs with 9EG7 resulted in significantly reduced total surface expression of β1 integrin, which was associated with reduced pMLC levels, supporting our in vivo findings. Genetic blockade of integrin activation by Tln1^ECKO enhanced bleeding and did not reduce CCM lesion burden in Krit1^ECKO mice. In sum, targeting β1 integrin with an activated-specific antibody reduces acute murine CCM lesion development, which we found to be associated with suppression of endothelial ROCK activity.

Keywords: cerebral cavernous malformation; endothelium; β1 integrin.

Figure 1.. Long-term (overnight) treatment with 9EG7 antibody preserves endothelial cell-cell junctions in Krit1-silenced endothelial cells.

HUVECs, grown on fibronectin-coated glass chambered slides, were transfected with KRIT1 (siKrit1) or scrambled control (siScr) siRNA for 48 hours. (A) The cells were treated overnight with 5 μg/mL 9EG7 or IgG isotype control antibody before they were fixed and stained for junctional proteins, VE-cadherin or zonula occludens (ZO)-1. (B,C) Staining intensity was measured along cell-cell junctions using ImageJ and is reported as a percent of IgG-treated, siRNA-transfected control. Silencing KRIT1 resulted in decreased VE-cadherin and ZO-1 abundance along cell-cell junctions (57.8±4.3% and 69±7% of control levels respectively). Treatment with 9EG7 rescued VE-cadherin and ZO-1 levels to 87.4±6.9% and 94.4±7% of control levels, respectively. (Scale bar 25 nm., one-way ANOVA, *p<0.05, **p<0.01, ***p<0.001, N=4)

Figure 2.. Long-term (overnight) 9EG7 antibody treatment leads to downregulation of surface β1 integrin and reduced pMLC.

(A) HUVECs were treated overnight with either IgG isotype control or 9EG7 antibody at a concentration of 5 μg/mL, or left untreated (Unt). Cells were trypsinized and stained on ice using an APC-conjugated anti-β1 integrin (clone TS2/16) antibody to assess surface β1 integrin levels by flow cytometry. (N=4). Mean fluorescence intensity (MFI) for each condition is represented as a percent of untreated control (ctl). (B) HUVECs were treated with 9EG7 or IgG as described above. Cells were lysed in RIPA buffer and total β1 integrin was analyzed by Western blot, with no significant differences observed between samples (N=3). (C) HUVEC lysates were analyzed by Western blot for phosphorylated myosin light chain (pMLC) and total MLC. β-actin was used as loading control. (N=5) (D) HUVECs were transfected with scrambled control (siScr Ctl) or KRIT1-targeting (siKrit1 KD) siRNA for 48 hours before overnight treatment with IgG or 9EG7 as described above. Silencing KRIT1 resulted in increased pMLC levels in IgG-treated cells, while treatment with 9EG7 significantly reduced pMLC levels in Krit1-silenced cells.(N=4) (*p<0.05, ***p<0.001).

Figure 3.. In vivo treatment with 9EG7 ameliorates cerebral cavernous malformations.

Schematic of treatment timeline is shown (A). Postnatal day 1 (P1) mice were administered 50μg 4-hydroxy-tamoxifen to induce deletion of Krit1. Following genotyping, Krit1^ECKO mice were divided into two groups and were administered 4mg/kg 9EG7 (n=7) or IgG isotype control (IgG Ctl, n=6) antibody via retroorbital injection at postnatal day 5 (P5). At postnatal day 7 (P7), the mice were sacrificed and brains were collected for analysis by micro-computed tomography (B,C) or histologically with H&E staining (D-F). Total lesion volume (C), and average lesion size (E), were significantly reduced in 9EG7-treated mice, while lesion number (F) was not significantly changed. (Student’s t-test, *p<0.05, **p<0.01)

Figure 4.. Administration of 9EG7 antibody suppresses myosin light chain phosphorylation in Krit1^ECKO mice.

Retinas and sectioned hindbrain tissue from 9EG7 or control IgG-treated Krit1^ECKO mice (treated as previously described in Figure 3A) were immunostained for phosphorylated-myosin light chain (pMLC). pMLC staining in retinal tissue was normalized to isolectin B₄ (IB4-FITC) and expressed as a percent of IgG control (A,B) (n=4). pMLC-positive endothelial cells within CCM lesions were detected in hindbrain sections and quantified as a percent of total endothelial cells within the lesion (C,D) (n=3). β1-integrin activation in vivo had a suppressive effect on pMLC levels in both retina and brain samples. (Scale bar (A): 50μm, (C): 75 μm. *p<0.05, **p<0.01).

Figure 5.. Endothelial loss of Talin1 does not rescue CCM1 lesions caused by Krit1^ECKO.

Postnatal day 1 mice were administered 50μg 4-hydroxy-tamoxifen to induce deletion of Krit1 and Tln1 (n=4), or Krit1 alone (n=3). At postnatal day 7, the mice were sacrificed and brains were collected for analysis by micro-computed tomography (A,B). Hindbrains were also cryosectioned and analyzed by immunofluorescent staining for CD31 (green) or red blood cells (red) (C). CCM lesions are indicated by solid teal arrows, while micro-hemorrhage, abundant in Krit1;Tln1^ECKO mice, is indicated by hollow teal arrows (C). (Scale bar: 100μm)