Inactivation of BACE2 stimulates release of endothelin-1 from human brain microvascular endothelial cells

Abstract

Beta-site amyloid precursor protein cleaving enzyme 2 (BACE2) is one of the most downregulated genes in the brain capillary endothelial cells derived from patients with Alzheimer's disease (AD). Endothelin-1 (ET-1) significantly contributes to the pathogenesis of AD. We hypothesized that loss of BACE2 increases production of ET-1 from human brain microvascular endothelial cells (BMECs). Genetic inactivation of BACE2 in cultured human BMECs significantly upregulated expression and release of ET-1. Mechanistic studies indicated that γ-aminobutyric acid type B receptor subunit 2/transforming growth factor beta 2 signaling pathway mediated the effect of BACE2 inhibition on ET-1 production.

Introduction

Endothelial dysfunction and cerebral hypoperfusion are early events in development of Alzheimer's disease (AD)^1,2; however, exact molecular mechanisms underlying vascular contribution to pathogenesis of AD remain incompletely understood. Beta-site amyloid precursor protein cleaving enzyme 2 (BACE2, a close homologue of BACE1) cleaves amyloid-β protein precursor (AβPP) at θ site, therefore inhibiting amyloid-β (Aβ) production, and suppressing development of AD pathology.^3,4 Previous study established that BACE2 causes proteolytic lavage of lymphangiogenic vascular endothelial growth factor receptor 3 ⁵ and protects expression and function of endothelial nitric oxide synthase. ⁶ BACE2 is highly expressed in the human brain endothelium and is one of the most downregulated genes in the capillary endothelial cells derived from brains of AD patients. ⁷ The downregulation of BACE2 is of special interest because the mechanistic consequences of impaired endothelial BACE2 function in AD have not been fully defined. In the present study we investigated the effects of endothelial BACE2 inactivation on expression, production, and release of endothelin-1 (ET-1) from the human brain microvascular endothelial cells (BMECs). The rationale for this approach was based on demonstrated ability of ET-1 to cause pathological vasoconstriction, oxidative stress, and inflammation.^8–10 Moreover, endothelial cells are major source of ET-1 in the human brain, ⁷ and mRNA and protein levels of ET-1 are significantly increased in the brains of AD patients. ¹¹ More recent study has demonstrated that ET-1 released form microvascular endothelium impairs function of neurovascular unit by disrupting blood-brain-barrier and by increasing tau phosphorylation. ¹² In the present study, we tested hypothesis that loss of BACE2 function promotes expression, production, and release of ET-1 from the human brain endothelium.

Methods

Cell culture

Primary human BMECs purchased from Applied Cell Biology Research Institute (Cell Systems, Kirkland, WA; cat. No. ACBRI 376 V) were cultured in endothelial growth medium (EGM2; Lonza, Allendale, NJ;) containing endothelial basal medium 2 (EBM2; cat. No. cc-3156), EGM2 bullet kit (cat. No. cc-3162, consisting of 2% fetal bovine serum, fibroblast growth factor, vascular endothelial growth factor, insulin-like growth factor, epidermal growth factor, ascorbic acid, hydrocortisone, and heparin), and Bac-off (Cell Systems; cat. No. 4Z0-644). We used cells of passages 4–5. In some experiments, cells were treated with recombinant human transforming growth factor beta 2 (TGF-β2, R&D system, Minneapolis, MN; cat. No. 302-B2) (2.5 ng/ml for 5 h), and then the cells were collected for measurement of mRNA.

Knockdown of BACE1, BACE2, γ-aminobutyric acid type B receptor subunit 1 (GABBR1), GABBR2, or a disintegrin and metalloproteinase domain-containing protein 10 (ADAM10) by small interfering RNA (siRNA)

Transfection of siRNA was performed on human BMECs as previously described in our study, ⁶ using Lipofectamine 2000 (cat. No. 11668027; Invitrogen/Thermo Fisher Scientific, Waltham, MA). BACE1siRNA targeting human BACE1 mRNA (ON-TARGETplus human BACE1siRNA #9, cat. No. J-003747-09), BACE2siRNA targeting human BACE2 mRNA (ON-TARGETplus human BACE2 siRNA #14, cat. No. J-003802-14), GABBR1 siRNA targeting human GABBR1mRNA (ON-TARGETplus human GABBR1siRNA mix of 4 siRNAs, cat. No. LQ-00509-01-0002), GABBR2siRNA targeting human GABBR2 mRNA (ON-TARGETplus human GABBR2siRNA #8, cat. No. J-005579-08), ADAM10siRNA targeting human ADAM10 mRNA (ON-TARGETplus human ADAM10siRNA #6, cat. No. J-004503-06), and ControlsiRNA (ON-TARGETplus Non-targeting siRNA #1, cat. No. D-001810-01) were obtained from Dharmacon, Horizon (Lafayette, CO). Cells were treated with siRNA (30 nM) and Lipofectamine 2000 (2.3 μl) in 1.5 ml EBM2 (for GABBR1siRNA, cells were treated with 60 nM of GABBR1siRNA or controlsiRNA and 4 μl of Lipofectamine 2000 in 1.5 ml EBM2) for 6 h, then 5 ml of EGM2 was added to the culture. After incubation for 20 h, cells were treated with EGM2 for another 24 or 48 h. For detection of phosphorylation of protein kinase B (AKT), cells were incubated with fresh EGM2 for 25 min before cell samples were collected.

Real time PCR (rt-qPCR)

RT-qPCR was performed as described in previous study.^6,13 Total RNA was isolated using RNeasy Plus Mini kit (Qiagen, Redwood City, CA; cat. No. 74137), and then reverse transcribed to cDNA using SuperScript III First-Strand Synthesis System kit (Invitrogen; cat. No. 18080-051), according to manufacturer's protocols. For quantification of mRNA levels of human BACE1, BACE2, TGF-β2, ET-1, GABBR1, GABBR2, and ADAM10, Bio-Rad CFX Connect Real-Time System (Bio-Rad, Hercules, CA) was used. Primers were PrimePCR SYBR Green Assay human BACE1 (cat. No. qHsaCID0012647), BACE2 (cat. No. qHsaCID0012156), TGF-β2 (cat. No. qHsaCID0018360), ET-1 (cat. No. qHsaCID0015579), GABBR1 (cat. No. qHsaCID0038341), GABBR2 (cat. No. qHsaCID0015507), ADAM10 (cat. No. qHsaCED0001377), GAPDH (cat. No. qHsaCED0038674), and β-actin (cat. No. qHsaCED0036269) (Bio-Rad, Coralville, IA). RT-qPCR was performed according to manufacturer's instruction with SsoAdvanced Universal SYBR Green Supermix (Bio-Rad) and Bio-Rad CFX manager 3.1. The primer sequences are intellectual property of Bio-Rad. All primers were validated by Bio-Rad to comply with MIQE guidelines. ¹⁴ Reaction conditions: 95⁰C, 45 s; (95⁰C, 10 s; 60⁰C, 20 s) x 39 cycles. The data were then quantified with the comparative C_T method for relative gene expression. ¹³ The amount of gene of interest relative to internal control GAPDH or β-actin was calculated as 2^−ΔCT. (where ΔC_T = C_T of targeted gene - C_T of internal control GAPDH or β-actin gene; expression levels of the internal control genes remained stable across different treatment conditions).

Western blot analysis

Western blot analysis was performed as described in our previous studies. ⁶ Protein samples were prepared by sonication in lysis buffer (containing 50 mmol/L NaCl, 50 mmol/L NaF, 50 mmol/L sodium pyrophosphate, 5 mmol/L EDTA, 5 mmol/L EGTA, 0.1 mmol/L Na₃VO₄, 1% Triton X-100, 10 mmol/L HEPES, pH 7.4) and protease inhibitor cocktail (Sigma-Aldrich, St Louis, MO; cat. No. P8340). Denatured protein samples were loaded on 4-15% mini-protein TGX gels (Bio-Rad; cat. No. 4561084). Nitrocellulose membrane (0.2 µM; Bio-Rad; cat. No. 1620112) was used for protein transferring. Primary antibodies (purchased from Cell Signaling Technology, Danvers, MA) were rabbit anti GABA_BR2 (cat. No. 3839S), rabbit anti AKT (cat. No. 9272S), and rabbit anti phospho-AKT (P-AKT) (cat. No. 9271S). Mouse anti BACE2 was obtained from Santa Curz Biotechnology (Dallas TX; cat. No. SC-271212). Mouse anti β-actin (Sigma-Aldrich, cat. No. A5316) was used as a loading control. After incubation with secondary antibody for 3 h at room temperature, and washing with TBS-T, the blots were imaged with LI-COR digital imaging analysis system (Odyssey Fc, Model 2800; LI-COR, Inc.) and analyzed using Image Studio Software Version 5.0; LI-COR, Inc.). Protein expression was normalized to β-actin. The ratio of p-AKT/AKT of each sample was calculated after p-AKT or AKT was normalized to β-actin.

Measurement of ET-1 in cell culture supernatant

Human BMECs were treated with BACE2siRNA or ControlsiRNA for 3 days, or with GABBR2siRNA or ControlsiRNA for 2 days. The cells were then cultured in EBM2 for 4 h and subsequently the supernatants (conditioned media) were collected. ET-1 protein concentrations in conditioned media were measured using ET-1 ELISA Kit (Invitrogen, cat. No. EIAET1), according to manufacturer's protocol.

Statistical analysis

Data are presented as mean ± SEM. All data sets underwent Shapiro-Wilk normality test and QQ plot analysis. Differences between mean values of two groups were compared using unpaired Student t-test. For non-parametric data, comparisons between two groups with non-normal distributed data were analyzed by the Mann-Whitney U-test (GraphPad Prism 9 software).

Results

Transfection of primary human BMECs with BACE2siRNA (30 nM) for 2 days significantly reduced BACE2 mRNA levels (Figure 1A), and increased ET-1 mRNA expression (Figure 1B). Furthermore, genetic inactivation of BACE2 for 3 days augmented production and release of ET-1 protein (Figure 1C). Moreover, BACE2siRNA elevated mRNA levels of TGF-β2 (Figure 1D). Importantly, treatment of human BMECs with TGF-β2 (2.5 ng/ml for 5 h) enhanced ET-1 mRNA expression (Figure 1E) thereby suggesting that increased TGF-β2 mediates stimulatory effect of BACE2 deletion on ET-1 production.

Figure 1.

Effect of BACE2 inhibition on ET-1 expression was mediated via GABA_BR2/TGF-β2 pathway. (A, B) BACE2siRNA transfection decreased BACE2 mRNA levels, and increased ET-1 mRNA expression in human BMECs. (C) After cells were treated with BACE2siRNA, ET-1 ELISA assay was performed to determine release of ET-1 into cell culture media. (D) Knockdown of BACE2 increased TGF-β2 mRNA expression. (E) Treatment of cells with recombinant human TGF-β2 enhanced ET-1 mRNA levels. (F, G) BACE2siRNA transfection inhibited the expression of GABBR2 mRNA and GABA_BR2 protein. (H-K) Knockdown of GABBR2 increased expression of TGF-β2 and EDN1, and augmented production and release of ET-1.

Our previous study revealed that genetic inactivation of BACE2 suppressed expression of GABBR2 (gene encoding GABA_BR2 protein). ⁶ Therefore, we examined the role of GABA_BR2 in mediation of the stimulatory effect of BACE2 inactivation on ET-1. As shown in Figure 1F and G, BACE2siRNA reduced GABBR2 mRNA expression and GABA_BR2 protein levels. Next, GABBR2siRNA treatment (30 nM, for 2 days) suppressed GABBR2 mRNA levels (Figure 1H), and enhanced mRNA expression of TGF-β2 (Figure 1I) and ET-1 (Figure 1J), thus leading to increased production and release of ET-1 (Figure 1K). These observations suggest that down-regulation of GABA_BR2 subunit mediates inhibitory effects of BACE2 on expression of TGFβ2 and ET-1.

Previous study on neurons has demonstrated that GABA_BR mediates taurine-induced phosphorylation of AKT thereby exerting neuroprotective effect. ¹⁵ Moreover, increased activity of p-AKT inhibits TGF-β2 mRNA expression. ¹⁶ Therefore, we tested whether AKT signaling is involved in BACE2/GABA_BR2/TGFβ2/ET-1 pathway. As shown in Supplemental Figure 1, knockdown of BACE2 or GABBR2 significantly decreased p-AKT levels, indicating that impairment of AKT signaling might mediate stimulatory effects of BACE2/GABA_BR2 inhibition on expression of TGF-β2 in BMECs.

In our prior study we showed that inactivation of BACE2 did not affect expression of GABBR1. ⁶ Interestingly, in the present study, suppression of GABBR1 by transfection of human BMECs with GABBR1siRNA (60 nM, for 2 days) elevated mRNA levels of TGF-β2 and ET-1 (Figure 2A-C). Thus, GABABR1 subunit may also participates in control of TGFβ2 and ET-1 levels. However, it is important to note that GABA_BR1/TGFβ/ET-1 signaling is not coupled to BACE2.

Figure 2.

Roles of GABBR1, ADAM10, and BACE1 in regulation of ET-1. (A-C) Genetic inactivation of GABBR1 increased mRNA expression of TGF-β2 and ET-1. (D-G) ADAM10siRNA or BACE1siRNA did not change ET-1 mRNA levels. Human BMECs were transfected with ADAM10siRNA (D, E), or BACE1siRNA (F, G), cell samples were collected for measurements of mRNA levels.

AβPP could be a substrate for α-like and/or β proteolytic cleavage by BACE2. Since our previous study demonstrated that BACE2siRNA treatment decreased production of soluble AβPPα (sAβPPα) but did not affect Aβ₁₋₄₀ levels in conditioned media, ⁶ we examined the effects of α- and β-secretase inhibition on ET-1 expression. Genetic inactivation of α-secretase ADAM10 (an α-secretase responsible for production of sAβPPα) by treatment with ADAM10siRNA (30 nM, for 2 days) did not affect ET-1 mRNA expression (Figure 2D, E). In addition, the inactivation of BACE1 by BACE1siRNA transfection (30 nM, for 2 days) also did not affect ET-1 mRNA levels (Figure 2F, G). Therefore, the stimulatory effect of BACE2 deletion on expression and secretion of ET-1 appears to be independent of α- and/or β-processing of AβPP.

Discussion

This is the first study to demonstrate that in the human brain endothelium, loss of BACE2 function stimulates production of ET-1. We also provide the first evidence that in the microvascular endothelium, inactivation of BACE2 causes down-regulation of GABBR2 and p-AKT, and augments expression of TGF-β2, thereby resulting in enhanced release of powerful vasoconstrictor, prooxidant, and proinflammatory protein ET-1 (Figure 3).

Figure 3.

Signaling pathway responsible for increased release of ET-1 from BMECs induced by inactivation of BACE2.

For many years, ET-1 has been recognized as a potent and long-acting vasoconstrictor. ¹⁷ While ET-1 is produced by vascular endothelial cells, it stimulates mitogenesis of smooth muscle cells, fibroblasts, and mesangial cells. ¹⁸ Recent study has demonstrated that ET-1 causes pericytes senescence. ¹⁹ In addition, ET-1 has prothrombotic and proinflammatory properties. ¹⁰ Importantly, existing literature supports the concept that ET-1 significantly contributes to the pathogenesis of AD.^8,10–12,20 Indeed, ET-1 causes capillary constriction,^8,9 disrupts blood-brain barrier, and increases tau phosphorylation. ¹² Importantly, prior studies demonstrated that ET-1 levels are increased in cerebral and temporal cortex as well as in cerebral blood vessels of AD patients.^11,21,22 Detrimental effects of ET-1 induced vasoconstriction and inflammation are most likely early events in pathogenesis of AD. Indeed, existing evidence suggests that reduction in cerebral blood flow occurs before the onset of dementia. ²³ Our observation that downregulation of BACE2 stimulates ET-1 production suggests that loss of endothelial BACE2 may contribute to the initiation and progression of AD pathology. The ability of BACE2 to prevent excessive increase in local concentration of ET-1 reinforces the concept that BACE2 is a vascular protective protein. ⁶

Bioavailability of ET-1 is determined by transcription rate, epigenetic regulation, mRNA stability, and proteolytic pathways responsible for ET-1 protein maturation. ²⁴ Furthermore, prior studies established that TGF-β activates ET-1 transcription via activin receptor-like kinases signaling pathway. ²⁴ In the present study we identified previously unrecognized signaling mechanism responsible for increased production and release of ET-1 from the human cerebrovascular endothelium. Inactivation of endothelial GABA_BR2 receptor subunit increased production of TGF-β2 and ET-1. The exact mechanism linking BACE2 function with GABA_BR2 subunit is currently unknown. In contrast to GABA_BR2, expression of GABA_BR1 subunit was not affected by the loss of BACE2. However, inactivation of GABA_BR1 increased expression of TGFβ2 and ET-1. Thus, GABA_BR1 subunit may participate in control of TGF-β2 and ET-1 levels independently of BACE2 activity. The signaling pathway that mediates GABA_BR-regulated production of TGF-β2 requires further studies. Interestingly, treatment with BACE2siRNA or GABBR2siRNA inhibited AKT phosphorylation, which is consistent with prior report that inactivation of GABA_BR suppressed p-AKT, ¹⁵ indicating that preservation of normal signaling activity of p-AKT may contribute in part to vascular protective function of endothelial BACE2.

It has been also demonstrated that increased activity of phosphorylated AKT inhibits TGF-β2 mRNA expression. ¹⁶ Thus, our study also suggests that AKT signaling could mediate the stimulatory effect of GABA_BR2 knockdown on TGF-β2 expression.

We do not have an explanation as to why endothelial GABA_BR1 and GABA_BR2 participate in different signaling pathways converging on control of TGFβ2 and ET-1 production in endothelium. However, detected increase in TGFβ2 expression is an important observation providing new insight into the BACE2 signaling in endothelium. TGFβ2 causes endothelial to mesenchymal transition thereby increasing permeability of endothelial cells.^25,26 Increased permeability of blood-brain-barrier may contribute to the development of AD pathology and cognitive impairment. It is also noteworthy that TGFβ2 levels are significantly increased in the brains of patients with AD. ²⁷ Most importantly, TGFβ2 causes neuronal cell death by binding to AβPP, therefore contributing to neuronal loss in patients with AD. ²⁸ Thus, we propose that preservation and protection of BACE2/GABA_B signaling in cerebrovascular endothelium may be viable therapeutic target in efforts to delay the onset and progression of AD.

It has been reported that excessive local concentrations of Aβ₁₋₄₀ and Aβ₁₋₄₂ cause contraction of capillaries resulting from stimulatory effects of Aβ peptides on release of ET-1 in AD brain. ⁸ Our previous study ⁶ demonstrated that down-regulation of BACE2 decreased AβPP protein expression but increased BACE1 protein levels. It is important to note that reduction of AβPP levels and increased expression of BACE1 have opposite effects on production of Aβ. This may explain our previous observation that levels of Aβ₁₋₄₀ were not altered by inactivation of BACE2. ⁶ To further rule out potential stimulatory effect of Aβ on ET-1 production, in the present study we knocked down BACE1. BACE1siRNA transfection did not affect ET-1 mRNA levels, indicating that in human BMECs basal production of Aβ does not affect ET-1 expression. Our previous study ⁶ in human BMECs revealed that BACE2siRNA treatment reduced production of neuroprotective molecule sAβPPα. In current study genetic inactivation of ADAM10 did not affect expression of ET-1 mRNA. Thus, our data suggest that under basal conditions, α or β cleavage of AβPP is not involved in the stimulatory effect of BACE2 inactivation on ET-1 production.

We speculate that severe loss of BACE2 in capillary endothelium of AD patients ⁷ may promote an increase in concentration of ET-1 leading to excessive vasoconstriction, oxidative stress, and inflammation. We also wish to point out that our findings reinforce concerns regarding potential adverse effects of BACE2 inhibition with nonselective BACE1 inhibitors employed in clinical trials on patients with AD.

Supplemental Material

sj-jpg-1-alr-10.1177_25424823251371040 - Supplemental material for Inactivation of BACE2 stimulates release of endothelin-1 from human brain microvascular endothelial cells

Supplemental material, sj-jpg-1-alr-10.1177_25424823251371040 for Inactivation of BACE2 stimulates release of endothelin-1 from human brain microvascular endothelial cells by Tongrong He and Zvonimir S Katusic in Journal of Alzheimer's Disease Reports