Photoreleasable ligands to study intracrine angiotensin II signalling

Abstract

Key points: The renin-angiotensin system plays a key role in cardiovascular physiology and its overactivation has been implicated in the pathogenesis of several major cardiovascular diseases. There is growing evidence that angiotensin II (Ang-II) may function as an intracellular peptide to activate intracellular/nuclear receptors and their downstream signalling effectors independently of cell surface receptors. Current methods used to study intracrine Ang-II signalling are limited to indirect approaches because of a lack of selective intracellularly-acting probes. Here, we present novel photoreleasable Ang-II analogues used to probe intracellular actions with spatial and temporal precision. The photorelease of intracellular Ang-II causes nuclear and cytosolic calcium mobilization and initiates the de novo synthesis of RNA in cardiac cells, demonstrating the application of the method.

Abstract: Several lines of evidence suggest that intracellular angiotensin II (Ang-II) contributes to the regulation of cardiac contractility, renal salt reabsorption, vascular tone and metabolism; however, work on intracrine Ang-II signalling has been limited to indirect approaches because of a lack of selective intracellularly-acting probes. Here, we aimed to synthesize and characterize cell-permeant Ang-II analogues that are inactive without uncaging, but release active Ang-II upon exposure to a flash of UV-light, and act as novel tools for use in the study of intracrine Ang-II physiology. We prepared three novel caged Ang-II analogues, [Tyr(DMNB)(4)]Ang-II, Ang-II-ODMNB and [Tyr(DMNB)(4)]Ang-II-ODMNB, based upon the incorporation of the photolabile moiety 4,5-dimethoxy-2-nitrobenzyl (DMNB). Compared to Ang-II, the caged Ang-II analogues showed 2-3 orders of magnitude reduced affinity toward both angiotensin type-1 (AT1R) and type-2 (AT2R) receptors in competition binding assays, and greatly-reduced potency in contraction assays of rat thoracic aorta. After receiving UV-irradiation, all three caged Ang-II analogues released Ang-II and potently induced the contraction of rat thoracic aorta. [Tyr(DMNB)(4)]Ang-II showed the most rapid photolysis upon UV-irradiation and was the focus of subsequent characterization. Whereas Ang-II and photolysed [Tyr(DMNB)(4)]Ang-II increased ERK1/2 phosphorylation (via AT1R) and cGMP production (AT2R), caged [Tyr(DMNB)(4)]Ang-II did not. Cellular uptake of [Tyr(DMNB)(4)]Ang-II was 4-fold greater than that of Ang-II and significantly greater than uptake driven by the positive-control HIV TAT(48-60) peptide. Intracellular photolysis of [Tyr(DMNB)(4)]Ang-II induced an increase in nucleoplasmic Ca(2+) ([Ca(2+)]n), and initiated 18S rRNA and nuclear factor kappa B mRNA synthesis in adult cardiac cells. We conclude that caged Ang-II analogues represent powerful new tools for use in the selective study of intracrine signalling via Ang-II.

Figure 1. Structures of the caged Ang-II analogues and competitive displacement of [¹²⁵I]Ang-II by caged Ang-II analogues

A, Ang-II. B, Ang-II-ODMNB. C, [Tyr(DMNB)⁴]Ang-II. D, [Tyr(DMNB)⁴]Ang-II-ODMNB. The position of the photosensitive DMNB moiety is indicated in brackets and was added either on the side chain of the tyrosine at position 4, the C-terminal carboxylic function of phenylalanine-8, or at both sites. Red and blue surfaces describe negative and positive electrostatic potentials (−3.5 k_BT, +3.5 k_BT), respectively. Three-dimensional structures were generated using PyMol visualization software. The electrostatic potentials were calculated using the Adaptive Poisson-Boltzmann Solver with the PyMol tool. (E, F, competitive displacement of [¹²⁵I]Ang-II by Ang-II or caged Ang-II analogues (n = 6/condition) in HEK 293 cells transfected with AT1R or AT2R. Data are percentage of specific radioligand binding in absence of competitors. Non-specific binding was determined in the presence of 1 μm Ang-II.

Figure 2. Concentration-dependent responses of rat thoracic aortic rings to various analogues without uncaging

Contraction-recordings were obtained with a curvilinear pen recorder for (A) Ang-II, (B) Ang-II-ODMNB, (C) [Tyr(DMNB)⁴]Ang-II and (D) [Tyr(DMNB)⁴]Ang-II-DMNB. E, immunoblotting for AT1R, AT2R and N-cadherin (positive control) in membranes isolated from rat thoracic aorta, as well as HEK 293, HEK 293-AT1R and HEK 293-AT2R cells. F, overall results expressed as percentage of contractile response induced by 40 mm KCl. Data are shown as the mean ± sem of at least 5 experiments per group performed on tissues isolated from separate animals for each experiment. Descending arrows on original recordings indicate baseline adjustment.

Figure 3. Photolysis kinetics of various analogues

Photolysis (100-W UV-lamp) of a 10⁻⁸ m solution of (A) [Tyr(DMNB)⁴]Ang-II, (B) Ang-II-ODMNB and (C) [Tyr(DMNB)⁴]Ang-II-ODMNB was performed in Krebs–Henseleit buffer and analysed by mass spectrometry (n = 3 per group). D, caged peptide concentration over time after UV-irradiation. E, signal-intensity (Y-axis) versus wavelength (X-axis) and HPLC elution-time (Z-axis) at different times after photorelease of [Tyr(DMNB)⁴]Ang-II as analysed by analytical HPLC.

Figure 4. Contractile responses of rat thoracic aortic rings after photolysis of various analogues

Isometric aortic tension recording was obtained with a curvilinear pen recorder after a 15 min incubation in the presence of caged analogues (10⁻⁸ m) followed by in situ photolysis with 30-W UV-lamp. A, control. B, [Tyr(DMNB)⁴]Ang-II. C, Ang-II-ODMNB. D, [Tyr(DMNB)⁴]Ang-II-DMNB. E, overall results as a percentage of contractile response induced by 40 mm KCl. Data are shown as the mean ± sem (n = 4 independent preparations from each of 4 animals per group).

Figure 5. AT1R-dependent ERK-phosphorylation after photolysis of [Tyr(DMNB)⁴]Ang-II

Effects of (A) Ang-II, (B) [Tyr(DMNB)⁴]Ang-II and (C) UV-irradiated [Tyr(DMNB)⁴]Ang-II on serum starved HEK 293 cells transfected with AT1R. Phosphorylated ERK1/2 (p-ERK) and total ERK1/2 (ERK) immunoreactivity was determined on cell-lysates. D, p-ERK immunoreactivity normalized to total ERK immunoreactivity (mean ± sem). E, p-ERK and ERK immunoreactivity in AT1R-transfected HEK 293 cells treated with vehicle, in the presence or absence of 10 nm Ang-II or [Tyr(DMNB)⁴]Ang-II (cAng-II), or valsartan (Val; 1 μm, 30 min) with or without UV-irradiation (1 min). F, mean ± sem data corresponding to the experiment in (E) (n = 3 per concentration). **P < 0.01, ***P < 0.001, ns, non-significant.

Figure 6. AT2R-dependent cGMP production after photolysis of [Tyr(DMNB)⁴]Ang-II

cGMP was measured in serum-starved HEK 293 cells transfected with AT2R after incubation with vehicle (control), Ang-II (10 nm) or [Tyr(DMNB)⁴]Ang-II, 10 nm) in the presence of PD123319 (PD, 1 μm), l-NAME (1 mm) and UV-irradiation, as indicated. Data are shown as the mean ± sem (n = 3 per condition). ***P < 0.001, ##P < 0.01 or ###P < 0.001 versus Ang-II, ^§§§P < 0.001 versus [Tyr(DMNB)⁴]Ang-II. ns, non-significant.

Figure 7. Cellular uptake of fluorescein-[Ahx⁰,Tyr(DMNB)⁴]Ang-II

A, intracellular distribution of fluorescein-[Ahx⁰]Ang-II, fluorescein-[Ahx⁰,Tyr(DMNB)⁴]Ang-II, fluorescein-[Ahx⁰]PACAP(28–38) and fluorescein-[Ahx⁰]TAT(48–60) upon confocal microscopy in live non-permeabilized HEK 293 cells. Cell Mask and DRAQ5 were used to delineate the plasma membrane and nuclei, respectively. B, intracellular fluorescence intensity (mean ± sem; ***P < 0.001). C, representative uptake efficiency of fluorescein-conjugated PACAP(28–38), TAT(48–60), Ang-II and [Tyr(DMNB)⁴]Ang-II in HEK 293 cells upon flow cytometry; quantified (D) with fluorescence-activated cell sorting analysis software FlowJo (mean ± sem; n = 3 per condition; *P < 0.05, ***P < 0.001).

Figure 8. Response of nucleoplasmic and cytosolic [Ca²⁺] to photolysis of intracellular [Tyr(DMNB)⁴]Ang-II

A, nucleoplasmic and cytosolic [Ca²⁺] recorded in canine cardiomyocytes before (t = 10 s) and after (t = 400 s) photolysis in cells loaded with vehicle (control), 20 nm [Tyr(DMNB)⁴]Ang-II (cAng-II), 20 nm cAng-II + 1 μm valsartan, 20 nm cAng-II + 100 μm 2-APB, 1 μm valsartan alone or extracellularly administered 20 nm Ang-II with or without 1 μm valsartan. Photolysis was induced by a pulse of UV-light (2–5 s, 70 μW) from a 405 nm/30 mW diode. Nucleoplasmic (B) and cytosolic (C) [Ca²⁺] recorded in ventricular cardiomyocytes before and after photolysis in cells preincubated with [Tyr(DMNB)⁴]Ang-II, vehicle (Ctl), valsartan, 2-APB along with non-photolysed [Tyr(DMNB)⁴]Ang-II or vehicle. Dotted bracket, UV-irradiation. DRAQ5 fluorescence was used to select the area corresponding to the nucleoplasm. Signals are presented as background-subtracted normalized fluorescence (%F/F₀), where F is the fluorescence intensity and F₀ is the resting fluorescence in the same cell prior to photolysis. D, [Ca²⁺] fluorescence in the nucleus and cytosol normalized to respective baseline (t = 10 s) values. Nuclear changes are indicated by dashed lines and cytosolic by continuous lines. E, for each condition (n = 8–12 cells), mean nuclear Fluo-4 fluorescence at baseline (t = 10 s; hatched bars) or after stimulation (t = 400 s) was quantified. Data are shown as the mean ± sem; ***P < 0.000. ns, non-significant.

Figure 9. Photolysis of intracellular [Tyr(DMNB)4]Ang-II regulates transcription

18S rRNA (A) and NF-κB mRNA (B) were quantified by quantitative PCR. A stimulation-control (scr) condition was performed where cardiomyocytes were loaded with a slowly-photolysing Ang-II analogue, Ang-II-ODMNB. Myocytes were incubated with [Tyr(DMNB)⁴]Ang-II or Ang-II-ODMNB (10 nm) for 30 min at room temperature. After incubation, cells were washed, placed on ice and exposed to a UV-lamp (30 W) for 1 min. In addition, cells were examined after treatment with buffer (Ctl) or extracellular Ang-II (Ang-II; 10 nm). Cells were then incubated at 37°C for 4 h prior to RNA extraction. Data are shown as the mean ± sem; *P < 0.05, **P < 0.01, ***P < 0.0001.