Soluble fusion proteins between single transmembrane photoreceptor guanylyl cyclases and their activators

Abstract

Among single-spanning transmembrane receptors (sTMRs), two guanylyl cyclase receptors, GC1 and GC2, are critically important during phototransduction in vertebrate retinal photoreceptor cells. Ca(2+)-free forms of guanylyl cyclase-activating proteins (GCAPs) stimulate GCs intracellularly by a molecular mechanism that is not fully understood. To gain further insight into the mechanism of activation and specificity among these proteins, for the first time, several soluble and active truncated GCs and fusion proteins between intracellular domains of GCs and full-length GCAPs were generated. The GC activity of myristoylated GCAP--(437-1054)GC displayed typical [Ca(2+)] dependence, and was further enhanced by ATP and inhibited by guanylyl cyclase inhibitor protein (GCIP). The myristoyl group of GCAP1 appeared to be critical for the inhibition of GCs at high [Ca(2+)], even without membranes. In contrast, calmodulin (CaM)--(437-1054)GC1 fusion protein was inactive, but could be stimulated by exogenous GCAP1. In a series of experiments, we showed that the activation of GCs by linked GCAPs involved intra- and intermolecular mechanisms. The catalytically productive GCAP1--(437-1054)GC1 complex can dissociate, allowing binding and stimulation of the GC1 fusion protein by free GCAP1. This suggests that the intramolecular interactions within the fusion protein have low affinity and are mimicking the native system. We present evidence that the mechanism of GC activation by GCAPs involves a dimeric form of GCs, involves direct interaction between GCs and GCAPs, and does not require membrane components. Thus, fusion proteins may provide an important advance for further structural studies of photoreceptor GCs and other sTMRs with and without different forms of regulatory proteins.

Figure 1

Schematic representation of truncated GCs and CBP–GC fusion proteins. The scheme represents Ext (extracellular), TM (transmembrane), KHD (kinase-homology), DD (dimerization), and Cat (catalytic) domains of GCs. The GC activity was assayed in the whole cell homogenate and in the detergent-free extracts in the presence of GCAP1 at 46 nM [Ca²⁺]_free.

Figure 2

Activity of GC constructs as a function of [Ca²⁺]_free. (A) Activity of GC for full-length GC1 and truncated ^437–1054GC1 (enzyme was stimulated by addition of exogenous, recombinant GCAP1). IC₅₀ for GC1 was 230 ± 40 and 225 ± 34 nM for ^437–1054GC1 (n = 3). (B) Activity of GC for GCAP1–^437–1054GC1 and His₆-GCAP1–^437–1054GC1 (arrow shows competition of GCIP with GCAP1 in GCAP1–^437–1054GC1 fusion protein at low [Ca²⁺]_free). IC₅₀ for His₆-GCAP1–^437–1054GC1 was 220 ± 34 nM. Inset: ATP titration of GC constructs (solid circles, GCAP1–^437–1054GC1; open circles, ^437–1054GC1 + GCAP1; open triangles, GC1 + GCAP1) in low [Ca²⁺]free. Similar results were obtained in three independent measurements. (C) Activity of GC for GCAP1–^437–1054GC2, GCAP2–^437–1054GC1, and GCAP2–^437–1054GC2. IC₅₀ values for GCAP2–^437–1054GC1 and GCAP2–^437–1054GC1 were 245 ± 74 and 255 ± 30 nM, respectively. (D) Activity of GC for CaM–^437–1054GC1 after stimulation by addition of GCAP1. Arrow shows recovery of GC1 activity by reconstitution with GCAP1 in low [Ca²⁺]_free. IC₅₀ for CaM–^437–1054GC1 was 195 ± 15 nM. The dashed line represents stimulation of ROS GCs by GCAP1. In all panels, the dashed line represents stimulation of ROS GCs by GCAP1.

Figure 3

Properties of GCAP1-GC1 fusion proteins in cell culture, biochemical assays, and protein purification. (A) Solubility of GC1 constructs. GC was prepared as described under Materials and Methods. Activity (as a percent of maximal activity) was measured in membrane and soluble cell fractions. Similar results were obtained in four independent measurements. Inset: Immunoblot analysis of GC1 solubility using polyclonal anti-GC1 antibodies. (B) Localization of GCAP1–^437–1054GC1 fusion protein in cytoplasm (1) and full-length GC1 in the membrane fraction (2) (upper panels corresponded to image in bright light, bottom panels show fluorescent image). (C) Gel filtration chromatography of GCAP1–^437–1054GC1 and His₆-GCAP1–^437–1054GC1 (Materials and Methods) and correlation with GC stimulation activity. Panel 1, GC activity profile; panel 2, incorporation of the myristoyl group into GCAP1–GC1 mutants. (D) Purification of PPE–GCAP1–^437–1054GC1. The GC1 fusion protein was purified as described under Materials and Methods. The purity was assessed by SDS–PAGE (lane 1) and by immunoblotting using anti-GC1 antibody (lane 2) and GCAP1 antibody (lane 3). The standards are (in kDa): phosphorylase B, 116; BSA, 80; ovalbumin, 52.5; carbonic anhydrase, 34.9; soybean trypsin inhibitor, 29.9; lysosyme, 21.8.

Figure 4

Intramolecular and dimeric nature of GCAP1–GC1-stimulated activity at low [Ca²⁺]_free. (A) Competition of GCAP1–^437–1054GC1 and ^437–1054GC1 with GCAP1(E75Q E111Q E155Q) at 2 μM [Ca²⁺]_free. The EC₅₀ values were 0.8 ± 0.1 and 0.85 ± 0.1 μM for GCAP1–^437–1054GC1 and ^437–1054GC1, respectively. (B) Dilution effect on GCAP1–^437–1054GC1 and GCAP1/ROS GC1 activities. Samples were diluted 0, 2, 4, and 8 times with increasing concentrations of all reagents in the assay at constant amounts of protein. Results are an average of two measurements. (C) Gel filtration chromatography. GCAP1–^437–1054GC1 was loaded at 5 μM [Ca²⁺]_free or with (gray circles and dashed line) addition of 1 mM EDTA (solid circles) on a Superose-6 column as described under Materials and Methods. Arrows show standards: α-amylase (200 kDa) and alcohol dehydrogenase (150 kDa). (D) The calibration curve for the gel filtration column. The standard proteins and compounds used were NaN₃, cytochrome c (12 kDa), carbonic anhydrase (29 kDa), bovine serum albumin (67 kDa), alcohol dehydrogenase (150 kDa), α-amylase (200 kDa), and blue dextran (void volume, 2000 kDa) as the high molecular mass standard. The protein standards were detected at 280 nm, blue dextran at 450 nm, and azide at 260 nm. ve/vo is a ratio of the elution volume to void volume. The GC fusion protein eluted at a volume that corresponded to 180 kDa. Similar relationships for the retention time and molecular mass were obtained in three independent experiments.