Development and Characterization of an 18F-labeled Ghrelin Peptidomimetic for Imaging the Cardiac Growth Hormone Secretagogue Receptor

Abstract

One-third of patients with heart disease develop heart failure, which is diagnosed through imaging and detection of circulating biomarkers. Imaging strategies reveal morphologic and functional changes but fall short of detecting molecular abnormalities that can lead to heart failure, and circulating biomarkers are not cardiac specific. Thus, there is critical need for biomarkers that are endogenous to myocardial tissues. The cardiac growth hormone secretagogue receptor 1a (GHSR1a), which binds the hormone ghrelin, is a potential biomarker for heart failure. We have synthesized and characterized a novel ghrelin peptidomimetic tracer, an ¹⁸F-labeled analogue of G-7039, for positron emission tomography (PET) imaging of cardiac GHSR1a. In vitro analysis showed enhanced serum stability compared to natural ghrelin and significantly increased cellular uptake in GHSR1a-expressing OVCAR cells. Biodistribution studies in mice showed that tissue uptake of the tracer was independent of circulating ghrelin levels, and there was negligible cardiac uptake and high uptake in the liver, intestines, and kidneys. Specificity of tracer uptake was assessed using ghsr ^-/- mice; both static and dynamic PET imaging revealed no difference in cardiac uptake, and there was no significant correlation between cardiac standardized uptake values and GHSR1a expression. Our study lays the groundwork for further refinement of peptidomimetic PET tracers targeting cardiac GHSR1a.

Keywords: GHSR1a; PET; biomarkers; cardiac; ghrelin; heart disease; heart failure; imaging.

Figure 1.

Radiosynthesis of [1-Nal⁴, Lys⁵(4-[¹⁸F]-FB)]G-7039. Reagents and conditions: (A) (i) ¹⁸F^-, K₂CO₃, Krypofix 222, DMSO, 120°C, 8 minutes, (ii) HCl (5 mol/L), 100°C, 3 minutes; (B) NHS, EDC, acetonitrile, RT, 15 minutes; (C) DIPEA, acetonitrile/h₂O, 85°C, 15 minutes.

Figure 2.

Chemical Structure of [1-Nal⁴, Lys⁵(4-[¹⁸F]-FB)]G-7039.

Figure 3.

Cell Uptake of [1-Nal⁴, Lys⁵(4-[¹⁸F]-FB)]G-7039. Both wild-type (wt) OVCAR and OVCAR/GHSR1a cells were incubated at the indicated time points with 0.1 to 0.5 MBq of [1-Nal⁴, Lys⁵(4-¹⁸FB)]G-7039. Values are means ± standard error of the mean (SEM; n = 6). *(P < .05; P = .0357) between transfected and wt for 20 minutes uptake and **(p<0.01; p = 0.0061) between transfected and wt for 60 minutes uptake.

Figure 4.

Biodistribution profile of [1-Nal4, Lys5(4-[18F]-FB)]G-7039. Fasted (A) and fed (B) female C57BL/6 mice were injected with 5.23-26.4 MBq [1-Nal⁴, Lys⁵(4-[¹⁸F]-FB)]G-7039 and killed 1, 2, and 4 hours postinjection. Insets show values from organs with low uptake on a more appropriate y axis. All values are means ± standard error of the mean (SEM). A, *P < .05 in liver at 2 hours versus all tissues (except spleen), ****P < .0001 in urine at 2 hours versus all other tissues. (B) *P < .05 in liver at 1 hour versus all tissues (except spleen, tail, and urine) and in urine at 1 hour versus all tissues (except liver), **P < .01 in Urine at 2 hours versus all tissues.

Figure 5.

Plasma concentration of ghrelin, GLP-1, glucagon, and insulin in fasted and fed mice. Blood plasma concentrations of (A) ghrelin, (B) glucagon-like peptide-1 (GLP-1), (C) glucagon, and (D) insulin in fasted and fed mice killed 1, 2, and 4 hours postinjection of [1-Nal⁴, Lys⁵(4-[¹⁸F]-FB)]G-7039. Values are shown as mean concentration in pg/mL ± standard error of the mean (SEM).

Figure 6.

Metabolic profiles in cardiac tissue from fasted and fed mice. Levels of GHSR1a, ghrelin, and the indicated metabolic markers were assessed in fasted and fed mice killed 1 hour postinjection by fluorescence microscopy. Values are given in raw integrated density (Arbitrary Units) of fluorescence. Each point represents average fluorescence intensity from 1 mouse. The middle line represents the mean value and the bars indicate standard error of the mean (SEM). *P < .05, **P < .01

Figure 7.

Positron emission tomography computed tomography (PET-CT) imaging in wild-type (wt) and ghsr⁻⁻ Mice. Top Panel: Uptake of [1-Nal⁴, Lys⁵(4-[¹⁸F]-FB)]G-7039 in wt (top) and ghsr^−/− (bottom) mice. All frames were integrated over a 5-minute time period except the 90-minute frame that was integrated over 30 minutes. Coronal slices through the mouse body at the section of liver are shown with view from anterior. The arrows indicated regions where volumes of interest (VOIs) were drawn. Bottom Panel: standardized uptake values (SUVs) Calculated from VOIs of spillover corrected heart (A), lung (B), and abdomen (C) 60 minutes postinjection and over a 30-minute time frame. Each point represents the average SUV calculated from an individual mouse VOI. No differences between wt and ghsr^−/− mice were observed in any VOIs.

Figure 8.

Time-activity curves for [1-Nal⁴, Lys⁵(4-[¹⁸F]-FB)]G-7039 uptake in wild-type (wt) and ghsr^−/− Mice. Spillover corrected heart (A), lung (B), and abdomen (C) uptake of [1-Nal⁴, Lys⁵(4-[¹⁸F]-FB)]G-7039. Values are means ± standard error of the mean (SEM; n = 6 for wt and n = 6 for ghsr^−/−). All significant differences were only within groups. (A) *P < .05 and ^# P < .05 at 5 and 10 minutes versus all time points (except 15 minutes). (B) *P < .05 and ^## P < .01 at 5 minutes versus all time points. (C) **P < .01 and ^#### P < .0001 at 5 minutes versus all time points.

Figure 9.

Correlations of heart SUV and plasma ghrelin with cardiac GHSR1a expression. (A) Static standardized uptake values (SUVs) from volumes of interest (VOIs) assigned to heart in wild-type (wt; n = 5) and ghsr^−/− (n = 6) mice plotted as a function of GHSR1a expression. (B) Static SUVs from VOIs assigned to heart in wt (n=5) and ghsr^−/− (n = 6) mice plotted as a function of plasma ghrelin concentrations. Each point represents a value from an individual mouse.