Introduction

Targeted molecular imaging agents are useful tools to understand protein expression patterns and molecular mechanisms of disease. For example, the clinical MRI agent gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid (Gd-EOB-DTPA) is used to image human liver cancer via its selective uptake primarily by OATP1B3 transport into healthy hepatocytes, but not liver cancer cells [1]. Gd-EOB-DTPA has additionally been established as a sensitive probe for cells engineered to express rat OATP1A1 or human OATP1B3, allowing for in vivo MRI tracking of these cells [2,3,4]. However, Gd is associated with nephrogenic systemic fibrosis in patients with renal disease, and Gd deposition has been found in the central nervous system, leading to increased efforts towards development and use of Gd-free agents [5]. Manganese(III) porphyrin constructs offer a safe, sensitive, and versatile platform for molecular MRI, but little work has been performed to target these agents towards specific biomarkers [6]. We hypothesize that addition of a phenyloxyethyl (PhOEt) group to manganese(III) porphyrin would facilitate its uptake by cells expressing OATP1A1 and OATP1B3 transporters.

Methods

Manganese(III) 5-ethoxybenzyl-10, 15, 20-tri(carboxyl) porphyrin (MnTriCP-PhOEt) was synthesized (Figure 1). UV-vis spectra at λ ∈ (300, 700) nm for MnTriCP-PhOEt and its precursors, and nuclear magnetic relaxation dispersion (NMRD) profiles at B₀ ∈ (0, 1) T, 1 mM, 37.0°C for MnTriCP-PhOEt, Gd-EOB-DTPA and Gd-DTPA were acquired. Spin-lattice relaxation rates (R₁) of 1 mM solutions were also measured at 1.5T and 3T (Figure 2). Next, cell lines (rat C6, human MDA-MB-231) were transduced with lentivirus to express rat Oatp1a1 and human Oatp1b3, respectively (Figure 3). Non-transduced control and transduced cells were each incubated with 1 mM MnTriCP-PhOEt, Gd-EOB-DTPA or Gd-DTPA for 60 minutes, washed, and R₁ relaxation rates of cell pellets were measured at 3T. Mice were administered 0.025 mmol/kg MnTriCP-PhOEt (n=3) via intraperitoneal injection and T₁-weighted imaging was performed at 3T for 60 minutes (Figure 4, Figure 5). An additional set of mice received intraperitoneal injections of 0.025 mmol/kg MnTriCP-PhOEt or an equivalent volume of saline (n=3), were sacrificed 44 minutes post-injection, and organs were harvested for relaxometry at B₀ ∈ (0.01, 1) T, 37.0°C.

Results

MnTriCP-PhOEt exhibited increased R₁ rates relative to Gd-EOB-DTPA at B₀ > 0.1005 T (Figure 2B). The R₁ rate of MnTriCP-PhOEt was 1.38-fold greater than Gd-EOB-DTPA at 1.5T (Figure 2B). At 3T, MnTriCP-PhOEt exhibited an increased R₁ rate (10.55 ± 0.4151 Hz) relative to Gd-EOB-DTPA (8.327 ± 0.3745 Hz) and Gd-DTPA (6.295 ± 0.3902 Hz) (Figure 2C). None of the transporter-expressing cells incubated with Gd-DTPA exhibited increased R₁ rates relative to treated control cells (Figure 3E, 3F). Oatp1a1- and Oatp1b3-expressing cells exhibited significantly increased R₁ rates following incubation with Gd-EOB-DTPA (2.082 ± 0.125, 1.548 ± 0.113 Hz), relative to controls not expressing these transporters (0.6933 ± 0.1329 Hz). Oatp1a1- and Oatp1b3-expressing cells also exhibited significantly increased R₁ rates following MnTriCP-PhOEt incubation (2.457 ± 0.198, 2.087 ± 0.132 Hz, respectively), relative to treated control cells (0.7675 ± 0.1466 Hz) (Figure 3E, 3F). Importantly, our data show increased R₁ values for transporter-expressing cells treated with MnTriCP-PhOEt, relative to the same cells treated with an equivalent concentration of Gd-EOB-DTPA, consistent with each agent’s relaxivity. Mice imaged in vivo exhibited liver uptake of MnTriCP-PhOEt over time, averaging a maximum 1.914-fold significant increase in liver signal intensity at 3T after a 0.025 mmol/kg dose, relative to pre-contrast images (n=3, p<0.05) (Figure 4A, 4B, 4C). When T₁ times of ex vivo organs were measured via relaxometry, liver tissue from mice administered 0.025 mmol/kg MnTriCP-PhOEt demonstrated significantly smaller T₁ times at field strengths above 0.3057 T, relative to liver tissue from mice administered saline (n=3, p<0.05), whereas no significant differences were observed in T₁ times of kidney and muscle tissue (Figure 4D, 4E).

Discussion

The manganese(III) porphyrin platform has much to offer for the field of molecular MR. Importantly, the planar geometry of the porphyrin construct allows for water coordination both above and below the plane of the molecule (q=2), thereby increasing its relaxivity relative to comparable constructs with access to only one water hydration site (Figure 1B, Figure 2B, Figure 2C). Further, the porphyrin offers four locations onto which different functional groups may be added to alter important properties such as circulation half-life and ligand activity. In this study, we show that molecular modification of manganese(III) porphyrin via addition of a PhOEt functional group enables transportation of this agent into cells through OATP1A1 and OATP1B3 transporters, with resultant relaxivities greater than cells treated with equivalent concentrations of Gd-EOB-DTPA. In animals, when 0.025 mmol/kg MnTriCP-PhOEt was administered, significant signal enhancement and T₁ shortening of liver tissue were observed, whereas previous manganese(III) porphyrin contrast agents were reported to be largely cleared through the kidneys [7]. Altogether, our research poses the possibility of replacing Gd-EOB-DTPA with MnTriCP-PhOEt, which exhibits a higher relaxivity at clinical field strengths for diagnostic purposes, and may be especially useful for patients with impaired renal function in which Gd-based contrast agents may be contraindicated. Future work will be focused on performing quantitative kinetic imaging to maximize MnTriCP-PhOEt detection at low doses in vivo.

Acknowledgements

No acknowledgement found.