PET CT static images of F CJ acquired at
PET/CT static images of F-CJ-042794 acquired at 1h p.i. from LNCaP tumour-bearing mice with/without co-injection of non-radioactive CJ-042794 (50μg) are shown in . The LNCaP tumours were clearly visualized in PET images with a moderate tumour-to-background contrast. High radioactivity accumulation was found in liver, kidneys and intestines, indicating that F-CJ-042794 was excreted through both renal and hepatobiliary pathways. Hearts (indicated by red arrows in the axial slices) and carotid hdac inhibitor (not shown) were also clearly visualized in PET images, suggesting that there was significant radioactivity retained in blood even at 1h p.i. There were no differences between the PET images of mice in baseline and blocked groups, which suggests that the uptake in LNCaP tumours was not mediated by the EP4 receptor.
The biodistribution data of F-CJ-042794 in mice at 1h p.i. are shown in , and are consistent with observations from dynamic and static PET images. High uptake was observed in blood (4.52±3.49%ID/g), intestines (37.5±12.5%ID/g), liver (6.71±0.75%ID/g) and kidneys (6.71±1.61%ID/g). The low uptake in bone (0.31±0.06%ID/g) indicates that F-CJ-042794 is stable against in vivo defluorination. The uptake in the heart without blood content was 1.58±0.40%ID/g, confirming the high heart uptake observed from PET images (, ) was due to radioactivity in blood. Modest uptake was observed in LNCaP tumour (1.12±0.06%ID/g). Due to an even lower uptake in muscle (0.41±0.05%ID/g), moderate tumour-to-muscle contrast ratio (2.73±0.22) was obtained resulting in the visualization of LNCaP tumours in PET images.
Consistent with the observations from PET images shown in , there were no significant differences (>0.05, Student’s -test) in uptake values for all collected tissues/organs shown in between baseline and blocked mice. The dosage of co-injected non-radioactive CJ-042794 used for the blocking study was 50μg, which was >1000-fold higher than the mass of F-CJ-042794 injected. However, no reduction in tumour uptake was observed in the blocked mice. Furthermore, several organs/tissues such as lung, heart, and blood cells (monocytes, macrophages and platelets) are all known to express the EP4 receptor., , , , , The lack of uptake reduction in LNCaP tumour and other collected organs/tissues in the blocked mice indicates that the uptake and retention of F-CJ-042794 were nonspecific, and not EP4 receptor mediated.
It is unclear at this stage what nonspecific mechanism(s) involved in the uptake of F-CJ-042794 into LNCaP tumours. However, our data emphasize the importance of specificity for the target of interest. Future work on the development of EP4-targeting PET tracers should focus on the use of highly specific pharmacophores such as CJ-023423 (). CJ-023423 is very selective (>200-fold) for the human EP4 receptor than other prostanoid receptors (EP1, EP2, EP3, FP, IP and TP). In addition, it has been reported by Nakao et al. that at 1µM concentration CJ-023423 did not exhibit any significant activity at all of >50 other major receptors tested.
In conclusion, we successfully synthesized and evaluated F-CJ-042794 as a PET imaging probe. Despite enabling visualization of LNCaP tumour xenografts in PET images, the uptake of F-CJ-042794 in tumours was nonspecific, precluding its application as a PET tracer for imaging EP4 receptor expression in cancer.
This work was supported in part by Prostate Cancer Canada and the RIX Family Leading Edge Foundation. The authors would like to thank Milan Vuckovic, Wade English and Baljit Singh for their technical assistance for 18F production.
Introduction In adult organisms the formation of new blood vessels from the pre-existing vasculature, or angiogenesis, is regulated by both activating and inhibiting factors, including prostanoids (Folkman, 1995, Liekens et al., 2001). Prostanoids are synthesized from the polyunsaturated fatty acid arachidonic acid by a series of metabolic pathways, including cyclooxygenases 1 or 2 (COX-1 or COX-2) and specific prostacyclin synthases, and are important in the regulation of a number of vascular processes, including angiogenesis (Alfranca et al., 2006). Vascular effects are mediated primarily by three prostanoids: prostaglandin E2 (PGE2), prostaglandin I2 (prostacyclin or PGI2) and thromboxane A2. In the vascular endothelium prostacyclin and PGE2 are the key prostanoids released during angiogenesis (Gately, 2000, Gately and Li, 2004) and PGE2 has also been reported to be an activator of angiogenesis (Nakanishi and Rosenberg, 2013, Wang and DuBois, 2004). PGE2 functions include increase in proliferation, angiogenesis, invasion and motility and the suppression of apoptosis of both tumor and endothelial cells (Wang and Dubois, 2006, Finetti et al., 2008, Salcedo et al., 2003). In addition, PGI2 functions in vasculature have also been studied extensively since its discovery in 1976, when it was identified as the major product of local arachidonic acid metabolism in vascular tissues (Moncada et al., 1976). Prostacyclin has many important functions including vasodilation (similar to endothelium-derived relaxing factor) and suppression of platelet aggregation (Shepherd and Katusic, 1991). Recently, prostacyclin – which is produced almost exclusively in the vascular endothelium – has been recognized for its participation in regenerative processes in the cardiovascular system, including angiogenesis and the repair of injured endothelium (Kawabe et al., 2010). Several studies have suggested that prostacyclin is important in angiogenesis. For example, vascular endothelial growth factor (VEGF) is a major activator of angiogenesis and also stimulates prostacyclin synthesis (He et al., 1999), and over-expression of the prostacyclin synthase gene induces angiogenesis in the mouse hind limb ischemia model (Hiraoka et al., 2003). Furthermore, the stable prostacyclin analogs SM-10902 and carbaprostacyclin induce angiogenesis and promote wound healing in animal models (Yamamoto et al., 1996, Liu et al., 2013).