Immune checkpoints, such as programmed death-ligand 1 (PD-L1), limit T-cell function and tumor cells use this ligand to escape the anti-tumor immune response. Treatments with monoclonal antibodies blocking these checkpoints have shown long-lasting responses, but only in a subset of patients. This study aims to develop a Nanobody (Nb)-based probe in order to assess human PD-L1 (hPD-L1) expression using positron emission tomography imaging, and to compare the influence of two different radiolabeling strategies, since the Nb has a lysine in its complementarity determining region (CDR), which may impact its affinity upon functionalization. The Nb has been conjugated with the NOTA chelator site-specifically via the Sortase-A enzyme or randomly on its lysines. [68Ga]Ga-NOTA-(hPD-L1) Nbs were obtained in >95% radiochemical purity. In vivo tumor targeting studies at 1 h 20 post-injection revealed specific tumor uptake of 1.89 ± 0.40%IA/g for the site-specific conjugate, 1.77 ± 0.29%IA/g for the random conjugate, no nonspecific organ targeting, and excretion via the kidneys and bladder. Both strategies allowed for easily obtaining 68Ga-labeled hPD-L1 Nbs in high yields. The two conjugates were stable and showed excellent in vivo targeting. Moreover, we proved that the random lysine-conjugation is a valid strategy for clinical translation of the hPD-L1 Nb, despite the lysine present in the CDR.

Prolyl hydroxylase domain-containing proteins (PHDs) regulate the adaptation of cells to hypoxia. Pan-hydroxylase inhibition is protective in experimental colitis, in which PHD1 plays a prominent role. However, it is currently unknown how PHD1 targeting regulates this protection and which cell type(s) are involved. Here, we demonstrated that Phd1 deletion in endothelial and haematopoietic cells (Phd1 Tie2:cre) protected mice from dextran sulphate sodium (DSS)-induced colitis, with reduced epithelial erosions, immune cell infiltration, and colonic microvascular dysfunction, whereas the response of Phd2 Tie2:cre and Phd3 Tie2:cre mice to DSS was similar to that of their littermate controls. Using bone marrow chimeras and cell-specific cre mice, we demonstrated that ablation of Phd1 in haematopoietic cells but not in endothelial cells was both necessary and sufficient to inhibit experimental colitis. This effect relied, at least in part, on skewing of Phd1-deficient bone marrow-derived macrophages towards an anti-inflammatory M2 phenotype. These cells showed an attenuated nuclear factor-κB-dependent response to lipopolysaccharide (LPS), which in turn diminished endothelial chemokine expression. In addition, Phd1 deficiency in dendritic cells significantly reduced interleukin-1β production in response to LPS. Taken together, our results further support the development of selective PHD1 inhibitors for ulcerative colitis, and identify haematopoietic cells as their primary target. Copyright © 2016 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.

Hepatic transport of 99m Tc-mebrofenin through organic anion transport protein 1a and 1b (Oatp1a/1b) and multidrug resistance protein 2 (Mrp2) was investigated by small-animal SPECT. On the basis of the results, a noninvasive method to visualize and quantify disturbances in hepatic transport is proposed. Methods: Friend virus B wild-type mice (untreated, bile ductligated, vehicle-or rifampicin-treated) and strain-matched knockout mice unable to express the uptake transporters Oatp1a/1b (Slco1a/1b 2/2 / 2/2 ) or the efflux transporter Mrp2 (Abcc2 2/2 ) were intravenously injected with 99m Tc-mebrofenin (n 5 3 per group). After dynamic small-animal SPECT and short CT acquisitions, time-activity curves of the liver and of the gallbladder and intestines were obtained and correlated with direct blood samples. Results: Normal hepatobiliary clearance of 99m Tcmebrofenin was severely impaired in the bile duct-ligated animal, as evidenced by elevated hepatic tracer levels. In Slco1a/ 1b 2/2 / 2/2 mice, a lower area under the curve (AUC) for the liver (P 5 0.014) was obtained and no activity was detected in the gallbladder and intestines. Renal rerouting was observed, along with an increase in the blood AUC (P 5 0.01). Abcc2 2/2 mice had a higher liver AUC (P 5 0.009), a delayed emergence time of 99m Tc-mebrofenin in the gallbladder (P 5 0.009), and a lower AUC for the gallbladder and intestines (P 5 0.001). The blood curve was similar to that of wild-type mice. 99m Tc-mebrofenin disposition was altered after rifampicin treatments. We observed a dose-dependent delayed time point at which tracer maximized in liver, an increased AUC for liver, and a lower AUC for gallbladder and intestines (P 5 0.042, 0.034, and 0.001, respectively, highest dose). Emergence in the gallbladder occurred later (P 5 0.009, highest dose), and blood AUC was higher (P 5 0.006). Conclusion: The current study visualized and quantified hepatic uptake and biliary efflux of 99m Tc-mebrofenin. Our results demonstrated the possibility of discriminating, on a quantitative level, between lack of functional activity of sinusoidal uptake versus that of biliary efflux transporters.

IntroductionDrug-induced cholestasis is a liver disorder that might be caused by interference of drugs with the hepatobiliary bile acid transporters. It is important to identify this interference early on in drug development. In this work, Positron Emission Tomography (PET)-imaging with a 18F labeled bile acid analogue was introduced to detect disturbed hepatobiliary transport of bile acids.Methods3β-[18F]fluorocholic acid ([18F]FCA) was prepared by nucleophilic substitution of a mesylated precursor with [18F]fluoride, followed by deprotection with sodium hydroxide. Transport of [18F]FCA was assessed in vitro using CHO-NTCP, HEK-OATP1B1, HEK-OATP1B3 transfected cells and BSEP & MRP2 membrane vesicles. Investigation of [18F]FCA metabolites was performed with primary mouse hepatocytes. Hepatobiliary transport of [18F]FCA was evaluated in vivo in wild-type, rifampicin and bosentan pretreated FVB-mice by dynamic μPET scanning.ResultsRadiosynthesis of [18F]FCA was achieved in a moderate radiochemical yield (8.11 ± 1.94%; non-decay corrected; n = 10) and high radiochemical purity (>99%). FCA was transported by the basolateral bile acid uptake transporters NTCP, OATP1B1 and OATP1B3. For canalicular efflux, BSEP and MRP2 are the relevant bile acid transporters. [18F]FCA was found to be metabolically stable. In vivo, [18F]FCA showed fast hepatic uptake (4.5 ± 0.5 min to reach 71.8 ± 1.2% maximum % ID) and subsequent efflux to the gallbladder and intestines (93.3 ± 6.0% ID after 1 hour). Hepatobiliary transport of [18F]FCA was significantly inhibited by both rifampicin and bosentan.ConclusionA 18F labeled bile acid analogue, [18F]FCA, has been developed that shows transport by NTCP, OATP, MRP2 and BSEP. [18F]FCA can be used as a probe to monitor disturbed hepatobiliary transport in vivo and accumulation of bile acids in blood and liver during drug development.

Since atherosclerotic plaques are small and sparse, their non-invasive detection via PET imaging requires both highly specific radiotracers as well as imaging systems with high sensitivity and resolution. This study aimed to assess the targeting and biodistribution of a novel fluorine-18 anti-VCAM-1 Nanobody (Nb), and to investigate whether sub-millimetre resolution PET imaging could improve detectability of plaques in mice. The anti-VCAM-1 Nb functionalised with the novel restrained complexing agent (RESCA) chelator was labelled with [18F]AlF with a high radiochemical yield (>75%) and radiochemical purity (>99%). Subsequently, [18F]AlF(RESCA)-cAbVCAM1-5 was injected in ApoE−/− mice, or co-injected with excess of unlabelled Nb (control group). Mice were imaged sequentially using a cross-over design on two different commercially available PET/CT systems and finally sacrificed for ex vivo analysis. Both the PET/CT images and ex vivo data showed specific uptake of [18F]AlF(RESCA)-cAbVCAM1-5 in atherosclerotic lesions. Non-specific bone uptake was also noticeable, most probably due to in vivo defluorination. Image analysis yielded higher target-to-heart and target-to-brain ratios with the β-CUBE (MOLECUBES) PET scanner, demonstrating that preclinical detection of atherosclerotic lesions could be improved using the latest PET technology.

No abstract

This first-in-dog study evaluates the use of the PET-radioligand [11C]DASB to image the density and availability of the serotonin transporter (SERT) in the canine brain. Imaging the serotonergic system could improve diagnosis and therapy of multiple canine behavioural disorders. Furthermore, as many similarities are reported between several human neuropsychiatric conditions and naturally occurring canine behavioural disorders, making this tracer available for use in dogs also provide researchers an interesting non-primate animal model to investigate human disorders. Five adult beagles underwent a 90 minutes dynamic PET scan and arterial whole blood was sampled throughout the scan. For each ROI, the distribution volume (VT), obtained via the one- and two- tissue compartment model (1-TC, 2-TC) and the Logan Plot, was calculated and the goodness-of-fit was evaluated by the Akaike Information Criterion (AIC). For the preferred compartmental model BPND values were estimated and compared with those derived by four reference tissue models: 4-parameter RTM, SRTM2, MRTM2 and the Logan reference tissue model. The 2-TC model indicated in 61% of the ROIs a better fit compared to the 1-TC model. The Logan plot produced almost identical VT values and can be used as an alternative. Compared with the 2-TC model, all investigated reference tissue models showed high correlations but small underestimations of the BPND-parameter. The highest correlation was achieved with the Logan reference tissue model (Y = 0.9266 x + 0.0257; R2 = 0.9722). Therefore, this model can be put forward as a non-invasive standard model for future PET-experiments with [11C]DASB in dogs.

BackgroundPrevious studies showed that radiolabeled murine monoclonal antibody (mAb) 14C5 and its Fab and F(ab')2 fragments, targeting αvβ5 integrin, have promising properties for diagnostic and therapeutic applications in cancer. To diminish the risk of generating a human anti-mouse antibody response in patients, chimeric variants were created. The purpose of this study was to recombinantly produce chimeric antibody (chAb) derivatives of the murine mAb 14C5 and to evaluate the in vitro and in vivo characteristics.MethodsIn vitro stability, specificity, and affinity of radioiodinated chAb and fragments (Iodo-Gen method) were examined on high-expressing αvβ5 A549 lung tumor cells. In vivo biodistribution and pharmacokinetic characteristics were studied in A549 lung tumor-bearing Swiss Nu/Nu mice.ResultsSaturation binding experiments revealed high in vitro affinity of radioiodinated chAb, F(ab')2, and Fab, with dissociation constants (KD) of 1.19 ± 0.19, 0.68 ± 0.10, and 2.11 ± 0.58 nM, respectively. ChAb 14C5 showed highest tumor uptake (approximately 10%ID/g) at 24 h post injection, corresponding with other high-affinity Abs. ChF(ab')2 and chFab fragments showed faster clearance from the blood compared to the intact Ab.ConclusionsThe chimerization of mAb 14C5 and its fragments has no or negligible effect on the properties of the antibody. In vitro and in vivo properties show that the chAb 14C5 is promising for radioimmunotherapy, due to its high maximum tumor uptake and its long retention in the tumor. The chF(ab')2 fragment shows a similar receptor affinity and a faster blood clearance, causing less non-specific retention than the chAb. Due to their fast blood clearance, the fragments show high potential for radioimmunodiagnosis.