Development and evaluation of HDAC and Hsp90 PET ligands

ic-school-black-48dp-14
Category
Ph D Defense
Date
2019-07-01 17:00
Venue
KU Leuven, Aula van de Tweede Hoofdwet, 1.02 - Kasteelpark Arenberg 41
3001 Heverlee, Belgie

Promovendus/a: Koen Vermeulen

Promotor(en): Prof. dr. Guy Bormans

Positron emission tomography (PET) is a translational, highly sensitive, in vivo molecular imaging technique that visualizes the dynamic biodistribution of positron-emitting radiotracers. Specific high-affinity interaction of PET radiopharmaceuticals thereby allows quantification of the expression of their target biological molecules serving as in vivo biomarkers for molecular processes in health and disease. Other imaging modalities such as Magnetic resonance imaging (MRI) or X-ray imaging provide mostly high-resolution anatomic information but limited information on specific molecular events.

PET radiotracers can be radiolabelled with positron-emitting isotopes of biologically relevant elements such as C, N, O and F which are frequently found in biomolecules and medicinal compounds that exhibit high affinity interactions with disease relevant biomarkers. It is thus theoretically possible to radiolabel these molecules without compromising the affinity. The resulting radiotracer is administered for a PET scan in ‘tracer’ amounts, avoiding pharmacological or toxicological effects.

In vivo visualisation of disease relevant biomarkers with PET allows to select patients that are likely to respond to a specific therapy and closely monitor the effect of the disease treatment in-line with the “personalized medicine” approach. Aberrancies in cellular homeostasis reflected in altered expression of the biomarker can be detected with PET in considerably earlier stages, before morphological changes occur. Furthermore, PET can be applied in drug development to demonstrate in vivo pharmacodynamics of new drug candidates, demonstrate their target engagement and quantify their dose-occupancy relation.

Epigenetics is the study of the changes in gene expression that are unrelated to the gene DNA sequence itself. Post-translational modifications (PTMs) of histones play an important role as epigenetic modulators in chromatin remodelling and DNA transcription. One of the most studied epigenetic PTMs is the acetylation of evolutionarily conserved lysine residues located at the N-termini of histone tails, controlled by the opposing forces of histone acetyltransferase (HAT) and histone deacetylases (HDAC). An acetylated histone lysine leads to an uncondensed, active chromatin state, whereas the deacetylated histone lysine yields a higher degree of interaction between the negatively charged DNA and positively charged histone amino acid sequences, forming a condensed, mostly inactive, chromatin state. This forward and backward catalysis of acetylation contributes to the acetylated histone homeostasis which in turn influences the cellular homeostasis. Epigenetic modifications are thus determinant factors in the control of downstream gene expression and gene expression variability. HDAC activity is not limited to histones, as cytosolic proteins (heat shock proteins (Hsp), cortactin, α-tubulin) have acetyl binding regions and can also be targeted by HDACs. There are 18 HDACs described in literature, divided in 4 classes, each with distinct or overlapping molecular functions. Their role in cancer is well documented with additional findings in other diseases including neurodegenerative disorders and cardiac diseases.

Hsp90 is a predominantly cytosolic, ATP-dependent, molecular chaperone important for folding, maturation and clearance of over 400 proteins. In a stress-free environment, Hsp90 makes up 1-3% of all the soluble cellular proteins, whereas this fraction can increase to 6-10% in cancer cells as a result of cellular stress. Many of its client proteins are involved in oncogenic pathways and contribute to several hallmarks of cancer. Oncogenic Hsp90 adopts a tumour-specific conformation in multi-chaperone complexes with a 100-fold affinity increase for Hsp90 inhibitors. Further, high Hsp90 cell surface (eHsp90) levels are observed in different tumour cells (fibrosarcoma, melanoma, breast cancer) and this has been related to tumour metastasis. Over 15 Hsp90 inhibitors, targeting the ATP-binding pocket, have been evaluated in clinical trials. The role of Hsp90 in cancer is well documented in literature, however recent publications also inquire a role of Hsp90 in neurodegenerative diseases.

In our search for selective HDAC2 PET tracers, we evaluated in vitro and in silico activities of Santacruzamate A and candidate radiotracer derivatives thereof, reported as potent HDAC2-selective inhibitors. In silico docking studies where highly in line with in vitro affinity assays but marked discrepancies between reported activity values and the values obtained in our assays were observed for these compounds. To understand the structural basis of the activity of some of these inhibitors, molecular docking studies were performed to understand their interaction patterns and binding modes with HDAC2. Here, key π-π interactions in the HDAC2 binding pocket were observed with high affinity HDAC inhibitors (Martinostat, SAHA and TSA) which did not occur for Santacruzamate A and derivatives. It was further observed in affinity assays that reported IC50 values could not be reproduced invalidating Santacruzamate A and derivatives for further development as PET tracers for visualisation of HDAC2.

Further we evaluated [11C]KB631, as a selective and high-affinity HDAC6 tracer in melanoma. HDAC6 specific binding of this tracer was evaluated in in vitro autoradiography studies on tumour and brain sections and ex vivo biodistribution studies in mice. Further, a pre-treatement biodistribution, with KB631, HDAC6-selective inhibitor Ricolinostat and Pan-HDAC inhibitor SAHA, was performed to assess the specificity of [11C]KB631 binding in healthy and B16.F10 melanoma bearing mice. The same mouse model was also used in in vivo µPET studies. In this proof of concept study, [11C]KB631 showed HDAC6-specific binding in vitro to PC3 and B16.F10 melanoma tumour sections and brain sections. In control and blocking ex vivo biodistribution studies in healthy and tumour mice, low brain uptake of the tracer was observed. In vivo HDAC6-specific binding of [11C]KB631 was also found in different organs of interest. In µPET studies in B16.10 melanoma inoculated mice, rather low tumour uptake was observed, however this uptake was HDAC6-specific as it could be blocked by pre-treatment with Ricolinostat. Further research is needed to explore the full potential of [11C]KB631 to study the role of HDAC6 in vivo in cancer.

Brasca et al. reported the potent and selective Hsp90 inhibitor NMS-E973, which showed strong cytostatic effects in tumour cell lines. We developed and evaluated [11C]NMS-E973 as a PET tracer for in vivo visualization of Hsp90. [11C]NMS-E973 shows strong binding to B16.F10 melanoma cells that can be inhibited with PU-H71 (200 µM), a non-structurally related Hsp90 inhibitor. [11C]NMS-E973 in vitro autoradiography showed Hsp90-specific binding on tissue slices of murine B16.F10 melanoma, LNCaP and PC3 prostate cancer, SKOV-3 ovary carcinoma and muscle tissue. Ex vivo autoradiography experiments were also conducted on muscle (skeletal muscle and myocardium) and B16.F10 melanoma tissue. Remarkable discrepancies were observed compared to in vitro data as in in vitro studies tracer binding to muscle was severely increased compared to the in vivo binding visualized by ex vivo autoradiography. This indicates the occurrence of changes in Hsp90 behaviour in dying/hypoxic and dead tissue. In a µPET study, fast and persistent in vivo tumour binding of [11C]NMS-E973 in B16.F10 melanoma was observed. Pre-treatment of B16.F10 melanoma mice with PU-H71 or Ganetespib completely blocked tumour accumulation of [11C]NMS-E973 and confirmed in vivo Hsp90 binding specificity. Hsp90-specific binding of [11C]NMS-E973 was also observed in blood, lungs and spleen of tumour-bearing animals but not in control animals. This suggests that Hsp90 is shedded from the tumour and spreads to the whole body, but this hypothesis needs to be evaluated further.

Although NMS-E973 was reported to enter brain, we did not observe significant brain uptake of [11C]NMS-E973. Therefore other Hsp90 inhibitors with physicochemical characteristics compatible with efficient brain uptake were explored. In cooperation with researchers from Novartis we identified [11C]YC-72-AB85 as a potential tracer for in vivo visualisation of brain Hsp90. [11C]YC-72-AB85 shows Hsp90-specific and saturable tumour binding, similar to [11C]NMS-E973 in B16.F10 melanoma mice. Hsp90-specific tracer binding is however also observed in brain in vitro (autoradiography) in rats and mice and in vivo (µPET) in rats. Compared to baseline, a clear increase of [11C]YC-72-AB85 rat brain uptake was observed after blocking with the non-brain permeable and non-structurally related Hsp90 inhibitor Onalespib, induced by tracer displacement from peripheral binding sites resulting in a higher fraction of tracer available to bind brain Hsp90. Pre-treatment with Onalespib followed by i.v. injection of YC-72-AB85 or SNX-0723 resulted in tracer displacement from brain, indicating Hsp90-specific tracer binding in brain.

In conclusion, in the search for a selective HDAC2 PET tracer based on Santacruzamate A and derivatives we found remarked differences between our in silico and in vitro assays and reported literature concerning activity/affinity towards HDAC2. Hence, it was concluded that Santacruzamate A and derivatives could not be used as HDAC2 PET tracers. However, we developed and evaluated PET tracers, which enabled visualisation of HDAC6 and Hsp90 in oncological malignancies. Furthermore, [11C]YC-72-AB85 allows Hsp90 imaging in brain. All of these tracers were evaluated in in vitro autoradiography studies on tumour and/or rodent brain sections where they expressed specific binding. Specific binding was also observed for all tracers in vivo in a B16.F10 melanoma mouse model as assessed by blocking studies with HDAC6 or Hsp90 structural unrelated inhibitors. Further, elevated Hsp90 levels could be detected with both Hsp90 tracers in blood of tumour mice, which was not observed in healthy counterparts. These pre-clinical results indicate that these PET tracers show promise to be used to quantify HDAC6 or Hsp90 expression in vivo in tumours or brain and may allow quantification of target occupancy after treatment with HDAC6 or Hsp90 inhibitors in a clinical setting.
 
 

All Dates

  • 2019-07-01 17:00

Powered by iCagenda