Final Report Summary - LUPAS (Luminescent polymers for in vivo imaging of amyloid signatures)
Executive summary:
The objective of LUPAS was to develop novel molecular agents and methods for use within diagnostic imaging of Alzheimer's disease (AD) and prion diseases. Our concept relied on reporter molecules based on luminescent conjugated polymers, luminescent conjugated polythiophenes (LCPs). The LCP molecules target pathogenic protein aggregates associated with these diseases and LCPs target these lesions with high selectivity and specificity. The objective was also to investigate potential therapeutic potency of the LCP imaging agents.
The work was organised in several interdependent layers of experimentation:
- synthesis of LCPs and nanoparticles;
- synthesis of pure amyloidotic molecular targets;
- assay development - fluorescence spectroscopy and MRI;
- development of disease models;
- assay of LCPs in tissue sections;
- in vivo imaging - two-photon (2P) fluorescence and MRI;
- therapeutic potential in prion model systems;
- therapeutic potential in vivo in AD models.
LUPAS was focused on providing new knowledge in chemical design and smart targeting of misfolded protein amyloid lesions in AD and prion diseases. The LUPAS partners were assembled from a wide range of areas ranging from experts within organic synthetic chemistry, synthetic nanochemistry, amyloid structure, prion disease AD, MRI, multi-photon physics and hyper spectral imaging. This group formed the critical mass of competences needed to reach the project's ambitious goals.
The main project outcomes affords visualisation of dynamics and biochemical activity of pathological processes of AD and prion diseases in real-time from the molecular level to the organ and full body scale. Most importantly, novel molecular insight regarding the pathological hallmarks of these diseases and the specific chemical requirements for optimal amyloid ligands was concluded.
Dissemination activities included scientific article publications, conference presentations, amyloid workshop for the pathology community and sample shipments to two dozens of labs worldwide.
Project context and objectives:
The European population is aging. As a consequence age-related neurodegenerative diseases such as AD are on a steady rise reaching epidemic proportions of serious concern to the afflicted patients and their relatives. These horrific diseases put a heavy burden on society including a decade of heavy dependence on caregivers culminating with care at nursing homes. If this epidemic is not sequestrated it is projected to threaten the core of the well-fare societies of all European nations. The culprit within these diseases is accumulated misfolded protein molecules, called amyloids, which corrupt normal cellular functions and cause disease. Amyloids are self-perpetuating fibrillar states, which can hide for decades within affected individuals and replicate with immense efficiency, especially in the elderly. We, as a community, still stand short in delivering definite diagnosis or benign disease-modifying treatment or cure to any neurodegenerative disease. The only way to fight these diseases is long-term focused high quality research. Within the LUPAS project we developed novel molecular agents and methods for use within diagnostic imaging that relies on reporter molecules based on luminescent conjugated polymers, LCPs. The LCP molecules target the pathogenic protein aggregates with high selectivity and specificity. Candidate lead molecules have also been shown to have therapeutic potential.
The LUPAS partners were assembled from a wide range of areas ranging from experts within organic synthetic chemistry, synthetic nanochemistry, amyloid structure, prion disease AD, MRI, multi-photon physics and hyper spectral imaging. This group formed the critical mass of competences needed to reach the project's ambitious goals. The partner organisations in the LUPAS consortium are: Linköping University (Sweden), Université Claude Bernard Lyon 1 (France), University of Tübingen (Germany), Norwegian University of Science and Technology (Norway), Zürich University Hospital (Switzerland), Charité Universitätsmedizin Berlin (Germany), Applied Spectral Imaging (Israel) and Genovis AB (Sweden). Most importantly the project was organised according to 7 work packages (WPs) that defined the work plan. The LUPAS project was described by 5 technical WPs aimed at developing novel optical and MRI (MRI) imaging agents, towards the objectives of proving smart tools for biomedical research, for prediction, diagnosis, monitoring and prognosis of diseases, and for support and guidance of therapeutic interventions within the field of chronic neurodegenerative diseases.
- WP1 entailed the management of the consortium.
- WP2 generated LCP molecular reagents, reaction protocols and defined amyloidotic molecular targets (AMTs) for implementation in WP5, WP6 and for grafting onto a magnetic nanoparticle (MNP) for MRI in WP3.
- WP3 generated novel imaging agents for MRI.
- WP4 constituted an arena for advanced optical spectroscopic characterisation and nuclear magnetic resonance (NMR) imaging of LCP or LCP-MNP stained material from WPs 2-3 and WPs 5-6.
- WP5 was dedicated to investigation of AD transgenic mouse and human AD material for LCP and LCP-MNP diagnosis, amyloid biomarkers, and experimental therapeutic intervention.
- WP6 was dedicated to investigation of prion diseases from mouse and human material for LCP and LCP-MNP diagnosis, amyloid biomarkers, and experimental therapeutic intervention.
- WP7 was dedicated to the execution of dissemination and exploitation activities of project results. Specifically, the LUPAS project will make available the new smart imaging agents for clinical use with external non-profit stakeholders such as university hospitals and related research groups.
LUPAS was a three-year focused research project under the Seventh Framework Programme (FP7). Significant achievements were reached which is well reflected by a surge of interest from the research community to use the LCP technology. Other indications of success are the high citation rate of publications from the LUPAS research groups. By combining the diversity of scientific knowledge available within the consortium, progress towards the development of multimodal amyloid ligands for non-invasive imaging was achieved. The outcome affords visualisation of dynamics and biochemical activity of pathological processes of AD and prion diseases in real-time from the molecular level to the organ and full body scale. Most importantly, novel molecular insight regarding the pathological hallmarks of these diseases and the specific chemical requirements for optimal amyloid ligands was concluded. One of the major tasks within LUPAS was to develop novel LCPs that can be utilised as tools for selective identification of protein deposits, the pathological hallmarks of AD and prion diseases. LCPs are fascinating molecules that behave like structural chameleons. When an LCP binds to its amyloid target, the LCP changes shape. The LCP thereby also alters its colour depending on the amyloid structure. Hence, it reports different outcome signals which can be detected. From a library of LCPs, we identified lead compounds with distinct chemical groups for detection and spectral assignment of protein deposits in AD and prion diseases. Amyloid structures are vastly heterogeneous. It is not known which structures are most relevant for disease. A number of LCPs displayed spectral discrimination of amyloid structures both from the same protein (prion protein (PrP), amyloid-beta (Abeta)) as well as for discrimination of different amyloids within the same patient (Abeta, Tau). In addition, we have shown that pre-fibrillar aggregates of PrP, and Abeta (oligomers) that goes undetectable by conventional amyloid ligands are readily detectable by LCPs. The ability of LCPs to discriminate between different structures of fibrillar and pre-fibrillar aggregates is of great importance for AD. These features render the LCPs superior to other small molecule probes for molecular diagnostics. Importantly, these LCPs can also be used for in vivo imaging of protein deposits in the brain of living animals. Hence, LCPs that are selective amyloid ligands and even capable of crossing the blood brain barrier were developed within the LUPAS project. Synthesis of LCPs were scaled up into gram scale amounts and are now being implemented within a variety of subprojects within LUPAS partner labs and within collaborations all over Europe and worldwide. Multiphoton imaging is a preferable technique for studying protein aggregation diseases in real time in transgenic mouse models and we have shown that the unique optical properties of LCPs make these dyes highly efficient for multiphoton in vivo imaging. Several LCPs cross the blood brain barrier which enables specific labelling of amyloid plaques (Abeta) and intraneuronal neurofibrillary tangles (Tau) in the parenchyma of the brain, as well as Abeta amyloid in blood vessels so called cerebral amyloid angiopathy. In vivo imaging of protein deposits in living mice was demonstrated using distinct LCPs and these LCP scaffolds are well tolerated by mice even during longer time periods and repetitive injections. The latter is imperative, because the final goal of LUPAS is to achieve an LCP-based agent for non-invasive clinical diagnostics of AD and prion diseases. From a diagnostic perspective, we have employed the LCPs on post-mortem tissue sections from patients with neurodegenerative diseases. The LCP selectively stain the protein deposits and expose these major pathological hallmarks of either AD or prion diseased brain. The identification can easily be done and distinguished due to the colour emitted from the LCP bound to the different entities. Hence, the LCP technique shows excellent promise for being implemented as a complementary technique in routine clinical diagnostics. Work towards fluorescence detection of aggregated protein as biomarkers in cerebral spinal fluid of AD and control patients has been performed, but results so far have not been conclusive. Optical probes are not the preferable agents for non-invasive imaging in humans, due to the limitations of optical imaging. In this regard, the LUPAS consortium aimed to broaden the scope to develop novel multimodal LCPs that can be used for both optical imaging and MRI. MRI is today a standard technique for imaging of pathological conditions in humans, but subtle pathological changes in tissue can be hard to detect due to limited effect on the MR signal. Contrast agents can be used to enhance the MR image contrast related to tissue pathology. Today only unspecific contrast agents are in use, but targeted contrast agents based on paramagnetic and superparamagnetic nanoparticles and nanocomplexes (MNPs) hold great promise for MRI. Within LUPAS we have synthesised a variety of nanoparticles for enhanced T1 and T2 relaxation dispersion potentially useful as contrast agents for MRI. In addition, the first prototype of a LCP-MNP conjugate can specifically target amyloid in vitro, in tissue samples, and promising results have also been obtained in vivo in transgenic mice.
Disease modifying treatments for neurodegenerative diseases are scarce and non-existent for prion diseases. Because the LCP molecules show selective binding towards the culprit - the aggregated proteins their potential as therapeutic agents against prion diseases was evaluated within LUPAS. Work on tissue sections and cell cultures show efficient LCP mediated reduction in prion infectivity as a result of unexpected stabilisation of prion aggregates. This mechanistic insight opens up completely new avenues for targeting both prion disease and other neurodegenerative diseases based on hyperstabilisation rather than clearance of the misfolded protein culprits.
Project results:
Monomeric and trimeric precursor thiophene compounds with distinct side functionalities, including a variety of amino acid derivatives, were synthesised. In the initial screen we included two molecules with ethylamine sidechains on thiophene rings 2 and 4 (5006, 5007), however, as indicated in earlier studies we focused on, anionic LCPs containing the monomer 3-thiophene acetic acid which have shown the best selectivity and specificity towards AMTs and protein aggregates in tissue under close to physiological conditions. Therefore our efforts in were focused on developing anionic LCPs with carboxylic acid functionalisation. From a selection of building blocks, we have utilised Suzuki coupling to synthesise a library of chemically defined oligothiophenes, ranging from tetrameric to heptameric thiophenes, having amine or carboxylic acid functionalities at distinct positions along the thiophene backbone. The synthetic routes for some of the LCPs are rather straightforward and have been optimised to render up to gram quantities of individual LCPs.
Early on in LUPAS we published a detailed protocol on the synthesis of p-FTAA. The straightforward synthesis of p-FTAA, including two Suzuki couplings makes this a highly tractable molecule for many labs with adequate synthetic expertise.
The method was published on the Nature Protocols website: 'Synthesis of a pentathiophene fluorescent probe, 4',3'''-bis-carboxymethyl-(2,2';5',2'';5'',2''';5''',2'''']quinquethiophene-5,5''' by Andreas Åslund, K. Peter Nilsson & Peter Konradsson Protocol Exchange doi:10.1038/nprot.2010.24 Nature Protocols website Community contributed.
During LUPAS more than 20 novel anionic LCPs were synthesised whereof several have been synthesised in gram scales. The LCPs have been used throughout the project and have been implemented at all partner sites.
Multifunctional amyloid specific LCP-nanoagents for MRI and fluorescence
The concept of enhancing the detection methodology of amyloid towards in vivo imaging using MRI requires paramagnetic species as contrast agents. Within LUPAS the amyloid specific anchor was based on LCP molecules. For MNP covalent attachment, asymmetrical variants of the anionic LCPs have also been synthesised. These amyloid specific ligands will anchor the MNP to amyloid plaque. A variety of spacer chemistry both in terms of chain length and composition was developed in LUPAS.
Nano synthesis
Nanoparticles are interesting platforms for such application since it concentrates magnetic ions in a limited volume and is susceptible to bear functional groups. In this WP the investigation on several types of nanomaterials based on Fe, Gd or Mn are performed. The influence of the chemical composition, size, and shape of the particles on the relaxivities and NMR properties was checked.
Nanoparticles for relaxivity enhancement in MRI
MRI is a non-invasive method that allows for longitudinal imaging of animals and patients. In dedicated animal MR-scanners, the desired organ or body part, for example the brain in rats and mice can be imaged in vivo with high quality and resolution. Spatial resolution in MRI is dependent on the signal to noise ratio (SNR) and the two main factors are image acquisition duration and magnetic field strength. When the object is put into the MR scanner, all the protons within the object act as small magnets and align along the magnetic field inside the MR-scanner spinning around the axis of the magnetic field. A short excitation pulse is then applied, shifting the alignment of the proton spins out of this direction. After the excitation pulse comes to an end, a signal is recorded in a coil. The signal created by the proton spins decays thereafter by two relaxation processes (called T1 and T2 relaxation) as the protons relax back to their original alignment. T1 and T2 vary with the tissue type, and by weighting the images by the T1 and T2 contrast can be created in the MR images. Commonly used contrast agents such as gadolinium or ironoxide complexes can be used to alter the relaxation properties of the tissue, leading to increased or reduced signal from the tissue where the contrast agents are present. Within LUPAS the relaxivity of 20 different particles have been measured on different field strength and 6 different conjugated MNP-LCPs have been measured at 7T. The relaxivity measurements indicate that the main properties of the particles are kept or enhanced after conjugation to LCP with different linkers. This is due to that most of the promising particles have high r2 relaxivity, and increased size will increase r2.
LCP-MNP retains amyloid fibril specificity
That the selective property of amyloid targeting was retained in LCP-MNP complexes was assessed by both fluorescence and electron microscopy of in vitro generated AMTs from Abeta 1-42.
Development of AMTs of Abeta and PrP for LCP screening
For screening of suitable photophysical LCP properties and binding parameters of LCPs we employed AMTs generated in vitro by recombinant protein preparations. Analyses of preparations were made both after reaching equilibrium and during kinetic measurements. The AMTs used were based on Abeta preparations and PrP employing quiescent and shaking conditions as well as different Abeta 1-40 and Abeta 1-42 mixtures in defined ratios. The spectroscopic screening with LCPs with AMT material was used for optimal settings for microscopy as well as for selection of reference spectra for tissue sections. From the kinetic screen we discovered that p-FTAA was a highly suitable molecule for generating LCPs sensitive for non-mature (ThT negative) prefibrillar amyloid, both from Abeta and human PrP as well as other proteins (Figure 6). This finding was published in Biochemistry 2010. 'A fluorescent pentameric thiophene derivative detects in vitro-formed prefibrillar protein aggregates' Hammarström P, Simon R, Nyström S, Konradsson P, Aslund A, Nilsson KP. Biochemistry. 2010 Aug 17;49(32):6838-45.
A focused library of LCPs followed this initial lead and allowed us to conclude that this feature of detection of prefibrillar aggregates is shared by anionic LCPs which should have a backbone consisting of five to seven thiophene units and carboxyl groups extending the conjugated thiophene backbone (Klingstedt et al. 2011). Importantly the carboxyl moiety extending the conjugation of the thiophene backbone also allowed for spectral discrimination of Abeta and Tau within human tissue samples. The library design and conclusions from the AMT screen was published in Organic and Biomolecular Chemistry 2011. 'Synthesis of a library of oligothiophenes and their utilisation as fluorescent ligands for spectral assignment of protein aggregates'. Klingstedt T, Aslund A, Simon RA, Johansson LB, Mason JJ, Nyström S, Hammarström P, Nilsson KP. Org Biomol Chem. 2011 Dec 21;9(24):8356-70.
Development of AMTs for human PrP instead of chemical preparations also entailed protein engineering. Within LUPAS we investigated the molecular basis for the dependency for misfolding of PrP as a consequence of variations in position 129. The 129 site is well known to be highly important for disease penetrance and clinical presentation, being either M (wt) or V (variant). Our data based on the in vitro behaviour of PrP in solution, showed that position 129 was an interesting site for recruitment of PrP into the amyloid state. Charge mutations (M129E and M129K) showed decreased spontaneous conversion propensity (nucleation) compared to the wild type. We also observed that M129C was totally resistant towards spontaneous engulfment of PrP into the amyloid fold, most likely due to the formation of an intermolecular disulfide bridge forming in the lag phase, rendering a non-convertible disulfide linked dimer effectuated by the 129CC interaction. In contrast to the rather large differences in fibrillation propensity found for extreme amino acid substitutions such as tryptophan, proline, cysteine and charge, quite minor changes were observed for conservative mutations M129L, M129A and as expected M129V. From these data we have formulated a molecular platform for experimental AMT assessment of e.g. synthetic prions and targets for LCP screening. This study was published in The Journal of Biological Chemistry 2012. 'Multiple substitutions of methionine 129 in human PrP reveal its importance in the amyloid fibrillation pathway' Nyström S, Mishra R, Hornemann S, Aguzzi A, Nilsson KP, Hammarström P. J Biol Chem. 2012 Jul 27;287(31):25975-84.
Use of LCPs for animal and human samples
LCPs with molecular scaffolds consisting of repetitive thiophene moieties have evolved as an interesting class of fluorescent probes, since this class of molecules offers the possibility to achieve an optical read-out of biomolecular interactions through their impact on the conformation and the geometry of the LCP backbone. These structurally induced optical changes of the LCP backbone also allow the tantalising possibility to detect conformational changes of aggregates, rendering opportunities to gain more information concerning the morphology of the protein deposits. The use of LCPs have been reviewed both for versatile and detailed usage by LUPAS researchers in Methods in Molecular Biology, 2012, 'A pentameric luminescent-conjugated oligothiophene for optical imaging of in vitro-formed amyloid fibrils and protein aggregates in tissue sections'. Nilsson KP, Lindgren M, Hammarström P. Methods Mol Biol. 2012;849:425-34 and Biochemical Society transactions 2012, 'Luminescent conjugated poly- and oligo-thiophenes: optical ligands for spectral assignment of a plethora of protein aggregates'. Klingstedt T, Nilsson KP. Biochem Soc Trans. 2012 Aug;40(4):704-10.
LCPs for AD
It is currently recognised that amyloid deposits in vivo are vastly heterogenous and appear polymorphic. LCPs have provided the opportunity to gain more information concerning the morphology of the protein deposits, and a greater understanding of the conformational phenotype encoded in distinct protein aggregates should be achieved. Instead of merely observing the total amount of protein aggregates, the LCPs allow spectral finger printing of heterogenic population of specific protein aggregates and these ligands have been efficient for studying protein aggregates in transgenic mouse models having AD Aß pathology. This was one of the core aims for LUPAS. Tissue sections mainly from mouse brain were herein stained with a variety of LCP probes and were used to assay amyloid binding, specificity and spectral features, mainly using different microscopy based techniques. Several amyloid pathological hallmarks of disease were studied in AD mouse models APP / PS1 and APP23 mice aged 18 months. Most studied LCPs identified amyloid plaque and cerebral amyloid angiopathy (CAA). The robustness of staining was also verified after prolonged storage of tissue sections (in the dark) by revisiting the sections after 2 years.
Compatibility of LCPs with other imaging modalities
Importantly the LCP molecules are highly compatible with new imaging technologies. In one study the LCP imaging tracer pFTAA was used for combined imaging using fluorescence tracing of amyloid plaque and time-of-flight secondary ion mass spectrometry (TOF-SIMS) was used to investigate cholesterol accumulations originating from surrounding astrocytes proximal to plaque. This study was recently published in Acta Neuropathologica 2013, 'Localisation of cholesterol, amyloid and glia in AD transgenic mouse brain tissue using TOF-SIMS) and immunofluorescence imaging' Solé-Domènech S, Sjövall P, Vukojevic V, Fernando R, Codita A, Salve S, Bogdanovic N, Mohammed AH, Hammarström P, Nilsson KP, LaFerla FM, Jacob S, Berggren PO, Giménez-Llort L, Schalling M, Terenius L, Johansson B. Acta Neuropathol. 2013 Jan;125(1):145-57.
In vivo imaging
Repetitive in vivo imaging in mice is an important tool for studying dynamic changes in structure and function of the brain. A methodology by LUPAS researchers describes a head fixation system, which allows rapid re-localisation of previously imaged regions of interest (ROIs) within the brain. Such ROIs can be automatically relocated and imaged over weeks to months with negligible rotational change and only minor translational errors. Previously stored imaging positions can be fully automated re-localised within a few seconds. This automated rapid and accurate relocation simplifies image acquisition and post-processing in longitudinal imaging experiments. The risk of laser induced phototoxicity is greatly reduced, because the laser is only used for data acquisition and not for finding previously imaged ROIs. Thus, here described head fixation device appears well suited for in vivo repetitive long-term imaging in mouse brain. The initial work of the above mentioned technology has been used to study Abeta-amyloid plaque formation and growth over time. Plaque growth in APP / PS1 were shown to grow radially by 0.3 µm per week (Hefendehl et al., 2011). The tracer used in these experiments were the methoxy-X04 dye developed by the Klunk laboratory (Klunk WE et al. J Neuropathol Exp Neurol. 2002 61(9):797-805), but can also be performed using LCPs. These developments and discoveries were published in two papers in Journal of neuroscience and Journal of Neuroscience Methods. 'Long-term in vivo imaging of beta-amyloid plaque appearance and growth in a mouse model of cerebral beta-amyloidosis', Hefendehl JK, Wegenast-Braun BM, Liebig C, Eicke D, Milford D, Calhoun ME, Kohsaka S, Eichner M, Jucker M. J Neurosci. 2011 Jan 12;31(2):624-9; 'Repeatable target localisation for long-term in vivo imaging of mice with 2-photon microscopy', Hefendehl JK, Milford D, Eicke D, Wegenast-Braun BM, Calhoun ME, Grathwohl SA, Jucker M, Liebig C. J Neurosci Methods. 2012 Apr 15;205(2):357-63.
In vivo imaging using LCPs
Initial experiments from LUPAS researchers revealed that LCPs pFTAA (5001) and pHTAA (5002) showed efficient penetrance of the BBB and readily stained amyloid plaque in APP / PS1 mice (Åslund 2009). Following this success within LUPAS we employed the larger molecule hFTAA (7001), which has a more red shifted fluorescence spectrum and equal or superior sentitivity than pFTAA. The LCP hFTAA passed the blood-brain barrier after systemic administration and was specifically bound to extracellular beta-amyloid deposits in the brain parenchyma (Abeta plaques) and in the vasculature (cerebral beta-amyloid angiopathy) of ß-amyloid precursor protein transgenic APP23 mice. Interestingly, peripheral application of hFTAA also stained intracellular lesions of hyperphosphorylated Tau protein in P301S Tau transgenic mice. In addition hyperspectral imaging of all three amyloid types was acquired ex vivo using two-photon excitation. Within this work an algorithm was developed to demonstrate the LCP ability to distinguish between two amyloid lesions and the background. hFTAA revealed a distinct shift in its emission spectra when bound to Abeta plaques versus Tau lesions. A spectral difference was also observed for Abeta plaques versus cerebral beta-amyloid angiopathy, indicating that different amyloid types and structural variances of a specific amyloid type can be distinguished. Hence, by adding spectral signatures to amyloid lesions, our results pave the way for a new biophotonic era of in vivo amyloid imaging, allowing in vivo differentiation of amyloid polymorphisms and monitoring changes of their structure / composition over time.
This study was recently published in American Journal of Pathology 2012, 'Spectral discrimination of cerebral amyloid lesions after peripheral application of luminescent conjugated oligothiophenes', Wegenast-Braun BM, Skodras A, Bayraktar G, Mahler J, Fritschi SK, Klingstedt T, Mason JJ, Hammarström P, Nilsson KP, Liebig C, Jucker M. Am J Pathol. 2012 Dec;181(6):1953-60.
LCP-MNP retain amyloid plaque specificity ex vivo
The developed LCP-MNP for MRI imaging was initially found to retain amyloid specificity against AMTs in vitro. A more complex sample such as tissue samples could post more challenging. Nevertheless, LCP-MNPs were highly fluorescent and showed specific binding and strong fluorescence when bound to amyloid plaque in tissue sections from transgenic mice. The specificity and selectivity in these samples were independent of attachment to MNPs reflecting the same fluorescence profiles as the corresponding LCP-spacer adducts. MRI data acquisition in vivo in APP / PS1 mice proved more challenging than in vivo fluorescence. Several application procedures and LCP-MNP contrast agents showed small amounts of penetrance of LCP-MNPs over the BBB following intravenous (i.v.) inoculations. Better success was achieved with direct administration of the contrast agents into the CSF.
Development of hyperspectral correlation analysis of multicolor, autofluorescence and LCP stained amyloid plaques
Multiphoton imaging is a preferable technique for studying protein aggregation diseases in real time in transgenic mouse models and within LUPAS it was further evident that the unique optical properties of LCPs make these dyes highly efficient for multiphoton in vivo imaging. In certain aspects there nevertheless appear problematic with autofluorescence. Green autofluorescence from amyloid plaques can be a problem in weakly stained samples, such as samples that have been i.v. injected at low levels e.g. LCP-MNP probes at low levels of LCP loading < 30 µM. Also, to analyse multi-staining experiments, one needs special attention to assess the spectral content. Researchers within LUPAS therefore developed a special algorithm that can accurately calculate the spectral similarity of a measured spectrum with a reference spectrum. The method is based on correlation, where the measured spectrum (X) of each pixel in a hyperspectral image is correlated with a reference spectrum (Y), using the following formula Where Xi is the i:th wavelength point and the bar over X denotes the average of the spectrum (and similar for Y). This makes it possible to give a mathematical number in the range of 0 to 1 where 1 is a measured spectrum identical to the reference. The formula is robust to background noise and does not suffer from the same problems as other linear spectral unmixing algorithms. The method of spectral correlation was recently published in 'Hyperspectral analysis using the correlation between image and reference'. PG Ellingsen, NK Reitan, BD Pedersen, M Lindgren J. Biomed. Opt. 2013. 18 (2), 020501.
Human AD diagnostics
Human brain tissue samples of sporadic AD patients were analysed within LUPAS. Tissues were screened with the anionic LCP library and from these studies LCPs that distinguish the two main pathological hallmarks of AD, A-beta plaques and neurofibrillary tangles (NFTs). The LCPs, pFTAA, hxFTAA and hFTAA, give a distinct spectroscopic signature from Abeta or tau aggregates. Notably these LCPs were five to seven membered oligothiophenes containing two additional carboxylic moieties terminating the conjugated backbone. These data were published in Organic and Biomolecular Chemistry 2011.
Therapeutics
LCPs as pharmacophores in AD mouse models
One aim with LUPAS was to test the potency of LCPs for possible pharmacological effects in transgenic mice with AD pathology. We tested LCOs as possible pharmocophores in vivo because pFTAA is known to target pre-fibrillar amyloid species, thereby possibly generating minor alterations in the molecular structure of amyloid fibrils that may interfere with amyloid fibril formation and their putative toxicity. It was demonstrated that p-FTAA (LCP 5001) is able to cross the blood brain barrier after i.v. and intraperitoneal (i.p.) injection and bind to cerebral amyloid plaques of APPPS1 mice, an early onset mouse model of cerebral amyloidosis. Repetitive injections of p-FTAA resulted in a robust staining of cerebral amyloid plaques in APPPS1 mice, and probes remained bound to plaques for up to 6 months. APPPS1 mice were treated for 12 weeks by weekly i.p. p-FTAA-injections and Abeta pathology and the general health was studied. Despite efficient binding of p-FTAA to cerebral amyloid, no significant p-FTAA dependent alteration in total brain amyloid burden, as judged by Congo Red staining and Abeta-immunohistochemistry was observed. These data indicate that persistent peripheral p-FTAA treatment neither causes obvious signs of toxicity, nor alters cerebral amyloid load in APPPS1 mice. To study whether continuous p-FTAA treatment beyond morphologically detectable alterations may modify amyloid toxicity on a functional level, long term potentiation and cognition in p-FTAA-treated APPPS1 mice will need to be assessed. This study was presented at the Foundation Ipsen conference in Paris February 2012.
S. Handrick, S. Prokop, S. Nystrom, J.J. Mason, N.R. Reitan, B.M. Wegenast-Braun, M. Lindgren, M. Jucker, K.P. Nilsson, P. Hammarstrom, F.L. Heppner Luminescent Conjugated Oligothiophenes (LCOs) as Pharmacophores in AD. Fondation Ipsen, Paris (2012)
Using LCP fluorescence as a follow up during treatment
For several years, the natural compound curcumin has been proposed to be a candidate for enhanced clearance of toxic Aß amyloid. Herein LUPAS researchers studied the potency of feeding curcumin as a drug candidate to alleviate Abeta toxicity in transgenic Drosophila. The longevity as well as the locomotor activity of five different AD model genotypes, measured relative to a control line, showed up to 75 % improved lifespan and activity for curcumin fed flies. In contrast to the majority of studies of curcumin effects on amyloid we did not observe any decrease in the amount of Abeta deposition following curcumin treatment. Importantly, conformation-dependent spectra from p-FTAA, LCP (5001) bound to Abeta deposits in different Drosophila genotypes over time, indicated accelerated pre-fibrillar to fibril conversion of Abeta (1-42) in curcumin treated flies. Our study shows that curcumin promotes amyloid fibril conversion by reducing the pre-fibrillar / oligomeric species of Abeta, resulting in a reduced neurotoxicity in Drosophila. Most importantly for LUPAS was that LCP fluorescence is an important tool for assessing conformational differences in treated versus untreated animals.
This study was published in PLoS One February 2012 Caesar I, Jonson M, Nilsson KP, Thor S, Hammarström P. Curcumin promotes A-beta fibrillation and reduces neurotoxicity in transgenic Drosophila. PLoS One. 2012;7(2):e31424. doi: 10.1371/journal.pone.0031424. This publication has been viewed online > 5 600 times and was featured in > 20 news articles worldwide.
Inflammatory modulation for new treatment strategies
Work within LUPAS also included assessment of alternative treatment strategies. The pathology of AD has an inflammatory component that is characterized by upregulation of proinflammatory cytokines, particularly in response to Abeta. The APPPS1 AD mouse model showed increased production of the common interleukin-12 (IL-12) and IL-23 subunit p40 by microglia. Genetic ablation of the IL-12/IL-23 signalling molecules by deficiency of p40 or its receptor complex, resulted in decreased cerebral amyloid load. In addition, peripheral administration of a neutralising p40-specific antibody also resulted in a reduction of cerebral amyloid load in APPPS1 mice. Furthermore, intracerebroventricular delivery of antibodies to p40 significantly reduced the concentration of soluble Abeta species and reversed cognitive deficits in aged APPPS1 mice. The concentration of p40 was also increased in the cerebrospinal fluid of subjects with AD, which suggests that inhibition of the IL-12/IL-23 pathway may attenuate AD pathology and cognitive deficits. The paper 'Inhibition of IL-12/IL-23 signalling reduces AD-like pathology and cognitive decline' by Vom Berg J, Prokop S, Miller KR, Obst J, Kälin RE, Lopategui-Cabezas I, Wegner A, Mair F, Schipke CG, Peters O, Winter Y, Becher B, Heppner FL.. Nat Med. 2012 Dec;18(12):1812-9 was recently published. This paper was highlighted in Nature Reviews Neuroscience 2013, Nature Reviews Drug Discovery 2013, and Nature Reviews Immunology 2013, New England Journal of Medicine, 2013. Follow-up of these mouse models by LCP fluorescence should add another level of analysis not only limited to plaque load but also LCP spectroscopy can delineate conformational differences within plaque. Such studies have so far been limited by the studied age groups of 4 and 8 months. Large differences in LCP spectroscopic differences in APP / PS1 mice occur after the age of 12 months (unpublished results), hence the time window needs to be adjusted for such studies to be meaningful.
Prions
Prion diseases or transmissible spongiform encephalopathy (TSE) have received unprecedented public interest despite its rareness in the human population. The main reason for this interest was the epidemic of bovine spongiform encephalopathy (mad cow disease) some 20 years ago and its evident transmissibility from cattle to humans. Prions can reside silent for decades in an infected individual and pose a threat to the human population due to its efficiency in jumping species barriers and through blood transfusions and transplants. Prion diseases are hence of outmost interest not only for basic research curiosity but rather it is imperative to possess sensitive detection and therapeutic strategies. The culprit protein in all prion diseases is called the PrP. The PrP is most abundant in mammalian neurons but is ubiquitously expressed throughout various cells and tissues. The functional role of native PrP is not fully understood. PrP is associated with a number of different prionoses: sporadic, inherited, and acquired, all of which are invariably fatal. The common molecular pathognomonic marker for prionoses is the presence of severe vacuolation within the CNS rendering a sponge-like tissue. Concomitant with the presence of spongiosis is the presence of a conformational isoform of PrP, which has converted from a largely helical globular protein cellular PrP (PrPC) that misfolds into an aggregation prone beta-sheet conformation, PrPSc, which often assemble into protein deposits with conspicuous similarities to amyloid.
The concept of prions as a novel pathogenic entity composed solely of corrupt protein has been extensively debated in several important review articles by LUPAS researchers. LUPAS researchers recently published 'Prion propagation, toxicity and degradation' by A. Aguzzi and J. Falsig in nature Neuroscience 2012. Nat Neurosci. 2012 Jun 26;15(7):936-9, to address this topic.
Development of prion assays
Lack of cytotoxicity in prion infected cell cultures has hampered mechanistic studies of prion-induced neurodegeneration. Recently, LUPAS researchers showed that prion-infected cultured organotypic cerebellar slices (COCS) experienced progressive spongiform neurodegeneration closely reproducing prion disease, with three different prion strains giving rise to three distinct patterns of PrP deposition. Neurodegeneration did not occur when PrP was genetically removed from neurons, and a pharmacological screen indicated that neurodegeneration was abrogated by compounds known to antagonise prion replication. Prion infection of COCS and mice led to enhanced fodrin cleavage, suggesting the involvement of calpains or caspases in pathogenesis. Accordingly, neurotoxicity and fodrin cleavage were prevented by calpain inhibitors but not by caspase inhibitors, whereas prion replication proceeded unimpeded. The article was published in PLoS Pathogens in 2012: 'Prion pathogenesis is faithfully reproduced in cerebellar organotypic slice cultures'. Falsig J, Sonati T, Herrmann US, Saban D, Li B, Arroyo K, Ballmer B, Liberski PP, Aguzzi A. PLoS Pathog. 2012 Nov;8(11):e1002985.
Therapeutics of prion diseases
Prion diseases in rodents faithfully reproduce the human disease, in contrast to AD in mice, which only show subtle pathophysiological phenotypes. This makes prion disease models a powerful assay for novel therapeutics that targets the culprit disease causing misfolding protein. LCPs interact with ordered protein aggregates and sensitively detect amyloids of many different proteins, suggesting that they may also possess antiprion properties. 8 different LCPs were analysed with a variety of side chain and chain lengths for antiprion activity. A variety of anionic, cationic, and zwitterionic LCPs reduced the infectivity of prion-containing brain homogenates and of prion-infected cerebellar organotypic cultured slices (COCS) and decreased the amount of scrapie isoform of PrPC (PrPSc) oligomers that could be captured in an avidity assay. Paradoxically, treatment enhanced the resistance of PrPSc to proteolysis, triggered the compaction, and enhanced the resistance to proteolysis of recombinant mouse PrP (23-231) fibres. These results suggest that LCPs act as antiprion agents by transitioning PrP aggregates into structures with reduced frangibility. Moreover, enzyme-linked immunosorbent assay (ELISA) on COCS and in vitro conversion assays with mouse PrP (23-231) indicated that poly(thiophene-3-acetic acid) (LCP 10001) may additionally interfere with the generation of PrPSc by stabilising the conformation of PrPC or of a transition intermediate. Therefore, LCPs represent a novel class of antiprion agents whose mode of action appears to rely on hyperstabilisation, rather than destabilisation, of PrPSc deposits.
Diagnostics of human prion diseases
From a diagnostic perspective, the LCPs / LCOs were also used to analyse post-mortem tissue sections from patients with neurodegenerative diseases. The fluorescence stains selectively bind to the protein deposits and expose these major pathological hallmarks of either AD or prion diseased brain. PrP deposits from Creutzfeldt-Jakob disease are notoriously complicated to visualise using conventional fluorescence dyes for amyloid such as Congo red and Thioflavin T. Several LCPs function well in this regard as was tested extensively on mouse material. To employ frozen material is especially important in this context. Within LUPAS this complicated procedure from a biosafety perspective was achieved. When this hurdle was overcome identification of PrP deposition in TSE brain can easily be done and distinguished due to the colour emitted from the LCP / LCO bound to the different entities for three distinct prion diseases. Hence, the LCP / LCO technique shows excellent promise for being implemented as a complementary technique in clinical diagnostics of TSEs.
Potential impact:
The European population is aging. As a consequence age-related neurodegenerative diseases such as AD are on a steady rise reaching epidemic proportions of serious concern to the afflicted patients and their relatives. These diseases put a heavy burden on society including a decade of profound dependence on caregivers culminating with care at nursing homes. If this epidemic is not sequestrated it is projected to threaten the core of the well-fare societies of all European nations.
The most thorough data on the ongoing and forthcoming Alzheimer epidemic comes from the Alzheimer Association is the USA (see http://www.alz.org online for further details). European national and transnational initiatives (see Alzheimer Europe http://www.alzheimer-europe.org online for further details) are emerging initiatives to tackle the rising incidence of dementias. In 2010 the European Commission (EC) funded the joint programming initiative on AD and neurodegenerative diseases (JPND) with a European Union (EU) contribution of close to EUR 2 million. From the Alzheimer Europe initiative it reads that 'In 2005, the total direct and informal care costs of AD and other dementias were estimated at EUR 130 billion in the EU-27 (EUR 21 000 per patient); 56 % of these costs was informal care. The most common forms of dementia in the EU are AD (about 70 % of cases), and vascular dementia (less than 30 %)'. These are staggering figures and are projected to rise.
A patient is diagnosed with AD mainly through the primary care where changes in short term memory and personality is based on cognitive tests. The cognitive tests are combined with MRI scans to rule out brain tumours or stroke as a primary cause for the symptoms. More developed centres allows for cerebrospinal fluid testing for biomarkers such as increased amounts of Tau and phosphorylated Tau and decreased amounts of soluble Abeta42. The culprit within these diseases is accumulated misfolded protein molecules, called amyloids of e.g. the mentioned biomarkers, which corrupt normal cellular functions and cause disease. Amyloids are self-perpetuating fibrillar states, which can hide for decades within affected individuals and replicate with immense efficiency, especially in the elderly. We, as a community, still stand short in delivering definite diagnosis or benign disease-modifying treatment or cure to any neurodegenerative disease. The only way to fight these diseases is long-term focused high quality research.
It is highly encouraging that the EC distributes increasing amounts of funding for tackling these diseases and hence work towards novel ideas on how to make disease modifying therapies and rapid diagnostics and follow up on treatments are ongoing in projects such as the described LUPAS project.
The LUPAS project is a testament to progress in cross-disciplinary research with a very impressive output of scientific impact. Over the years LUPAS researchers have been met with enthusiasm from the scientific community as feedback on published papers and conference presentations. LCP molecules have been sent to more than two dozen labs worldwide and a dozen collaborations have been initiated.
Final outcome and future perspectives
There is a tremendous need for quantitative diagnostic methods for early detection and evaluation of neurodegenerative disorders, such as AD and prion diseases. The need is underlined by the recent development of proposed therapeutically interventions targeting disease, so called disease modifiers, including immune therapy. Herein, quantitative physical outcome measures are urgently needed in terms of amyloid pathology within living patients.
Within the brief three-year time frame of LUPAS, the consortium developed important research tools for use in disease model systems (mouse models) in vivo and on histological ex vivo samples from humans. The LUPAS project exemplifies that broad cross-disciplinary expertise within diverse subjects can realise unmet needs within biomedicine. The realistic prognosis is that it will take a few more years to validate this technology in the preclinical phase prior to exploiting it in the clinic and hence lead to:
- improved diagnostics of AD;
- improved diagnostics of prion disease;
- new avenues for therapeutics in Alzheimer's disease;
- new avenues for therapeutics in prion disease;
- improved diagnostics of systemic amyloidosis.
The LUPAS consortium strives for continuing towards these goals beyond the project time frame pending future generous support from various stakeholders such as industry, academia, patients and politicians.
Economic impact of biomedical imaging
The value of LCP molecules becomes tangible and visible in the light of the various applications and proof-of-concept that has emerged as results from the LUPAS project. The potential business case needs to be developed and discussed from a strategic point of view. This brief market analysis is a first step towards understanding the market and market players. A market analysis should preferably cover the full potential of the LUPAS project outcome. That is all kinds of imaging applications from basic research to clinical applications as well as molecular diagnostics and therapeutic use. Herein we decided to focus on the imaging market. Today medical imaging technologies are widely used in clinical diagnosis to guide therapeutic and surgical intervention and to monitor disease progression, recurrence and treatment response and to improve surgical navigation. The clinical market is a multibillion market industry. The markets for PET tracers and MRI contrast agents, which are the most relevant markets in the context of LUPAS was estimated to USD 2.3 billion in 2011. The major players in the market are companies like GE, Bracco, Bayer and Guerbet. The market for radiopharmaceuticals and molecular imaging is expected to experience strong growth. Nuclear medicine, including single-photon emission computed tomography (SPECT) and positron emission tomography (PET), is revolutionizing how diseases are treated, such as AD. The FDA in early 2012 approved Amyvid (Florbetapir F 18 Injection) (Lilly / AvidRadiopharmaceuticals) for PET imaging of the brain in adults who are being evaluated for AD (AD) and other causes of cognitive decline. Neoprobe has licensed from AstraZeneca the worldwide rights to an imaging agent called AZD4694, which is a flourine-18 labelled radiotracer intended for use in the diagnosis of AD. The deal calls for Neoprobe to make a $5 million up-front payment to AstraZeneca. Terms of the deal also include up to USD 6.5 million in payment, contingent upon the achievement of clinical and regulatory milestones, and another payment of up to USD 11 million. The LUPAS project has shown that LCPs have a clear potential in imaging and for clinical use the first choice is a PET imaging agent. The combination of imaging and spinal fluid diagnostics is a novel possibility, which needs to be analysed further. Obviously, the LUPAS project has not at this stage performed a study for proof of concept for the purpose, however the data accumulated so far in combination with the size and growth of the target market is intriguing. Perhaps most importantly from a strategic point of view, is to define the competitive edge of the LCPs in this case. It will probably be more interesting to present a reagent that can move diagnostics of amyloid plaques one step further than the agents already approved for clinical use. It is of great importance to stress the benefit from a healthcare and patient perspective in using a LCP derived agent in comparison with the existing imaging methods. If there is a few distinct answers to that question, like for instance the possibility to image conformational differences in the plaques and correlate that output data to guidance for treatment, the probability to be able to get an approval and a significant market share increases rapidly. The preclinical contrast agent market is valuated to USD 200 million (2010) and it is expected to grow with a double digit on an annual basis the next couple of years. It is being dominated by optical imaging reagents (USD 70 million) followed by PET / SPECT radionucleotides (USD 60 million) and MRI (USD 40 million). It is important to notice that the preclinical market utilises contrast agents approved for clinical use as well as novel agents. The preclinical arena is therefore an important test for novel agents. Multimodal nanoparticles are one example as well as novel fluorescent biomarkers and PET/SPECT tracers. It is obvious that the LCPs recognition of amyloid and prion based deposits in combination with the fluorescent signature, radionucleotide labelling and the conjugation of LCPs to nanostructure open novel imaging possibilities. Optical imaging is already proven very well within the LUPAS consortium and the only limitation market wise may be the need for adequate animal models and technical skills that are required to perform optical imaging within the brain. An MRI contrast agent that can cross the blood-brain barrier and utilise LCP specificity would gain great interest and a field of research that is interesting to continue after the LUPAS project as such. In any preclinical imaging study the need for an orthogonal method to confirm your images is needed. In this case the LCPs inherent fluorescent properties make it ideal for post mortem tissue microscopy to confirm the findings from the in vivo imaging studies. Taken together these features are very promising in the perspective of commercialising a multimodal imaging toolbox for the preclinical arena for applications within AD and prion diseases. The market drivers and instrumentation available give on hand that it is interesting to market LCPs on the preclinical market.
For companies like Genovis and Applied Spectral Imaging, both partners within the LUPAS project, it may be valuable to further investigate the preclinical and tissue imaging markets in order to extend their respective business. To provide unique biomarkers and contrast agents in the field of Alzheimers research will add a competitive edge and a substantial market potential for reasonable investment. There are a few challenges in addition to performing a more in-depth competitive edge analysis. The most important hurdles to overcome initially is to set up a production source for the LCPs, move around technology barriers by providing application notes and reference studies, show that the LCPs are non-toxic and prove multimodality.
List of websites: http://www.lupas-amyloid.eu/
The objective of LUPAS was to develop novel molecular agents and methods for use within diagnostic imaging of Alzheimer's disease (AD) and prion diseases. Our concept relied on reporter molecules based on luminescent conjugated polymers, luminescent conjugated polythiophenes (LCPs). The LCP molecules target pathogenic protein aggregates associated with these diseases and LCPs target these lesions with high selectivity and specificity. The objective was also to investigate potential therapeutic potency of the LCP imaging agents.
The work was organised in several interdependent layers of experimentation:
- synthesis of LCPs and nanoparticles;
- synthesis of pure amyloidotic molecular targets;
- assay development - fluorescence spectroscopy and MRI;
- development of disease models;
- assay of LCPs in tissue sections;
- in vivo imaging - two-photon (2P) fluorescence and MRI;
- therapeutic potential in prion model systems;
- therapeutic potential in vivo in AD models.
LUPAS was focused on providing new knowledge in chemical design and smart targeting of misfolded protein amyloid lesions in AD and prion diseases. The LUPAS partners were assembled from a wide range of areas ranging from experts within organic synthetic chemistry, synthetic nanochemistry, amyloid structure, prion disease AD, MRI, multi-photon physics and hyper spectral imaging. This group formed the critical mass of competences needed to reach the project's ambitious goals.
The main project outcomes affords visualisation of dynamics and biochemical activity of pathological processes of AD and prion diseases in real-time from the molecular level to the organ and full body scale. Most importantly, novel molecular insight regarding the pathological hallmarks of these diseases and the specific chemical requirements for optimal amyloid ligands was concluded.
Dissemination activities included scientific article publications, conference presentations, amyloid workshop for the pathology community and sample shipments to two dozens of labs worldwide.
Project context and objectives:
The European population is aging. As a consequence age-related neurodegenerative diseases such as AD are on a steady rise reaching epidemic proportions of serious concern to the afflicted patients and their relatives. These horrific diseases put a heavy burden on society including a decade of heavy dependence on caregivers culminating with care at nursing homes. If this epidemic is not sequestrated it is projected to threaten the core of the well-fare societies of all European nations. The culprit within these diseases is accumulated misfolded protein molecules, called amyloids, which corrupt normal cellular functions and cause disease. Amyloids are self-perpetuating fibrillar states, which can hide for decades within affected individuals and replicate with immense efficiency, especially in the elderly. We, as a community, still stand short in delivering definite diagnosis or benign disease-modifying treatment or cure to any neurodegenerative disease. The only way to fight these diseases is long-term focused high quality research. Within the LUPAS project we developed novel molecular agents and methods for use within diagnostic imaging that relies on reporter molecules based on luminescent conjugated polymers, LCPs. The LCP molecules target the pathogenic protein aggregates with high selectivity and specificity. Candidate lead molecules have also been shown to have therapeutic potential.
The LUPAS partners were assembled from a wide range of areas ranging from experts within organic synthetic chemistry, synthetic nanochemistry, amyloid structure, prion disease AD, MRI, multi-photon physics and hyper spectral imaging. This group formed the critical mass of competences needed to reach the project's ambitious goals. The partner organisations in the LUPAS consortium are: Linköping University (Sweden), Université Claude Bernard Lyon 1 (France), University of Tübingen (Germany), Norwegian University of Science and Technology (Norway), Zürich University Hospital (Switzerland), Charité Universitätsmedizin Berlin (Germany), Applied Spectral Imaging (Israel) and Genovis AB (Sweden). Most importantly the project was organised according to 7 work packages (WPs) that defined the work plan. The LUPAS project was described by 5 technical WPs aimed at developing novel optical and MRI (MRI) imaging agents, towards the objectives of proving smart tools for biomedical research, for prediction, diagnosis, monitoring and prognosis of diseases, and for support and guidance of therapeutic interventions within the field of chronic neurodegenerative diseases.
- WP1 entailed the management of the consortium.
- WP2 generated LCP molecular reagents, reaction protocols and defined amyloidotic molecular targets (AMTs) for implementation in WP5, WP6 and for grafting onto a magnetic nanoparticle (MNP) for MRI in WP3.
- WP3 generated novel imaging agents for MRI.
- WP4 constituted an arena for advanced optical spectroscopic characterisation and nuclear magnetic resonance (NMR) imaging of LCP or LCP-MNP stained material from WPs 2-3 and WPs 5-6.
- WP5 was dedicated to investigation of AD transgenic mouse and human AD material for LCP and LCP-MNP diagnosis, amyloid biomarkers, and experimental therapeutic intervention.
- WP6 was dedicated to investigation of prion diseases from mouse and human material for LCP and LCP-MNP diagnosis, amyloid biomarkers, and experimental therapeutic intervention.
- WP7 was dedicated to the execution of dissemination and exploitation activities of project results. Specifically, the LUPAS project will make available the new smart imaging agents for clinical use with external non-profit stakeholders such as university hospitals and related research groups.
LUPAS was a three-year focused research project under the Seventh Framework Programme (FP7). Significant achievements were reached which is well reflected by a surge of interest from the research community to use the LCP technology. Other indications of success are the high citation rate of publications from the LUPAS research groups. By combining the diversity of scientific knowledge available within the consortium, progress towards the development of multimodal amyloid ligands for non-invasive imaging was achieved. The outcome affords visualisation of dynamics and biochemical activity of pathological processes of AD and prion diseases in real-time from the molecular level to the organ and full body scale. Most importantly, novel molecular insight regarding the pathological hallmarks of these diseases and the specific chemical requirements for optimal amyloid ligands was concluded. One of the major tasks within LUPAS was to develop novel LCPs that can be utilised as tools for selective identification of protein deposits, the pathological hallmarks of AD and prion diseases. LCPs are fascinating molecules that behave like structural chameleons. When an LCP binds to its amyloid target, the LCP changes shape. The LCP thereby also alters its colour depending on the amyloid structure. Hence, it reports different outcome signals which can be detected. From a library of LCPs, we identified lead compounds with distinct chemical groups for detection and spectral assignment of protein deposits in AD and prion diseases. Amyloid structures are vastly heterogeneous. It is not known which structures are most relevant for disease. A number of LCPs displayed spectral discrimination of amyloid structures both from the same protein (prion protein (PrP), amyloid-beta (Abeta)) as well as for discrimination of different amyloids within the same patient (Abeta, Tau). In addition, we have shown that pre-fibrillar aggregates of PrP, and Abeta (oligomers) that goes undetectable by conventional amyloid ligands are readily detectable by LCPs. The ability of LCPs to discriminate between different structures of fibrillar and pre-fibrillar aggregates is of great importance for AD. These features render the LCPs superior to other small molecule probes for molecular diagnostics. Importantly, these LCPs can also be used for in vivo imaging of protein deposits in the brain of living animals. Hence, LCPs that are selective amyloid ligands and even capable of crossing the blood brain barrier were developed within the LUPAS project. Synthesis of LCPs were scaled up into gram scale amounts and are now being implemented within a variety of subprojects within LUPAS partner labs and within collaborations all over Europe and worldwide. Multiphoton imaging is a preferable technique for studying protein aggregation diseases in real time in transgenic mouse models and we have shown that the unique optical properties of LCPs make these dyes highly efficient for multiphoton in vivo imaging. Several LCPs cross the blood brain barrier which enables specific labelling of amyloid plaques (Abeta) and intraneuronal neurofibrillary tangles (Tau) in the parenchyma of the brain, as well as Abeta amyloid in blood vessels so called cerebral amyloid angiopathy. In vivo imaging of protein deposits in living mice was demonstrated using distinct LCPs and these LCP scaffolds are well tolerated by mice even during longer time periods and repetitive injections. The latter is imperative, because the final goal of LUPAS is to achieve an LCP-based agent for non-invasive clinical diagnostics of AD and prion diseases. From a diagnostic perspective, we have employed the LCPs on post-mortem tissue sections from patients with neurodegenerative diseases. The LCP selectively stain the protein deposits and expose these major pathological hallmarks of either AD or prion diseased brain. The identification can easily be done and distinguished due to the colour emitted from the LCP bound to the different entities. Hence, the LCP technique shows excellent promise for being implemented as a complementary technique in routine clinical diagnostics. Work towards fluorescence detection of aggregated protein as biomarkers in cerebral spinal fluid of AD and control patients has been performed, but results so far have not been conclusive. Optical probes are not the preferable agents for non-invasive imaging in humans, due to the limitations of optical imaging. In this regard, the LUPAS consortium aimed to broaden the scope to develop novel multimodal LCPs that can be used for both optical imaging and MRI. MRI is today a standard technique for imaging of pathological conditions in humans, but subtle pathological changes in tissue can be hard to detect due to limited effect on the MR signal. Contrast agents can be used to enhance the MR image contrast related to tissue pathology. Today only unspecific contrast agents are in use, but targeted contrast agents based on paramagnetic and superparamagnetic nanoparticles and nanocomplexes (MNPs) hold great promise for MRI. Within LUPAS we have synthesised a variety of nanoparticles for enhanced T1 and T2 relaxation dispersion potentially useful as contrast agents for MRI. In addition, the first prototype of a LCP-MNP conjugate can specifically target amyloid in vitro, in tissue samples, and promising results have also been obtained in vivo in transgenic mice.
Disease modifying treatments for neurodegenerative diseases are scarce and non-existent for prion diseases. Because the LCP molecules show selective binding towards the culprit - the aggregated proteins their potential as therapeutic agents against prion diseases was evaluated within LUPAS. Work on tissue sections and cell cultures show efficient LCP mediated reduction in prion infectivity as a result of unexpected stabilisation of prion aggregates. This mechanistic insight opens up completely new avenues for targeting both prion disease and other neurodegenerative diseases based on hyperstabilisation rather than clearance of the misfolded protein culprits.
Project results:
Monomeric and trimeric precursor thiophene compounds with distinct side functionalities, including a variety of amino acid derivatives, were synthesised. In the initial screen we included two molecules with ethylamine sidechains on thiophene rings 2 and 4 (5006, 5007), however, as indicated in earlier studies we focused on, anionic LCPs containing the monomer 3-thiophene acetic acid which have shown the best selectivity and specificity towards AMTs and protein aggregates in tissue under close to physiological conditions. Therefore our efforts in were focused on developing anionic LCPs with carboxylic acid functionalisation. From a selection of building blocks, we have utilised Suzuki coupling to synthesise a library of chemically defined oligothiophenes, ranging from tetrameric to heptameric thiophenes, having amine or carboxylic acid functionalities at distinct positions along the thiophene backbone. The synthetic routes for some of the LCPs are rather straightforward and have been optimised to render up to gram quantities of individual LCPs.
Early on in LUPAS we published a detailed protocol on the synthesis of p-FTAA. The straightforward synthesis of p-FTAA, including two Suzuki couplings makes this a highly tractable molecule for many labs with adequate synthetic expertise.
The method was published on the Nature Protocols website: 'Synthesis of a pentathiophene fluorescent probe, 4',3'''-bis-carboxymethyl-(2,2';5',2'';5'',2''';5''',2'''']quinquethiophene-5,5''' by Andreas Åslund, K. Peter Nilsson & Peter Konradsson Protocol Exchange doi:10.1038/nprot.2010.24 Nature Protocols website Community contributed.
During LUPAS more than 20 novel anionic LCPs were synthesised whereof several have been synthesised in gram scales. The LCPs have been used throughout the project and have been implemented at all partner sites.
Multifunctional amyloid specific LCP-nanoagents for MRI and fluorescence
The concept of enhancing the detection methodology of amyloid towards in vivo imaging using MRI requires paramagnetic species as contrast agents. Within LUPAS the amyloid specific anchor was based on LCP molecules. For MNP covalent attachment, asymmetrical variants of the anionic LCPs have also been synthesised. These amyloid specific ligands will anchor the MNP to amyloid plaque. A variety of spacer chemistry both in terms of chain length and composition was developed in LUPAS.
Nano synthesis
Nanoparticles are interesting platforms for such application since it concentrates magnetic ions in a limited volume and is susceptible to bear functional groups. In this WP the investigation on several types of nanomaterials based on Fe, Gd or Mn are performed. The influence of the chemical composition, size, and shape of the particles on the relaxivities and NMR properties was checked.
Nanoparticles for relaxivity enhancement in MRI
MRI is a non-invasive method that allows for longitudinal imaging of animals and patients. In dedicated animal MR-scanners, the desired organ or body part, for example the brain in rats and mice can be imaged in vivo with high quality and resolution. Spatial resolution in MRI is dependent on the signal to noise ratio (SNR) and the two main factors are image acquisition duration and magnetic field strength. When the object is put into the MR scanner, all the protons within the object act as small magnets and align along the magnetic field inside the MR-scanner spinning around the axis of the magnetic field. A short excitation pulse is then applied, shifting the alignment of the proton spins out of this direction. After the excitation pulse comes to an end, a signal is recorded in a coil. The signal created by the proton spins decays thereafter by two relaxation processes (called T1 and T2 relaxation) as the protons relax back to their original alignment. T1 and T2 vary with the tissue type, and by weighting the images by the T1 and T2 contrast can be created in the MR images. Commonly used contrast agents such as gadolinium or ironoxide complexes can be used to alter the relaxation properties of the tissue, leading to increased or reduced signal from the tissue where the contrast agents are present. Within LUPAS the relaxivity of 20 different particles have been measured on different field strength and 6 different conjugated MNP-LCPs have been measured at 7T. The relaxivity measurements indicate that the main properties of the particles are kept or enhanced after conjugation to LCP with different linkers. This is due to that most of the promising particles have high r2 relaxivity, and increased size will increase r2.
LCP-MNP retains amyloid fibril specificity
That the selective property of amyloid targeting was retained in LCP-MNP complexes was assessed by both fluorescence and electron microscopy of in vitro generated AMTs from Abeta 1-42.
Development of AMTs of Abeta and PrP for LCP screening
For screening of suitable photophysical LCP properties and binding parameters of LCPs we employed AMTs generated in vitro by recombinant protein preparations. Analyses of preparations were made both after reaching equilibrium and during kinetic measurements. The AMTs used were based on Abeta preparations and PrP employing quiescent and shaking conditions as well as different Abeta 1-40 and Abeta 1-42 mixtures in defined ratios. The spectroscopic screening with LCPs with AMT material was used for optimal settings for microscopy as well as for selection of reference spectra for tissue sections. From the kinetic screen we discovered that p-FTAA was a highly suitable molecule for generating LCPs sensitive for non-mature (ThT negative) prefibrillar amyloid, both from Abeta and human PrP as well as other proteins (Figure 6). This finding was published in Biochemistry 2010. 'A fluorescent pentameric thiophene derivative detects in vitro-formed prefibrillar protein aggregates' Hammarström P, Simon R, Nyström S, Konradsson P, Aslund A, Nilsson KP. Biochemistry. 2010 Aug 17;49(32):6838-45.
A focused library of LCPs followed this initial lead and allowed us to conclude that this feature of detection of prefibrillar aggregates is shared by anionic LCPs which should have a backbone consisting of five to seven thiophene units and carboxyl groups extending the conjugated thiophene backbone (Klingstedt et al. 2011). Importantly the carboxyl moiety extending the conjugation of the thiophene backbone also allowed for spectral discrimination of Abeta and Tau within human tissue samples. The library design and conclusions from the AMT screen was published in Organic and Biomolecular Chemistry 2011. 'Synthesis of a library of oligothiophenes and their utilisation as fluorescent ligands for spectral assignment of protein aggregates'. Klingstedt T, Aslund A, Simon RA, Johansson LB, Mason JJ, Nyström S, Hammarström P, Nilsson KP. Org Biomol Chem. 2011 Dec 21;9(24):8356-70.
Development of AMTs for human PrP instead of chemical preparations also entailed protein engineering. Within LUPAS we investigated the molecular basis for the dependency for misfolding of PrP as a consequence of variations in position 129. The 129 site is well known to be highly important for disease penetrance and clinical presentation, being either M (wt) or V (variant). Our data based on the in vitro behaviour of PrP in solution, showed that position 129 was an interesting site for recruitment of PrP into the amyloid state. Charge mutations (M129E and M129K) showed decreased spontaneous conversion propensity (nucleation) compared to the wild type. We also observed that M129C was totally resistant towards spontaneous engulfment of PrP into the amyloid fold, most likely due to the formation of an intermolecular disulfide bridge forming in the lag phase, rendering a non-convertible disulfide linked dimer effectuated by the 129CC interaction. In contrast to the rather large differences in fibrillation propensity found for extreme amino acid substitutions such as tryptophan, proline, cysteine and charge, quite minor changes were observed for conservative mutations M129L, M129A and as expected M129V. From these data we have formulated a molecular platform for experimental AMT assessment of e.g. synthetic prions and targets for LCP screening. This study was published in The Journal of Biological Chemistry 2012. 'Multiple substitutions of methionine 129 in human PrP reveal its importance in the amyloid fibrillation pathway' Nyström S, Mishra R, Hornemann S, Aguzzi A, Nilsson KP, Hammarström P. J Biol Chem. 2012 Jul 27;287(31):25975-84.
Use of LCPs for animal and human samples
LCPs with molecular scaffolds consisting of repetitive thiophene moieties have evolved as an interesting class of fluorescent probes, since this class of molecules offers the possibility to achieve an optical read-out of biomolecular interactions through their impact on the conformation and the geometry of the LCP backbone. These structurally induced optical changes of the LCP backbone also allow the tantalising possibility to detect conformational changes of aggregates, rendering opportunities to gain more information concerning the morphology of the protein deposits. The use of LCPs have been reviewed both for versatile and detailed usage by LUPAS researchers in Methods in Molecular Biology, 2012, 'A pentameric luminescent-conjugated oligothiophene for optical imaging of in vitro-formed amyloid fibrils and protein aggregates in tissue sections'. Nilsson KP, Lindgren M, Hammarström P. Methods Mol Biol. 2012;849:425-34 and Biochemical Society transactions 2012, 'Luminescent conjugated poly- and oligo-thiophenes: optical ligands for spectral assignment of a plethora of protein aggregates'. Klingstedt T, Nilsson KP. Biochem Soc Trans. 2012 Aug;40(4):704-10.
LCPs for AD
It is currently recognised that amyloid deposits in vivo are vastly heterogenous and appear polymorphic. LCPs have provided the opportunity to gain more information concerning the morphology of the protein deposits, and a greater understanding of the conformational phenotype encoded in distinct protein aggregates should be achieved. Instead of merely observing the total amount of protein aggregates, the LCPs allow spectral finger printing of heterogenic population of specific protein aggregates and these ligands have been efficient for studying protein aggregates in transgenic mouse models having AD Aß pathology. This was one of the core aims for LUPAS. Tissue sections mainly from mouse brain were herein stained with a variety of LCP probes and were used to assay amyloid binding, specificity and spectral features, mainly using different microscopy based techniques. Several amyloid pathological hallmarks of disease were studied in AD mouse models APP / PS1 and APP23 mice aged 18 months. Most studied LCPs identified amyloid plaque and cerebral amyloid angiopathy (CAA). The robustness of staining was also verified after prolonged storage of tissue sections (in the dark) by revisiting the sections after 2 years.
Compatibility of LCPs with other imaging modalities
Importantly the LCP molecules are highly compatible with new imaging technologies. In one study the LCP imaging tracer pFTAA was used for combined imaging using fluorescence tracing of amyloid plaque and time-of-flight secondary ion mass spectrometry (TOF-SIMS) was used to investigate cholesterol accumulations originating from surrounding astrocytes proximal to plaque. This study was recently published in Acta Neuropathologica 2013, 'Localisation of cholesterol, amyloid and glia in AD transgenic mouse brain tissue using TOF-SIMS) and immunofluorescence imaging' Solé-Domènech S, Sjövall P, Vukojevic V, Fernando R, Codita A, Salve S, Bogdanovic N, Mohammed AH, Hammarström P, Nilsson KP, LaFerla FM, Jacob S, Berggren PO, Giménez-Llort L, Schalling M, Terenius L, Johansson B. Acta Neuropathol. 2013 Jan;125(1):145-57.
In vivo imaging
Repetitive in vivo imaging in mice is an important tool for studying dynamic changes in structure and function of the brain. A methodology by LUPAS researchers describes a head fixation system, which allows rapid re-localisation of previously imaged regions of interest (ROIs) within the brain. Such ROIs can be automatically relocated and imaged over weeks to months with negligible rotational change and only minor translational errors. Previously stored imaging positions can be fully automated re-localised within a few seconds. This automated rapid and accurate relocation simplifies image acquisition and post-processing in longitudinal imaging experiments. The risk of laser induced phototoxicity is greatly reduced, because the laser is only used for data acquisition and not for finding previously imaged ROIs. Thus, here described head fixation device appears well suited for in vivo repetitive long-term imaging in mouse brain. The initial work of the above mentioned technology has been used to study Abeta-amyloid plaque formation and growth over time. Plaque growth in APP / PS1 were shown to grow radially by 0.3 µm per week (Hefendehl et al., 2011). The tracer used in these experiments were the methoxy-X04 dye developed by the Klunk laboratory (Klunk WE et al. J Neuropathol Exp Neurol. 2002 61(9):797-805), but can also be performed using LCPs. These developments and discoveries were published in two papers in Journal of neuroscience and Journal of Neuroscience Methods. 'Long-term in vivo imaging of beta-amyloid plaque appearance and growth in a mouse model of cerebral beta-amyloidosis', Hefendehl JK, Wegenast-Braun BM, Liebig C, Eicke D, Milford D, Calhoun ME, Kohsaka S, Eichner M, Jucker M. J Neurosci. 2011 Jan 12;31(2):624-9; 'Repeatable target localisation for long-term in vivo imaging of mice with 2-photon microscopy', Hefendehl JK, Milford D, Eicke D, Wegenast-Braun BM, Calhoun ME, Grathwohl SA, Jucker M, Liebig C. J Neurosci Methods. 2012 Apr 15;205(2):357-63.
In vivo imaging using LCPs
Initial experiments from LUPAS researchers revealed that LCPs pFTAA (5001) and pHTAA (5002) showed efficient penetrance of the BBB and readily stained amyloid plaque in APP / PS1 mice (Åslund 2009). Following this success within LUPAS we employed the larger molecule hFTAA (7001), which has a more red shifted fluorescence spectrum and equal or superior sentitivity than pFTAA. The LCP hFTAA passed the blood-brain barrier after systemic administration and was specifically bound to extracellular beta-amyloid deposits in the brain parenchyma (Abeta plaques) and in the vasculature (cerebral beta-amyloid angiopathy) of ß-amyloid precursor protein transgenic APP23 mice. Interestingly, peripheral application of hFTAA also stained intracellular lesions of hyperphosphorylated Tau protein in P301S Tau transgenic mice. In addition hyperspectral imaging of all three amyloid types was acquired ex vivo using two-photon excitation. Within this work an algorithm was developed to demonstrate the LCP ability to distinguish between two amyloid lesions and the background. hFTAA revealed a distinct shift in its emission spectra when bound to Abeta plaques versus Tau lesions. A spectral difference was also observed for Abeta plaques versus cerebral beta-amyloid angiopathy, indicating that different amyloid types and structural variances of a specific amyloid type can be distinguished. Hence, by adding spectral signatures to amyloid lesions, our results pave the way for a new biophotonic era of in vivo amyloid imaging, allowing in vivo differentiation of amyloid polymorphisms and monitoring changes of their structure / composition over time.
This study was recently published in American Journal of Pathology 2012, 'Spectral discrimination of cerebral amyloid lesions after peripheral application of luminescent conjugated oligothiophenes', Wegenast-Braun BM, Skodras A, Bayraktar G, Mahler J, Fritschi SK, Klingstedt T, Mason JJ, Hammarström P, Nilsson KP, Liebig C, Jucker M. Am J Pathol. 2012 Dec;181(6):1953-60.
LCP-MNP retain amyloid plaque specificity ex vivo
The developed LCP-MNP for MRI imaging was initially found to retain amyloid specificity against AMTs in vitro. A more complex sample such as tissue samples could post more challenging. Nevertheless, LCP-MNPs were highly fluorescent and showed specific binding and strong fluorescence when bound to amyloid plaque in tissue sections from transgenic mice. The specificity and selectivity in these samples were independent of attachment to MNPs reflecting the same fluorescence profiles as the corresponding LCP-spacer adducts. MRI data acquisition in vivo in APP / PS1 mice proved more challenging than in vivo fluorescence. Several application procedures and LCP-MNP contrast agents showed small amounts of penetrance of LCP-MNPs over the BBB following intravenous (i.v.) inoculations. Better success was achieved with direct administration of the contrast agents into the CSF.
Development of hyperspectral correlation analysis of multicolor, autofluorescence and LCP stained amyloid plaques
Multiphoton imaging is a preferable technique for studying protein aggregation diseases in real time in transgenic mouse models and within LUPAS it was further evident that the unique optical properties of LCPs make these dyes highly efficient for multiphoton in vivo imaging. In certain aspects there nevertheless appear problematic with autofluorescence. Green autofluorescence from amyloid plaques can be a problem in weakly stained samples, such as samples that have been i.v. injected at low levels e.g. LCP-MNP probes at low levels of LCP loading < 30 µM. Also, to analyse multi-staining experiments, one needs special attention to assess the spectral content. Researchers within LUPAS therefore developed a special algorithm that can accurately calculate the spectral similarity of a measured spectrum with a reference spectrum. The method is based on correlation, where the measured spectrum (X) of each pixel in a hyperspectral image is correlated with a reference spectrum (Y), using the following formula Where Xi is the i:th wavelength point and the bar over X denotes the average of the spectrum (and similar for Y). This makes it possible to give a mathematical number in the range of 0 to 1 where 1 is a measured spectrum identical to the reference. The formula is robust to background noise and does not suffer from the same problems as other linear spectral unmixing algorithms. The method of spectral correlation was recently published in 'Hyperspectral analysis using the correlation between image and reference'. PG Ellingsen, NK Reitan, BD Pedersen, M Lindgren J. Biomed. Opt. 2013. 18 (2), 020501.
Human AD diagnostics
Human brain tissue samples of sporadic AD patients were analysed within LUPAS. Tissues were screened with the anionic LCP library and from these studies LCPs that distinguish the two main pathological hallmarks of AD, A-beta plaques and neurofibrillary tangles (NFTs). The LCPs, pFTAA, hxFTAA and hFTAA, give a distinct spectroscopic signature from Abeta or tau aggregates. Notably these LCPs were five to seven membered oligothiophenes containing two additional carboxylic moieties terminating the conjugated backbone. These data were published in Organic and Biomolecular Chemistry 2011.
Therapeutics
LCPs as pharmacophores in AD mouse models
One aim with LUPAS was to test the potency of LCPs for possible pharmacological effects in transgenic mice with AD pathology. We tested LCOs as possible pharmocophores in vivo because pFTAA is known to target pre-fibrillar amyloid species, thereby possibly generating minor alterations in the molecular structure of amyloid fibrils that may interfere with amyloid fibril formation and their putative toxicity. It was demonstrated that p-FTAA (LCP 5001) is able to cross the blood brain barrier after i.v. and intraperitoneal (i.p.) injection and bind to cerebral amyloid plaques of APPPS1 mice, an early onset mouse model of cerebral amyloidosis. Repetitive injections of p-FTAA resulted in a robust staining of cerebral amyloid plaques in APPPS1 mice, and probes remained bound to plaques for up to 6 months. APPPS1 mice were treated for 12 weeks by weekly i.p. p-FTAA-injections and Abeta pathology and the general health was studied. Despite efficient binding of p-FTAA to cerebral amyloid, no significant p-FTAA dependent alteration in total brain amyloid burden, as judged by Congo Red staining and Abeta-immunohistochemistry was observed. These data indicate that persistent peripheral p-FTAA treatment neither causes obvious signs of toxicity, nor alters cerebral amyloid load in APPPS1 mice. To study whether continuous p-FTAA treatment beyond morphologically detectable alterations may modify amyloid toxicity on a functional level, long term potentiation and cognition in p-FTAA-treated APPPS1 mice will need to be assessed. This study was presented at the Foundation Ipsen conference in Paris February 2012.
S. Handrick, S. Prokop, S. Nystrom, J.J. Mason, N.R. Reitan, B.M. Wegenast-Braun, M. Lindgren, M. Jucker, K.P. Nilsson, P. Hammarstrom, F.L. Heppner Luminescent Conjugated Oligothiophenes (LCOs) as Pharmacophores in AD. Fondation Ipsen, Paris (2012)
Using LCP fluorescence as a follow up during treatment
For several years, the natural compound curcumin has been proposed to be a candidate for enhanced clearance of toxic Aß amyloid. Herein LUPAS researchers studied the potency of feeding curcumin as a drug candidate to alleviate Abeta toxicity in transgenic Drosophila. The longevity as well as the locomotor activity of five different AD model genotypes, measured relative to a control line, showed up to 75 % improved lifespan and activity for curcumin fed flies. In contrast to the majority of studies of curcumin effects on amyloid we did not observe any decrease in the amount of Abeta deposition following curcumin treatment. Importantly, conformation-dependent spectra from p-FTAA, LCP (5001) bound to Abeta deposits in different Drosophila genotypes over time, indicated accelerated pre-fibrillar to fibril conversion of Abeta (1-42) in curcumin treated flies. Our study shows that curcumin promotes amyloid fibril conversion by reducing the pre-fibrillar / oligomeric species of Abeta, resulting in a reduced neurotoxicity in Drosophila. Most importantly for LUPAS was that LCP fluorescence is an important tool for assessing conformational differences in treated versus untreated animals.
This study was published in PLoS One February 2012 Caesar I, Jonson M, Nilsson KP, Thor S, Hammarström P. Curcumin promotes A-beta fibrillation and reduces neurotoxicity in transgenic Drosophila. PLoS One. 2012;7(2):e31424. doi: 10.1371/journal.pone.0031424. This publication has been viewed online > 5 600 times and was featured in > 20 news articles worldwide.
Inflammatory modulation for new treatment strategies
Work within LUPAS also included assessment of alternative treatment strategies. The pathology of AD has an inflammatory component that is characterized by upregulation of proinflammatory cytokines, particularly in response to Abeta. The APPPS1 AD mouse model showed increased production of the common interleukin-12 (IL-12) and IL-23 subunit p40 by microglia. Genetic ablation of the IL-12/IL-23 signalling molecules by deficiency of p40 or its receptor complex, resulted in decreased cerebral amyloid load. In addition, peripheral administration of a neutralising p40-specific antibody also resulted in a reduction of cerebral amyloid load in APPPS1 mice. Furthermore, intracerebroventricular delivery of antibodies to p40 significantly reduced the concentration of soluble Abeta species and reversed cognitive deficits in aged APPPS1 mice. The concentration of p40 was also increased in the cerebrospinal fluid of subjects with AD, which suggests that inhibition of the IL-12/IL-23 pathway may attenuate AD pathology and cognitive deficits. The paper 'Inhibition of IL-12/IL-23 signalling reduces AD-like pathology and cognitive decline' by Vom Berg J, Prokop S, Miller KR, Obst J, Kälin RE, Lopategui-Cabezas I, Wegner A, Mair F, Schipke CG, Peters O, Winter Y, Becher B, Heppner FL.. Nat Med. 2012 Dec;18(12):1812-9 was recently published. This paper was highlighted in Nature Reviews Neuroscience 2013, Nature Reviews Drug Discovery 2013, and Nature Reviews Immunology 2013, New England Journal of Medicine, 2013. Follow-up of these mouse models by LCP fluorescence should add another level of analysis not only limited to plaque load but also LCP spectroscopy can delineate conformational differences within plaque. Such studies have so far been limited by the studied age groups of 4 and 8 months. Large differences in LCP spectroscopic differences in APP / PS1 mice occur after the age of 12 months (unpublished results), hence the time window needs to be adjusted for such studies to be meaningful.
Prions
Prion diseases or transmissible spongiform encephalopathy (TSE) have received unprecedented public interest despite its rareness in the human population. The main reason for this interest was the epidemic of bovine spongiform encephalopathy (mad cow disease) some 20 years ago and its evident transmissibility from cattle to humans. Prions can reside silent for decades in an infected individual and pose a threat to the human population due to its efficiency in jumping species barriers and through blood transfusions and transplants. Prion diseases are hence of outmost interest not only for basic research curiosity but rather it is imperative to possess sensitive detection and therapeutic strategies. The culprit protein in all prion diseases is called the PrP. The PrP is most abundant in mammalian neurons but is ubiquitously expressed throughout various cells and tissues. The functional role of native PrP is not fully understood. PrP is associated with a number of different prionoses: sporadic, inherited, and acquired, all of which are invariably fatal. The common molecular pathognomonic marker for prionoses is the presence of severe vacuolation within the CNS rendering a sponge-like tissue. Concomitant with the presence of spongiosis is the presence of a conformational isoform of PrP, which has converted from a largely helical globular protein cellular PrP (PrPC) that misfolds into an aggregation prone beta-sheet conformation, PrPSc, which often assemble into protein deposits with conspicuous similarities to amyloid.
The concept of prions as a novel pathogenic entity composed solely of corrupt protein has been extensively debated in several important review articles by LUPAS researchers. LUPAS researchers recently published 'Prion propagation, toxicity and degradation' by A. Aguzzi and J. Falsig in nature Neuroscience 2012. Nat Neurosci. 2012 Jun 26;15(7):936-9, to address this topic.
Development of prion assays
Lack of cytotoxicity in prion infected cell cultures has hampered mechanistic studies of prion-induced neurodegeneration. Recently, LUPAS researchers showed that prion-infected cultured organotypic cerebellar slices (COCS) experienced progressive spongiform neurodegeneration closely reproducing prion disease, with three different prion strains giving rise to three distinct patterns of PrP deposition. Neurodegeneration did not occur when PrP was genetically removed from neurons, and a pharmacological screen indicated that neurodegeneration was abrogated by compounds known to antagonise prion replication. Prion infection of COCS and mice led to enhanced fodrin cleavage, suggesting the involvement of calpains or caspases in pathogenesis. Accordingly, neurotoxicity and fodrin cleavage were prevented by calpain inhibitors but not by caspase inhibitors, whereas prion replication proceeded unimpeded. The article was published in PLoS Pathogens in 2012: 'Prion pathogenesis is faithfully reproduced in cerebellar organotypic slice cultures'. Falsig J, Sonati T, Herrmann US, Saban D, Li B, Arroyo K, Ballmer B, Liberski PP, Aguzzi A. PLoS Pathog. 2012 Nov;8(11):e1002985.
Therapeutics of prion diseases
Prion diseases in rodents faithfully reproduce the human disease, in contrast to AD in mice, which only show subtle pathophysiological phenotypes. This makes prion disease models a powerful assay for novel therapeutics that targets the culprit disease causing misfolding protein. LCPs interact with ordered protein aggregates and sensitively detect amyloids of many different proteins, suggesting that they may also possess antiprion properties. 8 different LCPs were analysed with a variety of side chain and chain lengths for antiprion activity. A variety of anionic, cationic, and zwitterionic LCPs reduced the infectivity of prion-containing brain homogenates and of prion-infected cerebellar organotypic cultured slices (COCS) and decreased the amount of scrapie isoform of PrPC (PrPSc) oligomers that could be captured in an avidity assay. Paradoxically, treatment enhanced the resistance of PrPSc to proteolysis, triggered the compaction, and enhanced the resistance to proteolysis of recombinant mouse PrP (23-231) fibres. These results suggest that LCPs act as antiprion agents by transitioning PrP aggregates into structures with reduced frangibility. Moreover, enzyme-linked immunosorbent assay (ELISA) on COCS and in vitro conversion assays with mouse PrP (23-231) indicated that poly(thiophene-3-acetic acid) (LCP 10001) may additionally interfere with the generation of PrPSc by stabilising the conformation of PrPC or of a transition intermediate. Therefore, LCPs represent a novel class of antiprion agents whose mode of action appears to rely on hyperstabilisation, rather than destabilisation, of PrPSc deposits.
Diagnostics of human prion diseases
From a diagnostic perspective, the LCPs / LCOs were also used to analyse post-mortem tissue sections from patients with neurodegenerative diseases. The fluorescence stains selectively bind to the protein deposits and expose these major pathological hallmarks of either AD or prion diseased brain. PrP deposits from Creutzfeldt-Jakob disease are notoriously complicated to visualise using conventional fluorescence dyes for amyloid such as Congo red and Thioflavin T. Several LCPs function well in this regard as was tested extensively on mouse material. To employ frozen material is especially important in this context. Within LUPAS this complicated procedure from a biosafety perspective was achieved. When this hurdle was overcome identification of PrP deposition in TSE brain can easily be done and distinguished due to the colour emitted from the LCP / LCO bound to the different entities for three distinct prion diseases. Hence, the LCP / LCO technique shows excellent promise for being implemented as a complementary technique in clinical diagnostics of TSEs.
Potential impact:
The European population is aging. As a consequence age-related neurodegenerative diseases such as AD are on a steady rise reaching epidemic proportions of serious concern to the afflicted patients and their relatives. These diseases put a heavy burden on society including a decade of profound dependence on caregivers culminating with care at nursing homes. If this epidemic is not sequestrated it is projected to threaten the core of the well-fare societies of all European nations.
The most thorough data on the ongoing and forthcoming Alzheimer epidemic comes from the Alzheimer Association is the USA (see http://www.alz.org online for further details). European national and transnational initiatives (see Alzheimer Europe http://www.alzheimer-europe.org online for further details) are emerging initiatives to tackle the rising incidence of dementias. In 2010 the European Commission (EC) funded the joint programming initiative on AD and neurodegenerative diseases (JPND) with a European Union (EU) contribution of close to EUR 2 million. From the Alzheimer Europe initiative it reads that 'In 2005, the total direct and informal care costs of AD and other dementias were estimated at EUR 130 billion in the EU-27 (EUR 21 000 per patient); 56 % of these costs was informal care. The most common forms of dementia in the EU are AD (about 70 % of cases), and vascular dementia (less than 30 %)'. These are staggering figures and are projected to rise.
A patient is diagnosed with AD mainly through the primary care where changes in short term memory and personality is based on cognitive tests. The cognitive tests are combined with MRI scans to rule out brain tumours or stroke as a primary cause for the symptoms. More developed centres allows for cerebrospinal fluid testing for biomarkers such as increased amounts of Tau and phosphorylated Tau and decreased amounts of soluble Abeta42. The culprit within these diseases is accumulated misfolded protein molecules, called amyloids of e.g. the mentioned biomarkers, which corrupt normal cellular functions and cause disease. Amyloids are self-perpetuating fibrillar states, which can hide for decades within affected individuals and replicate with immense efficiency, especially in the elderly. We, as a community, still stand short in delivering definite diagnosis or benign disease-modifying treatment or cure to any neurodegenerative disease. The only way to fight these diseases is long-term focused high quality research.
It is highly encouraging that the EC distributes increasing amounts of funding for tackling these diseases and hence work towards novel ideas on how to make disease modifying therapies and rapid diagnostics and follow up on treatments are ongoing in projects such as the described LUPAS project.
The LUPAS project is a testament to progress in cross-disciplinary research with a very impressive output of scientific impact. Over the years LUPAS researchers have been met with enthusiasm from the scientific community as feedback on published papers and conference presentations. LCP molecules have been sent to more than two dozen labs worldwide and a dozen collaborations have been initiated.
Final outcome and future perspectives
There is a tremendous need for quantitative diagnostic methods for early detection and evaluation of neurodegenerative disorders, such as AD and prion diseases. The need is underlined by the recent development of proposed therapeutically interventions targeting disease, so called disease modifiers, including immune therapy. Herein, quantitative physical outcome measures are urgently needed in terms of amyloid pathology within living patients.
Within the brief three-year time frame of LUPAS, the consortium developed important research tools for use in disease model systems (mouse models) in vivo and on histological ex vivo samples from humans. The LUPAS project exemplifies that broad cross-disciplinary expertise within diverse subjects can realise unmet needs within biomedicine. The realistic prognosis is that it will take a few more years to validate this technology in the preclinical phase prior to exploiting it in the clinic and hence lead to:
- improved diagnostics of AD;
- improved diagnostics of prion disease;
- new avenues for therapeutics in Alzheimer's disease;
- new avenues for therapeutics in prion disease;
- improved diagnostics of systemic amyloidosis.
The LUPAS consortium strives for continuing towards these goals beyond the project time frame pending future generous support from various stakeholders such as industry, academia, patients and politicians.
Economic impact of biomedical imaging
The value of LCP molecules becomes tangible and visible in the light of the various applications and proof-of-concept that has emerged as results from the LUPAS project. The potential business case needs to be developed and discussed from a strategic point of view. This brief market analysis is a first step towards understanding the market and market players. A market analysis should preferably cover the full potential of the LUPAS project outcome. That is all kinds of imaging applications from basic research to clinical applications as well as molecular diagnostics and therapeutic use. Herein we decided to focus on the imaging market. Today medical imaging technologies are widely used in clinical diagnosis to guide therapeutic and surgical intervention and to monitor disease progression, recurrence and treatment response and to improve surgical navigation. The clinical market is a multibillion market industry. The markets for PET tracers and MRI contrast agents, which are the most relevant markets in the context of LUPAS was estimated to USD 2.3 billion in 2011. The major players in the market are companies like GE, Bracco, Bayer and Guerbet. The market for radiopharmaceuticals and molecular imaging is expected to experience strong growth. Nuclear medicine, including single-photon emission computed tomography (SPECT) and positron emission tomography (PET), is revolutionizing how diseases are treated, such as AD. The FDA in early 2012 approved Amyvid (Florbetapir F 18 Injection) (Lilly / AvidRadiopharmaceuticals) for PET imaging of the brain in adults who are being evaluated for AD (AD) and other causes of cognitive decline. Neoprobe has licensed from AstraZeneca the worldwide rights to an imaging agent called AZD4694, which is a flourine-18 labelled radiotracer intended for use in the diagnosis of AD. The deal calls for Neoprobe to make a $5 million up-front payment to AstraZeneca. Terms of the deal also include up to USD 6.5 million in payment, contingent upon the achievement of clinical and regulatory milestones, and another payment of up to USD 11 million. The LUPAS project has shown that LCPs have a clear potential in imaging and for clinical use the first choice is a PET imaging agent. The combination of imaging and spinal fluid diagnostics is a novel possibility, which needs to be analysed further. Obviously, the LUPAS project has not at this stage performed a study for proof of concept for the purpose, however the data accumulated so far in combination with the size and growth of the target market is intriguing. Perhaps most importantly from a strategic point of view, is to define the competitive edge of the LCPs in this case. It will probably be more interesting to present a reagent that can move diagnostics of amyloid plaques one step further than the agents already approved for clinical use. It is of great importance to stress the benefit from a healthcare and patient perspective in using a LCP derived agent in comparison with the existing imaging methods. If there is a few distinct answers to that question, like for instance the possibility to image conformational differences in the plaques and correlate that output data to guidance for treatment, the probability to be able to get an approval and a significant market share increases rapidly. The preclinical contrast agent market is valuated to USD 200 million (2010) and it is expected to grow with a double digit on an annual basis the next couple of years. It is being dominated by optical imaging reagents (USD 70 million) followed by PET / SPECT radionucleotides (USD 60 million) and MRI (USD 40 million). It is important to notice that the preclinical market utilises contrast agents approved for clinical use as well as novel agents. The preclinical arena is therefore an important test for novel agents. Multimodal nanoparticles are one example as well as novel fluorescent biomarkers and PET/SPECT tracers. It is obvious that the LCPs recognition of amyloid and prion based deposits in combination with the fluorescent signature, radionucleotide labelling and the conjugation of LCPs to nanostructure open novel imaging possibilities. Optical imaging is already proven very well within the LUPAS consortium and the only limitation market wise may be the need for adequate animal models and technical skills that are required to perform optical imaging within the brain. An MRI contrast agent that can cross the blood-brain barrier and utilise LCP specificity would gain great interest and a field of research that is interesting to continue after the LUPAS project as such. In any preclinical imaging study the need for an orthogonal method to confirm your images is needed. In this case the LCPs inherent fluorescent properties make it ideal for post mortem tissue microscopy to confirm the findings from the in vivo imaging studies. Taken together these features are very promising in the perspective of commercialising a multimodal imaging toolbox for the preclinical arena for applications within AD and prion diseases. The market drivers and instrumentation available give on hand that it is interesting to market LCPs on the preclinical market.
For companies like Genovis and Applied Spectral Imaging, both partners within the LUPAS project, it may be valuable to further investigate the preclinical and tissue imaging markets in order to extend their respective business. To provide unique biomarkers and contrast agents in the field of Alzheimers research will add a competitive edge and a substantial market potential for reasonable investment. There are a few challenges in addition to performing a more in-depth competitive edge analysis. The most important hurdles to overcome initially is to set up a production source for the LCPs, move around technology barriers by providing application notes and reference studies, show that the LCPs are non-toxic and prove multimodality.
List of websites: http://www.lupas-amyloid.eu/
