Please explain Near-Infrared Fluorescence Imaging of Apoptotic

| October 22, 2018

Letterpubs.acs.org/acschemicalneuroscience))Near-Infrared Fluorescence Imaging of Apoptotic NeuronalCell Death in a Live Animal Model of Prion DiseaseVictoria A. Lawson,,, Cathryn L. Haigh,,, Blaine Roberts,,#Vijaya B. Kenche,,,§ Helen M. J. Klemm, Colin L. Masters,,#Steven J. Collins,, Kevin J. Barnham,,§, and Simon C. Drew*,,§,,^Department of Pathology, The University of Melbourne, Victoria 3010, Australia, Mental Health Research Institute, Parkville, Victoria3052, Australia, #Centre for Neuroscience, The University of Melbourne, Victoria 3010, Australia, and §The Bio21 Molecular Science andBiotechnology Institute, The University of Melbourne, Victoria 3010, AustraliaAbstractencephalopathies (prion diseases). Current medical imaging paradigms focus strongly on the detection of amyloid (1), but deposition of protein aggregates may ariseafter significant neuronal dysfunction has occurred (2, 3).Therefore, imaging agents capable of identifying molecular events independent of protein deposition and priorto the onset of clinical symptoms are most desirable.Fluorescence imaging technology can provide a noninvasive means for disease detection and evaluation oftherapeutic strategies in animal models by reporting onthe activity of a host of biological events (4-6). Intrinsicand extrinsic fluorophores that are excited and detected inthe visible spectrum have been used extensively for in vitrobiological studies employing fluorescence reflectance imaging and microscopy (4, 7). However, their suitability forin vivo imaging applications, including brain imaging, islimited due to the strong attenuation of visible wavelengths by biological tissue (oxy- and deoxyhemoglobin,fat, melanin, water, bone), leading to poor tissue penetration depths (5, 8). Although an optically transparentwindow chamber can be surgically implanted to slightlyimprove penetration depth, this requires an invasivecraniotomy to be performed. Diseased or dying tissues,especially of the central nervous system, can also exhibithigher levels of autofluorescence in the visible region ascompared with healthy tissues, due to the build up oflipofuscin (9).Near-infrared (NIR) wavelengths (650-900 nm) passmore deeply through mammalian tissue and are thereforemore suitable for in vivo fluorescence imaging (6, 8). Invivo optical imaging of amyloid plaques in a murineAlzheimers model has already been demonstrated usingvarious NIR probes (10, 11). However, to monitor diseaseonset, progression, and response to treatment, contrastagents that are independent of amyloid deposits andsensitive to early markers of disease must be sought. Inthis study, we present an NIR imaging agent capable ofirreversibly binding to active caspases by conjugating ahydrophobic NIR cyanine dye to the broad spectrumApoptotic cell death via activation of the caspase family ofcysteine proteases is a common feature of many neurodegenerative diseases including Creutzfeldt-Jakob disease.Molecular imaging of cysteine protease activities at thepreclinical stage may provide valuable mechanistic information about pathophysiological pathways involved indisease evolution and in response to therapy. In this study,we report synthesis and characterization of a near-infrared(NIR) fluorescent contrast agent capable of noninvasivelyimaging neuronal apoptosis in vivo, by conjugating aNIR cyanine dye to Val-Ala-Asp-fluoromethylketone(VAD-fmk), a general inhibitor of active caspases. Following intravenous administration of the NIR-VAD-fmkcontrast agent, in vivo fluorescence reflectance imagingidentified significantly higher levels of active caspases inthe brain of mice with advanced but preclinical priondisease, when compared with healthy controls. The contrast agent and related analogues will enable the longitudinal study of disease progression and therapy in animalmodels of many neurodegenerative conditions.Keywords: Near infrared, neurodegeneration, prion, mice,apoptosis, caspase, fluorescence, optical imagingThe most common cause of dementia is neurodegeneration arising from the misfolding and aggregation of normal cellular proteins and theirdeposition, often as amyloid. These neurodegenerativeconditions include Alzheimers disease, Parkinsons disease, Huntingtons disease, and transmissible spongiformr 2010 American Chemical SocietyReceived Date: July 26, 2010Accepted Date: September 17, 2010Published on Web Date: September 30, 2010720DOI: 10.1021/cn100068x |ACS Chem. Neurosci. (2010), 1, 720727Letterpubs.acs.org/acschemicalneuroscienceside chain, indicating loss of the methyl ester during synthesis.[M þ H]þ: m/z (calc) = 1015.3, m/z (exp) = 1015.0; [M þ Na]þ:m/z (calc)=1038.3, m/z (exp) = 1039.0. The ester is not requiredfor cell permeability and is removed by intracellular esterasesprior to caspase inhibition. The excitation and emission maximaof NIR-VAD-fmk were comparable to the unconjugated DY750 (λex 740 nm, λem 770 nm). The concentration of purifiedNIR-VAD-fmk solutions was determined from the opticaldensity at 745 nm using an extinction coefficient of 270 000M-1 cm-1.Cell CultureOBL-21 mouse neuronal cells (12, 13) were cultured inDulbeccos Modified Eagles Media (DMEM; Gibco – Invitrogen,Victoria, Australia) supplemented with 10% fetal bovine serum(Invitrogen), 50 U mL-1 penicillin, and 50 μg mL-1 streptomycinsolution (Sigma-Aldrich; New South Wales, Australia). All celllines were maintained at 37 °C with 5% CO2 in a humidifiedincubator.Figure 1. Structure of the contrast agent NIR-VAD-fmk. Thefluorophore is a near-infrared cyanine dye (DY-750; Dyomics,GmbH). The VAD-fmk sequence irreversibly binds to the active siteof caspase enzymes and is a broad spectrum inhibitor of the caspasefamily.In Vitro ToxicityCells were plated to 20% confluency. Five microliters ofone-solution MTS reagent (Promega, Victoria, Australia)per 100 μL media was added to the media of test and controlcultures and incubated under normal culture conditions for90 min. Reaction product was quantified using absorbance at462 nm in a Fluostar Optima (BMG Labtech). All results werenormalized to cell density.caspase inhibitor VAD-fmk (Figure 1). To test the compound in vivo, we assessed its ability to detect neuronalapoptosis in a live murine model of prion disease. Compared with healthy controls, a significant increase in activecaspases was measured in the brain of prion-infected miceprior to onset of clinical signs such as ataxia and hind limbparesis. The ability to detect neuronal cell death in livingmice prior to development of clinical signs of neurodegenerative disease makes it suitable candidate forscreening potential therapeutics targeting early pathogenic pathways.In Vitro Active Caspase DetectionNIR-VAD-fmk (Figure 1) was solubilized in sterile phosphate buffered saline (PBS) (Gibco – Invitrogen) containing10% v/v high quality (cell culture tested) sterile-filteredDMSO (Sigma-Aldrich). Cells were incubated with 15 μMNIR-VAD-fmk for 30 min, washed, and imaged under 20Âmagnification with a Nikon Eclipse TE2000-E epi-fluorescence microscope using an excitation wavelength of 620 (30 nm and a 660 nm long-pass filter (Filter set 41008). Tocorrect for background, the average fluorescence intensity ofimages taken without the addition of fluorescent marker wassubtracted in an identical manner from all images.MethodsSynthesis of a Near-Infrared Fluorescent Inhibitorof Active CaspasesZ-Val-Ala-Asp(OMe)-fmk (Z=benzyloxycarbonyl; fmk=fluoromethylketone) was purchased from SM BiochemicalsLLC (Yorba Linda, USA). The peptide was dissolved in a33% v/v solution of HBr in AcOH (Sigma-Aldrich, Castle Hill,Australia) and stirred for 45 min to form the HBr salt ofNH2-VAD-fmk. The solution was evaporated to dryness andresidual HBr/AcOH was removed by washing with ethyl ether,then resuspended in DMF at a concentration of 100 mM. The NIRcyanine dye containing an amino-reactive N-hydroxysuccinimidyl-ester (DY-750 NHS-ester, Dyomics GmbH, Germany) wasdissolved in DMF and added to the peptide solution at a molarratio of 1.1:1 in the presence of 3 molar equiv of N,N-diisopropylethylamine. The reaction was allowed to proceed in the darkovernight at room temperature (RT) with stirring, before thesolution was again evaporated to dryness. The labeled product(denoted NIR-VAD-fmk) was purified by RP-HPLC using aShimadzu LCMS-2010EV LC/MS system with a 5 μm, 2.1 Â150 mm cyano column (Waters Corp., Massachusetts, USA)and a 40-45% MeCN/H2O gradient in the presence of 0.1%formic acid. Purity was g98% as determined from theRP-HPLC trace at 700 nm (Figure S1, Supporting Information).MS analysis indicated the conjugate contained a free aspartylr 2010 American Chemical SocietyMiceAll animal procedures conformed to National Health andMedical Research Council of Australia guidelines and wereapproved by the University of Melbourne Animal Experimentation Ethics Committee. The M1000 prion strain used inthis study was derived from the Fukuoka-1 strain of mouseadapted human prions (14). This strain was originally isolatedand maintained by passage in Balb/c mice (15). Five 6-weekold Tga 20 mice were inoculated under methoxyfluraneanesthetic in the left parietal region with 30 μL of 1% (w/v)homogenate prepared in PBS from a pool of brains derivedfrom Balb/c mice with clinical prion disease induced byM1000 infection. The Tga20 transgenic mouse line overexpresses murine PrPC and mice develop signs of clinical priondisease (bradykinesis, kyphosis, ataxia, and hind limb paresis)approximately 60 days after an intracerebral prion inoculation, with terminal disease occurring within days of symptomonset (16). Three 5-week old Tga20 mice received an intracerebral prion inoculation with 30 μL of 1% w/v brain homogenate prepared from uninfected Balb/c mice and served asage matched, sham inoculated controls.721DOI: 10.1021/cn100068x |ACS Chem. Neurosci. (2010), 1, 720727Letterpubs.acs.org/acschemicalneuroscienceIn Vivo Toxicitymethanol). Blots were blocked in 5% skim milk powder in PBS/0.1% Tween-20 (PBST) for 1 h at room temperature (RT) andprobed for 1 h at RT with rabbit polyclonal antibody 03R19raised to residues 89-103 of mouse PrP (19). The blot waswashed at RT 1 Â 10 min, and 3 Â 3 min, probed withantirabbit-HRP for 1 h at RT, washed as described, anddeveloped using ECL-plus (GE Healthcare). The preparedmembrane was imaged using the LAS-3000 imaging systemand analysis was performed using Multi Gauge V3.0 software(FujiFilm, Japan). Following normalization for total protein(BCA; Pierce) homogenates prepared from U1-3 and I1-3containing 50 μg of total protein were electrophoresed asdescribed above and probed overnight at 4 °C with an antibodyto cleaved caspase-3 (Asp175) (cat. no. 9661; Cell signalingTechnology), washed, incubated with antirabbit HRP for 1 h atRT, washed and developed with ECL-advance (GE Healthcare)before imaging and analysis as described above. Statisticalanalyses were carried out using Minitab 15.1.30.0 statisticalsoftware.NIR-VAD-fmk (Figure 1) was prepared in sterile phosphate buffered saline (PBS) (Gibco – Invitrogen) containing10% v/v high quality (cell culture tested) sterile-filteredDMSO (Sigma-Aldrich). To screen for possible toxicity ofthe new compound, two healthy 5-week-old Tga20 mice wereadministered a 100 μL injection of a 0.5 mM NIR-VAD-fmksolution via the lateral tail vein. They were monitored for 2 himmediately following treatment, and thereafter daily for1 week, whereupon a second dose was administered.In Vivo Active Caspase DetectionFifty-four days after inoculation, three prion-infected (I1,I2, I3) and three sham-inoculated Tga20 mice (U1, U2, U3)were administered 200 μL of a 0.25 mM NIR-VAD-fmksolution via the lateral tail vein. To assess possible contributionsfrom nonspecific autofluorescence of diseased tissue, an additional two prion-infected Tga20 mice (I4, I5) received 200 μL ofinactive PBS/DMSO diluent only. PBS was used in preference toa NIR labeled version of the well-known inactive substrate FAfmk. Although inactive against caspases, FA-fmk inhibits cathepsins (17), whose activities are up-regulated in prion-infectedmouse neuronal N2a cells (18). Moreover, it is not possible topredict whether a fluorescent NIR-FA-fmk analogue willexhibit blood-brain-barrier permeability comparable withNIR-VAD-fmk. Prior to imaging, NIR-VAD-fmk was permittedto circulate for 30 min following injection to allow circulationand excretion of any unbound probe. Mice were subsequentlyanaesthetised by the intraperitoneal administration of ketamine(80 mg kg-1)/xylazine (10 mg kg-1) and their heads shaved inpreparation for fluorescence reflectance imaging.Anesthetized mice were transferred to the dark box of aLAS-3000 imaging system (FujiFilm, Japan). Fluorescencereflectance images (16 bit, 36.7 Â 36.7 μm pixel size) wereobtained using a 710 nm LED epi-illuminator and a 785 nmlong-pass filter with an exposure time of 5 s and an aperture ofF0.85. These were superimposed upon monochrome imagesobtained using a white epi-illuminator with an exposure timeof 16.7 ms and an aperture of F2.8. Image analysis wasperformed using Multi Gauge V3.0 software (FujiFilm,Japan). Statistical analyses were carried out using Minitab15.1.30.0 statistical software.At the conclusion of imaging anaesthetized mice wereeuthanised by cardiac perfusion with PBS and brains sectioned through the sagittal plane. One half was snap frozen inliquid nitrogen and stored at -80 °C and the other half fixed in10% neutral buffered formalin (NBF).ImmunohistochemistryBrain hemispheres fixed in 10% NBF were immersed in 99%formic acid for 1 h prior to routine processing and immunostaining. Half-brains were paraffin embedded and 7 μm sectionscut and mounted on glass slides (Superfrost plus; Thermo).Sections were deparaffinized and rehydrated through a gradedethanol series to deionized water and stained with hematoxylinand eosin or immunostained with ICSM18 anti-PrP monoclonal antibody (D-Gen Ltd., London, UK) (20). For immunostaining, rehydrated sections were autoclaved for 20 min at132 °C and once cooled, washed in deionized water, exposed to4 M guanidine thiocyanate for 2 h at 4 °C, washed and treatedwith 96% formic acid for 5 min. After blocking with 20% fetalbovine serum for 30 min, sections were incubated with ICSM18overnight at 4 °C. Sections were then processed using a avidinbiotin immunohistochemical process (LSABþ; Dako) anddeveloped using diaminobenzadine (DABþ; DAKO) andcounterstained with hematoxylin.Results and DiscussionThe ability of labeled VAD-fmk compounds to detectactive caspases is well characterized (21), and their specificity for labeling apoptotic neuronal cells in live mice hasbeen clearly established (3). Nevertheless, prior to testingthe new NIR-VAD-fmk contrast agent in vivo, we firstexamined its toxicity and cell permeability in vitro using amouse neuronal (OBL-21) cell line. No significant reduction in cell viability (measured by MTS reduction) wasseen in response to treatments ranging in concentration by4 orders of magnitude, even after 3 days constant exposure to 75 μM. (This exceeds the nominal NIR-VADfmk blood concentration of 55.5 μM in mice, based upona typical blood volume of 800 μL and a 200 μL injection of0.25 mM NIR-VAD-fmk.) NIR-VAD-fmk (Figure S2,Supporting Information). To assess the cell permeabilityand the specificity of the NIR-VAD-fmk binding in response to apoptotic stimulus, the compound was appliedto the OBL-21 mouse neuronal cells following 10 minWestern BlottingBrain homogenates (20% w/v) were prepared in PBS usinga FastPrep Homogenizer (ThermoSavant) using a single cycle ofhomogenization before being further diluted to 6% (w/v) in PBSand final concentration of 0.1% SDS (to aid digestion of PrPC)and treated with proteinase K (100 μg mL-1, 1 h, 37 °C). PKdigestion was stopped by the addition of PefaBloc SC (Roche;4 mM) and 4 Â LDS loading dye (containing 12% v/v betamercaptoethanol) and heated to 100 °C for 10 min. Sampleswere electrophoresed on a 12% (bis/tris) gel (NuPAGE Invitrogen) at 200 V for 50 min in MES electrophoresis buffer (Invitrogen) and transferred to a PVDF membrane at 85 V for 60 min inTris/glycine transfer buffer (25 mM Tris, 200 mM glycine, 20%r 2010 American Chemical Society722DOI: 10.1021/cn100068x |ACS Chem. Neurosci. (2010), 1, 720727Letterpubs.acs.org/acschemicalneuroscienceFigure 2. NIR-VAD-fmk localization of caspase activation in OBL-21 mouse neuronal cells (a-c) 24 h post exposure to 10 min UVirradiation. (d-f) untreated control. After 24 h, cells were incubated with 15 μM NIR-VAD-fmk for 30 min, washed and imaged byfluorescence microscopy at 60Â magnification. Scale bar = 20 μm.yielded a negligible autofluorescent signal (Figure S5,Supporting Information), indicating that the increasedfluorescence was not attributable to accumulation oflipofuscin in prion disease (9).Following imaging, animals were euthanized and thebrain pathology was examined using traditional markersof disease neuropathology. Consistent with the in vivoobservation of NIR-VAD-fmk fluorescence in prioninfected mice, Western immunoblot analysis showed asignificant increase in cleaved caspase-3 (a central effectorcaspase) in prion-infected mice (Figure 5). Western immunoblot analysis confirmed the presence of PrPSc inprion infected mice and absence in sham-inoculated controls (Figure 6a). Hematoxylin and eosin staining wasconsistent with the preclinical state of the animals withlittle to no vacuolation apparent in the hippocampus,thalamus, pons or occipital cortex of infected mice I1, I2,and I3 that received NIR-VAD-fmk and mild to moderatevacuolation in infected mice I4 and I5 that by chancereceived the vehicle (Figure 6b). PrPSc deposition wassimilarly mild or undetectable in all prion infected mice(Figure 6).In hippocampal CA2 neurons, apoptosis has beenshown to occur prior to the accumulation of PrPSc andthe onset of clinical disease in a mouse-passaged 87 Vscrapie strain (2). However, confirmation of apoptosisrequired culling the mice to carry out terminal deoxynucleotidyl transferase-mediated uridine triphosphatenick end labeling (TUNEL) staining to show DNAfragmentation within the brain. In this study, we havenoninvasively measured caspase activation associatedwith apoptotic neuronal cell death in live prion-infectedmice 54 days postinoculation prior to development ofdisease (16). NIR imaging of molecular processes suchas caspase and other protease activities at the preclinicalstage promises to provide more valuable mechanisticultraviolet (UV) irradiation. UV exposure initiates apoptosis via an intrinsic cell death pathway involving mitchondrial release of cytochrome c, activation of the initiatorcaspase 9 and subsequent activation effector caspases3 and 6 (22). In comparison with the nonirradiated control,the UV irradiated cells displayed strong intracellular NIRfluorescence (Figure 2), demonstrating the cell-permeability of NIR-VAD-fmk and its ability to bind to activecaspases. Binding of the probe could be observed withinhours of UV insult (Figure S3, Supporting Information),and at 24 h a large number of cells could be demarcated byNIR-VAD-fmk binding (Figure S4, Supporting Information). Untreated OBL-21 cells showed only a lowlevel of intracellular binding, consistent with basal levels ofactive caspases associated with normal cellular processes,such as cell division and differentiation (23).Having established that NIR-VAD-fmk was nontoxic and able to permeate and bind to active caspasesin cultured cells, we examined its ability to label activecaspases in live animals using a murine model of priondisease. In healthy controls, no signs of toxicity oradverse reaction was observed following administrationof a single dose of NIR-VAD-fmk, nor after a seconddose one week later. NIR-VAD-fmk was then administered to Tga20 mice 54 days post inoculation withM1000 prions or a sham inoculation (Figure 3). At thistime there was no evidence of weight loss or overtsigns of clinical prion disease, which occurs approximately 60 days post inoculation in this transgenic mouseline. Nevertheless, quantification of NIR-VAD-fmkbinding by fluorescence reflectance imaging revealedincreased levels of active caspases in the brains of prioninfected mice compared with mice that received anintracerebral inoculation with uninfected brain homogenate (Figure 4). Prion-infected mice that received onlythe DMSO/PBS vehicle instead of NIR-VAD-fmkr 2010 American Chemical Society723DOI: 10.1021/cn100068x |ACS Chem. Neurosci. (2010), 1, 720727Letterpubs.acs.org/acschemicalneuroscienceFigure 3. Overlay of in vivo fluorescence and white-light images of (a-c) prion-infected (I1-I3) and (d-f) sham inoculated (U1-U3) Tga20mice, following administration of a NIR-VAD-fmk dose, 54 days postinoculation.Figure 4. Quantification of the images shown in Figure 3. Prioninfected mice that received the NIR-VAD-fmk exhibited a significantly higher fluorescence emission compared with the sham-inoculated healthy (uninfected) controls (Mann-Whitney, onetailed, W = 15.0, p = 0.040, n = 3). Black squares represent theaverage fluorescence intensity per unit area of each NIR image.Horizontal bars denote the group means.information about pathophysiological pathways involved in neurodegeneration compared with diagnosticimaging modalities that focus on detection of late-stagemarkers of disease such as amyloid formation.A number of other NIR contrast agents for detectingcell death have been developed in recent years. For example, a NIR (Cy5.5) fluorophore has been conjugated toannexin-V to detect externalization of phosphatidylserineat the cell membrane during tumor apoptosis (24). NIRVAD-fmk is demonstrably blood-brain-barrier permeablemaking it suitable for brain imaging applications and canbe further refined to detect specific caspases by changingthe caspase recognition sequence. Analogous NIR compounds based upon caspase substrates rather than inhibitorshave also been developed. One example is the enzymeactivatable Cy5.5-DEVD substrate tethered to cell-permeable nanoparticles, which has been shown to detect activecaspases 3 and 7 in cell culture (25). Another involves intracellular delivery of DEVD conjugated to Alexa Fluor 647r 2010 American Chemical SocietyFigure 5. (a) Western immunoblot analysis of brain homogenatesfrom prion-infected (I1-I3) and sham-inoculated (uninfected)(U1-U3) mice administered NIR-VAD-fmk. Molecular weightmarkers are shown. Consistent with in vivo imaging, quantificationof immunoreactivity against cleaved caspase-3 (b) indicated increasedlevels in prion-infected mice (Mann-Whitney, one-tailed, W =15.0, p = 0.040, n = 3). Horizontal bars denote the group means.and QSY 21 dyes via a cell-penetrating KKKRKV peptidesequence, which was demonstrated to detect NMDAinduced apoptosis of retinal ganglion cells following intravitreal injection in live rats (26). However, the bloodbrain-barrier permeability of such compounds for brainimaging applications remains unknown. Moreover, suchcompounds do not bind irreversibly to active caspases andtherefore must rely on self-quenching until the substrateis cleaved. Depending upon disease pathophysiology,the kinetics of uptake of the uncleaved (nonfluorescent)substrate and clearance of the cleaved (fluorescent) substrate may vary between healthy and diseased individuals,724DOI: 10.1021/cn100068x |ACS Chem. Neurosci. (2010), 1, 720727Letterpubs.acs.org/acschemicalneuroscienceFigure 6. Pathology associated with prion-infected Tga20 mice. (a) PK-treated (100 μg mL-1, 1 h, 37 °C) homogenates prepared from shaminoculated (U1-U3) and prion-infected (I1-I5) Tga20 mice were Western immunoblotted with 03R19. Homogenates prepared from a Tga20mouse with clinical disease are shown before (-) and after (þ) PK treatment for comparison. Molecular weight markers (kDa) are shown.(b) Immunostaining of PrPSc using ICSM18 and (c) hematoxylin and eosin staining of sham inoculated (U3) or prion infected mice with lowPrPSc (I3) and high PrPSc (I4) load. Images show the CA1 region of the hippocampus (HC), thalamus (Th), Pons and occipital cortex (OC).Images were taken at Â20 and Â40 magnification. Scale bars shown are 50 μm.target peptide sequence (21), while additional improvements in depth resolution and spatial localization maybe achieved using a fluorescence tomographic imagingapproach (6, 8, 27).making it difficult to compare relative concentrationsmeasured at the time of imaging. The use of the NIRVAD-fmk caspase inhibitor, rather than a substrate, can…

Order your essay today and save 30% with the discount code: ESSAYHELP
Order your essay today and save 30% with the discount code: ESSAYHELPOrder Now