Recent Studies of Ovine Neuronal Ceroid Lipofuscinoses from BARN, the Batten Animal Research Network☆

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Biochimica et Biophysica Acta 1852 (2015) 2279–2286

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta

journal homepage: www.elsevier.com/locate/bbadis

Recent studies of ovine neuronal ceroid lipofuscinoses from BARN, the Batten Animal Research Network☆

David N. Palmer a,e,⁎, Nicole J. Neverman b,e, Jarol Z. Chen a,e, Chia-Tien Chang c,e, Peter J. Houweling d,e, Lucy A. Barry a,e, Imke Tammen c,e, Stephanie M. Hughes b,e, Nadia L. Mitchell a,e

a Molecular Biosciences, Faculty of Agriculture and Life Sciences, PO Box 85084, Lincoln University, Lincoln 7647, New Zealand b Department of Biochemistry, Otago School of Medical Sciences, Brain Health Research Centre, University of Otago, PO Box 56, Dunedin 9054, New Zealand c Faculty of Veterinary Science, University of Sydney, PMB 4003, Narellan, NSW 2567, Australia d Murdoch Childrens Research Institute, Royal Children's Hospital, Flemington Road, Parkville, Melbourne, Victoria 3052, Australia e Batten Animal Research Network (BARN) (www.BARN.org.nz)

article info abstract

Article history: Studies on naturally occurring New Zealand and Australian ovine models of the neuronal ceroid-lipofuscinoses Received 19 March 2015 (Batten disease, NCLs) have greatly aided our understanding of these diseases. Close collaborations between Received in revised form 26 May 2015 the New Zealand groups at Lincoln University and the University of Otago, Dunedin, and a group at the University Accepted 8 June 2015 of Sydney, Australia, led to the formation of BARN, the Batten Animal Research Network. This review focusses on Available online 12 June 2015 presentations at the 14th International Conference on Neuronal Ceroid Lipofuscinoses (Batten Disease), recent

Keywords: relevant background work, and previews of work in preparation for publication. Themes include CLN5 and Neuronal ceroid lipofuscinoses CLN6 neuronal cell culture studies, studies on tissues from affected and control animals and whole animal NCLs in vivo studies. Topics include the effect of a CLN6 mutation on endoplasmic reticulum proteins, lysosomal func- Batten Disease tion and the interactions of CLN6 with other lysosomal activities and trafficking, scoping gene-based therapies, a CLN5 molecular dissection of neuroinflammation, identification of differentially expressed genes in brain tissue, an CLN6 attempted therapy with an anti-inflammatory drug in vivo and work towards gene therapy in ovine models of Animal models the NCLs. This article is part of a Special Issue entitled: "Current Research on the Neuronal Ceroid Lipofuscinoses fl Neuroin ammation (Batten Disease)". Neurodegeneration © 2015 Elsevier B.V. All rights reserved. Gene therapy Neural cultures Neuroinflammatory pathway Disease associated cardiology

1. Introduction under Australian and New Zealand pasture management. Management of sheep reproduction is well understood in our environments as are Studies on naturally occurring ovine models of the neuronal ceroid- the other husbandry skills for maintaining ﬂocks in good condition. lipofuscinoses (Batten disease, NCLs) have been central to our current Veterinary care and animal science advice are easily accessible. These understanding of many aspects of this group of devastating childhood three ovine models exhibit close clinical and pathological mimicry to neurodegenerative diseases. Currently mutations in 13 different genes the relevant human NCLs. Anatomically, sheep have a human-like are described that lead to different forms [NCL mutation database; gyrencephalic brain which, at 140 grammes by maturity, is larger than http://www.ucl.ac.uk/ncl, [1]], but there may be more. Of these, muta- that of many non-human primates, one-tenth that of a human adult tions in CLN6 lead to NCLs in New Zealand South Hampshire and and more comparable to that of a neonate or small child. Australian Merino sheep [2,3], and in CLN5 to an NCL in New Zealand Flocks of heterozygous ewes of each genotype have been established Borderdale sheep [4,5]. Sheep have been domesticated for about and are mated with heterozygous or homozygous affected rams to pro- 10,000 years, are bred for their amenability to human handling, mother- duce affected (CLN−/−), heterozygous (CLN+/−) and homozygous nor- ing properties and related factors that make them easy care, especially mal (CLN+/+) lambs. These sheep have been subject to much research over the last 40 years which has led to many advances in our understanding of Batten disease, reviewed in a number of publications ☆ This article is part of a Special Issue entitled: "Current Research on the Neuronal Ceroid [6–10]. Highlights include determining the nature of the so-called Lipofuscinoses (Batten Disease)". “lipofuscin-like” storage material, initially in what became the CLN6 ⁎ Corresponding author at: Molecular Biosciences, Faculty of Agriculture and Life Sciences, PO Box 85084, Lincoln University, Lincoln 7647, New Zealand. form, later extended to the CLN5 ovine and bovine forms and many E-mail address: [email protected] (D.N. Palmer). human and murine forms; determining the regional nature of cortical

http://dx.doi.org/10.1016/j.bbadis.2015.06.013 0925-4439/© 2015 Elsevier B.V. All rights reserved. 2280 D.N. Palmer et al. / Biochimica et Biophysica Acta 1852 (2015) 2279–2286 atrophy and its close association with neuroinflammation, developing bodies being removed in an autophagic process in the 6 days following neuron culture cell biology studies, and trialling therapeutic approaches. transduction [14]. Close collaborations have been a powerful driver, principally between Professor David Palmer's group at Lincoln University, New 2.2. Effect of a CLN6 mutation on endoplasmic reticulum proteins Zealand, where the CLN5 Borderdale and CLN6 South Hampshire flocks are housed; the group led by Dr Stephanie Hughes at the University of CLN6 is thought to be an endoplasmic reticulum (ER) resident pro- Otago, Dunedin, New Zealand who works mainly with neural cultures, tein [17,18] so its effect on other ER proteins was analysed using immu- viral vector and gene therapy development and have a colony of nclf nocytochemistry for a soluble ER marker protein (protein disulphide CLN6 affected mice, and that led by Associate Professor Imke Tammen isomerase, PDI) and a membrane-associated ER marker (calnexin). in Sydney, Australia, where the CLN6 affected Merinos reside. Long- Whilst PDI is present in both control and CLN6−/− cultures, calnexin is term associations with Professors Jon Cooper at Kings College, London; significantly reduced in affected cells (Fig. 1), perhaps indicating a de- Sir John Walker, MRC Mitochondrial Biology Unit, Cambridge and Bob crease in expression, or misprocessing or folding in the absence of func- Jolly and colleagues at Massey University, New Zealand are also invalu- tional CLN6. Further studies are required to resolve these possibilities. able. In order to make our collective intent clear we have formed BARN, the Batten Animal Research Network, which has proved to be an invaluable brand for planning and resourcing coordinated studies and financ- 2.3. Interactions of CLN6 with other lysosomal activities and trafficking ing them. Recent BARN initiatives were presented at a dedicated BARN session at the 14th International Conference on Neuronal Ceroid The broadly similar pathological and clinical profiles of the different Lipofuscinoses (Batten Disease) and Patients Organization Meeting in NCLs suggest that common molecular pathways might link the NCL dis- Córdoba, Argentina (October 22–26, 2014) and other BARN work was ease variants together, despite the differing primary genetic defects. presented in concurrent sessions at the meeting. This review focusses Several studies support the hypothesis that CLN-proteins interact and on those presentations, including relevant background work completed are involved in a common pathway but solid evidence for molecular since the last reviews of sheep model studies [9–11], some unpublished links between the NCLs is still scant, and in some cases the evidence is data and previews some of our current studies on ovine NCLs. contradictory. A co-localisation/co-immunoprecipitation study reported that CLN2, CLN3, CLN6, and CLN8 proteins co-localise and co- immunoprecipitate, and that the addition of the CLN2 protein (TPP1) 2. In vitro cell culture studies to cell lines deficient in CLN3, 6, or 8 corrected growth and apoptosis [21]. Another pull-down and co-immunoprecipitation study found evi- A method for culturing embryonic ovine neural cells, based on those dence to suggest that CLN5 is a central interaction partner and may play used for rodent cell cultures, was established over a decade ago [12]. an important role in connecting different NCL (CLN) proteins [22]. CLN5 Primary neural cells have been recovered from ~60-day old (and occa- was found to interact with CLN1, CLN2, CLN3, CLN6, and CLN8. These sionally up to 120 day old) foetal brains by tryptic dissociation from ho- data also suggest that CLN5 has the capacity to bind proteins from dif- mozygous CLN5 affected Borderdales, CLN6 affected South Hampshires, − ferent subcellular compartments as both lysosomal and ER proteins heterozygous CLN+/ sheep of both genotypes and control sheep, and are suggested as CLN5 binding partners. stored frozen [13,14]. These ovine primary cell cultures, consisting of a Although the function of CLN6 is yet to be elucidated, given the mixed population of neural precursors (neuroblasts), mature neurons, above data and the ER localisation of CLN6 protein, it is sensible to astrocytes and microglia [13], have been extensively characterised and hypothesise a role for CLN6 in protein trafficking pathways, particularly are used as in vitro cell biology models to investigate key issues in in a network involving other NCL-associated proteins and enzymes. It CLN5 and CLN6 disease pathology, as well as providing assay systems has been suggested that CLN6 plays a role in endocytosis of lysosomal to develop and test novel therapeutics. proteins. The endocytic pathway is reported to be reduced in neuronal cells of vLINCL patients [17,18], as is the synthesis rate of mutant 2.1. Storage bodies in cultured neural cells CLN6. It is also hypothesised that CLN6 mediates selective transport of lipids and proteins essential for the function and acidification of the ly- Once established from thawed cells, ovine CLN5 and CLN6 neural sosome [18]. This hypothesis arises from the reported increase in lyso- cells can be maintained in culture routinely for three to four weeks, somal pH in fibroblasts from a CLN6 patient [23]. although maintenance up to a few months has been achieved, by These hypotheses led to an investigation of the effect of the ovine judicious use of conditions to stop glial cell take-over of the cultures. A South Hampshire CLN6 mutation on NCL-associated soluble lysosomal high proportion of cells adhere as neurons [13]. These cultures have enzyme activities to determine whether this alters the activity of vari- been used to unravel a number of important changes in CLN6−/− cul- ous NCL-associated enzymes. Three NCL-associated lysosomal enzymes tured neurons, including time-dependent accumulation of subunit c were investigated; CLN1, CLN2, and CLN10, and a non-NCL associated containing fluorescent storage bodies [15] and a loss of GABAergic neu- lysosomal enzyme, β-hexosaminidase, was also included. Palmitoyl rons [16]. They provide a convenient platform in which to investigate protein thioesterase 1 (PPT1, CLN1), tripeptidyl proteinase 1 (TPP1, prenatal synaptic pathologies in ovine CLN5 and CLN6 NCLs, the mech- CLN2) and the lysosomal β-hexosaminidase enzymatic rates were com- anisms by which these alterations occur, and in which novel drug comparable across CLN6−/− and control derived neural cell lysates (Fig. 2), pounds and viral-mediated gene overexpression and correction indicating that this CLN6 mutation does not disrupt the trafficking of techniques can be investigated for effectiveness at correcting the these lysosomal enzymes to the lysosome. They are able to be processed numerous identified synaptic and lysosomal pathologies before pro- and trafficked correctly in the absence of functional CLN6 protein. How- gression into in vivo trials. ever, CLN6−/− primary cell lysates demonstrated a significant, albeit Recent experiments showed that accumulation of storage material small, decrease in Cathepsin D enzymatic activity, as compared to con- in CLN5−/− cultures was reversed following lentiviral-mediated over- trol cell lysates (Fig. 2c). expression of C-terminal myc-tagged CLN5 [14]. Dissociated mixed neu- These results indicate functional processing and trafficking of these ral cells from CLN5−/− foetal sheep brains had no obvious storage three lysosomal enzymes within CLN6−/− cells during the neural cell material at plating; however, storage material was obvious and wide- culture maintenance period. If these pathways were not functional the spread within microglia, neurons and astrocytes a week later. Transduc- enzyme activity would be greatly reduced, and the proteins must tion with the CLN5-containing lentiviral vector after this week in culture reach the lysosome in their fully modified and folded mature form in rapidly reversed the storage accumulation, more than half the storage order to function as enzymes. This would not be so if the CLN6 protein D.N. Palmer et al. / Biochimica et Biophysica Acta 1852 (2015) 2279–2286 2281

Fig. 1. Calnexin immunoreactivity is reduced in CLN6−/− cultures. Primary ovine neural cell cultures (embryonic day 68–70) were plated at 2 × 105 cells/cm2 as described [13,19] and fixed in 4% phosphate-buffered paraformaldehyde eight days later. a–b Calnexin was detected with an anti-calnexin antibody (GeneTex, USA #GTX109669; 1:200) and Alexa Fluor 488 goat anti-rabbit IgG (Life Technologies, USA, A-11008, 1:1000). Cells were counterstained with DAPI for nuclei (blue), scale bar 20 μm; or c–d, protein disulphide isomerase (PDI) using anti- bodies raised against human PDI [20], scale bar 100 μm. Single 0.2 μm confocal slices were acquired on a Zeiss 510 LSM confocal microscope (Zeiss, Germany). Note the decrease in anti- calnexin immunofluorescent intensity in CLN6−/− cultures compared to controls. Average fluorescence intensities were measured using the Measure RGB macro in Image J (NIH, USA). Five fields were analysed from three individual foetal preparations per genotype. The calnexin data in e) are significantly different, p b 0.05, whereas there is no significant difference between the PDI signals in f. is important in a common interaction or network involved in trafficking in when possible. Collections include perfusion fixed brain hemispheres of the newly synthesised proteins to their correct cellular destination. from CLN5 and CLN6 affected and age-matched control sheep at all stages of disease progression, cryopreserved, awaiting cryosectioning 2.4. In vitro testing of gene-based therapies or cut (sagittally and coronally) and catalogued ready for immunostain- ing. These include the cerebellum. Studies comparing the longitudinal The neural cell cultures are also extremely useful for scoping the pathology of all three genotypes (Borderdale CLN5, and the CLN6 muta- possibilities for gene based therapies. Lentiviruses have proven to be tions in South Hampshire and Merinos) are underway. Results so far in- versatile gene delivery vectors in many lysosomal storage diseases dicate that the courses of these diseases are remarkably similar, [24–26], and can transduce a wide range of cell types, integrating into especially so given the diversity of the underlying genetic defects. the host genome of both dividing and post-mitotic cells and resulting Also on hand is a collection of fresh frozen brain blocks for the isola- in long-term stable expression of the transgene. To explore their useful- tion of protein and total RNA/mRNA from selected brain regions at all ness in NCL gene therapy, lentiviral-mediated transduction efficiency, stages of disease. These are 6 mm thick sagittal sections, cut immediate- transgenic expression, and cell tropism were tested in cultures from ly post mortem, snap frozen in liquid nitrogen and stored at −80 °C after both affected South Hampshire (CLN6) and Borderdale (CLN5) sheep. being catalogued. Specific brain regions are dissected from these slabs Lentiviral vectors were pseudotyped with the vesticular stomatitis for RNA or protein extraction. Complementary DNA (cDNA) is generat- virus glycoprotein, VSV-g, with transgene expression under the control ed via reverse transcriptase (RT) reactions and then stored at −80 for of a constitutive myeloproliferative sarcoma virus promoter to prefer- future quantitative PCR analyses. entially target neurons, as tested in other model organisms [27]. Sheep are routinely diagnosed from blood samples taken at weaning Delivery of a green fluorescent protein (GFP) reporter gene was demon- or at the time of euthanasia [2,5], which are then stored as dry spots strated in both control and CLN6 affected neural cultures, with a greater catalogued for DNA extraction. Fresh, frozen and fixed samples of tropism for neuronal and neuroblast cells than glia [19]. This, along with various visceral tissues and organs, including eyes, complete these the significant diminution of storage body fluorescence observed in collections. CLN5-deficient neural cells after transduction with a CLN5 expressing lentiviral vector noted above [14], supports the use of lentiviral mediated gene transfer in ovine NCL in vivo. 3.1. A molecular dissection of neuroinflammation

3. Comparative CNS and visceral histopathology, Following earlier histological investigations, glial activation immunohistopathology and tissue extract studies characterised by astrocytosis and microglial activation has long been considered to be a generalised response in the NCLs [7,8,28].Thiswas Over the years we have been accumulating and replenishing stocks extended in comprehensive immunohistochemical studies using the for various tissues and extracts from the CLN5 and CLN6 affected ovine CLN6 sagittal sections above to show that glial activation begins sheep and controls at different stages of disease development. These perinatally, proceeds regionally, starting in regions associated with allow longitudinal comparisons of various changes to be made. Ideally early clinical symptoms, and precedes neurodegeneration [29–31]. tissues and extracts from three animals at each age (birth, 1, 2, 4, 6, 9, The parieto-occipital cortex is affected first, before neuroinflammation 12, 15, 18 months), as well as at an advanced disease (18 months and and subsequent neurodegeneration becomes more generalised within older) are retained, but sometimes there are shortfalls, which we fill other cortical regions. The cerebellum remains largely unaffected, 2282 D.N. Palmer et al. / Biochimica et Biophysica Acta 1852 (2015) 2279–2286

Fig. 2. The enzymatic activities of PPT1 (CLN1), TPP1 (CLN2), Cathepsin D (CLN10) and β-hexosaminidase in control and CLN6−/− cultured neural cell lysates. Sheep neural cell cultures from 3 control and 3 CLN6−/− foetuses were harvested into lysis buffer containing 0.1% Triton-X, 0.15 M NaCl, and protease inhibitors, aprotinin, leupeptin, pepstatin and PMSF. a. PPT1/ CLN1 enzymatic rates were measured (Moscerdam Substrates, the Netherlands), in samples containing 28 mg/mL protein from the hydrolysis of 4-methylumbelliferyl-6-thiopalmitoyl- −/− beta-D-glucoside using a 4-MU standard curve. Enzymatic rates did not differ significantly between control and CLN6 neural cell lysates, b. For TPP1/CLN2 activities, 100 mg/mL total protein lysate was incubated in citrate buffer, pH 4.5, at 37 °C for 60 min. to allow activation of TPP1, then 5 μM Ala-Ala-Phe 7-amido-4-methylcoumarin (AAF-AMC) added, fluorescent readings taken every 2 min. for 40 min. at 37 °C with 380 nm excitation and 460 nm emission and TTP1 activities quantified by comparison to an AMC standard curve. There was no significant difference in activities between control and CLN6−/− neural cell lysates. c. Cathepsin D/CLN10 assayed fluorometrically (Cathepsin D activity assay kit, Abcam, UK; ab65302), was slightly lower in affected cells than in controls, p b 0.05. d. β-hexosaminidase activity determined following a two-hour incubation of 25 mg/mL neural cell lysate with 25 μg/mL 4- methylumbelliferyl N-acetyl-β-D-glucosaminide dehydrate (4-MUGS) did not differ significantly between genotypes. In all data sets, the black bar values relate to individual foetal samples of each genotype, and red bars show the genotype mean value (+/−SEM).

Fig. 3. A proposed neuroinflammatory pathway in ovine CLN6 Batten disease, based on the literature [39–46]. D.N. Palmer et al. / Biochimica et Biophysica Acta 1852 (2015) 2279–2286 2283 even at end-stage disease. There was no neuroinflammation in the thal- 18 being up-regulated and one down-regulated in the affected animals. amus or hypothalamus [32]. When the gene expressions in the frontal regions of two affected sheep From these findings a hypothesis was developed, that since neuroin- at 3 months were compared to two 14 month old affected sheep, 135 flammation precedes neurodegeneration, it may be causative of it and genes were up-regulated in the affected animals, and 65 down- drug inhibition or suppression of the neuroinflammatory pathway regulated. Preliminary gene ontology analysis associated differential might provide a therapy. Whole animal experiments revealed that gene expression patterns with the induction of inflammation, cell targeting is required (see 5.2 below). Subsequently a molecular dissection surface receptor linked signal transduction, cellular and metal ion trans- of the neuroinflammatory cascade was initiated, to determine the points port, tissue remodelling (activation and inactivation of genes associated of lesion in the NCLs for effective and target-specific drug intervention with actin, myotubules, cell-adhesion) and apoptosis. [33]. This was based on a proposed neuroinflammatory pathway (Fig. 3) This study indicates that RNA-seq is a useful tool for studying disease which includes components of pro- and anti-inflammations, oxidative mechanisms in the NCLs, highlighting the diverse molecular changes ac- stress, neuronal survival via mediation of neurotrophic factors and companying neurodegeneration and the inflammatory mechanisms breaching of the blood–brain-barrier. that play an important role in disease progression. How variations in tis- The expression of a number of neuroinflammatory modulators sue storage times influence RNA integrity is a concern in longitudinal (TNF-α, TGF-β,IL-1β, IL-10, NF-ƙB, MnSOD, iNOS, JAK2, STAT 1 and 3, transcriptomic studies. Uncertainty was minimized in this study as all SOCS3, TrkB and BDNF) was investigated in different brain regions at samples were stored for similar lengths of time from sheep born over different ages of disease development [35]. Expressions of other genes two consecutive years. Future studies will use greater read depth that have been proposed to be involved in the pathogenesis of NCLs, RNA-seq analyses in tissues from all our ovine NCL and cell culture such as metallothioneins and progranulin [36–38], were also selected models, to develop a better understanding of the transcript profile of for investigation. Different inflammatory mediators were expressed CLN5 and CLN6, increase our understanding of the pathogenic cascade, differently in affected animals. Both pro- (TNF-α and IL-1β)and and evaluate the molecular effects of gene therapy agents. anti-inflammatory cytokines (TGF-β and IL-10) were up-regulated and suppression of cytokine signalling mediated by SOCS3 was activat- 3.3. The role of metals in NCL pathogenesis ed dramatically at the initiation of neurodegeneration. NF-ƙBandJAK2 activation followed whereas the oxidative responsive genes MnSOD BARN also provides tissues and organs for others to pursue indepen- and iNOS were not activated. TrkB expression increased at the advanced dent lines of enquiry. Recently there has been a renewed interest in the disease whilst BDNF expression remained unchanged. role of biometal disturbances, as they are now called, in the NCLs. Tis- Only SOCS3 was unequivocally upregulated prior to neuroinflamma- sues from both the Merino and South Hampshire sheep have been tion and many of the other changes in gene expression profiles are analysed for the possible involvement of disruptions of biometals in unlikely to be causative. This study also highlights the weakness of the pathogenesis of disease. Increased zinc and manganese concentra- building scenarios based on observed changes in single gene expres- tions in parallel with neurodegeneration, synaptic protein changes sions in isolation. A similar study of CLN5 ovine Batten disease under- and activation of Akt/GSK3 signalling in ovine CLN6 neuronal ceroid way will indicate the generality of these findings. A change of strategy lipofuscinosis was reported [37]. Deregulation of subcellular biometal is planned to continue these studies. Given the large number of changes homeostasis through loss of the metal transporter, Zip7, in CLN6 vari- observed, a transcriptomic approach is probably more appropriate. ants was also reported and a direct link between an apparent loss of control of biometal homeostasis and pathogenesis claimed [52]. 3.2. Transcriptomics These claims of links between changes in metal concentrations and pathogenesis should be treated with caution. An analysis of the concen- A preliminary study using RNA-sequencing (RNA-seq) has been con- trations of a wide range of metals in storage bodies from different tis- ducted to determine if a whole transcriptomic approach has the ability sues in CLN6 affected sheep indicated no such links, and that the to identify the principal/primary molecular changes during early dis- metals arise from the storage bodies having a history of active regula- ease and its progression [47]. RNA-seq is a next generation sequencing tion of metal homeostasis as part of their histories as active lysosome method that allows the quantification of gene expression as well as derived organelles [53]. the identification of alternative splicing isoforms and transcript start sites [48–50]. In this pilot study, eight brain samples from CLN6 affected 4. Whole animal in vivo studies Merino and control sheep were used to investigate differential gene expression between normal and affected frontal brain tissues at three and The flocks of animals at Lincoln and in Sydney provide 10–20 affect- fourteen months of age. Tissue samples collected at euthanasia were ed lambs of each breed each year, which are harvested at different times stored in RNAlater solution (Life Technologies, USA) at −80 °C for 16 for the in vitro and post mortem studies described above. They are also an to 17 months. RNA was extracted with TRIzol (Life Technologies), invaluable resource for whole animal in vivo studies. Sheep have excel- followed by on-column DNase treatment with the RNeasy Midi Cleanup lent facial recognition capability, long-term learning and spatial memo- Kit (Qiagen, Germany). RNA integrity was assessed using the 2100 ry skills [54–56] and a relatively long lifespan that can extend to teenage Bioanalyzer (Agilent Technologies, Inc., USA). Library preparation, years. Their anatomy, physiology and genome are more similar to those using the TruSeq Stranded Total RNA with Ribo-Zero Gold (Illumina, of humans than of mice. We have well defined records on the clinical USA), and paired-end sequencing using an Illumina HiSeqTM platform course of the ovine diseases, and have established protocols for perfor- were performed at the Australian Genome Research Facility (http:// mance testing in sheep for longitudinal in vivo assessment of therapeu- www.agrf.org.au). Sequencing reads were aligned to the sheep genome tic efficacy. These include behavioural, cognitive and neurological using TopHat (http://tophat.cbcb.umd.edu/) [49] and analysed with parameters and quantitative computer tomography (CT) and magnetic Cufflinks (http://cufflinks.cbcb.umd.edu). Cuffmerge and Cuffdiff resonance imaging (MRI). (http://cufflinks.cbcb.umd.edu/) [50] were used to identify differentially expressed genes. Genes that showed significant differential expres- 4.1. Behavioural, neurological and physiological studies sion (q-value b 0.05) were analysed further using the Database for Annotation, Visualization and Integrated Discovery (DAVID) 6.7 Biomarkers (sometimes called readouts) are extremely useful for (http://david.abcc.ncifcrf.gov/) [51]. measuring the progress of disease-modifying therapies, be they func- Nineteen genes showed significant differential expression between tional, behavioural or physiological. Practical in vivo biomarkers have the frontal regions of two affected and two normal 3 month old sheep, been established for all three sheep models for assessing therapeutic 2284 D.N. Palmer et al. / Biochimica et Biophysica Acta 1852 (2015) 2279–2286 efficacy. In one study, Australian Merino CLN6 affected, heterozygous 4.3. Gene therapy in ovine models of the NCLs and homozygous normal sheep were compared at 3–5 week inter- vals, between 26 and 60 weeks of age for progressive postural, be- Gene therapy is an exciting prospect for the treatment of lysosomal havioural and live-weight changes [57]. Significant differences storage diseases, including the NCLs. Approaches described in murine were found between weight gains (affected sheep gained weight models of NCLs [69,70] have been extended to clinical trials in humans more slowly) grazing behaviour (affected sheep spent less time for the CLN2 form [71,72]. Because of their closer similarity to humans, grazing), ambulatory behaviour (affected sheep spent more time therapy development in sheep should offer better data than mice; on walking), steps taken per day (affected sheep took more steps), vector titres, doses and target sites and should be more predictive of ac- and responsiveness to visual and auditory stimuli (affected sheep tual clinical outcomes. showed a progressive reduction in responsiveness). A composite variable of key behaviours was identified in a principal components analysis (scores for affected sheep increased more rapidly than con- 4.3.1. Development of gene therapies trols). In the future, frequent monitoring of relevant indicators using Following efficient lentiviral mediated gene transfer to neurons in automatic recording technology, would facilitate the assessment of culture [19], a programme of direct injections into sheep brains in vivo disease progression in therapy trials. was initiated. No transduction of cells followed intracisternal injections Automatic recording technology has already been used to re- of lentiviral GFP constructs in sheep, however columnar infusion into cord electroencephalography (EEG) changes in CLN5 affected the cortical parenchyma resulted in stable yet localised GFP transduc- Borderdale sheep in a longitudinal study spanning many months tion evident up to 80 day post-injection [19]. Of importance was the [58]. Twenty-four hour recordings were collected via surgically im- targeted transduction of ependymal and subependymal cells along the planted electrodes and vigilance states and neurophysiological extent of the ventricular surface, including type B astrocytic cells, hallmarks, such as sleep spindles, quantified. Mild sleep abnormal- thought to be the bona fide adult neural stem cells. These findings ities, decreased amplitude and abnormal epileptiform waveforms were encouraging for human translation. Correction of the ependyma in the EEGs of affected sheep resemble those seen in children lining the ventricles could result in protein secretion into the cerebral with Batten disease, demonstrating the translational relevance of spinal fluid (CSF) for circulation to the entire CNS whilst transduced the NCL sheep models. neuroblasts could provide an alternative route for spread of the gene Electrocardiography has also been explored in ovine NCL. Various product to the degenerating cortex in NCL patients via the chain migra- studies have reported cardiac abnormalities such as repolarization dis- tion of neuroblastic granule cells. turbances, ventricular hypertrophy and sinus node dysfunction in pa- In the next set of experiments, juvenile CLN5 and CLN6 affected tients with various forms of NCL [59–64]. Possible cardiac pathology sheep received lentiviral vectors containing the respective CLN5 was investigated in the CLN6 affected Merino sheep. Electrocardiograms and CLN6 transgenes with attached myc tags, with concurrent deliv- (ECGs) were conducted on affected and clinically normal control sheep ery of GFP-containing vector, injected directly into the brain paren- at approximately 13 and 20 months of age, corresponding to mid-stage chyma and the ventricular system. Animals were pre-clinical when and late-stage diseases. Analysis of the ECG and echocardiographic data injected but the doses were low (100- to 1000-fold less than in did not reveal any abnormalities in the affected sheep, nor was storage subsequent trials). Successful delivery of CLN5 and CLN6 to the material found in their heart valves, myocardium, or the sinoatrial or affected sheep brain was demonstrated; however, no neurologic or atrioventricular nodes by histology. Most sections from all sheep behavioural amelioration resulted; an outcome attributed to the showed some degree of myocardial sarcocystosis which was considered sub-therapeutic vector doses. an incidental finding. This study did not indicate any cardiac involve- Adeno-associated viral (AAV) vectors, now often preferred for gene ment in CLN6 NCL in Merino sheep. therapy studies because of their marked long-term transduction efficiency and spread, have been recently trialled in sheep [73]. A broad transgene distribution throughout the CNS was evident a month after 4.2. Attempted therapy with an anti-inflammatory drug in vivo intraventricular delivery of AAV9-GFP to the normal sheep brain, clus- ters of GFP-positive cells extending from the prefrontal to occipital cor- The finding that neurodegeneration is consequent to neuroin- tices and cerebellum, as well as the full extent of the spinal cord [73].In flammation and likely to be caused by it [29,31], led to a study to contrast, unilateral columnar intraparenchymal infusions of AAV9-GFP determine if early chronic administration of minocycline, an anti- resulted in a more localised GFP distribution, but with evidence of distal inflammatory drug, would halt or reverse the development of transduction. A predominant neurotropism resulted from either route disease [65]. Minocycline is a tetracycline with broad anti- of injection. inflammatory properties [66–68]. It was administered orally to Subsequently therapeutic doses of both lentiviral and AAV9 derived CLN6 affected South Hampshire lambs from three to 14 months of vectors, coding for untagged native proteins, were injected into both the age. The minocycline was absorbed without significant rumen bio- ventricles and brain parenchyma of pre-clinical CLN5 and CLN6 affected transformation to maintain pharmacological concentrations of lambs. All the CLN5 affected injected sheep were indistinguishable from 1 μMinplasmaand400nMincerebrospinalfluid, but these did normal age-matched control sheep at 18 months of age, clinically and not result in inhibition of microglial activation or astrocytosis and using the behavioural tests mentioned above, whilst the uninjected af- did not change the neuronal loss or clinical course of the disease. fected sheep were at end-stage disease. Only one of the six injected This indicates that minocycline does not inhibit a critical step in CLN6 affected sheep has had an outcome similar to the CLN5 injected the neuroinflammatory cascade in the CLN6 form, that the animals. Monitoring of the injected sheep continues. Because of their re- neuroinflammatory cascade is complex and that targeted selection markable health, previous plans to post-mortem these sheep 18 months of candidate drugs is required for this strategy to be effective. How- after injection have been suspended, to allow us to gain an insight into ever the mode of anti-inflammatory action of minocycline is unclear. how long the treatment is effective for. Identification of the critical steps in the neuroinflammatory cascade will greatly increase the likelihood of success this strategy, stimulat- ing a molecular dissection of the neuroinflammatory cascade in Transparency Document these diseases (see Section 4.1). They also highlight the wisdom of animal model studies preceding highly speculative procedures in The Transparency document associated with this article can be humans. found, in the online version. D.N. Palmer et al. / Biochimica et Biophysica Acta 1852 (2015) 2279–2286 2285

Acknowledgements [20] S.D. Shnyder, M.J. Hubbard, ERp29 is a ubiquitous resident of the endoplasmic reticulum with a distinct role in secretory protein production, J. Histochem. Cytochem. 50 (2002) 557–566. We thank the many people who also contributed to the work pre- [21] D.A. Persaud-Sawin, T. Mousallem, C. Wang, A. Zucker, E. Kominami, R.-M. Boustany, sented here. Janet Xu, Katharina Russell, Huibing Jiang, Martin Wellby, Neuronal ceroid lipofuscinosis: a common pathway? Pediatr. Res. 61 (2007) 146–152. [22] A. Lyly, C. von Schantz, C. Heine, M.L. Schmiedt, T. Sipila, A. Jalanko, A. Kyttälä, Novel Amy Smaill and Drs Graham Barrell and Graham Kay; all contributed interactions of CLN5 support molecular networking between neuronal ceroid to the work at Lincoln University whilst Hannah Best, Hollie Wicky lipofuscinosis proteins, BMC Cell Biol. 10 (2009) 83. and Dr Lucia Schoderböck contributed to the work at the University of [23] J.M. Holopainen, J. Saarikoski, P.K.J. Kinnunen, I. Järvelä, Elevated lysosomal pH in – Otago, Dunedin. The PDI antibody was a gift from Professor Michael neuronal ceroid lipofuscinoses (NCLs), Eur. J. Biochem. 268 (2001) 5851 5856. [24] A.I. Brooks, C.S. Stein, S.M. Hughes, J.A. Heth, P.M. McCray Jr., S.L. Sauter, J.C. Johnson, D.A. Hubbard, University of Melbourne. Cory-Slechta, H.J. Federoff, B.L. Davidson, Functional correction of established central Dr Greg Cronin and Dr Peter Thomson, Amanda Sutton and Danai nervous system deficits in an animal model of lysosomal storage disease with feline im- fi – Beganovic performed the Merino behavioural studies and Drs Susan munode ciency based vectors, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 6216 6221. [25] A. Bosch, E. Perret, N. Desmaris, D. Trono, J.M. Heard, Reversal of pathology in the en- Piripi and Niek Beijerink performed the cardiology studies all at the Fac- tire brain of mucopolysaccharidosis type VII mice after lentivirus-mediated gene ulty of Veterinary Science, University of Sydney. David J. Palmer provid- transfer, Hum. Gene Ther. 11 (2000) 1139–1150. ed invaluable technical support to the Merino flock. [26] V. Gieselmann, Metachromatic leukodystrophy: genetics, pathogenesis and therapeutic options, Acta Paediatr. 97 (Suppl. 457) (2008) 15–21. The work was supported by grants from the US National Institutes of [27] J. Jakobsson, C. Ericson, M. Jansson, E. Björk, C. Lundberg, Targeted transgene expres- Health, (NS053559), the Batten Disease Support and Research Associa- sion in rat brain using lentiviral vectors, J. Neurosci. Res. 73 (2003) 876–885. tion, the Neurological Foundation of New Zealand, Cure Kids New [28] R.D. Jolly, A. Shimada, I. Dopfmer, P.M. Slack, M.J. Birtles, D.N. Palmer, Ceroid- lipofuscinosis (Batten's disease): pathogenesis and sequential neuropathological Zealand, Batten Disease New Zealand, Lysosomal Diseases New Zealand, changes in the ovine model, Neuropathol. Appl. Neurobiol. 15 (1989) 371–383. Pub Charities New Zealand, the American New Zealand Association, Lin- [29] M.J. Oswald, D.N. Palmer, G.W. Kay, S.J. Shemilt, P. Rezaie, J.D. Cooper, Glial activa- coln University, Otago Research Committee, Otago Medical Research tion spreads from specific cerebral foci and precedes neurodegeneration in pre- symptomatic ovine neuronal ceroid lipofuscinosis (CLN6), Neurobiol. Dis. 20 Foundation and the University of Sydney. (2005) 49–63. [30] M.J. Oswald, D.N. Palmer, G.W. Kay, K.J. Barwell, J.D. Cooper, Location and connectiv- ity determine GABAergic interneuron survival in the brains of South Hampshire References sheep with CLN6 neuronal ceroid lipofuscinosis, Neurobiol. Dis. 32 (2008) 50–65. [31] G.W. Kay, D.N. Palmer, P. Rezaie, J.D. Cooper, Activation of non-neuronal cells within [1] M. Kousi, A.E. Lehesjoki, S.E. Mole, Update of the mutation spectrum and clinical cor- the prenatal developing brain of sheep with neuronal ceroid lipofuscinosis, Brain relations of over 360 mutations in eight genes that underlie the neuronal ceroid Pathol.16(2006)110–116. lipofuscinoses, Hum. Mutat. 33 (2012) 42–63. [32] G.W. Kay, N.P. Jay, D.N. Palmer, The specific loss of GnRH-positive neurons from the [2] I. Tammen, P.J. Houweling, T. Frugier, N.L. Mitchell, G.W. Kay, J.A.L. Cavanagh, R.W. hypothalamus of sheep with CLN6 neuronal ceroid lipofuscinosis occurs without Cook, H.W. Raadsma, D.N. Palmer, A missense mutation (c. 184C N T) in ovine glial activation and has only minor effects on reproduction, Neurobiol. Dis. 41 CLN6 causes neuronal ceroid lipofuscinosis in Merino sheep whereas affected (2011) 614–623. South Hampshire sheep have reduced levels of CLN6 mRNA, Biochim. Biophys. [34] J. Chen, L.A. Barry, N.L. Mitchell, D.N. Palmer, Progress in the molecular dissection of Acta Mol. basis Dis. 1762 (2006) 898–905. neuroinflammation in ovine Batten disease, Medicina 74 (supl. II) (2014) 17. [3] R.W. Cook, R.D. Jolly, D.N. Palmer, I. Tammen, M.F. Broom, R. McKinnon, Neuronal [35] J. Chen, L.A. Barry, N.L. Mitchell, D.N. Palmer, A molecular dissection of neuroinflam- ceroid lipofuscinosis in Merino sheep, Aust. Vet. J. 80 (2002) 292–297. mation in ovine NCLs, Manuscript in preparation. [4] R.D. Jolly, D.G. Arthur, G.W. Kay, D.N. Palmer, Neuronal ceroid-lipofuscinosis in [36] J.K. Götzl, K. Mori, M. Damme, K. Fellerer, S. Tahirovic, G. Kleinberger, J. Janssens, J. Borderdale sheep, N. Z. Vet. J. 50 (2002) 199–202. van der Zee, C.M. Lang, K. Kremmer, J.J. Martin, S. Engelborghs, H.A. Kretzschmar, [5] T. Frugier, N.L. Mitchell, I. Tammen, P.J. Houweling, D.G. Arthur, G.W. Kay, O.P. van T. Arzberger, C. Van Broeckhoven, C. Haass, A. Capell, Common pathobiochemical Diggelen, R.D. Jolly, D.N. Palmer, A new large animal model of CLN5 neuronal ceroid hallmarks of progranulin-associated frontotemporal lobar degeneration and neuro- lipofuscinosis in Borderdale sheep is caused by a nucleotide substitution at a con- nal ceroid lipofuscinoses, Acta Neuropathol. 127 (2014) 845–860. sensus splice site (c.571 + 1G N A) leading to excision of exon 3, Neurobiol. Dis. [37] K.M. Kanninen, A. Grubman, J. Meyerowitz, C. Duncan, J.L. Tan, S.J. Parker, P.J. 29 (2008) 306–315. Crouch, B.M. Paterson, J.L. Hickey, P.S. Donnelly, I. Volitakis, I. Tammen, D.N. [6] R.D. Jolly, D.N. Palmer, The neuronal ceroid-lipofuscinoses (Batten disease): compar- Palmer, A.R. White, Increased zinc and manganese in parallel with neurodegenera- ative aspects, Neuropathol. Appl. Neurobiol. 21 (1995) (1995) 50–60. tion, synaptic protein changes and activation of Akt/GSK3 signaling in ovine CLN6 [7] M. Haltia, The neuronal ceroid-lipofuscinoses, J. Neuropathol. Exp. Neurol. 62 (2003) neuronal ceroid lipofuscinosis, PLoS One 8 (2013) e58644. 1–13. [38] K.R. Smith, J. Damiano, S. Franceschetti, S. Carpenter,L.Canafoglia,M.Morbin,G.Rossi, [8] M. Haltia, The neuronal ceroid-lipofuscinoses: from past to present, Biochim. D.Pareyson,S.E.Mole,J.F.Staropoli,K.B.Sims,J.Lewis,W.L.Lin,D.W.Dickson,H.H.Dahl, Biophys. Acta Mol. basis Dis. 1762 (2006) 850–856. M. Bahlo, S.F. Berkovic, Strikingly different clinicopathological phenotypes determined [9] D.N. Palmer, I. Tammen, C. Drögemüller, G.S. Johnson, M.L. Katz, F. Lingaas, Large an- by progranulin-mutation dosage, Am. J. Hum. Genet. 90 (2012) 1102–1107. imal models, in: S.E. Mole, R.E. Williams, H.H. Goebel (Eds.), The Neuronal Ceroid [39] C.K. Glass, K. Saijo, B. Winner, M.C. Marchetto, F.H. Gage, Mechanisms underlying in- Lipofuscinoses (Batten Disease), 2nd editionOxford University Press Inc., New flammation in neurodegeneration, Cell 140 (2010) 918–934. York 2011, pp. 284–320. [40] C. Heine, J. Tyynelä, J.D. Cooper, D.N. Palmer, M. Elleder, A. Kohlschütter, T. Braulke, [10] M.Bond,S.M.Holthaus,I.Tammen,G.Tear,C. Russell, Use of model organisms for the Enhanced expression of manganese-dependent superoxide dismutase in human study of neuronal ceroid lipofuscinosis, Biochim. Biophys. Acta 1832 (2013) 1842–1865. and sheep CLN6 tissues, Biochem. J. 376 (2003) 369–376. [11] D.N. Palmer, L.A. Barry, J. Tyynelä, J.D. Cooper, NCL disease mechanisms, Biochim. [41] H.M. Schipper, W. Song, H. Zukor, J.R. Hascalovici, D. Zeligman, Heme oxygenase-1 Biophys. Acta Mol. basis Dis. 1832 (2013) 1882–1893. and neurodegeneration: expanding frontiers of engagement, J. Neurochem. 110 [12] G.W. Kay, S.M. Hughes, D.N. Palmer, In vitro culture of neurons from sheep with Bat- (2009) 469–485. tendisease,Mol.Genet.Metab.67(1999)83–88. [42] R.N. Saha, K. Pahan, Regulation of inducible nitric oxide synthase gene in glial cells, [13] G.W. Kay, M.J. Oswald, D.N. Palmer, The development and characterisation of com- Antioxid. Redox Signal. 8 (2006) 929–947. plex ovine neuron cultures from fresh and frozen foetal neurons, J. Neurosci. [43] M.J. Lim, N. Alexander, J.W. Benedict, S. Chattopadhyay, S.J. Shemilt, C.J. Guérin, J.D. Methods 155 (2006) 98–108. Cooper, D.A. Pearce, IgG entry and deposition are components of the neuroimmune [14] S.M. Hughes, K.M. Hope, J.B. Xu, N.L. Mitchell, D.N. Palmer, Inhibition of storage pa- response in Batten disease, Neurobiol. Dis. 25 (2007) 239–251. thology in prenatal CLN5-deficient sheep neural cultures by lentiviral gene therapy, [44] A. Saha, C. Sarkar, S.P. Singh, Z. Zhang, J. Munasinghe, S. Peng, G. Chandra, E. Kong, Neurobiol. Dis. 62 (2014) 543–550. A.B. Mukherjee, The blood–brain barrier is disrupted in a mouse model of infantile [15] S.M. Hughes, G.W. Kay, T.W. Jordan, G.K. Rickards, D.N. Palmer, Disease-specificpa- neuronal ceroid lipofuscinosis: amelioration by resveratrol, Hum. Mol. Genet. 21 thology in neurons cultured from sheep affected with ceroid lipofuscinosis, Mol. (2012) 2233–2244. Genet. Metab. 66 (1999) 381–386. [45] S. Kopan, S. Uthayatharsini, M.J. Warburton, The lysosomal degradation of [16] M.J. Oswald, G.W. Kay, D.N. Palmer, Changes in GABAergic neuron distribution in neuromedin B is dependent on tripeptidyl peptidase-I: evidence for the impairment situ and in neuron cultures in ovine (OCL6) Batten disease, Eur. J. Paediatr. Neurol. of neuropeptide degradation in late-infantile neuronal ceroid lipofuscinosis, 5 (Suppl. A) (2001) 135–142. Biochem. Biophys. Res. Commun. 319 (2004) 58–65. [17] S.E. Mole, G. Michaux, S. Codlin, R.B. Wheeler, J.D. Sharp, D.F. Cutler, CLN6, which is [46] L.P. Henderson, L. Lin, A. Prasad, C.A. Paul, T.Y. Chang, R.A. Maue, Embryonic striatal associated with a lysosomal storage disease, is an endoplasmic reticulum protein, neurons from Niemann–Pick type C mice exhibit defects in cholesterol metabolism Exp. Cell Res. 298 (2004) 399–406. and neurotrophin responsiveness, J. Biol. Chem. 275 (2000) 20179–20187. [18] C. Heine, B. Koch, S. Storch, A. Kohlschütter, D.N. Palmer, T. Braulke, Defective endo- [47] C.T. Chang, P.J. Houweling, I. Tammen, RNA-sequencing and differential expression plasmic reticulum-resident membrane protein CLN6 affects lysosomal degradation analysis in the CLN6 Merino sheep model, Medicina 74 (Suppl. 11) (2014) 24. of endocytosed arylsulfatase A, J. Biol. Chem. 279 (2004) 22347–22352. [48] Z. Wang, M. Gerstein, M. Snyder, RNA-Seq: a revolutionary tool for transcriptomics, [19] K.S.Linterman,D.N.Palmer,G.W.Kay,L.A.Barry,N.L.Mitchell,R.G.McFarlane,M.Black, Nat. Rev. Genet. 10 (2009) 57–63. M.S. Sands, S.M. Hughes, Lentiviral-mediated gene transfer to the sheep brain: implica- [49] C. Trapnell, L. Pachter, S.L. Salzberg, TopHat: discovering splice junctions with RNA- tions for gene therapy in Batten disease, Hum. Gene Ther. 22 (2011) 1011–1020. Seq, Bioinformatics 25 (2009) 1105–1111. 2286 D.N. Palmer et al. / Biochimica et Biophysica Acta 1852 (2015) 2279–2286

[50] C. Trapnell, B.A. Williams, G. Pertea, A. Mortazavi, G. Kwan, M.J. van Baren, S.L. [63] S. Fukumura, Y. Saito, T. Saito, H. Komaki, E. Nakagawa, K. Sugai, M. Sasaki, A. Oka, I. Salzberg, B.J. Wold, L. Pachter, Transcript assembly and quantification by RNA-Seq Takamisawa, Progressive conduction defects and cardiac death in late infantile neu- reveals unannotated transcripts and isoform switching during cell differentiation, ronal ceroid lipofuscinosis, Dev. Med. Child Neurol. 54 (2012) 663–666. Nat. Biotechnol. 28 (2010) 511–515. [64] J.R. Østergaard, T.B. Rasmussen, H. Mølgaard, Cardiac involvement in juvenile neuro- [51] G. Dennis, B.T. Sherman, D.A. Hosack, J. Yang, W. Gao, H.C. Lane, R.A. Lempicki, nal ceroid lipofuscinosis (Batten disease), Neurology 76 (2011) 1245–1251. DAVID: Database for annotation, visualization, and integrated discovery, Genome [65] G.W. Kay, D.N. Palmer, Chronic oral administration of minocycline to sheep with Biol. 4 (2003) 3. ovine CLN6 neuronal ceroid lipofuscinosis maintains pharmacological concentra- [52] A. Grubman, G.E. Lidgerwood, C. Duncan, L. Bica, J.L. Tan, S.J. Parker, A. Caragounis, J. tions in the brain but does not suppress neuroinflammation or disease progression, Meyerowitz, I. Volitakis, D. Moujalled, J.R. Liddell, J.L. Hickey, M. Horne, S. Longmuir, J. Neuroinflammation 10 (2013) 97. J. Koistinaho, P.S. Donnelly, P.J. Crouch, I. Tammen, A.R. White, K.M. Kanninen, De- [66] S.M. Lee, T.Y. Yune, S.J. Kim, Y.C. Kim, Y.J. Oh, G.J. Markelonis, T.H. Oh, Minocycline regulation of subcellular biometal homeostasis through loss of the metal transport- inhibits apoptotic cell death via attenuation of TNF-alpha expression following er, Zip7, in a childhood neurodegenerative disorder, Acta Neuropathol. Commun. 2 iNOS/NO induction by lipopolysaccharide in neuron/glia co-cultures, J. Neurochem. (2014) 25, http://dx.doi.org/10.1186/2051-5960-2-25. 91 (2004) 568–578. [53] D.N.Palmer,R.D.Martinus,G.Barns,R.D.Reeves,R.D.Jolly,Ovineceroid-lipofuscinosis. [67] M. Nikodemova, J.J. Watters, S.J. Jackson, S.K. Yang, I.D. Duncan, Minocycline down- I. Lipopigment composition is indicative of a lysosomal proteinoses, Am. J. Med. Genet. regulates MHC II expression in microglia and macrophages through inhibition of (Supp. 5) (1988) 141–158. IRF-1 and protein kinase C (PKC)α/βII, J. Biol. Chem. 282 (2007) 15208–15216. [54] K.M. Kendrick, A.P. da Costa, A.E. Leigh, M.R. Hinton, J.W. Peirce, Sheep don't forget a [68] S. Kumar, J. Boehm, J.C. Lee, p38 MAP kinases: key signalling molecules as therapeu- face, Nature 414 (2001) 165–166. tic targets for inflammatory diseases, Nat. Rev. Drug Discov. 2 (2003) 717–726. [55] C. Lee, S. Colegate, A.D. Fisher, Development of a maze test and its application to assess [69] R.E. Haskell, S.M. Hughes, J.A. Chiorini, J.M. Alisky, B.L. Davidson, Viral-mediated de- spatial learning and memory in Merino sheep, Appl. Anim. Behav. Sci. 96 (2006) 43–51. livery of the late-infantile neuronal ceroid lipofuscinosis gene, TPP-I to the mouse [56] A.J. Morton, L. Avanzo, Executive decision-making in the domestic sheep, PLoS One 6 central nervous system, Gene Ther. 10 (2003) 34–42. (2011) e15752. [70] M. Griffey, E. Bible, C. Vogler, B. Levy, P. Gupta, J. Cooper, M.S. Sands, Adeno- [57] G.M. Cronin, D.F. Beganovic, A.L. Sutton, D.J. Palmer, P.C. Thomson, I. Tammen, associated virus 2-mediated gene therapy decreases autofluorescent storage mate- Manifestation of neuronal ceroid lipofuscinosis in Australian Merino sheep: obser- rial and increases brain mass in a murine model of infantile neuronal ceroid vations on altered behaviour and growth, Appl. Anim. Behav. Sci. (2015) (Accepted lipofuscinosis, Neurobiol. Dis. 16 (2004) 360–369. with emendations). [71] R.G. Crystal, D. Sondhi, N.R. Hackett, S.M. Kaminsky, S. Worgall, P. Steig, M. [58] N. Perentos, A.Q. Martins, T.C. Watson, U. Bartsch, N.L. Mitchell, D.N. Palmer, M.W. Souweidane, S. Hosain, L. Heier, M. Ballon, M. Dinner, K. Wisniewski, M. Kaplitt, Jones, A.J. Morton, Translational neurophysiology in sheep: measuring sleep and B.M. Greenwald, J.D. Howell, K. Strybing, J. Dyke, H. Voss, Administration of a repli- neurological dysfunction in CLN5 Batten disease affected sheep, Brain 138 (2015) cation deficient adeno-associated virus gene transfer vector expressing the human 862–874. CLN2 cDNA to the brain of children with late infantile neuronal ceroid [59] E. Reske-Nielsen, U. Baandrup, P. Bjerregaard, I. Bruun, Cardiac involvement in juve- lipofuscinosis, Hum. Gene Ther. 15 (2004) 1131–1154. nile amaurotic idiocy — aspecific heart muscle disorder. Histological findings in 13 [72] S. Worgall, D. Sondhi, N.R. Hackett, B. Kosofsky, M.V. Kekatpure, N. Neyzi, J.P. Dyke, autopsied patients, Acta Pathol. Microbiol. Scand. 89 (1981) 357–365. D. Ballon, L. Heier, B.M. Greenwald, P. Christos, M. Mazumdar, M.M. Souweidane, [60] P. Michielsen, J.J. Martin, E. Vanagt, C. Vrints, T. Gillebert, J. Snoeck, Cardiac involve- M.G. Kaplitt, R.G. Crystal, Treatment of late infantile neuronal ceroid lipofuscinosis ment in juvenile ceroid lipofuscinosis of the Spielmeyer–Vogt–Sjögren type: pro- by CNS administration of a serotype 2 adeno-associated virus expressing the CLN2 spective noninvasive findings in two siblings, Eur. Neurol. 23 (1984) 166–172. cDNA, Hum. Gene Ther. 19 (2008) 463–474. [61] K. Sakajiri, N. Matsubara, T. Nakajima, N. Fukuhara, T. Makifuchi, M. Wakabayashi, S. [73] N.L. Mitchell, G.K. Barrell, M. Wellby, H.E. Wicky, D.N. Palmer, S.M. Hughes, Gene Oyanagi, E. Kominami, A family with adult type ceroid lipofuscinosis (Kufs' disease) therapy using adeno-associated virus serotype 9 in the sheep brain, Medicina 74 and heart muscle disease: report of two autopsy cases, Intern. Med. 34 (1995) (Supl. II) (2014) 23. 1158–1163. [62] H. Tomiyasu, W. Takahashi, T. Ohta, F. Yoshii, M. Shibuya, Y. Shinohara, An autopsy case of juvenile neuronal ceroid-lipofuscinosis with dilated cardiomyopathy, Rinsho Shinkeigaku 40 (2000) 350–357.