Gene Therapy (2009) 16, 303–308 & 2009 Macmillan Publishers Limited All rights reserved 0969-7128/09 $32.00 www.nature.com/gt SHORT COMMUNICATION Mechanisms of distribution of mouse b-galactosidase in the adult GM1-gangliosidosis brain

MLD Broekman1,3,4, LA Tierney1, C Benn2, P Chawla2, JH Cha2 and M Sena-Esteves1 1Department of Neurology, Molecular Neurogenetics Unit, Massachusetts General Hospital, Charlestown, MA, USA; 2Department of Neurology, MassGeneral Institute for Neurodegenerative Disease, Massachusetts General Hospital/B114-2000, Charlestown, MA, USA; 3Department of Neurosurgery, University Medical Center Utrecht, Utrecht, The Netherlands and 4Rudolf Magnus Institute of Neuroscience, University Medical Center Utrecht, Utrecht, The Netherlands

GM1-gangliosidosis is a lysosomal storage disease (LSD) metabolically active enzyme can be produced, released and caused by an autosomal recessive deficiency of lysosomal distributed throughout the CNS, is necessary. The goal of acid b-galactosidase (bgal). This leads to accumulation of this study was to investigate the pattern and mechanism of GM1-ganglioside and its asialo derivative GA1 in the central distribution of bgal in the adult GM1-gangliosidosis mouse nervous system (CNS), and progressive neurodegeneration. brain upon hippocampal injection of an AAV vector-encoding Therapeutic AAV-mediated gene delivery to the brain for bgal. We found evidence that three different mechanisms LSDs has proven very successful in several animal models. contribute to its distribution in the brain: (1) diffusion; (2) GM1-gangliosidosis is also a prime candidate for AAV- axonal transport within neurons from the site of production; mediated gene therapy in the CNS. As global neuropathol- (3) CSF flow in the perivascular space of Virchow–Robin. ogy characterizes the most severe forms of this disease, In addition, we found evidence of axonal transport of vector- therapeutic interventions need to achieve distribution of bgal encoded mRNA. throughout the entire CNS. Therefore, careful consideration Gene Therapy (2009) 16, 303–308; doi:10.1038/gt.2008.149; of routes of administration and target structures from where published online 25 September 2008

Keywords: GM1-gangliosidosis; AAV; b-galactosidase; axonal transport mRNA; lysosomal storage disease

Introduction importance to understand the mechanisms by which these enzymes can be distributed in the brain. In The lysosomal storage disease GM1-gangliosidosis is addition to diffusion in the brain parenchyma,10,11 it caused by an autosomal recessive deficiency of lysoso- has been shown that at least two of these enzymes, mal acid b-galactosidase (bgal).1 The disease affects b-glucuronidase and arylsulfatase A, can be transported many organs, including the central nervous system by axonal transport within neurons from the site of (CNS). There, deficiency of bgal results in the accumula- production.12–14 Moreover, CSF flow has been shown to tion of the ganglioside GM1 and its asialo derivative contribute to the distribution of both b-glucuronidase, 2 GA1, which causes progressive neurodegeneration. a-L-iduronidase and tripeptidyl peptidase 1 throughout Recently, direct gene delivery to the CNS by adeno- the CNS.11,15,16 associated viral (AAV) vectors has emerged as a strategy Given our continued interest in developing an AAV- to engineer an in situ source of lysosomal enzyme for mediated gene therapy approach to treat GM1-gang- several neuronopathic lysosomal storage diseases.3 This liosidosis, we set out to investigate whether the has been possible because lysosomal enzymes can be mechanisms of distribution earlier described for GUSB, secreted at high levels by genetically modified cells in the Arylsulfatase A and tripeptidyl peptidase 1 are also brain, and subsequently taken up by enzyme-deficient functional for bgal. This has obvious implications for cells through receptor-mediated endocytosis resulting in choosing target structures to be engineered by AAV- the correction of lysosomal storage in large regions of the mediated gene delivery to achieve global distribution of 4–9 murine CNS. Translation of these findings to clinical enzymes in the brain with the minimum number of trials will require careful consideration of routes of injections possible. The hippocampus has been pre- administration and target structures from where meta- viously used to map the distribution of lysosomal bolically active enzymes can be produced, released and enzymes by axonal transport12,14 because its circuitry is distributed throughout the CNS. Thus it is of paramount well characterized. Therefore, we characterized the distribution of bgal in adult GM1-gangliosidosis mouse Correspondence: Dr M Sena-Esteves, Department of Neurology, brain after AAV-mediated gene delivery to the hippo- Massachusetts General Hospital Building 149, 13th Street, Room campus. To this aim, we injected 1 ml of AAV2/1-CBA- 6309, Charlestown, MA 02129, USA. 10 17 E-mail: [email protected] MBG-W vector [4.1 Â10 g.c. (genome copies)] into the Received 27 April 2008; revised 27 August 2008; accepted 27 August left hippocampus of 6-week-old GM1-gangliosidosis 2008; published online 25 September 2008 mice.18 One month postinjection, we observed mouse Mechanisms of distribution of mouse b-galactosidase MLD Broekman et al 304 bgal in physically and synaptically connected areas the entorhinal cortex (arrowheads in Figures 2c and d), consistent with three main mechanisms of lysosomal we conclude that bgal activity in cortex is mostly the enzyme distribution in the brain. result of diffusion, a mechanism shown to be involved in the distribution of several other lysosomal enzymes.11,20 The significance of vector mRNA in the entorhinal cortex Diffusion of bgal will be addressed below. Upon injection of AAV vector into the hippocampus, the overlaying cerebral cortex was stained in a pattern that appears to radiate from the hippocampus (Figures 1a Retrograde axonal transport within and b). Staining in the cortex expands from a region neurons from the site of production spanning the retrosplenial cortex to the primary soma- tosensory cortex barrel field in rostral sections The contralateral hippocampal region was strongly (Figure 1a3) to a region spanning the retrosplenial cortex positive for bgal, with the highest activity (darkest stain) to the temporal association cortex in caudal sections associated with the CA3 pyramidal cell layer (Figure 1a4, (Figure 1a5). This expanding staining pattern seems to arrow in Figure 1e). This is consistent with retrograde mirror the progression of the hippocampal formation axonal transport along the commissural collaterals from a medial position in rostral sections (Figure 1a2) to that project from this region to the oriens and radiatum a medial lateral position in caudal sections (Figure 1a5). layers in the injected hippocampus.21,22 This notion is As in situ hybridization for vector-associated mRNA supported by the fact that the ventral hippocampal showed no signal in bgal-positive cortical regions except commissure was intensely stained (gray arrow in

Figure 1 Histochemical analysis of b-gal expression by X-gal staining upon hippocampal injection. One ml of AAV2/1-CBA-MBG-W vector stock (4.1 Â1010 g.c.) was infused at 0.1 ml minÀ1 into the left hippocampus (stereotactic coordinates from bregma: AP À2 mm, ML +1.5 mm, DV À2 mm) of 6-week-old GM1-gangliosidosis mice (n ¼ 5). Mice were killed at 1 month postinfusion and bgal distribution in the brain analyzed by X-gal staining as described.17 Cells distant from the injection site (white arrow in a4) were enzyme-positive (a1–6). Staining indicative of the presence of the lysosomal enzyme was observed in the ipsilateral (b), and contralateral cortex (c), the entorhinal cortex (Ect) (d), the ipsilateral and contralateral hippocampus and DG (a3–5 and e), medial nucleus (MS) and both ipsilateral and contralateral dorsal lateral nuclei (LSD) of the septum (f), perivascular space (arrow in g), mammillary body (MB) (h) and leptomeninges (arrow in i). Scale bars in (a) represent 1 mm. Magnification (b-i): Â 40.

Gene Therapy Mechanisms of distribution of mouse b-galactosidase MLD Broekman et al 305 in this manner can be released at the axonal terminals. Also the pattern of bgal distribution in the medial and dorsolateral septal nuclei (Figure 1f) suggests that anterograde transport of bgal is an important component of its axonal transport. Most bgal staining in the septal area is associated with dorsolateral nuclei that receive projections from the CA3 pyramidal cell layer.23,24 Another example of axonal transport of bgal is the staining present in the mammillary nucleus (Figures 1a5 and h). This nucleus receives afferents from and sends projections to the subicular region of the hippocampal formation that travels along the postcommissural fornix.25,26 Although most in vivo studies to date have indicated that lysosomal enzymes are retrogradely transported in axons,12,13,27 our data strongly suggests that anterograde transport of bgal is an important component of axonal distribution of this enzyme in the adult GM1-gang- liosidosis brain. This notion is supported by a recent study showing identical frequency of anterograde and retrograde movement of GFP-labeled a-L-iduronidase in neurites of primary rat neurons.28

CSF flow in perivascular space of Virchow–Robin In addition to bgal distribution in the hippocampus, we observed strong staining of the leptomeninges and perivascular space in the brain (Figures 1g and i). This suggests that a considerable amount of bgal is released from transduced cells into the interstitial space and is further distributed by CSF flow in the perivascular space of Virchow–Robin. An alternative explanation is that AAV vector transduced those bgal-positive regions, and the apparent absence of ISH signal in leptomeninges is because of low spatial resolution associated with the radioactive technique employed here. Presently, we are unable to exclude this possibility. However, the first interpretation is consistent with data available for 11 15 Figure 2 Detection of vector mRNA by ISH. Radioactive in situ b-glucuronidase, a-L-iduronidase, and tripeptidyl hybridization was performed in two animals as described19 using peptidase16 and is likely the basis for the success in the following [35S] end-labeled antisense WPRE probe: 50-CAT using intracerebroventricular (i.c.v.)11 or intrathecal infu- TAAAGCAGCGTATCCACATAGCGTAAAAGGAGCAACATAGTT 15 16 0 sion of AAV vectors or purified recombinant enzymes AA-3 . Intense signal was observed in the ipsilateral ventral in adult animals to achieve widespread correction of hippocampal commissure (arrow in a). In the ventral hippocampus (CA3 and DG) (b and c), high signal was detected as well. Some lysosomal storage in the brain. Previously we have signal was associated with the ipsilateral CA1-CA2 regions of the shown i.c.v. infusion of AAV vectors in neonatal GM1- hippocampus (asterisk in c and d). In non-injected areas as the gangliosidosis mice is highly effective in correcting ipsilateral entorhinal cortex (arrowheads in c and d) and the neurochemical pathology.17 The effectiveness of this contralateral CA1-CA3 region of the hippocampus (arrows in b–d), strategy in adult GM1-gangliosidosis mice is likely to a considerable signal was detected as well. Tissue sections in a–d are depend on achieving efficient transduction of choroid parallel sections to those shown in Figures 1a2–a5 stained with X-gal. plexus, ependymal cells and/or periventricular regions with a particular AAV serotype.11

Figure 1a2). In the rest of the contralateral hippocampus, most bgal activity is associated with the oriens and Axonal transport of mRNA radiatum layers of CA1-CA2, and not the pyramidal cell layer per se (Figure 1e). This pattern of enzyme distribu- We investigated the distribution of AAV-transduced cells tion is consistent with anterograde transport of bgal in the hippocampus by in situ hybridization using a along commissural collaterals from the AAV-injected WPRE-specific antisense probe (Figures 2a–d). In addi- CA3 region, or Schaffer collaterals from the non-injected tion to a strong signal in the injected structure (ventral CA3 pyramidal cell layer. Moreover, the fact that bgal hippocampus), we found a considerable signal in the appears to diffuse to the contralateral overlaying cortex contralateral hippocampus (arrows in Figures 2b–d), in a radial gradient emanating from the hippocampus ipsilateral CA1 region (asterisks in Figures 2c and d) and (Figures 1a3–a4 and c) suggests that enzyme transported ipsilateral entorhinal cortex (arrow head Figures 2c and d).

Gene Therapy Mechanisms of distribution of mouse b-galactosidase MLD Broekman et al 306 No signal was present using antisense probe in unin- These results confirm our interpretation of the ISH jected GM1-KO brains or sense probe in AAV-injected results that vector-derived mRNA was distributed to GM1-KO brains. The signal in the contralateral hippo- projection fields of CA3 pyramidal neurons in the campus and ipsilateral CA1 region is consistent with contralateral hippocampus. Luca et al. found similar mRNA presence in commissural and Schaffer collaterals, results with apparent distribution of vector-encoded respectively, emanating from AAV-transduced CA3 mRNA to axonally connected structures, but discounted pyramidal neurons. The intense signal in the ventral it as background.14 Axonal transport of specific hippocampal commissure (arrow Figure 2a) supports the mRNAs to growth cones is a highly regulated process notion that vector-derived mRNA is present in commis- involving RNA ‘zip codes’ and transporter proteins.32 sural collaterals. It is possible that the WPRE element present in the In the entorhinal cortex, stellate neurons in layer II vectors used in both studies reporting axonal localization send projections to the outer two-thirds of the molecular of mRNA, contains a cryptic or real axonal mRNA layer of the DG through the perforant pathway.29,30 As localization signal. At this point we do not have the entorhinal cortex in adult mice does not receive evidence suggesting a functional role for axonal vector projections from the hippocampus, we postulate that the mRNA in the distribution of lysosomal enzymes in ISH signal observed in this structure (arrowhead in the brain. It is unlikely for this mRNA to be translated Figures 2c and d) is because of AAV vector transduction into functional bgal in axonal terminals as lysosomal after retrograde transport along the perforant pathway, enzymes require endoplasmic reticulum translation, as reported earlier.31 Transduction of the ipsilateral glycosylation and maturation in the Golgi compartment. entorhinal cortex is consistent with the fact that this To our knowledge none of these structures are present in region presented the highest bgal staining intensity in the axonal terminals, although there is evidence suggesting ipsilateral cortex (Figures 1a4–5 and 1d). that polyribosomes are present in certain areas of Next we investigated whether the ISH signal in axons,33 and that translation takes place in growth contralateral hippocampus was because of axonal cones.32 mRNA ‘transport’ or the presence of AAV-transduced cells. We isolated DNA and RNA from laser capture microdissected ipsi-, and contralateral hippocampi and performed PCR on DNA and cDNA using WPRE- Correction of lysosomal storage in specific primers. This analysis revealed the presence of enzyme-positive regions AAV vector genomes and vector-derived mRNA in the ipsilateral hippocampus, but only vector-derived mRNA We have previously shown that lysosomal storage in in the contralateral hippocampus (Figures 3a and b). GM1-gangliosidosis mouse brain is accompanied Spiking DNA samples with 500 pg of control plasmid by cholesterol storage/trapping in the endosomal/ showed that the absence of PCR amplification from the lysosomal compartment as revealed by Filipin staining contralateral hippocampus was most likely because of of histological sections of the brain.17 Thus, Filipin the absence of AAV vector genomes and not the presence staining can be used as a surrogate marker of lysosomal of a PCR inhibitor in the DNA preparations (Figure 3c). storage in GM1 brain. To test whether bgal was effective in correcting lysosomal storage in distal sites, we analyzed the pattern of Filipin staining in the brains of both treated and untreated GM1-KO mice. Strong perikaryal staining, consistent with endosomal/lysoso- mal accumulation of cholesterol, was observed through- out the entire brain in untreated GM1-KO mice (Figure 4h). In AAV1-mbgal treated bgalÀ/À mice we Figure 3 Vector mRNA not DNA present in contralateral observed weak, diffuse Filipin staining in all enzyme- hippocampus. Sections from two AAV2/1-CBA-MBG-W injected positive regions of the brain as the cortex above both brains designated for LCM, were mounted on RNase-free, hippocampi (Figures 4a and b), ipsilateral and contral- membrane slides (Molecular Machines and Industries-MMI, Glatt- ateral hippocampi (Figures 4c and d), and ipsilateral brugg, Switzerland), dried at room temperature and stored at thalamus (Figure 4e). In the contralateral thalamus 1 À80 C. Next, they were dried in serial ethanols and residual TBS- (Figure 4f), the ipsilateral striatum (Figure 4g), contral- freezing media was removed. The sections were viewed under a Nikon TE2000 inverted microscope connected to CellCut LCM ateral striatum (data not shown) and other regions of the equipment (MMI). The ipsilateral and contralateral hipppocampi cortex (data not shown), there was no evidence of storage were identified and dissected (UVCut software package; MMI). The reduction. We conclude from this histological analysis of samples were picked up on adhesive lids of 500 ml mini Isolation- lysosomal storage that bgal present in distal sites is Caps (MMI) and used for DNA and RNA extraction using PicoPure functional. DNA and RNA isolation kits (Arcturus-Molecular Devices, Moun- Here, we describe the mechanisms that contribute to tain View, CA, USA). RNA was subjected to an additional DNase digestion (Qiagen, Valencia, CA, USA), according to the manufac- distribution of lysosomal acid b-galactosidase in the turer’s instructions. DNA and RNA concentrations were deter- GM1-gangliosidosis brain upon AAV-mediated delivery: mined using a NanoDrop ND-1000 Spectrophotometer (NanoDrop, diffusion, axonal transport and CSF flow. In addition, we Wilmington, DE, USA). cDNA was prepared using Omniscript found evidence of axonal transport of vector-encoded reverse transcription kit (Qiagen). PCR on DNA (a), cDNA (b), and mRNA. These results will guide our selection of target DNA spiked with 500 pg of plasmid control (c) for WPRE was structures and delivery modalities to achieve widespread performed in triplicate using the following primers: forward 50-ATGTTGCTCCTTTTACGCTATGTGG-30 and reverse 50-GAAAGC distribution of bgal throughout the GM1-gangliosidosis GAAAGTCCCGGAAAGG-30. Samples: 1- Ipsilateral hippocampus; central nervous system after AAV-mediated gene 2- water control; 3- Contralateral hippocampus; 4- Plasmid control. delivery.

Gene Therapy Mechanisms of distribution of mouse b-galactosidase MLD Broekman et al 307

Figure 4 Correction of storage in enzyme-positive areas. One series of sections representing the entire brain were stained with Filipin (Sigma, St Louis, MO, USA) and counterstained for nuclei with TO-PRO3 as described earlier.17 Untreated bgalÀ/À mice showed strong perikaryal staining throughout the brain, consistent with endosomal/lysosomal accumulation of cholesterol as in the somatosensory cortex (h). As in normal mouse brain (not shown), weak diffuse Filipin staining was observed in the ipsilateral and contralateral cerebral cortex overlaying the hippocampus (a and b), the ipsi- and contralateral hippocampus (c and d) and the ipsi- and contralateral thalamus (e and f). Strong staining was detected in the contralateral thalamus (f), and both ipsi- (g) and contralateral (not shown) striatum. Scale bars represent 200 mm.

Acknowledgements References

This work was supported by NIH Grant R21NS053993. 1 Okada S, O’Brien JS. Generalized gangliosidosis: beta-galactosi- We thank Deidre McCarthy for her help with laser dase deficiency. Science 1968; 160: 1002–1004. capture microdissection, and the Massachusetts General 2 O’Brien JS, Stern MB, Landing BH, O’Brien JK, Donnell GN. Hospital Neuroscience Center Microscopy and Image Generalized gangliosidosis: another inborn error of ganglioside Analysis Core (NIH Grant P30NS045776) for use of the metabolism? Am J Dis Child 1965; 109: 338–346. laser capture microdissection system and Nikon Super- 3 Sands MS, Davidson BL. Gene therapy for lysosomal storage coolscan 9000 slide scanner. diseases. Mol Ther 2006; 13: 839–849.

Gene Therapy Mechanisms of distribution of mouse b-galactosidase MLD Broekman et al 308 4 Skorupa AF, Fisher KJ, Wilson JM, Parente MK, Wolfe JH. disease phenotypes in a mouse model of late infantile neuronal Sustained production of beta-glucuronidase from localized sites ceroid lipofuscinosis. Mol Ther 2008; 16: 649–656. after AAV vector gene transfer results in widespread distribu- 17 Broekman ML, Baek RC, Comer LA, Fernandez JL, Seyfried TN, tion of enzyme and reversal of lysosomal storage lesions in a Sena-Esteves M. Complete correction of enzymatic deficiency large volume of brain in mucopolysaccharidosis VII mice. and neurochemistry in the GM1-gangliosidosis mouse brain by Exp Neurol 1999; 160: 17–27. neonatal adeno-associated virus-mediated gene delivery. 5 Sferra TJ, Qu G, McNeely D, Rennard R, Clark KR, Lo WD et al. Mol Ther 2007; 15: 30–37. Recombinant adeno-associated virus-mediated correction of 18 Hahn CN, del Pilar Martin M, Schroder M, Vanier MT, Hara Y, lysosomal storage within the central nervous system of the Suzuki K et al. Generalized CNS disease and massive GM1- adult mucopolysaccharidosis type VII mouse. Hum Gene Ther ganglioside accumulation in mice defective in lysosomal acid 2000; 11: 507–519. beta-galactosidase. Hum Mol Genet 1997; 6: 205–211. 6 Passini MA, Watson DJ, Vite CH, Landsburg DJ, Feigenbaum AL, 19 Benn CL, Farrell LA, Cha JH. Neurotransmitter receptor analysis Wolfe JH. Intraventricular brain injection of adeno-associated in transgenic mouse models. Methods Mol Biol 2004; 277: 231–260. virus type 1 (AAV1) in neonatal mice results in complementary 20 Taylor RM, Wolfe JH. Decreased lysosomal storage in the adult patterns of neuronal transduction to AAV2 and total long-term MPS VII mouse brain in the vicinity of grafts of retroviral vector- correction of storage lesions in the brains of beta-glucuronidase- corrected fibroblasts secreting high levels of beta-glucuronidase. deficient mice. JVirol2003; 77: 7034–7040. Nat Med 1997; 3: 771–774. 7 Bosch A, Perret E, Desmaris N, Heard JM. Long-term and 21 Swanson LW, Wyss JM, Cowan WM. An autoradiographic study significant correction of brain lesions in adult mucopolysacchar- of the organization of intrahippocampal association pathways in idosis type VII mice using recombinant AAV vectors. Mol Ther the rat. J Comp Neurol 1978; 181: 681–715. 2000; 1: 63–70. 22 Amaral DG, Witter MP. The three-dimensional organization of 8 Bosch A, Perret E, Desmaris N, Trono D, Heard JM. Reversal of the hippocampal formation: a review of anatomical data. pathology in the entire brain of mucopolysaccharidosis type VII Neuroscience 1989; 31: 571–591. mice after lentivirus-mediated gene transfer. Hum Gene Ther 23 Swanson LW. The anatomical organization of septo-hippocam- 2000; 11: 1139–1150. pal projections. Ciba Found Symp 1977; 58: 25–48. 9 Consiglio A, Quattrini A, Martino S, Bensadoun JC, Dolcetta D, 24 Swanson LW, Sawchenko PE, Cowan WM. Evidence for Trojani A et al. In vivo gene therapy of metachromatic collateral projections by neurons in Ammon’s horn, the dentate leukodystrophy by lentiviral vectors: correction of neuropathol- gyrus, and the subiculum: a multiple retrograde labeling study ogy and protection against learning impairments in affected in the rat. J Neurosci 1981; 1: 548–559. mice. Nat Med 2001; 7: 310–316. 25 van Groen T, Wyss JM. The connections of presubiculum and 10 Taylor RM, Wolfe JH. Decreased lysosomal storage in the adult parasubiculum in the rat. Brain Res 1990; 518: 227–243. MPS VII mouse brain in the vicinity of grafts of retroviral vector- 26 Gonzalo-Ruiz A, Alonso A, Sanz JM, Llinas RR. Afferent corrected fibroblasts secreting high levels of beta-glucuronidase. projections to the mammillary complex of the rat, with special Nat Med 1997; 3: 771–774. reference to those from surrounding hypothalamic regions. 11 Liu G, Martins I, Wemmie JA, Chiorini JA, Davidson BL. J Comp Neurol 1992; 321: 277–299. Functional correction of CNS phenotypes in a lysosomal storage 27 Luca T, Givogri MI, Perani L, Galbiati F, Follenzi A, Naldini L et al. disease model using adeno-associated virus type 4 vectors. Axons mediate the distribution of arylsulfatase A within the mouse J Neurosci 2005; 25: 9321–9327. hippocampus upon gene delivery. Mol Ther 2005; 12: 669–679. 12 Passini MA, Lee EB, Heuer GG, Wolfe JH. Distribution of a 28 Chen F, Vitry S, Hocquemiller M, Desmaris N, Ausseil J, lysosomal enzyme in the adult brain by axonal transport and Heard JM. Alpha-L-Iduronidase transport in neurites. Mol Genet by cells of the rostral migratory stream. J Neurosci 2002; 22: Metab 2006; 87: 349–358. 6437–6446. 29 Blackstad TW. On the termination of some afferents to the 13 Hennig AK, Levy B, Ogilvie JM, Vogler CA, Galvin N, Bassnett S hippocampus and fascia dentata; an experimental study in the et al. Intravitreal gene therapy reduces lysosomal storage in rat. Acta Anat (Basel) 1958; 35: 202–214. specific areas of the CNS in mucopolysaccharidosis VII mice. 30 Raisman G, Cowan WM, Powell TP. An experimental analysis J Neurosci 2003; 23: 3302–3307. of the efferent projection of the hippocampus. Brain 1966; 89: 14 Luca T, Givogri MI, Perani L, Galbiati F, Follenzi A, Naldini L 83–108. et al. Axons mediate the distribution of arylsulfatase A within 31 Passini MA, Macauley SL, Huff MR, Taksir TV, Bu J, Wu IH et al. the mouse hippocampus upon gene delivery. Mol Ther 2005; 12: AAV vector-mediated correction of brain pathology in a mouse 669–679. model of Niemann-Pick A disease. Mol Ther 2005; 11: 754–762. 15 Watson G, Bastacky J, Belichenko P, Buddhikot M, Jungles S, 32 Sotelo-Silveira JR, Calliari A, Kun A, Koenig E, Sotelo JR. RNA Vellard M et al. Intrathecal administration of AAV vectors for the trafficking in axons. Traffic 2006; 7: 508–515. treatment of lysosomal storage in the brains of MPS I mice. Gene 33 Steward O, Ribak CE. Polyribosomes associated with synaptic Therapy 2006; 13: 917–925. specializations on axon initial segments: localization of protein- 16 Chang M, Cooper JD, Sleat DE, Cheng SH, Dodge JC, synthetic machinery at inhibitory synapses. J Neurosci 1986; 6: Passini MA et al. Intraventricular enzyme replacement improves 3079–3085.

Gene Therapy