Neuropil and neuronal changes in hippocampal NADPH- diaphorase ...

Blackwell Science, LtdOxford, UKNANNeuropathology and Applied Neurobiology0305-18462004 Blackwell Publishing Ltd? 200430?292303Original ArticleNOS in ME7-induced prion diseaseC. W. Picanço-Diniz et al.

Neuropathology and Applied Neurobiology (2004), 30, 292–303

doi: 10.1111/j.1365-2990.2004.00537.x

Neuropil and neuronal changes in hippocampal NADPHdiaphorase histochemistry in the ME7 model of murine prion disease C. W. Picanço-Diniz*, D. Boche†, W. Gomes-Leal*, V. H. Perry† and C. Cunningham† *Universidade Federal do Pará, Centro de Ciências Biológicas, Departamento de Morfologia, Laboratório de Neuroanatomia Funcional, Belém, Brazil, and †CNS Inflammation group, Southampton Neuroscience Group, School of Biological Sciences, University of Southampton, UK

C. W. Picanço-Diniz, D. Boche, W. Gomes-Leal, V. H. Perry and C. Cunningham (2004) Neuropathology and Applied Neurobiology, doi: 10.1111/j.1365-2990.2004.00537.x Neuropil and neuronal changes in hippocampal NADPH-diaphorase histochemistry in the ME7 model of murine prion disease Nitric oxide (NO) has been implicated in neurotoxicity and cerebral blood flow changes in chronic neurodegeneration, but its activity in the mammalian prion diseases has not been studied in detail. Nicotine adenine dinucleotide phosphate (NADPH)-diaphorase (NADPH-d) histochemistry is a simple and robust histochemical procedure that allows localization of the tissue distribution of NO synthases. The aim of the present study is to assess whether NADPH-d histochemical activity is altered in the hippocampus in the ME7 model of prion disease in C57BL/6J mice. At early and late stages after the initiation of the disease we assessed features of the NADPH-d positive cells and the neuropil histochemical activity in CA1 and den-

tate gyrus using densitometric analysis. In C57BL/6J mice 13 weeks postinjection of the prion agent ME7, when behavioural changes first become apparent, neuropil NADPH-d histochemical staining increases, whereas at late stages it decreases dramatically. Both type I and type II NADPH-d positive cells were found to survive throughout the hippocampal formation into the late stages of the disease, but diaphorase activity was reduced in dendritic branches and abnormal varicosities were present in both dendritic and axonal processes of NADPH-d positive type I cells. The pathophysiological implications of the results remain to be investigated but both blood flow alteration and NO neurotoxicity may be features of the disease.

Keywords: blood flow, neurodegeneration, nitric oxide, peroxynitrite, scrapie

Introduction Prion diseases are chronic and fatal transmissible degenerative disorders afflicting mammals. They are characterized by amyloid deposits, neuronal loss, gliosis and vacuolation of the neuropil [1]. Prion disease is associc ated with a conversion of soluble prion protein (PrP ) to sc an insoluble protease-resistant form (PrP ), which is Correspondence: Colm Cunningham, CNS Inflammation Group, School of Biological Sciences, University of Southampton, SO16 7PX, UK. Tel: 0044 23 80597642; Fax: 0044 23 80592711; E-mail: [email protected]

292

deposited in the parenchyma of the central nervous system [2]. There are studies in murine models describing the early neuropathological [3], behavioural [4], biochemical [5] and electrophysiological [6] changes in the disease but the direct cause of this damage remains unclear. Recently, evidence for oxidative stress late in the disease has been presented through the detection of oxidative damage to proteins [5], lipids [7,8] and DNA [9] and through altered activities of anti-oxidative enzymes [10,11]. Peroxynitrite formation is a consistent feature of oxidative stress and results from the reaction of nitric oxide (NO) with super© 2004 Blackwell Publishing Ltd

NOS in ME7-induced prion disease

oxide anions [12]. At physiological levels NO is an important molecule responsible for a variety of functions such as long-term potentiation and memory formation in the hippocampus [13,14] as well as in the control of blood flow [15–17]. There have been few studies on the role of NO in prion disease [5,18–22]. Because NO is short-lived, freely diffusible and extremely membrane permeant, attempts to localize its production site are best achieved by the localization of its synthetic enzyme, NO synthase (NOS), and nicotine adenine dinucleotide phosphate (NADPH)-diaphorase (NADPH-d) histochemistry is a robust histochemical procedure that allows this. Some of the above studies described NADPH-d histochemical activity in prion disease in the late stages of disease. Two of these reports suggest that NOS activity was markedly decreased in the cerebellum of mice and hamsters infected with an unspecified scrapie strain as compared to normal brains [19,20]. A further study reported the NADPH-d staining intensity of neurones in septum, thalamus, hypothalamus and amygdala of 139H- and 263K-infected hamsters to be greater than in control hamsters [22] although this was not compared with other features of the pathology. None of these studies investigated either changes in NOS activity early in the disease process or changes in the neuropil. The ME7 strain is a well-characterized murine prion disease strain with well-described hippocampal and thalamic pathology, for which behavioural data are also available [4]. Given the importance of NO in the regulation of synaptic activity and of local blood flow, and the neurotoxic potential of its uncontrolled production, we have investigated whether there are changes in NADPH-d histochemical activity in the neuropil and in type I neurones of the mouse hippocampus both early and late in the ME7 model of prion disease.

Materials and methods

Animals and surgery All experiments were carried out on C57BL/6J mice in accordance with the UK Animals (Scientific procedures) Act 1986. All efforts were made to minimize the number and suffering of animals used. Female mice aged 2– 3 months were anaesthetized with Avertin (2,2,2 tribromoethanol in tertiary amyl alcohol) at a dose of approximately 0.1 ml/5 g body weight and placed in a stereotaxic frame (Kopf Instruments, Tujunga, CA, USA).

293

1 ml of 10% w/v of ME7-derived brain homogenate or normal brain homogenate (NBH) was injected into the right hippocampus via a fine glass micropipette at the following stereotaxic coordinates relative to bregma: anteroposterior: -2.0 mm, medio-lateral: 1.7 mm and depth: 1.6 mm. After each injection the pipette was left in situ for 2 min to avoid reflux along the injection tract. NBH was derived from the brains of age-matched normal C57BL/6J mice as described elsewhere [21]. At specified times, early (13 weeks) or late (19–21 weeks), animals were killed by terminal anaesthesia with sodium pentobarbitone and transcardially perfused with heparinized 0.9% saline followed by 4% paraformaldehyde in 0.1 M phosphate buffer pH 7.4 or by 10% formalin. Brains were postfixed in the same fixative overnight. These times were chosen in order to coincide with the first behavioural changes as reported by Guenther et al. [4] and with the terminal stages of the disease. Four animals from each group (13 weeks, 19 weeks, NBH) were paraformaldehyde fixed for diaphorase staining, three from each group were formalin fixed for analysis sc of synaptophysin and PrP , and four from each group were fresh frozen (perfused with 0.9% saline, but unfixed) for analysis of microglial labelling.

Histochemical procedures All chemicals we have used in this investigation were supplied by Sigma (Poole, UK). The NADPH-d reaction requires aldehyde fixation in order to inhibit other types of NADPH-dependent oxido-reductases that can also convert tetrazolium salts to coloured and insoluble formazan deposition upon reduction. Four per cent paraformaldehdye has been demonstrated to inhibit other diaphorase activities while enhancing formazan deposition in NOS-positive populations [23]. Coronal or parasagital sections of 100 mm thickness were cut on a vibratome and all sections processed for free-floating histochemistry following the malic indirect method [24]. This was performed on four animals for each of the NBH and early and late stage ME7 animals. In brief, sections were collected in 0.1 M TRIS buffer pH 8.0, washed twice and then incubated at 37∞C in the histochemical reaction mixture. This contained: 0.6% malic acid in 0.1 M TRIS buffer, final pH 8.0, to which was added bNADP (0.1% w/v), manganese chloride (0.04% w/v) and nitroblue tetrazolium (0.03% w/v, previously solubilized in DMSO (0.3% w/v). Triton X-100 at 0.1% v/v was added to the final mixture and sections

© 2004 Blackwell Publishing Ltd, Neuropathology and Applied Neurobiology, 30, 292–303

294

C. W. Picanço-Diniz et al.

were incubated in this solution at 37∞C, with continual agitation for 2 h. Microscope inspection was used to assess when the reaction product was visible in the distal dendrites (second and third order) and the reaction was terminated at this time. The reaction was terminated by washing the sections for 10 min in 0.1 M TRIS buffer, pH 8.0.

Immunocytochemical procedures Some sections were used to detect neuronal NOS (nNOS) by immunocytochemistry. In the case of the double staining reaction the immunohistochemistry was performed before the NADPH-d histochemistry. In brief, sections were washed in 0.1 M phosphate-buffered saline (PBS) with Triton X-100 (0.1%) pH 7.2–7.4 (PBST), plus bovine serum albumin (BSA) at 0.25% w/v, and kept at 4∞C overnight. After permeabilization, sections were incubated for 3 h in 100% normal goat serum. This high blocking serum concentration was used to prevent nonspecific binding of the antibody to activated astrocytes in the hippocampus. The primary antibody (polyclonal Rabbit antiNOS I, AHP477, SEROTEC, Oxford, UK) was diluted in 0.1 M PBS pH 7.2–7.4 at the concentration of 3.3 mg/ml (1 : 150) and incubated with the sections for 3 days at 4∞C. The sections were washed in PBST/BSA and transferred to the secondary polyclonal antibody (biotinylated goat anti-rabbit, Vector Laboratories Berlinger, CA, USA) diluted in PBS at 1 : 200 concentration, overnight at 4∞C. After PBS washing, sections were transferred to avidinbiotin-peroxidase complex (ABC) solution (Vector Laboratories) for 1 h and washed again before peroxidase reaction using diaminobenzidine (DAB, 0.5 mg/ml) as substrate and intensified using Nickel (0.6 mg/ml of Ammonium Nickel chloride). All steps were carried out with gentle and continued agitation. Sections of 100 mm were chosen over thinner ones in order to preserve higher order dendritic branches in the same section. For microglial labelling, transcardially saline-perfused brains were frozen without fixation in optimal cutting temperature (OCT) embedding medium over isopentane and sectioned (10 mm) on a cryostat. These sections were stained using an antibody against the macrophage marker CD68 (FA11, Serotec, Oxford, UK, at a dilution of 1/20). Briefly, sections were dried at 37∞C, fixed in ethanol, washed, blocked with 10% normal serum and incubated with primary antibody for 2 h. Sections were then washed and incubated with the appropriate secondary

antibody before addition of the ABC and peroxidase reaction with DAB. Formalin-fixed tissue was sectioned on a microtome (10 mm) and stained for PrP as follows: sections were incubated with 6H4 (Prionics, Zurich, Switzerland) overnight after hydrated autoclaving at 121∞C for 20 min and blocking with a specially formulated blocking kit to allow labelling of mouse tissue with a mouse primary antibody (Mouse-on-mouse, Vector Laboratories). Subsequent steps were performed as for FA11. Synaptophysin staining was performed on formalin-fixed tissue using SY38 (Chemicon, CA, USA). Sections were pretreated with 0.2 M boric acid, pH 9, at 65∞C for 30 min and cooled to room temperature to improve antigen retrieval, washed in PBS and blocked using 10% normal horse serum. Sections were incubated with SY38 (1 : 100), at 19∞C, overnight before incubation with biotinylated horse anti-mouse secondary antibody (1 : 200) followed by ABC. The DAB reaction was carried out in the presence of ammonium nickel chloride (0.06% w/v) to intensify labelling.

Cell counting and densitometric analysis Counting procedure In brain sections reacted for NADPH-d, the hippocampal lamination was easily identified and facilitated the definition of the areas to be analysed. Because type I cells are relatively rare it was simple to count them all in the whole set of sections containing the hippocampal area. Therefore, the cell counts represent the total number of type I cells in the entire hippocampus. Type II cells cannot be reliably counted using diaphorase staining because of the much less intense staining and thus were not included in the quantitative analysis. Relative histochemical activity Seven

hippocampal NADPH-d reacted sections were selected from each control and prion-diseased brain. Digital images were captured using a video camera attached to the microscope to measure the density of the reaction product. The transmitted light in different dorsal hippocampal layers, cell soma or dendrites was assayed using LEICA-QWin system for densitometry to assess changes in the neuropil. Rectangular windows containing the maximum area possible without passing into other hippocampal strata were densitometrically analysed. Analyses of seven sections were performed to obtain an average for each animal and then the four animals from each experimental group were averaged in order to calculate a mean for n = 4.



Changes in the diaphorase formazan deposition within the dendritic tree was assessed by comparing transmittance in secondary and tertiary dendritic branches of at least 30 neurones, from each of four animals in each experimental group. In order to minimize sample bias in the cells chosen for transmittance measurements, we have chosen sections at random and analysed all type I neurones with well-preserved dendritic trees within the layers of CA1. The average of at least 30 cells was used to calculate averages for each animal and these averages were used to calculate a mean for n = 4 to compare experimental groups. The group of sections used for dendritic transmittance measurements was also used for the neuropil measurements. NADPH-d staining is subject to inter-run variability and decreasing activity with time. Thus, for the sake of the comparative analysis of tissues that have not been reacted at the same time and thus have experienced slightly different experimental conditions, we have adopted a normalized relative scale based on a choice of an internal control measurement in each section or neurone that is not affected significantly by the disease. This allows us to make comparisons between sections stained at different times. For neuropil measurements, the transmitted light through the granular layer of the dentate gyrus was chosen as an internal control and was designated Tmax. The transmittance of the layer under study was designated Tmin and a contrast index was calculated according to the previously published equation [25]: C = (Tmax - Tmin)/(Tmax + Tmin). Contrast indices are estimated for each layer in each section. For dendritic measurements Tmax is the transmittance through the dendritic branch and Tmin is that through the cell soma (internal standard). Contrast indices were also calculated for every chosen cell. Because the dendrites show a higher transmittance than the internal standard (the soma) the contrast index was calculated in order to graphically represent the data as a decrease in density of staining rather than as an increase in transmittance. Thus, for dendritic changes the contrast is calculated according to the equation: 1/C = (D + S)/(D - S), where D = transmittance S = transmittance in soma.

295

The calculation of contrast indices has the advantage that it normalizes the relative scales for each morphological parameter, and we have found that this procedure reduces the effect of the staining variability associated with different histological parameters (e.g. degree of tissue fixation, thickness, inter-run variability, etc.), between different sections both within and between groups.

Statistical analysis Parametric statistical analyses were performed by ANOVA with Newman Keuls post hoc test and the differences between groups were accepted as significant at the 99% confidence level (P < 0.01). In all cases multiple analyses for each animal were averaged before using these averages to calculate the group mean to be used in calculations for statistical significance. Thus, for all statistical calculations n = 4.

Results Several significant changes in the distribution of hippocampal NOS enzymes both in neuropil and NADPH-d type I neurones were observed in prion-diseased mice. In the hippocampus NADPH-d staining shows a clear laminar pattern that was coincident with Nissl-stained layers of the dorsal hippocampus both in control and prion-infected animals. The entire hippocampal region was permeated by a network of fine varicose NADPH-d stained fibres, which branched extensively throughout the tissue, as well as by a diffuse blue pattern in the neuropil in all layers, easily seen in both control and prion-infected animals at the early stages of the disease. This staining is dramatically reduced at late stages. The histochemical reaction also revealed two different types of NADPH-d stained cells: a slightly stained subpopulation (type II) and a subgroup of deeply stained neurones resembling a Golgi impregnation (type I). These nonpyramidal neurones were typically smooth or very sparsely spiny. Their dendritic arbors were very well stained but their axons were not always evident. These heavily stained neurones showed variable dendritic arbor morphologies and based on these morphologies they are likely to be local circuit neurones [26].

Neuronal changes in

dendrite

and

Both type I and type II diaphorase-positive neurones were found at all stages of the disease throughout the hippoc-


296


ampal formation. The type I population of the hippocampal formation is variable and relatively small with the majority of cells occurring in the subiculum and in the most ventral and caudal part of the CA3. Type I neurones were not found in the dentate gyrus. We did not detect any selective loss of the type I cells throughout the disease. In fact there was a nonstatistically significant increase in these cells in early and late prion disease with respect to NBH-injected animals. The numbers of type I cells counted in the entire hippocampal formation are shown in Table 1. The evolution of the NADPH-d staining in primary and secondary dendrites of type I neurones is shown in

Blate scrapie

A normal

Figure 1. Diaphorase staining intensity in the secondary and tertiary dendrites of type I neurones was decreased at late stages of disease (Figure 1B) relative to normal animals (Figure 1A). There was relative sparing of staining of Table 1. Number of type I NADPH-diaphorase positive neurones in entire hippocampal formation (± SD) 220 ± 25 261 ± 45 261 ± 48

NBH 13 weeks 19–21 weeks

NADPH, nicotine adenine dinucleotide phosphate; NBH, normal brain homogenate.

G normal SO Py SR SL SM Gr Po

C late scrapie

D early scrapie

H early scrapie

E normal

F late scrapie

I late scrapie

S. Lac

Figures 1. A,B. High power pictures of NADPH-d positive neurones in control (A), and late stage (B) illustrating decreased histochemical staining in dendrites. Arrows indicate the decreased staining in secondary dendrites. C,D. High power illustrations of late stage dendritic (C) and early stage axonal spheroids (D); arrows indicate the spheroids (A–D scale bar = 10 mm). E,F. Double staining for NADPH-d and nNOS immunocytochemistry in control (E) and late stage prion disease (F) (scale bar = 0.3 mm). Diaphorase staining is shown in blue, DAB staining appears black/dark brown, and arrowhead indicates the stratum lacunosum molecular. G–I. Low power pictures of NADPH-d histochemically stained sections of the mouse hippocampus in control (G), early (H) and late (I) stages of the disease (scale bar = 1 mm). SO, stratum oriens; Py, pyramidal cell layer; SR, stratum radiatum; SL, stratum lacunosum moleculare; SM, stratum moleculare of the dentate gyrus; Gr, granule cell layer; Po, polymorphic layer. NADPH, nicotine adenine dinucleotide phosphate; NADPH-d, NADPH-diaphorase; nNOS, neuronal nitric oxide synthase; DAB, diaminobenzidine. © 2004 Blackwell Publishing Ltd, Neuropathology and Applied Neurobiology, 30, 292–303


the cell soma. This was quantified by densitometric analysis and analysis of variance revealed statistically significant decreases in intensity at both early and late stages (ANOVA with Newman-Keuls post hoc test, P < 0.01). These data are shown in Figure 2A. There was also evidence for dendritic and axonal pathology in the type I diaphorase-

A

50

297

positive populations. Abnormal varicosities were found in both dendrites and axons of prion-diseased animals at early and late stages of the disease but not in control animals. These changes are more frequent at late than at early stages. Figure 1C depicts a dendritic swelling at the late stage of disease whereas Figure 1D shows an example of an axonal spheroid at 13 weeks postinjection.

Neuropil changes

45 40 1/C=(D+S)/(D–S)

35 30 25 20 15 10 5 0 Control

B

Early

Late

0.3

Contrast=(GrDG – L)/(GrDG + L)

PoDG MolDG

0.25

Lmol Rad

0.2

Or 0.15

0.1

0.05

0

Control

Early

Late

Figure 2. (A) Densitometric analysis of the dendritic tree in control, early stage and late stage of the disease. The staining for the secondary and tertiary dendrites is represented by ([U25CF]) and (), respectively. Decreased staining was found to be statistically significant at the late stages (P < 0.01, ANOVA with Newman-Keuls post hoc test). (B) Densitometric analysis of neuropil NADPHdiaphorase staining in control, early and late stage disease. Experimental groups were found to be statistically significantly different on neuropil staining in the Lmol (P < 0.01) and MolDG (P < 0.05) by ANOVA with Newman-Keuls post hoc test. Error bars are not shown for the other layers to preserve clarity. NADPH, nicotine adenine dinucleotide phosphate; PoDG, polymorphic layer of dentate gyrus; MolDG, molecular layer of dentate gyrus; Lmol, lacunosum molecular; GrDG, granular layer of dentate gyrus; D, transmittance in dendrite; S, transmittance in soma; L, transmittance in layer under study.

The dorsal hippocampal layers of control and prioninfected animals after NADPH-d staining are shown in Figure 1G,H,I. Distinct layers of staining of different intensity define the lamina of the hippocampal formation: from top to bottom these are the CA1 (stratum oriens, pyramidal cell layer, stratum radiatum, and stratum lacunosum molecular), and the dentate gyrus (strata molecular, granular and polymorphic). The pyramidal cell layer of CA1 and the granular cell layer of the dentate gyrus show very low NADPH-d staining. The lacunosum molecular of CA1 and the molecular layer of the dentate gyrus are the most densely NADPH-d stained layers of the hippocampus, and together these layers define the margins of the hippocampal fissure where larger blood vessels of the hippocampus are present (Figure 1G). In animals with prion disease, at the point at which behavioural changes first become apparent [4], NADPH-d positive layers tend to show somewhat increased staining (Figure 1H) with respect to NBHinjected animals (Figure 1G), whereas at late stages (19– 21 weeks) staining decreases dramatically (Figure 1I). The results of densitometric analysis of diaphorase staining of the neuropil are shown in Figure 2B. As is apparent in Figure 2B, neuropil NADPH-d histochemical activity in all layers, except the pyramidal cell layer, exhibits an increase at early stages, followed by a decrease at late stage. This was apparent in all layers but diaphorase staining in the lacunosum molecular layer of CA1 and the dorsal molecular layer of the dentate gyrus was found to be significantly increased at 13 weeks and dramatically decreased at 19–21 weeks (lacunosum molecular ANOVA with Newman-Keuls post hoc test, P < 0.01; molecular layer of dentate gyrus P < 0.05). A dark blue band of abnormal intense histochemical activity (not illustrated) was found near the injection site in the pyramidal cell layer at early stage both in ME7- and NBH-injected animals but not at later stages. Double staining for nNOS immunocytochemistry and NADPH-d histochemistry in control sections (Figure 1E)


298


revealed that all layers except the lacunosum molecular present strong double staining, recognized by the dominant brown colour produced by the DAB immunohistochemistry reaction superimposed on the blue formazan product. In addition to the diffuse brown immunostaining extended throughout the hippocampus a dark brown band in the inner third of the molecular layer of the dentate gyrus was also apparent. This layer reflects increased density of nNOS in the inner molecular layer. The stratum lacunosum moleculare, which is clearly stained for diaphorase, is almost completely unstained for nNOS. This is apparent as an isolated blue layer defining the superior margin of the hippocampal fissure, and is marked by an arrowhead (Figure 1E). Thin layers of nNOS immunolabelling were apparent at both the superior limit and to a lesser degree the inferior limit of the stratum radiatum in control animals. The nNOS-specific staining is almost completely lost in the late stage of disease (Figure 1F) as was also shown for diaphorase staining, shown in Figure 1I. The dense staining in the inner molecular layer of the dentate gyrus is also destroyed at the late stage of disease, while the dense staining at the superior and inferior limits of the stratum radiatum is also diminished. NOS-immunopositive cells were most numerous in the subiculum and in the hilar region of the dentate gyrus. In general the pattern of the immunostaining was somatodendritic and spared the nucleus of the interneurones. In contrast, somata of pyramidal and granular neurones were immunonegative or very weakly stained. Type II diaphorase-positive cells become more obvious upon double labelling with anti-nNOS antibody in animals lacking significant neuropil staining (Figure 1F). In the CA1 and dentate gyrus, sparse somatodendritic immunostaining was observed in a small number of interneurones scattered throughout the layers. The double labelling revealed that the majority of these cells were both NADPH-d and nNOS positive. These cells have not been quantified as diaphorase staining is not sufficiently intense for reproducible counting of this cell population. However, it appears that the numbers of nNOS-stained cells may increase at the late stage of disease.

sc

disease. PrP deposition is visible throughout the hippocampus and is most obvious in the hilus of the dentate gyrus and along the mossy fibres that innervate CA3. This labelling is markedly increased at 20 weeks relative to 13 weeks (Figure 3A–C). Microglial activation (Figure 3D–F) is apparent throughout the hippocampus at 13 weeks but is most conspicuous in the region of the strata oriens, radiatum and lacunosum molecular of CA1 (Figure 3E). This microglial staining at 13 weeks corresponds more closely with the layers of most dense innervation of the CA1 of the hippocampus as illustrated by synaptophysin labelling (Figure 3G) than with the deposisc tion of PrP (Figure 3B). There is no overt cell death at 13 weeks but decreases in synaptophysin immunolabelling in the stratum radiatum are evident at this time (Figure 3H) (data published elsewhere; [27]). Synaptophysin labelling is much reduced and completely disorganized at 20 weeks (Figure 3I).

Discussion In this study we have carried out a detailed examination of the NADPH-d histochemical activity in normal and priondiseased C57BL/6J mice using the ME7 model. The distribution of NADPH-d staining and the colocalization of nNOS-positive cells with type I and type II diaphorasepositive neurones qualitatively reproduces the results described elsewhere [2,5]. Several significant changes were observed in hippocampal NOS enzyme distribution in both the neuropil and NADPH-d type I neurones in priondiseased animals. The neuropil histochemical activity in the lacunosum molecular layer of CA1 and in the molecular layer of the dentate gyrus was significantly increased at early stages and all layers were dramatically reduced in the late stages of disease. In addition, we observed a continuous decrease in the histochemical activity in secondary and tertiary dendrites of type I cells throughout the disease. However, the hippocampal type I NADPH-d neurone population was not decreased in number even at the late stages of disease. Abnormal varicosities were detected in both axons and dendrites at early and more frequently at late stages.

Pathological features at 13 and 20 weeks. A summary of some pathological features visible at 13 and 20 weeks postinoculation is shown in Figure 3. These time points are chosen to represent the time at which behavioural changes first become apparent and the late stage of

Technical considerations NADPH-d has been characterized biochemically and immunochemically as being identical to NOS [28,29]. Three different isoforms of this enzyme have been recog-



A

PrP

D

microglia

G

299

synaptophysin

Normal

B

E

H

F

I

Early scrapie

C

Late scrapie sc

Figure 3. Immunnostaining for PrP , FA11 and SY38 in normal and prion-diseased animals at 13 and 20 weeks postinoculation with ME7. sc A–C. PrP labelling is visible in the hippocampus of 13 week animals, particularly in the dentate gyrus, while late stage animals show greatly sc increased levels of PrP throughout the hippocampus. Activated microglia (D–F) are found predominantly in the stratum radiatum of the CA1 at 13 weeks (E), becoming more generalized by 20 weeks (F). G–I. The regions of most dense innervation of the CA1 are shown by synaptophysin labelling in the normal animal (G). Labelling in the stratum radiatum is decreased at 13 weeks (H) and hippocampal synaptophysin labelling becomes highly disorganized by 20 weeks (I). (scale bar = 500 mm.) sc PrP , insoluble prion protein of protease-resistant form.

nized: neuronal NOS (nNOS), inducible NOS (iNOS) and endothelial NOS (eNOS). There are other enzymes with diaphorase activity in the brain but it has been shown that aldehyde fixation of the tissue inhibits other NADPH-d activities while enhancing histochemical staining in the nNOS-positive population [23]. Thus, it is reasonable to conclude that the NADPH-d histochemical changes observed here reflect changes in NOS activity. However, this does not inform on which NOS isoform is altered. The double staining for NADPH-d and nNOS shows that NADPH-d activity colocalizes with nNOS immunostaining in most hippocampal layers, although the strongly positive NADPH-d histochemical staining in the lacunosum

molecular does not colocalize with immunopositivity for nNOS. Previous findings in the mouse hippocampus demonstrate that the lacunosum molecular layer of CA1 is immunonegative for nNOS [30]. The isoform responsible for the activity in the lacunosum molecular in the current study remains to be identified. However, it is clear that in all other hippocampal layers where NADPH-d activity is destroyed late in the disease, the nNOS neuropil staining is also destroyed. Thus, it is unlikely that the diaphorase activity quantified in this study is the result of diaphorase activities other than the NOS-associated enzymes. With respect to the localization of the enzyme activity, previous electron-microscopy reports suggest that hippoc-


300


ampal NADPH-d activity is predominantly presynaptic [31]. Additionally, NADPH-d is an activity-dependent enzyme; synaptic input has a greater effect on nNOS activity than on its protein expression [32]. This is consistent with the data of Ovadia et al. [20], which demonstrate that decreased diaphorase activity in the cerebellum of priondiseased mice correlates with decreased citrulline production rather than with decreased nNOS transcription/ translation. These data indicate that NADPH-d staining, which examines activity, is more likely to reveal dysfunction than methods examining expression only.

NADPH-d in prion disease: neuronal and neuropil changes in NADPH-d Three previous studies describe NADPH-d histochemical activity in prion disease but these studies focused only on the late stage of disease. In the current study the type I NADPH-d positive neurones were not decreased in number through the disease. NADPH-d type I cells are also reported to be resistant to acute ischaemic death and to be spared in chronic neurodegenerative diseases [33–35]. Although this cell population survives late into the disease, we demonstrate a continuous decrease in the activity of the histochemical activity in secondary and tertiary dendrites of type I cell throughout the disease. This is consistent with previous reports of diminished and abnormally localized neuronal NOS in the cerebellum of scrapieinfected mice [19,20] and with Alzheimer’s patients showing sparing of NADPH-d type I neurones but reduced expression of endothelial NOS and neuronal NOS [36,37]. However, our data are not consistent with reported elevations of nNOS in the septum, thalamus, hypothalamus and amygdala of infected hamsters [22]. These differences between mouse and hamster could be related to interspecies differences and regional differences within the brain, as well as differences in the prion strain used: some hippocampal and hypothalamic pathology has been described in 139H and 263K, respectively [22], but these strains are not sufficiently well characterized to allow detailed comparison here. However, previous studies of NADPH-d activity in prion disease did not examine changes in the neuropil and the dramatic decreases seen at late stages in the current study are consistent with decreased dendritic staining also reported here and with the decreased activities reported by Ovadia et al. [20] and Keshet et al. [19].

Abnormal varicosities have previously been reported in CA1 pyramidal cells in similar models of prion disease using fluorescent dye injections in brain slices [38,39]. In this study we report, for the first time, abnormal varicosities in both axons and dendrites in the NADPH-d population. These morphological changes were observed at early and more frequently at late stages and could suggest early impairment in metabolite transport on NADPH-d positive interneurones. Proteins such as kinesin and cytoplasmic dynein, two major molecules responsible for axonal transport, are found in axon spheroids of chronic neurodegenerative disorders [40] and their accumulation in spheroids in motor neurone disease impairs fast axonal transport.

Functional consequences of altered NO production The strong diaphorase staining near the arteries in the hippocampal fissure suggests a role for NADPH-d in these layers in local blood flow regulation of the CA1 and dentate gyrus, and indeed there is some evidence that NO is involved in hippocampal blood flow control [15,17]. NADPH-d staining on vascular endothelial cells of cerebral blood vessels and innervation of smooth muscle cells of the arterial wall has been demonstrated [41,42]. In the hippocampus the highest densities of microvessels, arterioles and of the NADPH-d positive nerve fibre network are in the lacunosum molecular layer of CA1 [16]. Because NO contributes to local blood flow in the hippocampus and we describe statistically significant changes for NOS in the lacunosum molecular of CA1 and the molecular layer of the dentate gyrus, both at early and late stages of the disease, we predict that hippocampal blood flow may be slightly increased at early stages, and substantially decreased at late stages of the disease. In this study we have made a quantitative assessment of NADPH-d activity in the hippocampus. This decrease in activity may be secondary to the well-described neurodegenerative changes in this region, but even in this case the dramatic nature of the decrease is likely to have consequences nonetheless. Similar alterations also occur elsewhere in the brain and a major disturbance to cerebral blood flow in the late stages could account for the rapid decline observed in human and experimental prion diseases. This remains a key issue to address. There are also deleterious consequences of increased NO production early in disease. There is evidence that NO



is involved in synaptic plasticity in the stratum radiatum of CA1 [30] and thus abnormal NO levels may have a role in the previously described behavioural changes [4]. It is also known that NO can be directly neurotoxic [17] and can contribute to oxidative damage. There are now a number of studies that present evidence for altered oxidative and anti-oxidant activities in models of prion disease [5,8– 10,43]. These studies include increased mitochondrial oxygen free radical production [10] and decreased activity of anti-oxidative enzymes superoxide dismutase and glutathione peroxidase [11,43], Evidence for increased nitrotyrosine staining, a marker of peroxynitrite-associated protein damage [9], and elevated isoprostane F2, a marker of oxidative damage to membranes [8], demonstrate that altered oxidant balance has deleterious consequences. Because increased NOS activity will lead to elevated NO, it is likely that increased NADPH-d activity in the neuropil will contribute to peroxynitrite generation.

Conclusion The demonstration here of increased NADPH-d activity in the hippocampus at early stages of ME7-induced prion disease in the C57BL/6J mouse suggests that NOS may play a role in the disease from an early stage. The major latestage decrease in NOS activity in the hippocampus described here may result in significant changes in blood flow that may explain the rapid behavioural/neurological decline that characterizes the clinical phase of disease. Similarly, it is possible that the early increases in NOS activity may play a role in the first synaptic, axonal and dendritic damage reported in this model and together these possibilities highlight the need for a more detailed study of the functions of NO in this chronic degenerative disease.

Acknowledgements This work was supported by the Wellcome Trust. C.W. Picanço-Diniz and W. Gomes-Leal were supported by a grant from CNPq, Brazil.

References 1 Collinge J. Prion diseases of humans and animals: their causes and molecular basis. Annu Rev Neurosci 2001; 24: 519–50

301

2 Prusiner SB. Molecular biology of prion diseases. Science 1991; 252: 1515–22 3 Jeffrey M, Fraser JR, Halliday WG, Fowler N, Goodsir CM, Brown DA. Early unsuspected neuron and axon terminal loss in scrapie-infected mice revealed by morphometry and immunocytochemistry. Neuropathol Appl Neurobiol 1995; 21: 41–9 4 Guenther K, Deacon RM, Perry VH, Rawlins JN. Early behavioural changes in scrapie-affected mice and the influence of dapsone. Eur J Neurosci 2001; 14: 401–9 5 Kim JI, Ju WK, Choi JH, Choi E, Carp RI, Wisniewski HM, Kim YS. Expression of cytokine genes and increased nuclear factor-kappa B activity in the brains of scrapie-infected mice. Brain Res Mol Brain Res 1999; 73: 17–27 6 Johnston AR, Fraser JR, Jeffrey M, MacLeod N. Synaptic plasticity in the CA1 area of the hippocampus of scrapieinfected mice. Neurobiol Dis 1998; 5: 188–95 7 Kim NH, Park SJ, Jin JK, Kwon MS, Choi EK, Carp RI, Kim YS. Increased ferric iron content and iron-induced oxidative stress in the brains of scrapie-infected mice. Brain Res 2000; 884: 98–103 8 Minghetti L, Greco A, Cardone F, Puopolo M, Ladogana A, Almonti S, Cunningham C, Perry VH, Pocchiari M, Levi G. Increased brain synthesis of prostaglandin E2 and F2-isoprostane in human and experimental transmissible spongiform encephalopathies. J Neuropathol Exp Neurol 2000; 59: 866–71 9 Guentchev M, Voigtlander T, Haberler C, Groschup MH, Budka H. Evidence for oxidative stress in experimental prion disease. Neurobiol Dis 2000; 7: 270–73 10 Choi YG, Kim JI, Lee HP, Jin JK, Choi EK, Carp RI, Kim YS. Induction of heme oxygenase-1 in the brains of scrapieinfected mice. Neurosci Lett 2000; 289: 173–6 11 Thackray AM, Knight R, Haswell SJ, Bujdoso R, Brown DR. Metal imbalance and compromised antioxidant function are early changes in prion disease. Biochem J 2002; 362: 253–8 12 Law A, Gauthier S, Quirion R. Say NO to Alzheimer’s disease. the putative links between nitric oxide and dementia of the Alzheimer’s type. Brain Res Brain Res Rev 2001; 35: 73–96 13 Bohme GA, Bon C, Lemaire M, Reibaud M, Piot O, Stutzmann JM, Doble A, Blanchard JC. Altered synaptic plasticity and memory formation in nitric oxide synthase inhibitor-treated rats. Proceedings of the Natl Acad Sci USA 1993; 90: 9191–4 14 Haley JE, Wilcox GL, Chapman PF. The role of nitric oxide in hippocampal long-term potentiation. Neuron 1992; 8: 211–6 15 Jiang MH, Kaku T, Hada J, Hayashi Y. Different effects of eNOS and nNOS inhibition on transient forebrain ischemia. Brain Res 2002; 946: 139–47 16 Lovick TA, Brown LA, Key BJ. Neurovascular relationships in hippocampal slices: physiological and anatomical studies of mechanisms underlying flow-metabolism cou-


302

17

18

19

20

21

22

23

24

25

26

27

28


pling in intraparenchymal microvessels. Neuroscience 1999; 92: 47–60 Montecot C, Rondi-Reig L, Springhetti V, Seylaz J, Pinard E. Inhibition of neuronal (type 1) nitric oxide synthase prevents hyperaemia and hippocampal lesions resulting from kainate-induced seizures. Neuroscience 1998; 84: 791–800 Ju WK, Park KJ, Choi EK, Kim J, Carp RI, Wisniewski HM, Kim YS. Expression of inducible nitric oxide synthase in the brains of scrapie-infected mice. J Neurovirol 1998; 4: 445–50 Keshet GI, Ovadia H, Taraboulos A, Gabizon R. Scrapieinfected mice and PrP knockout mice share abnormal localization and activity of neuronal nitric oxide synthase. J Neurochem 1999; 72: 1224–31 Ovadia H, Rosenmann H, Shezen E, Halimi M, Ofran I, Gabizon R. Effect of scrapie infection on the activity of neuronal nitric-oxide synthase in brain and neuroblastoma cells. J Biol Chem 1996; 271: 16856–61 Walsh DT, Betmouni S, Perry VH. Absence of detectable IL-1beta production in murine prion disease: a model of chronic neurodegeneration. J Neuropathol Exp Neurol 2001; 60: 173–82 YeX, Meeker HC, Scallet AC, Carp RI. Comparison of NADPH diaphorase activity in the brains of hamsters infected with scrapie strains 139H or 263K or with normal hamster brain homogenate. Histol Histopathol 2001; 16: 997–1004 Buwalda B, Nyakas C, Gast J, Luiten PG, Schmidt HH. Aldehyde fixation differentially affects distribution of diaphorase activity but not of nitric oxide synthase immunoreactivity in rat brain. Brain Res Bull 1995; 38: 467– 73 Scherer-Singler U, Vincent SR, Kimura H, McGeer EG. Demonstration of a unique population of neurons with NADPH-diaphorase histochemistry. J Neurosci Methods 1983; 9: 229–34 Borba JM, Araujo MS, Picanco-Diniz CW, Manhaes-deCastro R, Guedes RC. Permanent and transitory morphometric changes of NADPH-diaphorase-containing neurons in the rat visual cortex after early malnutrition. Brain Res Bull 2000; 53: 193–201 Gabbott PL, Bacon SJ. Co-localisation of NADPH diaphorase activity and GABA immunoreactivity in local circuit neurones in the medial prefrontal cortex (mPFC) of the rat. Brain Res 1995; 699: 321–8 Cunningham C, Deacon R, Wells H, Boche D, Waters S, Diniz CP, Scott H, Rawlins JN, Perry VH. Synaptic changes characterize early behavioural signs in the ME7 model of murine prion disease. Eur J Neurosci 2003; 17: 2147–55 Bredt DS, Glatt CE, Hwang PM, Fotuhi M, Dawson TM, Snyder SH. Nitric oxide synthase protein and mRNA are discretely localized in neuronal populations of the mammalian CNS together with NADPH diaphorase. Neuron 1991; 7: 615–24

29 Dawson TM, Bredt DS, Fotuhi M, Hwang PM, Snyder SH. Nitric oxide synthase and neuronal NADPH diaphorase are identical in brain and peripheral tissues. Proceedings of the Natl Acad Sci USA 1991; 88: 7797– 801 30 Burette A, Zabel U, Weinberg RJ, Schmidt HH, Valtschanoff JG. Synaptic localization of nitric oxide synthase and soluble guanylyl cyclase in the hippocampus. J Neurosci 2002; 22: 8961–70 31 Faber-Zuschratter H, Wolf G. Ultrastructural distribution of NADPH-diaphorase in cortical synapses. Neuroreport 1994; 5: 2029–32 32 Weruaga E, Brinon JG, Porteros A, Arevalo R, Aijon J, Alonso JR. Expression of neuronal nitric oxide synthase/NADPH-diaphorase during olfactory deafferentation and regeneration. Eur J Neurosci 2000; 12: 1177– 93 33 Ferrante RJ, Kowall NW, Beal MF, Richardson EP Jr, Bird ED, Martin JB. Selective sparing of a class of striatal neurons in Huntington’s disease. Science 1985; 230: 561–3 34 Ferriero DM, Arcavi LJ, Sagar SM, McIntosh TK, Simon RP. Selective sparing of NADPH-diaphorase neurons in neonatal hypoxia-ischemia. Ann Neurol 1988; 24: 670– 6 35 Hyman BT, Marzloff K, Wenniger JJ, Dawson TM, Bredt DS, Snyder SH. Relative sparing of nitric oxide synthase-containing neurons in the hippocampal formation in Alzheimer’s disease. Ann Neurol 1992; 32: 818– 20 36 de la Monte SM, Bloch KD. Aberrant expression of the constitutive endothelial nitric oxide synthase gene in Alzheimer disease. Mol Chem Neuropathol 1997; 30: 139– 59 37 Norris PJ, Faull RL, Emson PC. Neuronal nitric oxide synthase (nNOS) mRNA expression and NADPH-diaphorase staining in the frontal cortex, visual cortex and hippocampus of control and Alzheimer’s disease brains. Brain Res Mol Brain Res 1996; 41: 36–49 38 Belichenko PV, Brown D, Jeffrey M, Fraser JR. Dendritic and synaptic alterations of hippocampal pyramidal neurones in scrapie-infected mice. Neuropathol Appl Neurobiol 2000; 26: 143–9 39 Brown D, Belichenko P, Sales J, Jeffrey M, Fraser JR. Early loss of dendritic spines in murine scrapie revealed by confocal analysis. Neuroreport 2001; 12: 179–83 40 Toyoshima I, Sugawara M, Kato K, Wada C, Hirota K, Hasegawa K, Kowa H, Sheetz MP, Masamune O. Kinesin and cytoplasmic dynein in spinal spheroids with motor neuron disease. J Neurol Sci 1998; 159: 38–44 41 Gabbott PL, Bacon SJ. Histochemical localization of NADPH-dependent diaphorase (nitric oxide synthase) activity in vascular endothelial cells in the rat brain. Neuroscience 1993; 57: 79–5 42 Tsuchida A, Handa Y, Nojyo Y, Kubota T. Ultrastructure of NADPH diaphorase-positive nerve fibers and their ter-



minals in the rat cerebral arterial system. J Chem Neuroanat 2001; 21: 267–75 43 Milhavet O, McMahon HE, Rachidi W, Nishida N, Katamine S, Mange A, Arlotto M, Casanova D, Riondel J, Favier A, Lehmann S. Prion infection impairs the cellular

303

response to oxidative stress. Proceedings of the Natl Acad Sci USA 2000; 97: 13937–42


Received 16 July 2003 Accepted after revision 5 October 2003