Differential Effect of the Route of Inoculation of Rabies Virus on NeuN Immunoreactivity in the Cerebral Cortex of Mice/ Efecto Diferencial de la Via de Inoculacion del Virus de la Rabia sobre la Inmunorreactividad de NeuN en la Corteza Cerebral de Ratones.
Rabies remains a public health problem in developing countries. Rabies can be prevented by the vaccination of pets and people at high risk of contact with the virus. The entire population is not vaccinated because of the high cost associated with the vaccine and other factors. In addition, it is almost impossible to eradicate a virus that has wild vectors. Despite the availability of preventive controls, there are still approximately 55.000 deaths reported yearly due to rabies worldwide. The destruction of natural habitats and climate change are factors that cause vector displacement to areas that are inhabited by humans and domestic animals, increasing the risk of disease transmission (WHO, 2013).
The rabies virus spreads through retrograde axonal transport from the bite of an infected animal. Once the virus enters into the nerve terminals of the new host, it is transported to the motor neurons of the spinal cord and advances to the brain, causing marked neuronal dysfunction with fatal results. Despite the severity of clinical symptoms, rabies produces only a low inflammatory and microglial reaction in the nervous tissue. Only small brain abnormalities and the presence of eosinophilic cytoplasmic inclusions called Negri bodies are observed with conventional anatomical and histopathological analysis methods. Negri bodies are formed by a granular matrix surrounded by viral particles (Iwasaki & Tobita, 2002).
Rabies does not appear to induce massive neuronal loss. However, in some experimental studies that have been performed, apoptosis were reported in some neurons principally when the virus was inoculated intracerebrally (Jackson & Fu, 2013). Therefore, it is important to perform studies about the dynamics of cellular infection using alternative tools as the neuronal markers. These molecules have allowed evaluating parameters as changes in the structure, number and distribution of neurons in the cerebral cortex of rabies-infected mice. (Torres-Fernandez et al, 2004; Torres-Fernandez et al, 2005; Rengifo & TorresFernandez, 2007; Hurtado et al, 2015).
The neuronal nuclear protein (NeuN) has become a widely used neuronal marker for the histopathological diagnosis of nervous system diseases and for research on embryonic development, adult neurogenesis and neuronal loss studies in natural and experimental conditions (Wolf et al., 1996; Sarnat et al., 1998; Gittins & Harrison, 2004; Yang et al., 2011). These features make this protein an excellent marker for the study of neurological diseases that induce morphological changes that are difficult to detect, such as rabies disease. However we don't know previous studies about NeuN in rabies. The distribution of NeuN in the cerebral cortex of mice under normal conditions and the NeuN immunoreactivity in rabies-infected mice was assessed in this work. In addition, the effects of two different pathogen inoculation routes were compared to contribute knowledge about rabies pathogenesis.
MATERIAL AND METHOD
Laboratory animals and viral inoculation. A total of 20 female ICR (Institute of Cancer Research) mice at 28 days old were used. Mice were maintained in the animal facility of the National Institute of Health (Instituto Nacional de Salud, INS) in the high-security area under environmental and nutritional conditions in accordance with the ethical and legal standards required for animal laboratory research approved by the Ethics Committee of the INS and Colombian law (Law 86 of 1989 and resolution No. 008430 of 1993 of the Ministry of Health). Two groups of 5 mice were inoculated with 0.03 ml (each animal) of a solution of the challenge virus standard (CVS) strain of rabies virus by two different infection routes. The first group was inoculated intramuscularly with a viral dilution of 10 (9.1) (LD50), and the second group was inoculated intracerebrally with a dilution of 10 (4.1) (LD50). For each group of infected animals, an equivalent number of control animals inoculated under the same conditions with only the viral carrier (virus-free solution containing 2 % standard horse serum and the antibiotics penicillin and streptomycin at 200 UI/ml and 4 mg/ml, respectively) was used.
Procedure to obtain brain tissue sections. Mice showing signs associated with the rabies terminal phase were selected. Signs were observed for approximately 5 to 7 days after inoculation. Animals were anesthetized by injecting 0.2 ml of 30% chloral hydrate intraperitoneally (350 mg/kg). The animals were then fixed by intracardiac perfusion first with phosphate buffered saline (PBS) solution at pH 7.3 for 5 minutes and then with a 4 % paraformaldehyde fixative solution prepared in phosphate buffer (70 ml of solution for 10 minutes). PBS was allowed to circulate again for one minute after perfusion when the fixation was complete. Brains were carefully extracted and were immersed in the same fixing solution that was used for perfusion. Samples were maintained at 4[degrees]C for a minimum of 20 hours up to two weeks before processing for immunohistochemistry.
Hemispheres were separated, and tissue slices with a maximum thickness of 1 cm were obtained from each. Slices were made on the coronal plane from a tissue brain fragment between the corpus callosum knee and the dorsal fornix; this area mainly covers the motor cortex according to the Valverde's mouse atlas (Valverde, 1998). These tissue blocks were placed on a vibratome to obtain 50-[micro]m-thick coronal sections that were then transferred to small circular glass boxes with a 1.5 cm radius containing PBS. In each case, the sections were placed and allowed to stir (in a horizontal shaker-rotator) at room temperature (20 [degrees]C) overnight.
Immunohistochemistry protocol. The sections were processed in suspension during all stages of the immunohistochemical protocol to detect NeuN and with constant stirring at room temperature (20[degrees]C). PBS at pH 7.3 was used for the initial wash and for all washes after each stage of the process. After the first wash, the sections were treated with 0.05 M ammonium chloride to counteract the effect of aldehydes and with 3 % hydrogen peroxide to inactivate endogenous peroxidases. The sections were then incubated in normal horse serum and bovine serum albumin.
Three different dilutions of the primary antibody (monoclonal anti-NeuN from Chemicon-Millipore; Germany, Darmstadt) were tested: 1:625, 1:1250 and 1:2500. Incubation with primary antibody was performed overnight (20 hours) due to the thickness of the sections. The next day, after washing with PBS, sections were incubated with the secondary antibody (biotinylated polyclonal anti-mouse IgG from Sigma) for two hours. Two dilutions of the secondary antibody were tested, 1:400 and 1:600, which were prepared in PBS. After washing, the sections were treated with ABC Vector[R] solution (avidin- biotin complex) for two hours. Then, the immunostain were developed using a solution of the diaminobenzidine (DAB) chromogen. Different developing times were tested. Sections were rapidly washed with distilled water and then with PBS to end the reaction. Finally, sections were extended on 1 % gelatin-pretreated glass slides, air dried and mounted with Entellan[R].
Histological and digital image analysis. Initially, qualitative observations of immunohistochemical preparations obtained from control animals were performed to determine the presence of NeuN-immunoreactive (NeuN+) cells and to observe the distribution of NeuN under normal conditions. The same procedure was applied to samples taken from groups of animals inoculated with rabies virus. Panoramic and detailed pictures of histological sections of normal and infected samples were taken to perform quantitative analysis. Three neuron counts from the motor cortex of each sample from control and virusinfected animals were performed using a Zeiss-Netzmiier microscope with a 1 [mm.sup.2] mesh in 10x fields to locate the area of the frontal cortex. Counts were made through all cortical layers of the motor area.
Images of the cerebral cortex from control and infected samples were digitized, and cellular optical densitometry was performed using Q-Capture Pro software. Cerebral cortex images were captured in 40x fields (three sections per sample). Subsequently, the Image J program was used to quantify protein immunostaining, and diameter measurements (in micrometers) were performed in NeuN+ cells with Q-Capture Software following similar protocols to those we published previously. The quantitative analysis was performed on 5 samples of infected animals and their respective controls for each study group. The statistical Wilcoxon-Mann-Whitney test (nonparametric) suitable for a small number of high-variability samples was used (Torres-Fernandez et al, 2004; Rengifo & Torres-Fernandez).
Distribution of neurons immunoreactive for NeuN in the cortex of normal mice. Antibody dilutions for the best obtained immunostaining were 1:1250 for the primary antibody and 1:400 for the secondary antibody. NeuN+ neurons were observed in all cortical layers with an apparently uniform distribution, except for layer I, where a small number of NeuN+ neurons (Figs. 1a and 2 a) was noted. However, the counts performed on each of the layers of the cortex showed a higher concentration of immunoreactive cells in layers III, V and VI in all groups of animals. The morphology exhibited by labeled neurons revealed highly homogeneous cell bodies and no labeling of cellular processes (dendrites) (Fig. 1c).
The effect of rabies virus on the number of NeuN+ cortical neurons. A panoramic view of the microscopically observed histological preparations showed decreased staining of NeuN+ cells in the cerebral cortex of infected animals (Figs. 1b and 1d); in contrast, a loss in the number of NeuN+ neurons was not evident. However, it was found that infection with rabies by either of the two routes of infection decreased the total number of NeuN+ cells in the inoculated animals, but this decrease was only statistically significant for the group inoculated intracerebrally.
The number of neurons NeuN+ was 1754 [+ or -] 69 in the control group and 1568 [+ or -] 72 in the group infected intracerebrally (p = 0.0317) (Table I). In the group infected intramuscularly, the number of neurons NeuN+ was 1407 [+ or -] 144, whereas in controls, it was 1458 [+ or -] 172 (p = 0.7937) (Table II). The loss of neurons (NeuN+) was more pronounced in cortical layer V of animals inoculated intracerebrally, especially in cortical layer Vc (p = 0.008) (Table I, Figs. 1d and 2b), whereas the difference was not statistically significant for any of the cortical layers in animals inoculated intramuscularly (Table II).
Optical densitometry and NeuN+ cell size. A loss of immunostaining was observed in all cortical layers and especially in layer V (Figs. 1 and 2). The optical densitometry analysis showed decreased cell immunolabeling was only statistically significant for the tissues of animals inoculated intracerebrally (p = 0.0079) (Table III), whereas in mice inoculated intramuscularly, the difference was not statistically significant. These results are consistent with the observed neuronal counts. In addition, a decreasing trend in the size of the neurons profiles in both groups of infected animals was found after measuring the diameter of NeuN-immunoreactive neurons (intracerebral, p = 0.0317; intramuscular, p = 0.0546) (Table IV).
Standardization of dilutions for NeuN immunohistochemistry. In reviewing the literature, there was no consistent correlation between the methods used and the dilution of anti-NeuN antibody chosen in each case. In the original reference describing NeuN, protein samples were embedded in plastic resin, and the dilution used was 1:100 (Mullen et al., 1992). In most cases, nervous tissue samples were processed by paraffin embedding, and the dilutions used were very diverse: 1: 100 (Gittins & Harrison), 1:500 (Wolf et al.; Unal-Cevik et al, 2004; Collombet et al, 2006), 1:1000 (Gill et al, 2005), 1:1500 (Sarnat et al.) and 1:2000 (Salamon et al., 2006). Additionally, the work performed on sections obtained by cryotomy (sections from frozen tissue), which better preserves the immunoreactivity of antigens, reported dilutions ranging from 1:100 to 1:5000 (Jongen-Relo & Feldon, 2002; Lee et al., 2003; McPhail et al., 2004; Yang et al.). Sections obtained with the vibratome used in this study are also suitable for good preservation of antigens; however, there are few previous reports of using vibratome sections for the study of NeuN (Escobar et al., 2008; Torres-Fernandez et al., 2008), in these studies, a dilution of 1:2500 was reported. The variability in the dilutions cannot be explained by the origin of the antibodies because most studies were performed with a single commercially available antibody; only in recent years have we seen anti-NeuN antibodies available from other brands. Given the above information, it was necessary to perform the assays that led us to establish the dilution of 1:1250 as that with which the best results were observed. The selection criterion was to obtain good labelling of neuronal cell bodies and lower nonspecific background staining.
Distribution of NeuN+ neurons in the normal mouse cortex. The results of this study are consistent with our previous study (Torres-Fernandez et al., 2008) in which NeuN+ neurons were observed to be uniformly distributed in dense layers thorough the mouse cortex, except for in layer I, where there were only a small number of neurons. This distribution of NeuN+ neurons in the cortex of mice is similar to that observed in other studies conducted on mice (Unal-Cevik et al.) and rats (Jongen-Relo & Feldon) and in the human cerebral cortex (Escobar et al.). The shortage of NeuN+ neurons in layer I of the cortex simply reflects the known fact that this layer has few neurons. There are two types of neurons in cortical layer I: Cajal horizontal cells (also called Cajal-Retzius) and other types of smaller neurons (Marin-Padilla, 1998). Some authors have described the lack of NeuN staining in Cajal-Retzius cells. However, in this work and in our previous work the presence of NeuN+ cells in layer I was evident. The results are similar to those reported in mice, rats and humans (Unal-Cevik et al.; Jongen-Relo & Feldon; Escobar et al.). Moreover, these cells are also immunoreactive for the protein calretinin (DeFelipe, 1997). In a previous study, we found the presence of calretinin immunoreactive cells in layer I of the mouse cortex (Torres- Fernandez et al., 2004), and its distribution was highly similar to that observed for NeuN immunostaining. Therefore, at least for the mouse cerebral cortex, cells of the molecular layer (layer I) have both neuronal markers, i.e., NeuN and calretinin.
Effect of rabies virus on NeuN immunoreactivity in mouse cortical neurons. Infection by intracerebral inoculation of rabies virus decreased NeuN immunoreactivity in the cerebral cortex of mice, as established by fewer NeuN+ cells as well as a decrease in immunostaining calculated by optical densitometry and the size in the profile of neurons. Whether this may be related to neuronal death or whether infection with rabies virus affects the expression of the protein or only its immunoreactivity is still undefined. The use of NeuN immunohistochemistry in studies that require neuronal counts is the more reliable alternative to the traditional method based on Nissl staining with cresyl violet or thionine because NeuN more readily distinguishes between small neurons and glia (Gittins & Harrison). However, the usefulness of NeuN staining was interrogated for this purpose in a study which showed loss of immunoreactivity of the protein, but no neuronal death (Unal-Cevik et al.). A subsequent investigation criticized this earlier study and supported the usefulness of NeuN for neuronal counts, suggesting that a reasonable time should elapse before the effect of a disease on a neuronal population was assessed (Collombet et al.).
The loss of NeuN immunoreactivity induced by rabies virus intracerebral inoculation could occur because this type of experimental infection in mice induces neuronal death by apoptosis (Jackson & Fu). However, studies that correlate neuronal count with Nissl staining, quantify neuronal death by apoptosis and label cells with NeuN are needed to definitively establish whether this protein is a good marker for studies of rabies-induced neuronal loss in different areas of the nervous system. Moreover, no loss in the number of NeuN+ cells in the cortex of animals intramuscularly inoculated with rabies virus was found. This observation could indicate that in this condition, there is no neuronal death and no effect on protein expression. We should keep in mind that this experimental model more closely simulates what happens during natural infection. It has been established in rabies infections of both humans and other animals that under natural conditions, this viral infection does not induce neuronal death by apoptosis (Jackson et al., 2008; Suja et al., 2011). Similarly, apoptosis was not observed in mice intramuscularly inoculated with rabies virus except in special experimental conditions (Jackson & Fu).
The decrease of NeuN+ immunoreactivity in layer V was more dramatic in animals inoculated intracerebrally. Studies on other diseases have reported similar effects on pyramidal neurons. In patients with head trauma, a decrease in NeuN+ neurons was found in cortical layers III and V (Escobar et al.). This susceptibility may occur because the pyramidal cells in layers III and V are part of large interconnected networks with other areas of the nervous system (DeFelipe). Finally, the susceptibility of pyramidal neurons to rabies infection has also been demonstrated in other studies we have conducted (Hurtado et al.). On the other hand, the difference in NeuN immunoreactivity in animals infected by both routes of inoculation confirms that it is necessary to interpret with caution the findings obtained in experiments with rabies where the intracerebral inoculation route was used. The results obtained by intramuscular infection with rabies better reflect the natural conditions of the disease and thus offer more realistic approaches to understanding the physiopathological mechanisms of this viral disease.
We thank to our colleague Jorge Alonso Rivera for advice and support in the processing and presentation of images.
Collombet, J. M.; Masqueliez, C.; Four, E.; Burckhart, M. F.; Bernabe, D.; Baubichon, D. & Lallement, G. Early reduction of NeuN antigenicity induced by soman poisoning in mice can be used to predict delayed neuronal degeneration in the hippocampus. Neurosci. Lett., 398(3):337-42, 2006.
DeFelipe, J. Types of neurons, synaptic connections and chemical characteristics of cells immunoreactive for calbindin-D28K, parvalbumin and calretinin in the neocortex. J. Chem. Neuroanat., 14(1):1-19, 1997.
Escobar, M. I.; Guzman, F.; Buritica, E.; Riascos, D.; Villamil, L. & Pimienta, H. Alteracion de la organizacion laminar y de la dendroarquitectura de la corteza cerebral del humano post-trauma craneoencefalico. Colomb. Med, 39(3 Supl. 3):51-9, 2008.
Gill, S. K.; Ishak, M. & Rylett, R. J. Exposure of nuclear antigens in formalinfixed, paraffin-embedded necropsy human spinal cord tissue: detection of NeuN. J. Neurosci. Methods, 148(1):26-35, 2005.
Gittins, R. & Harrison, P. J. Neuronal density, size and shape in the human anterior cingulate cortex: a comparison of Nissl and NeuN staining. Brain Res. Bull., 63(2):155-60, 2004.
Hurtado, A. P.; Rengifo, A. C. & Torres-Fernandez, O. Immunohistochemical overexpression of MAP-2 in the cerebral cortex of rabies-infected mice. Int. J. Morphol., 33(2):465-70, 2015.
Iwasaki, Y. & Tobita, M. Pathology. In: Jackson, A. C. & Wunner, H. W. (Eds.). Rabies. San Diego, Academic Press, 2002. pp.283-306.
Jackson, A. C. & Fu, Z. F. Pathogenesis. In: Jackson, A. C. (Ed.). Rabies. San Diego, Academic Press, 2013. pp.299-349.
Jackson, A. C.; Randle, E.; Lawrance, G. & Rossiter, J. P. Neuronal apoptosis does not play an important role in human rabies encephalitis. J. Neurovirol., 14(5):368-75, 2008.
Jongen-Relo, A. L. & Feldon, J. Specific neuronal protein: a new tool for histological evaluation of excitotoxic lesions. Physiol. Behav., 76(4-5):449-56, 2002.
Lee, D. R.; Helps, S. C.; Gibbins, I. L.; Nilsson, M. & Sims, N. R. Losses of NG2 and NeuN immunoreactivity but not astrocytic markers during early reperfusion following severe focal cerebral ischemia. Brain Res., 989(2):221-30, 2003.
Marin-Padilla, M. Cajal-Retzius cells and the development of the neocortex. Trends Neurosci., 21(2):64-71, 1998.
McPhail, L. T.; McBride, C. B.; McGraw, J.; Steeves, J. D. & Tetzlaff, W. Axotomy abolishes NeuN expression in facial but not rubrospinal neurons. Exp. Neurol., 185(1):182-90, 2004.
Mullen, R. J.; Buck, C. R. & Smith, A. M. NeuN, a neuronal specific nuclear protein in vertebrates. Development, 116(1):201-11, 1992.
Rengifo, A. C. & Torres-Fernandez, O. Decreased number neurons expressing GABA in the cerebral cortex of rabies-infected mice. Biomedica, 27(4):548-58, 2007.
Salamon, N.; Andres, M.; Chute, D. J.; Nguyen, S. T.; Chang, J. W.; Huynh, M. N.; Chandra, P. S.; Andre, V. M.; Cepeda, C.; Levine, M. S.; Leite, J. P.; Neder, L.; Vinters, H. V. & Mathern, G. W. Contralateral hemimicrencephaly and clinical-pathological correlations in children with hemimegalencephaly. Brain, 129(Pt. 2):352-65, 2006.
Sarnat, H. B.; Nochlin, D. & Born, D. E. Neuronal nuclear antigen (NeuN): a marker of neuronal maturation in early human fetal nervous system. Brain Dev., 20(2):88-94, 1998.
Suja, M. S.; Mahadevan, A.; Madhusudana, S. N. & Shankar, S. K. Role of apoptosis in rabies viral encephalitis: a comparative study in mice, canine, and human brain with a review of literature. Patholog. Res. Int., 2011:374286, 2011.
Torres-Fernandez, O.; Lamprea, N. P. & Santamaria, G. Expresion de la proteina nuclear neuronal NeuN en el cerebro de ratones. Rev. Asoc. Colomb. Cienc. Biol., 20(1):194-95, 2008.
Torres-Fernandez, O.; Yepes, G. E.; Gomez, J. E. & Pimienta, H. J. Effect of rabies virus infection on the expression of parvalbumin, calbindin and calretinin in mouse cerebral cortex. Biomedica, 24(1):63-78, 2004.
Torres-Fernandez, O.; Yepes, G. E.; Gomez, J. E. & Pimienta, H. J. Calbindin distribution in cortical and subcortical brain structures of normal and rabies-infected mice. Int. J. Neurosci., 115(10):1375-82, 2005.
Unal-Cevik, I.; Kilinc, M.; Gursoy-Ozdemir, Y.; Gurer, G. & Dalkara, T. Loss of NeuN immunoreactivity after cerebral ischemia does not indicate neuronal cell loss: a cautionary note. Brain Res., 1015(1-2):169-74, 2004.
Valverde, F. The Golgi Atlas of the Postnatal Mo-use Brain. New York, Springer-Verlag, 1998.
Wolf, H. K.; Buslei, R.; Schmidt-Kastner, R.; Schmidt-Kastner, P. K.; Pietsch, T.; Wiestler, O. D. & Blumcke, I. NeuN: a useful neuronal marker for diagnostic histopathology. J. Histochem. Cytochem., 44(10):1167-71, 1996.
World Health Organization (WHO). WHO Expert Consultation on Rabies. In: World Health Organization (WHO). Technical Report Series No. 982. Geneva, WHO Press, 2013.
Yang, Y.; Fung, S. J.; Rothwell, A.; Tianmei, S. & Weickert, C. S. Increased interstitial white matter neuron density in the dorsolateral prefrontal cortex of people with schizophrenia. Biol. Psychiatry, 69(1):63-70, 2011.
Grupo de Morfologia Celular
Instituto Nacional de Salud (INS)
Av. Calle 26 No. 51-20
Aura Caterine Rengifo *; Vanessa Jazmin Umbarila *; Mary Janeth Garzon * & Orlando Torres-Fernandez *
* Grupo de Morfologia Celular, Instituto Nacional de Salud (INS), Bogota, Colombia. This work was funded by Departamento Administrativo de Ciencia, Tecnologia e Innovacion COLCIENCIAS and Instituto Nacional de Salud (INS, Colombia). Code of Grant 210465740573, Contract 639/2014.
Caption: Fig. 1. Distribution of NeuN immunoreactivity in the cerebral cortex of mice. (a) Panoramic view of the cerebral motor cortex of a control mouse showing NeuN labeling. (b) Panoramic view of the cerebral cortex of a rabies-infected mouse inoculated intracerebrally. Note the decrease in NeuN immunoreactivity. (c) Magnification of layer V from the image in (a), where a high density of neurons with strong staining is observed. (d) Magnification of layer V of the image in (b) shows the lower cell density. Layer IV is not included in this picture because this layer is thought to be absent in the agranular motor cortex (Valverde, 1998). DAB stain. Scale bar, 60 [micro]m (figs. a y b); scale bar, 15 [micro]m (figs. c y d).
Caption Fig. 2. Distribution of NeuN immunoreactivity in the mouse cortex. (a) NeuN+ cells in a sample of the motor cortex from a control mouse; note the strong staining through all cortical layers. (b) NeuN+ cells in the motor cortex of a mouse inoculated with rabies virus intracerebrally; note the decreased staining, especially in layer V. The loss of NeuN-immunoreactive neurons is evident, and the immunostaining also exhibits a lower intensity. DAB stain. Scale bar, 200 mm.
Table I. Distribution of the number of NeuN+ cells observed in a 1-mm cortical column in different layers of the motor area. Mice intracerebrally inoculated with rabies virus and their controls. For each sample, the mean of the number of neurons per layer is given. Layer Control samples Mean # of cells per layer 1 2 3 4 5 I 21 32 16 26 19 23 [+ or -] 6 II 149 162 136 135 158 148 [+ or -] 12 III 430 444 468 459 475 455 [+ or -] 18 Va 146 143 101 138 123 130 [+ or -] 19 Vb 134 121 108 143 133 128 [+ or -] 14 Vc 131 126 118 159 151 137 [+ or -] 17 Vd 146 127 125 162 171 146 [+ or -] 21 VIa 151 124 142 170 184 154 [+ or -] 23 VIb 176 140 148 162 112 148 [+ or -] 24 VIc 170 138 132 114 119 135 [+ or -] 22 VId 156 167 153 122 156 151 [+ or -] 17 Total 1810 1724 1647 1790 1801 1754 [+ or -] 69 Layer Infected samples Mean # of cells per layer p value 1 2 3 4 5 I 24 19 13 28 26 22 [+ or -] 6 0.999 II 138 115 134 135 137 132 [+ or -] 10 0.111 III 429 474 486 464 449 460 [+ or -] 22 0.310 Va 106 99 98 99 131 107 [+ or -] 14 Vb 115 106 113 96 114 109 [+ or -] 8 0.056 Vc 108 104 112 110 114 110 [+ or -] 4 Vd 113 112 105 113 137 116 [+ or -] 12 VIa 116 118 101 142 161 128 [+ or -] 24 0.111 VIb 136 126 118 144 153 135 [+ or -] 14 0.421 VIc 119 108 118 124 119 118 [+ or -] 6 0.222 VId 121 132 130 137 143 133 [+ or -] 8 0.087 Total 1525 1513 1528 1592 1684 1568 [+ or -] 72 0.0317** (**) Statistically significant; ([+ or -]) Standard deviation. Table II. Distribution of the number of NeuN+ cells observed in a 1 mm cortical column in different layers of the motor area. Mice intramuscularly inoculated with rabies virus and their controls. In each sample, the mean of the number of neurons per layer is given. Layer Control samples Mean # of cells per layer 1 2 3 4 5 I 17 16 59 13 19 25 [+ or -] 19 II 108 94 156 118 117 119 [+ or -] 23 III 322 332 329 377 321 336 [+ or -] 23 Va 100 106 140 110 86 108 [+ or -] 20 Vb 99 95 140 117 100 110 [+ or -] 19 Vc 111 120 145 127 114 123 [+ or -] 14 Vd 141 130 153 156 141 144 [+ or -] 10 VIa 140 140 164 158 147 150 [+ or -] 11 VIb 132 80 147 142 101 120 [+ or -] 29 VIc 106 120 166 94 100 117 [+ or -] 29 VId 114 95 136 101 80 105 [+ or -] 21 Total 1390 1328 1735 1513 1326 1458 [+ or -] 172 Layer Infected samples Mean # of cells P value per layer 1 2 3 4 5 I 24 28 15 16 22 21 [+ or -] 5 0.7381 II 142 140 122 142 137 137 [+ or -] 8 0.1349 III 373 339 330 313 344 340 [+ or -] 22 0.6905 Va 134 105 88 88 102 103 [+ or -] 19 0.6508 Vb 123 105 84 89 110 102 [+ or -] 16 0.6905 Vc 134 95 90 97 101 103 [+ or -] 18 0.0952 Vd 156 100 95 136 134 124 [+ or -] 26 0.2302 VIa 158 93 98 137 138 125 [+ or -] 28 0.0714 VIb 124 102 111 150 134 124 [+ or -] 19 0.8413 VIc 116 117 118 114 112 115 [+ or -] 2 0.6905 VId 151 129 96 83 103 112 [+ or -] 27 0.6905 Total 1635 1353 1247 1365 1437 1407 [+ or -] 144 0.7937 Table III. Optical densitometry of NeuN immunoreactivity per cell in cerebral cortex from mice inoculated with rabies virus intracerebrally or intramuscularly. Each value corresponds to the mean of the reading of 20 cells. Reading scale (0-255 grey levels); the higher the value, greater light transmittance and, lower optical density, therefore lower immunoreactivity. Measurements at 40X. Sample Optical density. Mice inoculated intracerebrally Controls Infected 1 104.12 [+ or -] 12.99 125.65 [+ or -] 6.73 2 108.38 [+ or -] 1.30 127.67 [+ or -] 13.17 3 106.11 [+ or -] 10.58 127.62 [+ or -] 6.20 4 114.07 [+ or -] 3.89 127.77 [+ or -] 2.11 5 112.54 [+ or -] 2.52 125.43 [+ or -] 8.76 Mean 109.05 [+ or -] 4.21 126.83 [+ or -] 1.18 p = 0.0079** Sample Optical density. Mice inoculated intramuscularly Controls Infected 1 117.17 [+ or -] 5.68 97.08 [+ or -] 9.94 2 101.51 [+ or -] 5.95 114.04 [+ or -] 12.19 3 97.92 [+ or -] 5.56 122.61 [+ or -] 1.75 4 11.95 [+ or -] 4.46 109.19 [+ or -] 12.56 5 103.80 [+ or -] 3.19 102.15 [+ or -] 9.68 Mean 106.47 [+ or -] 7.90 109.01 [+ or -] 10.0 p = 0.8413 (**) Statistically significant; ([+ or -]) Standard deviation. Table IV Diameter (in micrometers) of NeuN+ cells in cerebral cortex of mice inoculated with rabies virus by intracerebral or intramuscular route and their controls. Each value corresponds to the mean of 20 cells per section in three sections. Measurements were performed with a 40X objective. Diameter of NeuN+ cells. Sample Mice inoculated intracerebrally Controls Infected 1 10.92 [+ or -] 0.08 8.70 [+ or -] 0.38 2 12.35 [+ or -] 0.19 9.80 [+ or -] 0.61 3 12.17 [+ or -] 0.21 10.78 [+ or -] 1.20 4 10.26 [+ or -] 2.39 9.93 [+ or -] 0.10 5 10.38 [+ or -] 0.19 9.36 [+ or -] 0.10 Mean 11.21 [+ or -] 0.10 9.71 [+ or -] 0.76 p = 0.0317** Diameter of NeuN+ cells. Sample Mice inoculated intramuscularly Controls Infected 1 10.39 [+ or -] 0.38 7.91 [+ or -] 0.63 2 9.42 [+ or -] 1.50 7.44 [+ or -] 0.05 3 10.57 [+ or -] 0.40 8.72 [+ or -] 0.37 4 9.66 [+ or -] 2.09 10.11 [+ or -] 2.06 5 9.53 [+ or -] 0.88 8.77 [+ or -] 0.05 Mean 9.92 [+ or -] 0.53 8.59 [+ or -] 1.02 p = 0.0546 (**) Statistically significant; [+ or -] Standard deviation.