Spatial learning during the course of autoimmune ... - Science Direct

Behavioural Bm#~ Research, 54 (1993) 57-66 @ 1993 Elsevier Science Publishers B.V. All rights reserved. 0166-4328/93/$06.00

57

BBR 01414

Spatial learning during the course of autoimmune disease in MRL mice Boris Saki6 a, Henry Szechtman a, Susan Denburg b, Ramona Carbotte b and Judah A. Denburg c Departments ql " Biomedical Sciences. ~'PsvchiatO, and "Medicine, McMaster UniversiO,, Hamilton, Ont. (Canada)

(Received 1 September 1992) (Revised version received 10 December 1992) (Accepted 17 December 1992) Key words: Spatial learning; Emotionality; Thigmotaxis; Autoimmunity; Brain-reactive antibody; Longitudinal study; MRL mouse

The present study examines whether autoimmune MRL-Ipr mice develop impairments in learning and memory that correlate with changing severity of lupus-like disease. MRL-lpr mice (n = 20) were tested in the Morris water-maze at 12, 14, 16 and 18 weeks of age. Age-matched controls were congenic MRL + / + mice ( n - 20) that develop the disease much later. Immune status was assessed by the presence of anti-nuclear antibodies (ANA), brain-reactive antibodies, proteinuria, and haematocrit. Learning rates and memory retention did not differ between the substrains, and did not correlate or deteriorate with advancing age and autoimmunity. However, the baseline performance level in autoimmune MRL-Ipr mice was shifted, as evidenced by a consistently longer task-solving latencies. Thigmotaxic swimming (along the pool wall) was pronounced in the MRL-Ipr group, and was associated with the observed difference in performance. The present study does not support the notion that learning/memory abilities of autoimmune MRL-Ipr mice are impaired per se, but may support the hypothesis that the rapid progress of humoral autoimmunity affects the emotionality of lupus-prone mice.

INTRODUCTION

Cognitive and emotional disturbances are common clinical symptoms observed during the course of the autoimmune disease, systemic lupus erythematosus (SLE). Up to 70°J/o of lupus patients develop major neuropsychiatric manifestations such as stroke, seizure, psychosis or depression 4. Many experience so-called 'minor' neuropsychiatric symptomatology such as mood swings, paresthesiae; moreover, a large subset of lupus patients have subclinical cognitive dysfunction apparent on neuropsychological testing 5'6. One hypothesis for the etiology of behavioural symptomatology in SLE is that a defect in the blood-brain barrier permits brain-reactive antibodies (BRA) to react with neuronal antigens, disturbing normal central nervous system function i ~,12 There are several autoimmune strains of mice (NZB, BXSB, MRL, NZB/W) in which one can assess interactions between autoimmunity and behavioural syrupCorrespondence: B. Saki6, Department of Biomedical Sciences, HSC 4N7, McMaster University, 1200 Main St. West, Hamilton, Ont. Canada LSN 3Z5.

tomatology. Indeed, several behavioural studies indicate that during the course of autoimmune disease these mice show impaired performance on learning tasks involving active or passive avoidance 7'2-~. Other studies suggest that the presence of BRA in circulation is correlated with learning problems in the aforementioned t a s k s 13,16.

Similarly, we observed impaired performance of autoimmune MRL-lpr mice in a swimming task, but the deficit was relatively mild 2°. Performance of MRL-lpr mice was worse during reversal testing in the Morris water-maze but the mice had no difficulty in the initial acquisition of an escape response using extramaze cues z°. However, in that study animals were tested at 10 weeks of age, when overt manifestations of SLE are relatively mild. Therefore, we hypothesized that behavioural deterioration would be more evident by 18 weeks of age, when a more severe form of lupus-like disease is apparent. For this purpose, the present study tested learning and memory of diseased MRL-lpr mice at different phases of lupus-like disease, in comparison to age-matched M R L + / + congenic controls that develop the severe form of disease in the second year of life.

58 MATERIALS AND METHODS

Animals Twenty MRL/MpJ-lpr/lpr (MRL-lpr) and 20 M R L MpJ- + / + ) (MTL + / ~- ) mice of both substrains (10/ sex) were purchased from The Jackson Laboratory (Bar Harbor, ME) at 28 days of age ( + 3 days; day of birth = Day 0; Days 0 to 6 = 0 weeks of age) and assigned to four groups (according to substrain and sex). Upon arrival, they were left to habituate for 10 days in home cages (5 mice/cage) in a mouse room maintained on a 12-h light cycle (8 AM-8 PM). Several days preceding testing all mice were handled daily for a week by the experimenter, as described previously2°. One MRLlpr female died between 16 and 18 weeks of age.

Spatial learning Spatial learning was tested in a large circular fish tank (diameter, 1.83 m; height, 61 c m ) m a d e of natural (white) polyethylene (Canbar Products, Waterloo, Ont.. Canada). The pool was filled with 21 °C water to a height of 13 cm. The escape platform was made entirely of transparent Plexiglas and consisted of a circular top (diameter, 15 cm; thickness. 0.4cm) connected by a rod (diameter 1.2 cm) to a heavy base (diameter 14 cm: thickness 3 cm). The surface of the platform contained several concentric grooves to facilitate mounting of the platform by the animal. The top of the platform was submerged 0.5 cm below the surface of the water in the middle of the southwest quadrant, and was not readily visible to the experimenter. More importantly, ~t was invisible to the animal as evidenced by the fact that mice could swim right by it, without changing direction. An overhead video camera recorded the animal's activity in the pool for subsequent analysis. The pool room was rectangular with no windows and with illumination provided by fluorescent ceiling lights. A number of fixed, distal cues surrounded the pool: a white (western) wall to which was fastened the camera cable, and a blue (eastern) wall which had a vent on it. A blue door, adjacent to a sink, was at the southern white wall. The experimenter sat in the northwest corner, close to a stand with video equipment and a pale blue curtain hanging on a metal pole along the length of the north side. Pool edges were situated approximately 40 cm from 3 walls. Animals were transported to the pool room in their home cages and allowed 5 min before the start of testing. Procedure. At 12 weeks of age mice were trained during 4 consecutive days (4 trials/day) to escape onto a submerged platform located in the middle of the southwest (SW) quadrant. A 2-min trial was begun by lowering the mouse into the pool along the wall and

facing it at one of the four designated starting positions (North, South, East or West/. The sequence of these four starting positions varied randomly on each day but the last position on a given day was the same as the first starting position on the next testing day. Trials were spaced approximately 30 min apart. On Day 1, animals that escaped onto the platform were permitted to stay for 30 s: those that failed to locate it by the end of 2 min were placed on the platform and allowed to stay for 30 s. After the trial, each mouse was let1 to dry out under a heat lamp. On Training Days 2, 3, and 4. animals were removed from the pool as soon as lhev located the platform or the trial ended. "Fen days after completion of 4 sessions of 4 trials, long-term retention was assessed in a block of four trials, using the same procedure as on Days 2-4 of training. On the next davL the platform was relocated to thc northeast (NEI quadrant and each mouse was placed directly onto it for 30 to experience the new platform position. Ten to 15 min later, testing of acquisition begun as at 12 weeks of age (Days 2-4 of training). The same procedure was repeated at 16 and 18 weeks of age, when the platform was relocated into NW and SE quadrant, respectively. Measures. The latency to locate the platform was recorded at time of testing using a stopwatch. Twentyfour-h retention was assessed as the difference between the latencies in the first trial on the given day and the last trial on preceding day (when starting positions are the same). Ten-day retention was assessed as the d i f ference between mean latencies on 2 days set apart by a 10-day interval. Thigmotaxis was measured as the time spent swimming in an annulus confined by the edge of the pool and the outer edge of the platform. Only the initial (first) daily trials at different ages were measured. Swimming speed was calculated by measuring the length (Sigma-Scan. Jandel Scientific. CAI of continuous, non-circling swimming paths recorded during initial 10-15 s of exposure to the pool and divided by the actual time spent m swimming.

Immunological testing Urine and blood samples were collected after behavioural testing was completed at 12. 14. 16 and 18 weeks of age. Drops of fresh urine were obtained by a slight abdominal pressure. Blood samples (approx. 0.2 ml~ were collected under light ether anaesthesia using blood micropipettes (50 #1. Drummond Scientific Company, Broomall, PAl to bleed the mouse from the retro-orbital sinus of the eye. The following immunological tests were performed: (a) Anti-nuclear antibodies 1ANAl were assayed using a fluorescent assay. Frozen rat liver tissue was sectioned to 6/~m in a cryostat and mounted on 12-well

59 teflon-masked slides. Serum samples were double diluted (up to 1:16 384) in PBS. Twenty-#l serum samples were layered on each well and incubated further for 30 rain in a humid chamber, at room temperature. Slides were washed for 10 rain in PBS, layered with 20/~1 FITC-conjugated sheep anti-mouse IgG (H + L) (Jackson Immunoresearch Labs), and incubated further for 20 min in a humid chamber, at room temperature. The conjugate was diluted in PBS to 1:30 and centrifuged before use to remove any precipitate. Slides were washed in PBS again, cover slipped and read with an epifluorescent microscope at 480 nM. (b) Fluorescent Brain Reactive Antibody Assay. The murine neuroblastoma cell line Neuro-2A (CCL 131; ATCC, Rockville, MD) was cultured in Eagle's Minimal Essential Medium (Sigma, St. Louis, MO) with non-essential amino acids and Earle's BSS, supplemented with 10 °,,o FCS. Cells were dislodged from confluent monolayer, with 25 °/o trypsin and resuspended in medium to an approximate concentration of 3 x 10 4 cells/ml. The suspension was added to 60-well microtiter plates at 10 #l/well to give approximately 300 cells per well. Microtiter plates were precoated with poly-Llysine, diluted in PBS to 10/~g/ml, to improve adhesion of cells to the surface of the wells. Plates were incubated overnight at 37 °C. Sera were stored at -70 °C and thawed at room temperature for 1 h. Samples (10/~1) were diluted by PBS up to 1:4096 and incubated with slight agitation at room temperature for 1 h. Plates were rinsed 3 times with medium, layered with FITC-conjugated sheep anti-mouse lgG ( H + L ; Jackson Immunoresearch Labs.), diluted in PBS to 1:50, and incubated with agitation at room temperature for 1 h. After rinsing, the plates were covered with 8011o glycerin in PBS, and evaluated under an inverted epifluorescent Nikon microscope. Sera were tested 3 times under identical conditions, and assigned positive if were fluorescent in at least 2 independent assays. (c) Proteinuria was measured with tetrabromphenol paper (Albustix, Ames Co.) on freshly expressed urine samples. This colorimetric assay is specific for albumin, with approximate protein concentrations as follows: trace (10 mg'~'o), 30, 100, and 300 mg~o. For the purposes of multiple regression and ANOVA analysis these ordinal values were recoded into interval values: 1, 2, 3 and 4, respectively. (d) Haematocrit was measured only at 18 weeks of age, in order to minimize the time under anaesthesia and amount of blood taken in repeated sampling. The samples were taken from the retro-orbital sinus by heparinized Fisher micro-haematocrit capillary tubes. Sealed tubes were centrifuged for 10 rain in standard

micro-haematocrit centrifuge (Clay-Adams, Parsipaany, N J) and read in a Critocaps reader. Previous studies have revealed that low haematrocrit ( < 40 ~o) is a reliable measure of autoimmune haemolytic anemia in autoimmune mice 9.

Body weights To test whether malnutrition occurred in autoimmune MRL-lpr mice and whether body size was different between the substrains, all mice were weighted at the end of study. Statistics Each dependent measure was analyzed by an analysis of variance (ANOVA) in a Substrain by Sex design, with repeated measure factors Age, Day or Trial as appropriate. In cases where homogeneity of variance or normal distribution of data was violated, logarithmic transformation or non-parametric tests were employed. The relationship between behavioural variables (acquisition latency, memory retention, learning rate and thigmotaxis) and immunological variables (ANA titre, BRA titre, and proteinuria) in MRL-lpr strain was assessed by Multiple Regression, with immune parameters assigned as predictors in Stepwise and Enter methods. Values presented are means and S.E.M. The criterion for statistical significance was set at P