Progressive impairment of directional and spatially precise trajectories by TgF344-AD Rats in the Morris Water Task

Spatial navigation is impaired in early stages of Alzheimer’s disease (AD), and may be a defining behavioral marker of preclinical AD. Nevertheless, limitations of diagnostic criteria for AD and within animal models of AD make characterization of preclinical AD difficult. A new rat model (TgF344-AD) of AD overcomes many of these limitations, though spatial navigation has not been comprehensively assessed. Using the hidden and cued platform variants of the Morris water task, a longitudinal assessment of spatial navigation was conducted on TgF344-AD (n=16) and Fischer 344 (n=12) male and female rats at three age ranges: 4 to 5 months, 7 to 8, and 10 to 11 months of age. TgF344-AD rats exhibited largely intact navigation at 4-5 and 7-8 months of age, with deficits in the hidden platform task emerging at 10-11 months of age. In general, TgF344-AD rats displayed less accurate swim trajectories to the platform and a wider search area around the platform region compared to wildtype rats. Impaired navigation occurred in the absence of deficits in acquiring the procedural task demands or navigation to the cued platform location. Together, the results indicate that TgF344-AD rats exhibit comparable deficits to those found in individuals in the early stages of AD.

Introduction therefore serve as an early marker of AD, with some studies indicating that disoriented 70 patients are more likely to convert from aMCI to an AD diagnosis (Bird et al., 2010; each time-point. However, the cues, cue layout, pool, and platform location was modified 162 at each time-point (Fig. 1B). An overhead camera was fastened above the pool to record 163 swim behavior for subsequent analysis. wall. Once the platform was found, the subject remained on the platform for 10 seconds. 173 Drop location varied between time point/trials/days, but was maintained between 174 subjects. At the end of the trial, the rat was returned to a holding cage, during which time 175 the remaining rats in the cohort were tested (10-20min duration). Subjects were run in 176 groups of seven for each task. At the end of the 4 trials, rats were returned to their home 177 cages in the colony room and the same procedure was repeated the next day. Time to 178 reach the platform was recorded by an experimenter at the end of each trial. The mean 179 escape latencies (seconds) for each animal were calculated for each day (every 4 trials).  in a Linux bash shell using FFMPEG (https://www.ffmpeg.org). Image files were then 207 imported into Fiji (https://imagej.nih.gov/ij/) and x-y-coordinates of the animal's nose 208 was acquired for each video frame (10 frames/sec) using the Manual Tracking plugin.

209
Custom Matlab (R2017, The MathWorks, Natick, MA) scripts were designed to smooth 210 the tracked swim paths using the runline function from the Chronux toolbox 211 (www.chronux.org). The smoothed paths were then analyzed for path length (cm), swim 212 speed (cm/sec), platform proximity (cm), and search area. Platform proximity was 213 measured by summing the distance between the subject's location and the center of the

236
For clarity of analysis, the movements were further divided into 4 broad categories.  non-spatial errors included diving behavior (diving below the surface of the water during 256 a trial), floating (the absence of swimming for more than 3 sec), platform deflections (the 257 failure to obtain purchase onto the platform after contact), mounting errors (the failure to 258 climb the platform after 1 sec of contacting the platform), and jumping off the platform.

259
Each error was assigned a score of 1 and a total was obtained by summing the errors

278
One subject was excluded from the analyses at 10-11 months of age after 279 developing glaucoma. Furthermore, one subject died during testing at 10-11 months of 280 age and thus was not included in the analysis below. suggest that 10-11-month-old Tg rats may express fewer of these directed swims. To 320 address this possibility, we classified swim paths into 11 movement categories (Fig. 4).

321
These swim movements were then merged into 4 categories based on their directedness 322 to the platform (target-direct), a lack of directedness but proximity to the platform (target-indirect), or a tendency to organize movements in non-platform locations (spatial-324 indirect), or in a random search (non-spatial). other groups (X 2 s >5.96, ps<.014). Additionally, inspection of target-direct trajectories at 332 earlier ages suggest that Tg rats made fewer of these movements relative to WT animals.

333
Indeed, group comparisons reached significance at 7-8 months of age (X 2 (1) = 5.82, 334 p=.015), and although group differences were not significantly different at 4-5 months of 335 age, Tg females made fewer target-direct trajectories compared to WT females (X 2 (1) = 336 4.07, p=.043). Together, these results support the conclusion that Tg rats make less 337 directionally precise movements at 7-8 and 10-11 months of age, with a significantly 338 lower proportion of these movements by female Tg rats.

339
The reduced frequency of target-direct paths by Tg rats suggest that other 340 movements, including those directed near the platform (target-indirect) or in other non-341 platform locations (spatial-indirect and non-spatial), may be favored by Tg animals (Fig.   342 6B-D). Consistent with this hypothesis, Tg rats were found to express a greater 343 proportion of target-indirect movements at 10-11 months of age (X 2 (1) = 5.93, p=.015). 344 Furthermore, with respect to target-indirect movements, there was a clear sex difference 345 with Tg males showing a significantly greater proportion of these paths relative to all 346 other groups (X 2 s>6.75, ps<.009). Interestingly, this finding was apparent at 4-5 months, 347 however; at that age Tg females had a greater proportion of target-indirect paths relative 348 to all other groups (X 2 s>6.82, ps<.009). Additionally, Tg rats performed a greater number 349 of spatial-indirect trajectories (X 2 (1) = 14.31, p<.001) and fewer non-spatial movements 350 at 10-11 months (X 2 (1)=14.61, p<.001). In sum, the reduced frequency of direct 351 trajectories by Tg rats at 10-11 months of age corresponds with an increase in spatially  Given Tg rats perform a lower proportion of directed movements (either target-365 direct or target-indirect movements) at 10-11 months, we tested the hypothesis that Tg rat 366 performance is more highly explained by becoming more efficient at non-direct 367 movements rather than switching between non-direct to direct movement categories. To takes up 27% less of the variance for Tg males and 16% less of the variance for Tg 388 females relative to WT males and WT females, respectively (Fig. 9). Interestingly, Tg 389 male and females exhibited differences in the predictive power of each factor.

390
Specifically, at 10-11 months, switching between movements is a stronger predictor of 391 performance for Tg males (β=4.12, t(18)=4.406, p<.001) relative to Tg females (β=1.77, t(18)=4.36, p<.001). Furthermore, although changes in efficiency take up a larger 393 proportion of the variance for task performance in Tg females relative to Tg males (67% 394 versus 45%, respectively), changes in efficiency hold less predictive weight for Tg 395 females (β=.993, t(18)= 5.72, p<.001) relative to Tg males (β=1.80, t(18)= 3.67, p<.001). 396 Overall, these results indicate that the performance of Tg animals is less associated with 397 switching between movement categories than WT animals, supporting our hypothesis 398 that Tg animals are more likely the improve the efficiency of their movements than 399 switch to more direct trajectories. animals demonstrated similar swim speeds at 10-11 months of age. 408 We also analyzed the number of non-spatial errors per rat at each age point. The 409 number of errors were summed across days to produce a non-spatial error score (Harker 410 and Whishaw, 2002). Overall, animals in both groups displayed near zero non-spatial 411 error scores at each age of testing (Fig. S2). By 10-11 months of age, only half of the 412 animals in each group expressed 1 or 2 errors over the 5 days of hidden platform testing, 413 indicating an absence of procedural learning deficits in the Tg group (X 2 (2)= 7.54, p=.37). Thus, given the absence of clear group differences in non-spatial behaviors, it is 415 unlikely that procedural errors contributed to the deficits described in the sections above.

436
In the cued platform task, Tg and WT rats showed similar performance at each testing age (Fig. S3). Mixed ANOVAs conducted on escape latencies at each time-point 438 indicated all animals had decreased escape latencies in the second trial block versus the 439 first trial block at 4-5 months and 7-5 months (Fs≥ 4.46, ps≤.0.045), but only trending 440 differences were observed at 10-11 months (F(1,23)=3.81, p=.063). Importantly, there 441 were no group differences detected between Tg and WT groups at the three test ages.

442
Further, there were no main effects of sex at each testing age. At 4-5 months of age, there 443 was a significant sex by block effect, whereby female animals were slightly slower to 444 reach the platform relative to male animals in the first trial block. Furthermore, Tg 445 females had slightly elevated latencies to reach the platform relative to Tg males, though 446 this genotype by sex effect only trended towards significance (F(1,24)=2.97,p=.098).

448
The primary conclusion of the present study is that clear spatial navigation 449 impairments by TgF344-AD rats were identified at 10-11 months of age. Specifically,

450
TgF344-AD rats demonstrated increased escape latencies and path lengths, and they 451 searched a wider area of the pool and were less precise in their search for the platform 452 location (Figs 2 and 3). In addition, by 10-11 months of age, the directionality of their 453 trajectories to the platform (Fig. 5 and 6) and switching from less direct to more direct 454 trajectory types was attenuated in Tg rats ( Fig. 8 and 9). While navigation impairments 455 were detected during training in the hidden platform task, a 60 second no-platform probe 456 test conducted on the 6 th day indicated that both Tg and WT groups displayed a similar 457 preference for the platform quadrant (Fig. 10). This pattern of impairments at 10-11 458 months supports the conclusion that although Tg animals are impaired at executing an optimal trajectory and search pattern near the hidden platform region, Tg rats can express 460 a preference for that location by the end of the experiment. 461 The deficits reported in the present study were observed in the absence of group 462 differences in sensorimotor or procedural learning. Several analyses support this and WT animals were equivalent in standard measures of water task performance at 7-8 481 months of age (i.e., escape latency and path length). One possible explanation for the latter finding is that repeated spatial training and procedural learning may have allowed 483 Tg animals to utilize strategies that result in similar performance on standard measures. 484 Indeed, a convolution analysis indicated that Tg rats get better at less direct movements 485 to find the platform than switching to more direct movements. Finally, it is important to 486 note that previous studies report that Tg rats tend to have greater spatial difficulties in 487 manipulations involving "reversal" tests in which the goal is moved to a novel location.  The present study found that TgF344-AD rats express clear navigation 543 impairments at 10-11 months of age. A detailed path analysis indicated that subtle 544 deficits in the directedness of trajectories to the hidden platform can be detected at earlier 545 ages and can be sensitive to sex differences. The latter observations may underlie the 546 subtle differences between Tg and WT rats that were found using classic measures of 547 water maze (escape latency, path length, search proximity and search area) at these ages, 548 and are likely not due to factors associated with non-spatial task demands. Furthermore,        Figure S1. Average swim speed (cm/s) is plotted for each age of testing. Groups distinguished by color: black=Wild Type males (WtM), red=Transgenic males (TgM), gray=Wild Type females (WtF), blue=Transgenic females (TgF). Note that males consistently swam faster than females across all age points (ps<.05, ANOVAs). Figure S2.