Raúl Tudela^{1}, Emma Muñoz-Moreno^{2}, Xavier López-Gil^{2}, and Guadalupe Soria^{1,2}

The TgF344-AD rats represent the most suitable and promising animal model for Alzheimer’s disease (AD) research. Resting-state functional MRI was longitudinally acquired every 3 months in a cohort of transgenic Tg344-AD and control Fisher rats between 5 and 15 months of age, together with cognitive task evaluation. Independent component analysis was applied to rs-fMRI volumes and 10 networks were anatomically identified. Spearman correlation coefficients between functional and cognitive parameters were computed. Our results show that while no differences were observed in the cognitive task between both groups, significant differences were found in the functional networks.

**INTRODUCTION**

**METHODS**

rs-fMRI was longitudinally acquired in a cohort of transgenic Tg344-AD (n=9) and control Fisher rats (n=10) between 5 and 15 months of age in a 7T scanner at four different time-points by using a single-shot gradient-echo EPI sequence (TR/TE = 2000/10,75 ms). 600 volumes of 64x64x34 voxels and 0.4x0.4x0.6 mm³/voxel were acquired.

Before
each scan the working memory performance of both groups was evaluated
by means of a delayed non-matching-to-sample (DNMS) task, following a
procedure modified from Callaghan et al., 2012^{2}.

Image
preprocessing included: slice-timing, motion correction,
skull-stripping, spatial normalization, spatial smoothing, detrending
and regression by motion parameters, and temporal filtering (0.01 -
0.1 Hz). 30 independent components were obtained using FSL MELODIC^{3}
considering the whole cohort. 10 components were selected based on
the anatomical structures comprising the network and their
correspondence to the literature^{4-8}.

Dual
regression was performed using these 10 components to find the
subject-specific time-series and spatial maps for each network. The
standard deviation of the time-series (Amplitude) of each component
and the mean of the Z-values in the spatial maps where each network
was localized (Shape Variability)^{9} were computed for each
subject and the differences evaluated using Kruskal-Wallis test.

The results of the DNMS test were evaluated considering the total number of trials and the ratio of correct responses per group and time-point. Spearman correlation coefficient was computed between these cognitive function parameters and the Amplitude and Shape Variability for the 10 components of the two groups and the four time-points.

**RESULTS**

Between t1 and t4 significant differences (uncorrected p<0.05) appeared in the Amplitude for sensorimotor I and II, and somatosensory networks for the control group. There were also significant differences between both groups at t1 in the visual-auditory, default and thalamus-hypothalamus networks and in the somatosensory and sensorymotor II networks at t4. Time-points 2 and 3 were more stable and not significant differences were found in any of the comparisons.

Considering Shape variability there were significant differences between t1 and t4 time-points in the sensorimotor II network for the control group and in the visual-auditory and thalamus-hypothalamus networks for the TgF344-AD group. There were significant differences between both groups at t1 in the default and thalamus-hypothalamus networks and at t4 in the somatosensory and default networks. Again, time-point t2 and t3 showed fewer differences between groups or time-points.

To evaluate the working memory, as assessed by the DNMS task, the number of trials and the ratio of correct responses were considered (Figure 2). At t1 the transgenic group performed significantly less number of trials compared to the controls. Then the number of trials slightly increased over time while in the control group was more stable. The stability between time-points is clearer in the ratio of correct responses where there are not significant differences between groups or time-points.

Figure 3 shows the matrix of Spearman correlation coefficients between the Amplitude and Shape variability of the different networks and the results of the cognitive function. Both Amplitude and Shape variability correlates similarly with the cognitive results. However, at t4 there were more significant Spearman coefficients in the control group, such as in the somatosensory and sensorimotor II, than in the TgF344-AD group.

**DISCUSSION**

**CONCLUSION**

1. Cohen, R. M., Rezai-Zadeh, K., Weitz, T. M., Rentsendorj, A., Gate, D., Spivak, I., et al. A transgenic Alzheimer rat with plaques, tau pathology, behavioral impairment, oligomeric Aβ and frank neuronal loss. J. Neurosci. 2013; 33:6245–6256.

2. Callaghan, C. K., Hok, V., Della-Chiesa, A., Virley, D. J., Upton, N., and O’Mara, S. M. Age-related declines in delayed non-match-to-sample performance (DNMS) are reversed by the novel 5HT6 receptor antagonist SB742457. Neuropharmacology 2012; 63:890–7.

3. Beckmann, C.F., Smith, S.M. Tensorial extensions of independent component analysis for multisubject FMRI analysis. NeuroImage 2005; 25:294-311

4. Henckens, M.J.A.G., van der Marel, K., vsn der Toorn, A., Pillai, A.G:, Fernández, G., Dijkhuizen, R.M., Joëls, M. Stress-induced alterations in large-scale functional networks of the rodent brain. NeuroImage 2015; 105:312-322.

5. Sierakowiak, A., Monnot, C., Aski, S.N., Uppman, M., Li, T.-Q., Damberg, P., Brené, S. Default Mode Network, Motor Network, Dorsal and Ventral Basal Ganglia Networks in the Rat Brain: Comparison to Human Networks Using Resting State-fMRI. PLoS ONE 2015; 10(3):e0120345-1:20.

6. Hsu, L.-M., Liang, X., Gu, H., Brynildsen, J.K., Stark, J.A., Ash, J.A., Lin, C.-P., Lu, H., Rapp, P.R., Stein, E.A., Yang, Y. Constituents and functional implications of the rat default mode network. PNAS 2016; 113(31):E4541-E4547.

7. Bajic, D., Craig, M.M., Borsook, D., Becerra L. Probing Intrinsic Resting-State Networks in the Infant Rat Brain. Front. Behav. Neurosci. 2016; 10:192.

8. Ma, Z., Perez, P., Ma, Z., Liu, Y., Hamilton, C., Liang, Z., Zhang, N. Functional atlas of the awake rat brain: A neuroimaging study of rat brain specialization and integration. NeuroImage 2016; https://doi.org/10.1016/j.neuroimage.2016.07.007.

9. Nickerson, L.D., Smith, S.M., Öngür, D., Beckmann, C.F. Using Dual Regression to Investigate Network Shape and Amplitude in Functional Connectivity Analyses. Front. Neurosci. 2017; 11:115.

Figure
1. (A) Coronal views of the default network (Z>2.3). Boxplots for
(B) the Amplitude and (C) the Shape Variability for control and
TgF344-AD groups for the four time-points (T1-T4). The box extends
from the lower to upper quartile values of the data, with a red line
at the median. The whiskers extend from the box to show the range of
the data. Flier points are those past the end of the whiskers.
Asterisk indicates significant difference (uncorrected p<0.05)
between groups connected by the line under it.

Figure
2. Boxplots for (A) the number of trials and (B) the ratio of correct
responses in the DNMS tests performed by both control and TgF344-AD
groups at the four time-points. The box extends from the lower to
upper quartile values of the data, with a red line at the median. The
whiskers extend from the box to show the range of the data. Flier
points are those past the end of the whiskers. Asterisk indicates
significant difference (uncorrected p<0.05) between groups
connected by the line under it.

Figure
3. Spearman correlation coefficients between the Amplitude for the 10
selected networks and the number of trials and ratio of correct
responses of the DNMS test at the four time-points for (A) the
control and (B) the TgF344-AD groups. And Spearman correlation
coefficients between Shape variability and the number of trials and
ratio of correct responses of the DNMS test at the four time-points
for (C ) the control and (D) the TgF344-AD groups.