Additional Supporting Information may be found in the online version of this article.

Figure S1: Summary of the genetically modified H-Ras alleles used in this study. The wild-type H-Ras allele (H-Ras+) is also included for comparison. Black boxes: H-Ras coding sequences labeled with the corresponding exon number; white boxes: H-Ras noncoding exons; gray boxes: polyadenylation sequences; dotted boxes: IRES-β-geo cassette; checkered box: PGK-Hygromycin (Hyg) resistance cassette; octogonal box (STOP): transcriptional inhibitory sequences; filled triangles: loxP sites. Asterisk: G12V oncogenic mutation. LSL: loxP-STOP-loxP cassette. The proteins expressed by each allele are indicated in the right column. H-RasG12V allele results from crossing H-Ras+/LSLG12V mice to EIIaCre transgenic animals to excise the PGK-Hyg-STOP cassette flanked by loxP sites.

Figure S2: Acoustic startle response and PPI in H-Ras+/G12V mice. (A) Normal startle response profile to a variety of acoustic intensities in H-Ras+/G12V mice (n = 10) and control siblings (n = 7). Startle reflex, defined as a response significantly different from the response to background noise, was detected for intensities higher than 100 dB. (B) PPI of startle response in H-Ras+/G12V (n = 14) and H-Ras+/+ mice (n = 10). Only the 90 dB stimulus produced significant PPI (%pp90 against 0: H-Ras+/+, t(9) = 3.392, P = 0.008; H-Ras+/G12V, t(13) = 2.419, P = 0.031). No significant effect of genotype or genotype × prepulse interaction was found.

Figure S3: Supplemental information on locomotor and exploratory activity in H-RasG12V mice. (A) Open field task: Resting time, average speed and maximum speed were normal in heterozygous mice (upper panels, H-Ras&plus;/G12V: n &equals; 11, H-Ras&plus;/&plus;: n &equals; 11). Homozygous mice rested more time, but had similar average and maximum speed than wild-type littermates (lower panels, H-RasG12V/G12V: n &equals; 9, H2 Ras&plus;/&plus;: n &equals; 10). (B) Locomotor and exploratory activity was also assessed using the automatic system IntelliCage for a period of 72 h. Both heterozygous and homozygous mice exhibited more activity during the dark phase of the light cycle (F(5,95) phase &equals; 11.915, P < 0.001 and F(5,50) phase &equals; 20.111, P < 0.001, respectively). We did not observe difference in the number of visit in heterozygous mice compared to their control littermates, but, in agreement with our observations in the open field, homozygous mice showed a lower number of visits than their wild-type siblings (lower panel: F(1,10)genotype &equals; 11.267, P &equals; 0.007). It should be noted that homozygous mice and littermates were tested in the IntelliCage shortly after concluding the battery of cognitive tasks, which included the highly stressful fear conditioning task. This may explain the overall lower activity of this cohort of mice in this novel environment, especially during the first day. In subsequent days the control siblings increased their activity to a level similar to that observed on the experiments on naïve animals (see upper panel), but homozygous mice still presented a low number of visits.

Figure S4: Supplemental information on water maze experiments. (A) No difference in average swimming speed during the three phases of the water maze experiment was found between heterozygous (upper panels) and homozygous mice (lower panels) and their respective control littermates. (B) Thigmotaxis in heterozygous mice was not different from control values. (C) The analysis of quadrant occupancy during the three probe trials confirmed the normal performance of heterozygous mice and revealed a reduced preference for the target quadrant in homozygous mice during the first probe trial. In agreement with the results for annulus crossings presented in Fig. 5, in the first probe trial there was a significant quadrant × genotype interaction for the percentage of time spent in quadrants (F(3,51) &equals; 3.317, P &equals; 0.027), and an almost significant effect of genotype in the percentage of time spent in the target quadrant wild-type mice showed a significant preference for the target quadrant (t(9) &equals; 5.324, P < 0.001), but mutant mice did not (t(10) &equals; 1.924, P &equals; 0.083). (&plus;): significantly different from chance level (25%).

Figure S5: Supplemental information on object recognition memory experiments. H-Ras&plus;/G12V (n &equals; 11) and in H-RasG12V/G12V mice (n &equals; 11) spent similar time exploring the objects than their respective wild-type siblings (n &equals; 11 and n &equals; 10) both during the training and the test session.

Please note: Wiley–Blackwell are not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author.

GBB_443_sm_SuppInfo.pdf267KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.