Donald et al. 10.1073/pnas.0712324105. |
Fig. 7. Insertion site of the pGT1Lxf targeting vector into mouse P-Rex2 gene in ES cell line RRD186 and genotyping methods. (A) Four DNA probes (A-D, purple bars) were generated by PCR from E14-cell Wt genomic mouse DNA against a 33,284-bp region of interest around exons 34 and 35 (black bars) of the mouse P-Rex2 gene on chromosome 1 [region 11.19-11.22 Mb according to National Center for Biotechnology Information (NCBI) assembly M36] to determine the insertion site (red frame) of gene-trap targeting vector pGT1Lxf in ES cell line RRD186, which was generated by BayGenomics. The four probes were used on DNA digested with either one of the listed restriction enzymes. Probe A was used routinely on EcoR1-digested DNA for genotyping. The figure was assembled by using Vector NTI. (B) Primer sequences for the generation of Southern blotting probe A and for genotyping of P-Rex2-/- mice by PCR.
Fig. 8. Generation and general characterization of P-Rex2-/- mice. (A) Schematic domain structure of the mouse P-Rex2 protein, exon organization of the P-Rex2 cDNA (NCBI accession no. AM109952), and genomic exon/intron organization of the mouse P-Rex2 gene on chromosome one. Red stripes indicate the insertion site of the BayGenomics pGT1Lxf gene-trap targeting vector into the intron between exons 34 and 35, which results in the disruption of the P-Rex2 gene and deletion of P-Rex2 expression in the mouse. (B) Southern blot using probe A with EcoR1-digested genomic DNA to analyze the targeting of the mouse P-Rex2 gene. (C) Organ weights of five male and five female 3-month-old P-Rex2-/- (Ko) and P-Rex2+/+ (Wt) mice expressed as percentage of total body weight. Data are mean ± SD. (D) Body weights of male and female P-Rex2-/- (red) and P-Rex2+/+ (blue) mice (12-15 animals per group). Data are mean ± SE, statistics are ANOVA test.
Fig. 9. Characterization of P-Rex2 antibodies and time course of P-Rex2 expression in mouse cerebellum. (A) two milligrams of total cerebellar lysate from P-Rex2-/- (Ko) and P-Rex2+/+ (Wt) mice and 100 ng of recombinant EE-tagged P-Rex1 or P-Rex2 were separated by SDS/PAGE and Western blotted with rabbit polyclonal antiserum 78 against amino acids 717-799 of mouse P-Rex2 or with the same antiserum after affinity purification on full-length recombinant human P-Rex2, as indicated. (B Left) Western blot of P-Rex2 expression in total cerebellar lysates from Wt mouse pups at the indicated age (in days). Blot is representative of three blots. (B Right) Coomassie staining of the Western blot membrane on the left.
Fig. 10. High-resolution imaging of EGFP-P-Rex2 Purkinje cells. Cerebellar slices from 18-day-old EGFP-P-Rex2-/- (Ko) and EGFP-P-Rex2+/+ (Wt) mice were imaged by high-resolution confocal microscopy as described in Materials and Methods. (A) Typical examples of dendritic spines of EGFP-P-Rex2-/- and EGFP-P-Rex2+/+ Purkinje cells. (Scale bars, 5 mm.) (B) Method of measuring the width and length of the main Purkinje cell dendrite. High-resolution image stacks of EGFP-P-Rex2-/- and EGFP-P-Rex2+/+ Purkinje cells were analyzed by using Volocity software. Main dendrite width was measured at a distance of one cell-body diameter away from the cell body. Main dendrite length was measured from the middle of the cell body to the first major branch point of the dendrite or, in the absence of major branching, as far as traceable. (Scale bar, 20 mm.)
Fig. 11. Basic locomotor activity and gait of P-Rex2-/- mice. (A) The cohort of 53 male and female P-Rex2+/+ (Wt, blue) and P-Rex2-/- (Ko, red) mice was assessed for their basic locomotor activity by keeping them separately in infrared-beam-equipped "activity boxes" in the dark for two hours and measuring the total number of infrared-beam breaks (Upper), or successive beam breaks at opposite ends of the box (runs; Lower). (B) The gait of the mice was assessed through analysis of footprints on a paper course after painting their front feet red and their hind feet black, as detailed in Materials and Methods. Stride length is calculated from the mean of three steps in the middle of the course for each animal. Data are mean ± SE; statistics are ANOVA. The key to the black and purple P values is as in the legend to Fig. 3.
Fig. 12. Acoustic startle response, prepulse inhibition, and elevated plus-maze behavior of P-Rex2-/- mice. (A) The cohort of 53 male and female P-Rex2+/+ (Wt, blue) and P-Rex2-/- (Ko, red) mice was assessed for their acoustic startle response to a set of 36 randomized 30-msec bursts of 120-db white noise over a period of seven minutes in a chamber rigged to motion sensors. Prepulses of 2-16 db preceded 24 of 36 bursts. The mean startle response to pulses without prepulse (Upper) and the inhibition of the startle response by 16-db prepulses (Lower) are plotted. (B) The animals were left to explore an elevated plus-maze with open and enclosed arms for five minutes as detailed in Materials and Methods. Data are mean ± SE, statistics are ANOVA, done as described in Material and Methods. The key to the black and purple P values is as in the legend to Fig. 3.
Fig. 13. Purkinje cell morphology and motor behavior of P-Rex1-/- mice. (A) Cerebellar slices from P-Rex1+/+ (Wt) and P-Rex1-/- mice (Ko) (two 2-month-old and two 4-month-old mice per genotype) were stained for Calbindin (Scale bars, 50 mm.) and analyzed by microscopy as in Figs. 2 and 5 for percentage of Purkinje cells with visible dendritic trunks in the plane of focus. Data are mean + SE, statistics are t test. (B-D) A cohort of 20 P-Rex1+/+ (Wt; blue) and P-Rex1-/- mice (Ko; green) (at least four females and four males per group) were tested at two months of age for their motor functions. (B) Rotarod performance. Fall time: time before falling off at 42 rpm within 60-s trial. Active running: time actively running at 42 rpm within 60-s trial. Passive turns: average of turns clinging passively to the rotarod at three high speeds (26, 35, and 42 rpm). Slips: foot slips per animal throughout the trial. Displacement activities: standing up, looking up, reversing direction, or grooming at three lowest speed settings (10, 15, and 20 rpm). Statistics are nonparametric Mann-Whitney U test, except Fall time, which was categorized into falling and nonfalling animals and analyzed by Chi2 test. (C) Basic locomotor activity. Beam breaks and runs in activity boxes were recorded as detailed in Materials and Methods. (D) Limb clasping. Mice were scored for limb clasping (mean ± SE; occurrence was insufficient for meaningful statistical analysis) as described in Materials and Methods on three separate days.
Fig. 14. Generation of P-Rex1-/-/P-Rex2-/- mice. (A) P-Rex1 and P-Rex2 Western blots showing that P-Rex1 protein levels are not crudely up-regulated in total-brain lysates from P-Rex2-/- mice and, vice versa, that P-Rex2 is not up-regulated in total brain lysates from P-Rex1-/- mice. Equal amounts of lysed tissue are loaded in each lane. (B) PCR genotyping for Wt and targeted alleles of P-Rex1 and P-Rex2, done as described in Materials and Methods, showing a Wt (lane 1), a double-heterozygous (lane 2), and a P-Rex1-/-/P-Rex2-/- mouse (lane 3) of the same litter from double-heterozygous parents. (C) Body weights of male and female P-Rex1-/-/P-Rex2-/- mice (Dko, orange) compared with P-Rex2-/- (Ko, red) and P-Rex1+/+/P-Rex2+/+ mice (Wt, blue). Data are mean ± SE; statistics are ANOVA, done as described in Material and Methods.
Movie 1. Rotarod performance of 2-month-old male P-Rex2-/- mice. The mouse on the right is a P-Rex2-/- mouse and the mouse on the left is a P-Rex2+/+ mouse of the same age (2 months) and sex (male), running on the rotarod at a speed of 42 rpm. The P-Rex2-/- mouse can be seen doing passive turns.
SI Materials and Methods
Generation of P-Rex2-/- mice, EGFP-P-Rex2 mice, and P-Rex1-/-/P-Rex2-/- Mice.
The mouse P-Rex2 gene on chromosome 1 was targeted by the gene-trap method by BayGenomics (baygenomics.ucsf.edu) by using the pGT1Lxf targeting vector. We obtained 129Ola ES cell clone RRD186, injected it into C57Bl6 mouse blastocysts, and raised chimeric mice. Germ-line transmission was achieved and heterozygous mice generated by crossing the best chimeric males to C57Bl6 females. Mice were screened by PCR by using primers designed to detect the presence of an 824-bp fragment of the targeting vector (SI Fig. 7). Four different Southern probes were designed within a 33-kb region of interest including exons 34 and 35 and part of the neighboring introns (SI Fig. 7). By using these probes together with a panel of nine restriction enzymes, the insertion site of the targeting vector was determined to be in the intron between exons 34 and 35, »5 kb 3' of exon 34. Once the insertion site had been determined, genotyping was done as follows: Southern probe A, corresponding to a 645-bp stretch »2 kb 5' of the insertion site, was used with EcoR1-digested genomic DNA, resulting in a 6.2-kb fragment for the targeted allele and a 4.2-kb fragment for the Wt allele. Alternatively, PCR genotyping was performed by using primers listed in SI Fig. 7. Heterozygous P-Rex2± mice were bred together to generate homozygous P-Rex2-/- animals, which were backcrossed at least six times to C57Bl6 genetic background before analysis of the strain. EGFP-P-Rex2 mice were generated by crossing EGFP-Pcp2 transgenic mice from The Jackson Laboratory, which have green fluorescent Purkinje cells (1), with our P-Rex2-/- animals. EGFP-positive P-Rex2± F1 animals were bred together to obtain offspring with EGFP-Purkinje cells that were P-Rex2-/- or P-Rex2+/+ littermates, which we named EGFP-P-Rex2-/- and EGFP-P-Rex2+/+ mice. Generation of P-Rex1-/-/P-Rex2-/- mice was achieved by breeding together our P-Rex1-/- mice (as in ref. 2, except backcrossed seven times to C57Bl6 background) and the P-Rex2-/- mice. P-Rex1-/-/P-Rex2-/- mice were obtained with the expected Mendelian frequency. Mice were housed and bred in open cages in a shower-in barrier unit. Mice used for behavior tests were kept in a standard barrier unit from the age of six weeks onward.P-Rex1 and P-Rex2 Western Blots.
P-Rex1 Western blots were done by using mAb 6F12 as described in ref.2. For P-Rex2 Western blots, sheep polyclonal antibodies were raised against amino acids 794-805 of mouse P-Rex2 (antigen was KLH-conjugated synthetic peptide), and rabbit polyclonal antiserum 78 was raised against amino acids 717-799. The construct for production of the latter antigen was cloned by PCR from a mouse brain cDNA library and recombinant protein was produced in and purified from Escherichia coli as a GST-fusion (SI Fig. 8). Region 717-799 was chosen for antibody production for its low homology between P-Rex1 and P-Rex2. Both P-Rex2 antibodies were affinity-purified before use, the sheep antibody by using peptide 894-805 and the rabbit antibody by using purified recombinant full-length human EE-P-Rex2 as bait.In Situ
Hybridization. Mice were killed by CO2 asphyxiation and brains removed rapidly and snap frozen on dry ice. Fresh-frozen 10-mm cryostat sections of brain were cut and stored at -80°C until processed. Tissue sections were brought to room temperature before being immersion-fixed for 10 min in 4% paraformaldehyde in 0.1 M phosphate buffer. Sections were then rinsed briefly in 0.1 M PBS (PBS) twice and dehydrated through graded alcohols. Four different oligonucleotide DNA probes complementary to base pairs 770-809, 2903-2942, 4561-4600, and 4765-4804 of mouse P-Rex2 cDNA (accession no. AM109952.1) were synthesized. The probes (concentration, 100 ng/ml) were 3' end-labeled with 1 pmol/ml [35S]dATP (NEN) by using 50 units of terminal deoxynucleotide transferase (Amersham Pharmacia) for 30 min at 37°C, and purified by using a Qiagen nucleotide removal kit. Hybridizations were carried out at 37°C overnight in a humidified incubator: each labeled probe (specific activity of >1 ´ 107 dpm/mg) was diluted in 4´ standard saline citrate (SSC), 50% formamide, salmon testis DNA (250 mg/ml), 10% dextran sulfate, 1´ Denhardt's solution, and 0.3% b-mercaptoethanol. After hybridization, sections were washed in 1´ SSC/0.2% Na2S203 at 55° C twice for 30 min and then in 1´ SSC/ 0.2% Na2S203 at room temperature for 2 min and 0.1´ SSC/0.2% Na2S203 at room temperature for 2 min. Sections were dehydrated through graded alcohols and air-dried before exposure to Hyperfilm (Kodak) for 2 to 4 weeks. Subsequently, sections were coated with K5 nuclear emulsion (Ilford), exposed in the dark for 5-12 weeks, developed, and counterstained with methylene blue.FACS-Sorting of Purkinje Neurons from Other Cerebellar Cells.
Purkinje cells were isolated from EGFP-Pcp2 transgenic mice (The Jackson Laboratory) as described in ref. 1. In brief, cerebella were removed from 3-week-old mice. Cubes (0.5-mm3) of cerebella were digested for 10 min at 37°C with 0.025% trypsin (type I; Sigma-Aldrich) in dissociation solution consisting of Ca2+-free Hanks' balanced salt solution (HBSS) containing 3 mg/ml BSA, 15 mM Hepes, 1.5 mM MgSO4, and 3 mg/ml glucose (pH 7.4). The enzymatic reaction was stopped by the addition of one volume of dissociation solution containing 0.25 mg/ml soybean trypsin inhibitor and 40 mg/ml DNase I (both from Sigma-Aldrich). Tissues were triturated mildly by sequential passage through wide-bore and fine-tipped pipettes. After the cells were filtered through a 35-mm nylon mesh, they were resuspended in Ca2+- and Mg2+-free dissociation solution at a final concentration of 5 ´ 106 cells per ml. Cell sorting was performed by using a FACsAria machine. The sort decision was based on forward scatter and EGFP fluorescence.Cytochemistry, Immunocytochemistry and Imaging.
Mice were anesthetized and killed by transcardial perfusion with PBS followed by 4% paraformaldehyde in PBS. Brains were removed, postfixed in 4% paraformaldehyde in PBS, and cryoprotected by immersion in 30% sucrose in PBS at 4°C. For EGFP-P-Rex2 mice, 100-mm sections were cut by using a vibratome and mounted onto slides by using Vectashield mounting medium (Vector Laboratories). Sections were analyzed by confocal microscopy on the basis of the green fluorescence of the Purkinje cells. For P-Rex1-/-, P-Rex2-/-, and P-Rex1-/-/P-Rex2-/- mice, 200-mm cerebellar sections were cut by using a cryostat and stored briefly in PBS at 4°C. Sections were incubated in blocking buffer (PBS with 2% BSA, 5% goat serum, 0.1% Tween 20, and 0.02% Triton X-100) for one hour, then incubated with rabbit Calbindin antibody (3) in blocking buffer at 4°C overnight. After washing in PBS, sections were incubated with goat anti-rabbit Alexa Fluor 568 (Molecular Probes) and Hoechst stain (Sigma) in blocking buffer for 3 h at room temperature, washed in PBS, rinsed in water, mounted by using Vectashield medium, and analyzed by confocal microscopy. High-resolution confocal images of GFP-labeled Purkinje cells were taken with an Olympus FV1000 with a 60 ´ 1.4 NA oil immersion lens. Lower-resolution confocal images of Calbindin-stained Purkinje cells were taken with a Zeiss LSM 510 META with a 20 ´ 0.5 NA air objective. Quantitative measurements were obtained from confocal image stacks by using Volocity software (Improvision).Behavioral Tests
. Behavioral tests were carried out with 53 P-Rex2-/- and P-Rex2+/+ mice (at least 12 females and 12 males per group), 20 P-Rex1-/- and P-Rex1+/+ mice (at least 4 females and 4 males per group), and 43 P-Rex1-/-/P-Rex2-/- and P-Rex1+/+/P-Rex2+/+ mice (at least 8 females and 10 males per group). Periods of behavioral testing were for 2-3 weeks at the ages of 2, 6, 9, 12, and 15 months for P-Rex2-/- mice and at the age of 2 months for P-Rex1-/- and P-Rex1-/-/P-Rex2-/- mice. For P-Rex2-/- animals, all tests were performed at all ages except for the elevated plus-maze test, which was done at 2, 6, and 11 months. Two other cohorts of P-Rex2-/- and P-Rex2+/+ mice were tested at 2 months and 9 months of age, respectively, to further validate results. Mice were housed (3-5 per cage) with a 12-h light/dark cycle (lights on at 7:00 am) and had access to food and water ad libitum at all times. All procedures on live animals were conducted in accordance with the requirements of the U.K. Animals (Scientific Procedures) Act 1986. Before commencing behavior tests, 6-week old mice were habituated to being handled for at least one minute twice daily for 2 weeks. Tails were marked with permanent pen for identification. During periods of behavior testing, vaginal smears were performed at least once weekly on all females to establish stages in the estrus cycle. Between testing periods, animals were weighed and tail-marked twice weekly.Rotarod tests were performed essentially as described in ref. 4 to measure learning of skilled movements and motor coordination. The animals were trained on two successive days to run on a slowly accelerating rotarod (5-42 rpm within 5 min) for 6 min each. The following day, animals were tested at set speeds of 5, 10, 15, 20, 26, 35, and 42 rpm for 60 s twice each. The following behaviors were scored: falling-off time, number of foot-slips, number of full passive turns clinging to the rod, time of first passive turn, and displacement behaviors (number of look-ups, stand-ups, reversing of direction, and grooming). In this article, we used data from the second run throughout, because data obtained in the first run reflects the ability to adapt to new speed settings as well as the ability for motor coordination. Rotarod training was repeated at each age tested.
Basic locomotor activity (LMA) was measured in infrared-beam-equipped "activity boxes" over a two-hour period in the dark (5). Infrared-beam breaks and runs (successive breaks of infrared beams at opposite ends of the box) were recorded.
Selected tests from the SHIRPA battery were performed to assay basic motor functions and included tests for negative geotaxis (animal placed head-down onto the bottom of a vertical grid; ability to right itself and climb up the grid is scored), wire maneuver (animal allowed to grip narrow bar by front paws; ability to climb onto the bar is scored), visual placing (animal suspended by tail is slowly approached to a surface; point at which front paws stretch out is measured), and grip strength (animal suspended by tail is allowed to grip narrow bar with front paws; grip strength scored by gently pulling animal back by tail) (6).
Beam walking was used as an additional test for motor coordination (4). Mice were placed onto a 1-m-long horizontal wooden beam and motivated to walk along it by shining a 60W light at the front and placing a "safe-house" (dark plastic box containing litter material from their cage) at the end. Five beam types of increasing difficulty (16 and 8 mm width flat; 12, 8, and 6 mm round) were used successively, twice each. Average time taken in two trials to run along the beam, slips, and falling-off were recorded. Animals that did not complete the course because of "freezing" along the way or falling off were scored as taking 150 s. Training was done the previous day by placing the mice three times onto the beam, each time increasingly further from the "safe-house."
Footprint analysis was performed essentially as described (4, 7). In brief, mice were trained three times to be scruffed and then run along a 1.2-m-long, 15-cm-wide paper-lined course (18-cm-high walls). They were motivated to walk the course by shining a 60W lamp at the start and placing a "safe-house" (see above) at the end. Immediately after the third training run, mice were scruffed to have their front paws painted red and hind paws black (nontoxic, water-soluble paint) and then ran the course. This was repeated once. Time taken to complete the course and number of stops along the way were recorded. Footprint patterns were analyzed for stride length, front-leg gait width, back-leg gait width, and front/hind paw overlap (4, 7).
P-Rex1-/-/P-Rex2-/- mice and old P-Rex2-/- mice tended to clasp hind legs instead of showing normal escape posture when lifted by the tail. To quantify limb clasping, each animal was scored on two to three separate days for signs of limb clasping in one or both hind legs.
The acoustic startle response was measured exactly as described in ref. 4. In brief, animals were placed into an acoustic startle chamber (San Diego Instruments) for 15 min. After a 5-min acclimatization period with continuous white background noise (65 db), 72 30-ms pulses of white noise were given at random intervals, the first half at 120 db and the second at 105 db, with two-thirds of the pulses preceded by prepulses of 2, 4, 8, or 16db above background. The whole-body startle response to pulse-alone trials and the gating (i.e., inhibition of response due to presentation of prepulses) were recorded as the mean startle during a 50-ms time window from the onset of the 120-db pulse. Prepulse inhibition was calculated as percent reduction between prepulse and pulse-alone trials.
An elevated plus-maze was used to evaluate anxiety (8). In brief, the plus-maze is an elevated plus-shaped platform of two open and two enclosed arms (10-cm-high walls). Animals were placed on an open arm and their movement around the maze monitored by a computerized motion-tracking system (Ethovision) for 5 min. Behaviors recorded included head-dips over the edge, full-body stretches to peak over the edge, or around corners and grooming. Time spent in enclosed and open areas, speed and duration of movement, and distance traveled were analyzed by using Ethovision software.
Statistical Analysis.
Before statistical analysis, quantitative data were tested for normality (Shapiro-Wilk test). When the variables displayed a large range, indicating the possibility of heterogeneous random error, a logarithmic transformation of the data was applied. Difference between genotypes for the quantitative variables (e.g., body weight or fall time), accounting for age, was tested by using ANOVA for repeated measures. Difference between genotypes for qualitative data (e.g., limb clasping) was assessed by using Chi2 tests.1. Tomomura M, Rice DS, Morgan JI, Yuzaki M (2001) Eur J Neurosci 14:57-63.
2. Welch HC, Condliffe AM, Milne LJ, Ferguson GJ, Hill K, Webb LM, Okkenhaug K, Coadwell WJ, Andrews SR, Thelen M, et al. (2005) Curr Biol 15:1867-1873.
3. Heppelmann B, Senaris R, Emson PC (1994) Brain Res 635:293-299.
4. Carter RJ, Lione LA, Humby T, Mangiarini L, Mahal A, Bates GP, Dunnett SB, Morton AJ (1999) J Neurosci 19:3248-3257.
5. Colebrooke RE, Humby T, Lynch PJ, McGowan DP, Xia J, Emson PC (2006) Eur J Neurosci 24:2622-2630.
6. Rogers DC, Fisher EM, Brown SD, Peters J, Hunter AJ, Martin JE (1997) Mamm Genome 8:711-713.
7. Brunskill EW, Ehrman LA, Williams MT, Klanke J, Hammer D, Schaefer TL, Sah R, Dorn GW, II, Potter SS, Vorhees CV (2005) Eur J Neurosci 22:1265-1276.
8. Plagge A, Isles AR, Gordon E, Humby T, Dean W, Gritsch S, Fischer-Colbrie R, Wilkinson LS, Kelsey G (2005) Mol Cell Biol 25:3019-3026.