Supporting Information

Files in this Data Supplement:

Supporting Figure 6
Supporting Table 1
Supporting Figure 7
Supporting Figure 8
Supporting Table 2
Supporting Figure 9
Supporting Figure 10
Supporting Figure 11
Supporting Table 3
Supporting Figure 12

Fig. 6. Strategy for developing dopamine-deficient floxed-stop (DDfs) mice, which carry two Tyrosine hydroxylase floxed-stop (Th^fs) alleles, one Dbh allele, and one Dbh^Th allele. (A) Creation of Th^fs allele. A targeting construct was prepared by inserting a Pgk-Neo^R cassette flanked by loxP sites into the KpnI site within the first intron of the Th gene. We predicted that the Pgk-Neo^R cassette would disrupt transcription of Th mRNA (1, 2). The targeting construct has »6 kb of 5' flanking region and »1.7 kb of 3' flanking region. A BglII site located »200 bp 5' of the Th translation initiation codon, which also corresponds to a putative binding site for the Oct-2 transcription factor (4, 5), was modified to an XbaI site using PCR-based mutagenesis. The targeting construct was introduced into AK18.1 embryonic stem (ES) cells (derived from 129S4 mice) by electroporation. Correctly targeted ES cells were identified with a PCR-based screening strategy using primer A (5'-GCAGCGCATCGCCTTCTATCG), which is located within the NeoR gene, and primer B (5'-CTCAAACACTTTCAAAGCCCGAG), which is located in exon 2 downstream of the SacII site, to amplify a »2-kb product. One correctly targeted ES clone was identified (out of 400 ES cell clones screened). PCR amplification using primer C (5'CTCCTGTCCCAGAACACC) and primer D (5'-TGCTCCAGATACCCCTGG), which flank the engineered XbaI site, followed by XbaI digestion of the PCR product, confirmed the presence of targeted allele. Chimeras harboring the Th^fs allele were generated by injecting correctly targeted ES cells into C57BL/6 (B6) blastocysts. Five male chimeras were generated and bred with C57BL/6 mice; two yielded germline transmission. Mice were genotyped using primers C and D. Breeding of Th^+/fs mice with each other produced no Th^fs/fs mice (n = 50 births). This finding is consistent with our prediction that Th^fs/fs mice would be Th-deficient and exhibit an embryonic lethal phenotype as previously described for Th-null mice (3). To evaluate whether this embryonic lethality was related to insertion of Pgk-Neo^R, the stop cassette was removed from the germ line of Th^+/fs mice by breeding to Rosa26-Cre mice, a general deletor mouse line, to generate Th^+/lox mice. Recombination and deletion of the floxed Pgk-Neo^R cassette was confirmed with PCR by using a set of primers flanking the stop cassette. An intercross of Th^+/lox mice successfully yielded Th^lox/lox mice that were viable and demonstrated no gross dysfunction of feeding, locomotor, or thermoregulatory behaviors. In addition, brain catecholamine levels of Th^lox/lox mice were normal as measured by HPLC with electrochemical detection. (B) Breeding strategy for generating DDfs mice. Th^+/fs mice were bred with Dbh^Th/+ mice (6) to generate double heterozygotes. The double heterozygotes were then bred with each other to generate Th^fs/fs,Dbh^+/Th mice ( = DDfs); their littermates with at least one intact Dbh allele and one intact Th allele were used as controls because they have nearly normal levels of all catecholamines. DDfs mice were identified at 10 days of age by their small size and responsiveness to L-dopa. They were injected daily with L-dopa (»50 mg/kg i.p.) before viral rescue.

1. Pham, C. T., MacIvor, D. M., Hug, B. A., Heusel, J. W. & Ley, T. J. (1996) Proc. Natl. Acad. Sci. USA 93, 13090–13095.

2. Ren, S. Y., Angrand, P. O. & Rijli, F. M. (2002) Dev. Dyn. 225, 305–315.

3. Zhou, Q. Y., Quaife, C. J. & Palmiter, R. D. (1995) Nature 374, 640–643.

4. Yang, C., Kim, H. S., Seo, H. & Kim, K. S. (1998) J. Neurochem. 71, 1358–1368.

5. Dawson, S. J., Yoon, S. O., Chikaraishi, D. M., Lillycrop, K. A. & Latchman, D. S. (1994) Nucleic Acids Res. 22, 1023–1028.

6. Zhou, Q. Y. & Palmiter, R. D. (1995) Cell 83, 1197–1209.

Fig. 7. Control and virally rescued dopamine-deficient floxed-stop (vrDDfs) images showing Tyrosine hydroxylase (Th) (green) but with dopamine transporter immunostaining removed (supplemental to Fig. 2). (A–D) Coronal sections through the striatum, showing projections of the midbrain dopamine neurons. Th staining was present throughout the caudate putamen (CPu) and ventrolateral striatum of vrDDfs mice. The medial shell and core of the nucleus accumbens (NAc) were devoid of Th signal in vrDDfs mice.

Fig. 8. Coronal sections through the nucleus accumbens (NAc) of mice treated with 2 mg/kg i.p. SKF81297 1 h prior to perfusion and immunostained for Th (green) and c-Fos (red). (A) DDfs mice have no detectable TH signal but abundant c-Fos positive nuclei throughout the striatum because of hypersensitive D1R signaling in the absence of dopamine. (B) Control mice have abundant Th staining but very few cFos- positive cells in the NAc. (C) vrDDfs mice have strong Th staining in the caudate putamen (CPu) and ventrolateral striatum but not in the NAc; c-Fos positive nuclei are abundant in the NAc (but not in CPu) because of the lack of dopamine production in this region. Immunostaining was performed on 2-4 mice per genotype; representative samples are shown.

Fig. 9. Body weight (BW) of control (n=12) and dopamine-deficient floxed-stop (DDfs) (n=8) mice after bilateral injection of canine adenovirus expressing Cre recombinase (CAVCre) into the caudate putamen (CPu). Black bar signifies daily L-dopa treatment; 1 week after surgery (S), DDfs mice stopped receiving daily injections of L-dopa. Those mice that continue to gain weight are designated virally rescued DDfs mice (vrDDfs). Untreated DDfs mice stop eating, lose body weight, and die within »3 days without L-dopa (data not shown). BW vrDDfs typically approach that of control littermates 1-3 months after surgery. Data are means ± SEM.

Fig. 10. Lickometer analysis of feeding behavior (supplemental to Fig. 3). (A) The total number of meals during the A=acclimation, B=baseline, or R=recovery phases of the experiment were not statistically different. (B) There were no statistically significant differences in the distribution of meal durations between the genotypes. Data are presented as means ± SEM.

Fig. 11. Lickometer analysis of mice given 2 h of restricted access to highly palatable liquid diet for 10 consecutive days (supplemental to Fig. 3). (A) Cumulative licks over 2 h. (B) Liquid diet consumed (by weight) during the 2-h interval. Data are presented as means ± SEM.

Fig. 12. Activity in the elevated plus maze (Columbus Instruments) was examined by placing control (n=7) and vrDDfs (n=8) mice in the center of the plus maze and video recording for 5 min. ETHOVISION (Noldus) software was used for tracking and analysis. The time spent in the closed arms versus the open arms, the total distance traveled, and the number of entries into each compartment were determined. There were no significant differences between genotypes. Data are presented as means ± SEM.