Neuron, Volume 80 Supplemental Information Fezf2 Expression Identifies a Multipotent Progenitor for Neocortical Projection Neurons, Astrocytes, and Oligodendrocytes Chao Guo, Matthew J. Eckler, William L. McKenna, Gabriel L. McKinsey, John L. R. Rubenstein, and Bin Chen 1 Supplementary Figure 1. Related to Figure 1. Characterization of Fezf2-CreERT2 mice. (A) In situ hybridization showing Fezf2 expression in the VZ of P0 wildtype brains (black arrow). (B) GFP-expressing cells in the VZ of P0 Fezf2-GFP BAC transgenic mouse (white arrow). (C-D) In situ hybridization for Fezf2 (C) and Cre (D) at P7 demonstrating that Cre recapitulates Fezf2 expression. (E-F’) No recombination was observed in Fezf2-CreERT2; RCE-GFP (E-E’) or Fezf2-CreERT2; R26R-LacZ (F-F’) animals after vehicle administration. (G) Recombination in Fezf2-CreERT2; R26R-LacZ mice after tamoxifen administration. (H-I) Recombination was observed in deep-layer postmitotic neurons of Fezf2-CreERT2; TauR-mGFP mice. (J,J’) GFP+ cells in the brains of P21 Fezf2-CreERT2; RCE-GFP mice that received tamoxifen at E18.5. (J’) All GFP+ cells displayed astrocyte or oligodendrocyte morphology, however GFP+ projection neurons in layer 5 were not present, demonstrating the lack of recombination in postmitotic neurons. Ctx, cerebral cortex; Hip, hippocampus; Str, striatum; SVZ, subventricular zone; Thal, thalamus; TM, tamoxifen; WM, white matter. Scale bars: (B, D, E’, F’, J, J’) 250 µm, (E, F) 500 µm, (G) 100 µm, (H) 50 µm, (I) 25 µm. 2 3 4 Supplementary Figure 2. Related to Figure 2. Fezf2+ RGCs generate all major projection neuron subtypes. (A-C) Enlarged inserts from Figure 2 F(A), K(B) and P(C). (D-F) GFP labeled axons tracts demonstrating that E12.5 Fezf2+ RGCs generate callosal projections (D), corticothalamic projections (E) and subcerebral projections (F). (G-O) E14.5 Fezf2+ RGCs generate callosal projection neurons. (G) Whole mount brain indicating injection site (red arrow) of Ctß into one cortical hemisphere. (H-O) Upperlayer GFP+ neurons (arrowheads) that were labeled by Ctß injection into the contralateral cortical hemisphere (H, J-L) and expressed SATB2 (I, M-O). Ctx, cerebral cortex; OB, olfactory bulb; TM tamoxifen. Scale bars: (C) 25 µm, (F) 500 µm, (G) 100 µm, (I, O) 200 µm. 5 6 Supplementary Figure 3. Related to Figure 3. Clonal analysis of Fezf2+ RGCs. (A) GFP+ clone from a P21 Fezf2-CreERT2; RCE-GFP brain that received TM at E12.5, indicating that early Fezf2+ RGCs generate deep- and upper-layer neurons, astrocytes and oligodendrocytes. (B-E) Additional examples of clones identified using Fezf2-CreERT2; Confetti mice that received tamoxifien at E12.5 (B-C) or E14.5 (D-E). TM, tamoxifen. Scale Bars: (A, C, E) 100 µm, (A4) 25 µm, (C4, E4) 10 µm. 7 8 Supplementary Figure 4. Related to Figure 4. Characterization of CUX2 expression during early to mid corticogenesis. (A-C) Immunohistochemistry at E10.5 demonstrating robust CUX2 expression within the cortical hem. (D-F) Immunohistochemistry at E10.5 did not detect CUX2 expression in the neocortex. (G-I) At E13.5 CUX2 immunoreactivity was mainly detected in the marginal zone of the cortical plate. (J) CUX2 protein expression at E14.5. The majority of CUX2+ cells were in the SVZ and cortical plate and exhibited a nuclear morphology reminiscent of migrating interneurons. (K-M) At E14.5 many CUX2+ cells in the VZ, SVZ and cortical plate expressed the interneuron marker GAD67. Ctx, cerebral cortex; CH, cortical hem; CP; cortical plate; LV, lateral ventricle; PP, preplate; SVZ, subventricular zone; TM, tamoxifen; VZ, ventricular zone. Scale Bars: (A) 250 µm, (C, F, I) 25 µm, (J) 50 µm, (M) 25 µm. 9 10 Supplementary Figure 5. Related to Figure 4. Characterization of Cux2-Cre and Cux2CreERT2 mice. (A-D) In situ hybridization for Cux2 at E11.5 (A-B) and E15.5 (C-D). (EK) E15.5 Cux2-Cre brains. (E-F) GFP+ (E) or TdTomato+ (F) cells were present throughout the brains. (E) RCE-GFP reporter. (F) Ai14 reporter. (G-K) Characterization of cortical cell types labeled by GFP in Cux2-Cre; RCE-GFP mice. At E15.5 GFP+ cells expressed the RGC marker PAX6 (G) as well as the basal progenitor marker TBR2 (H). GFP+ cells also occupied the cortical plate and expressed the subcortical neuron marker CTIP2 (I). (J-K) Many GFP+ cells in the cortical plate and IZ expressed CUX2 (J) and CRE (K). However, most GFP+ cells in the VZ/SVZ were not labeled by CUX2 or CRE antibodies. (L-N) Cux2-CreERT2 lineage traced cells occupied deep cortical layers and expressed the deep-layer markers NFIB (L), SOX5 (M) and TBR1 (N). CH, cortical hem; CP, cortical plate; Ctx, cerebral cortex; IZ, intermediate zone; LV, lateral ventricle; PP, preplate; SVZ, sub-ventricular zone; TM, tamoxifen; VZ, ventricular zone. Scale bars: (A, C) 200 µm, (B) 10 µm, (D) 25 µm, (F) 500 µm, (G-K) 25 µm, (L-M) 50 µm, (N) 25 µm. 11 Supplemental Experimental Procedures Mice Generation of RCE-GFP (Sousa et al., 2009), R26R-LacZ (Friedrich and Soriano, 1991), tauR-mGFP (Hippenmeyer et al., 2005), Ai9 (Madisen et al., 2010), Ai14 (Madisen et al., 2010), Confetti (Snippert et al., 2010), Dlx1/2-Cre (Potter et al., 2009), Nkx2.1-Cre (Xu et al., 2008), Cux2-Cre (Franco et al., 2011) and Cux2-CreERT2 (Franco et al., 2012) mice was previously described. Fezf2-CreERT2 mice were generated by modifying the RP23-141E17 BAC which contains the mouse Fezf2 gene plus 36 kb of 5′ flanking sequence and 160 kb of 3′ flanking sequence. It was targeted with the pERFN vector containing a CreERT2-frt-neo-frt cassette flanked by approximately 200-400 bp homology A and B boxes. CreERT2 was inserted in front of the Fezf2 open reading frame through homologous recombination by a defective prophage system (Lee et al., 2001). The start codon of the Fezf2 ORF was mutated to prevent expression of FEZF2 from the transgene. The A box was amplified from the RP23-141E17 BAC by PCR using primers 5’-GGTACCCTTGGGTGCACTTTGTTCTG-3’ and 5’GGCGCGCCAAGCTGGGCCAGGCTGGGC-3’. The B box was amplified using primers 5’- GTTTAAACGCCAGCTCAGCTTCCCTGGAG-3’ and 5’GCGGCCGCGCAGCACACGCCACAGTTGGT-3’. The modified BAC was purified and injected into zygotes derived from FVB mice according to established protocols (Gong and Yang, 2005). Tamoxifen induction of Fezf2-CreERT2 or Cux2-CreERT2 was achieved by single intraperitoneal injection (4-6mg/40g body weight) of tamoxifen (Sigma) dissolved in 12 corn oil into pregnant mothers at indicated gestation stages. Alternatively, for high efficiency labeling 1mg/40g body weight of 4-hydroxytamoxifen (Sigma) dissolved in 100% EtOH was injected intraperitoneally. The day on which a vaginal plug was detected was designated as embryonic day 0.5 (E0.5). The day of birth was designated as postnatal day 0 (P0). Identification of single RGC clones Clonal analysis was preformed on Fezf2-CreERT2; RCE-GFP or Fezf2-CreERT2; Confetti brains using low-efficiency tamoxifien induction. XFP+ cell bodies in one cortical hemisphere were present in 2 or 3 neighboring sections. Based on the low number of cells labeled and the proximity of the cells, we identified these cells as the progeny from a single RGC. In situ hybridization Generation of the Fezf2 probe was described previously (Eckler et al., 2011). The Cre probe was generated by PCR using primers 5’TGCCACGACCAAGTGACAGCAATG-3’ and 5’ACCAGAGACGGAAATCCATCGCTC-3’. Generation of the Cux2 probe was described previously (Zimmer et al., 2004). Immunohistochemistry The following additional primary antibodies were used: chicken anti-ß-gal (Abcam ab9361), goat anti-SOX2 (Santa Cruz Biotech sc-17320), rabbit anti-TBR2 13 (Abcam ab23345), rat anti-CTIP2 (Abcam ab18465), rabbit anti-SATB2 (Abcam ab34735), chicken anti-GFP (Aves GFP-1020), rabbit anti-GFAP (Dako Z0334), rabbit anti-OLIG2 (Abcam ab33427), rabbit anti-SOX5 (Abcam ab26041), rabbit anti-NFIB (Active Motif 39091), rabbit anti-CUX1 (Santa Cruz Biotech sc-13024), rabbit antiPAX6 (Covance PRB-278P), rabbit anti-TBR1 (abcam ab31940), rabbit anti-CRE (Covance PRB-106P) and mouse anti-GAD67 (Millipore MAB5406). Fluorescent secondary antibodies were from Jackson laboratories. Nuclei were visualized with DAPI (Molecular Probes). For cell counting a minimum of 3 sections each were counted from 3 experimental replicates (minimum of nine sections total per group). To determine statistical significance, GraphPad was used to preform an unpaired t-test between experimental groups. Retrograde Tracing Retrograde tracing was performed using AlexaFluor 594-conjugated Cholera Toxin β subunit (CT-β) (Invitrogen) injections into one cortical hemisphere of P1 mice. The brains were collected at P21 for analysis. Microscopy Bright field and epiflourescence images were captured on an Olympus BX51 microscope using a Q Imaging Retiga EXj camera or a Keyence BZ-9000 microscope. Confocal images were captured on a Leica TCS SP5 confocal microscope. Images were processed using Adobe Photoshop CS5 to adjust brightness and contrast. 14 References Gong, S., and Yang, X.W. (2005). Modification of bacterial artificial chromosomes (BACs) and preparation of intact BAC DNA for generation of transgenic mice. Current protocols in neuroscience / editorial board, Jacqueline N Crawley [et al] Chapter 5, Unit 5 21. 15