Suppression of interneuron programs and maintenance of selected spinal fates by the factor AML1/Runx1

Nicolas Stifani*, Adriana R. O. Freitas*, Anna Liakhovitskaia†, Alexander Medvinsky†, Artur Kania‡§¶, and Stefano Stifani*ʈ

*Center for Neuronal Survival, Montreal Neurological Institute, McGill University, Montreal, QC, Canada H3A 2B4; †Institute for Stem Cell Research/Medical Research Council Centre for Stem Cell Biology, University of Edinburgh, Edinburgh EH9 3JQ, United Kingdom; ‡Department of Anatomy and Cell Biology, Division of Experimental Medicine, McGill University, Montreal, QC, Canada H3A 2B2; §De´partement de Me´decine, Universite´de Montreal, Montreal, QC, Canada H3C 3JY; and ¶Neural Circuit Development Laboratory, Institut de recherches cliniques de Montreal, Montreal, QC, Canada H2W 1R7

Edited by Charles F. Stevens, Salk Institute for Biological Studies, La Jolla, CA, and approved February 29, 2008 (received for review November 29, 2007) Individual spinal motor neuron identities are specified in large part pattern by examining a previously characterized Runx1lacZ/ϩ by the intrinsic repertoire of transcription factors expressed by knockin mouse line in which ␤-galactosidase (␤-gal) expression undifferentiated progenitors and maturing neurons. It is shown recapitulates the expression of endogenous Runx1 (13, 18, 19). In here that the AML1/Runx1 (Runx1) is expressed embryo day (E)9.5 Runx1lacZ/ϩ embryos, ␤-gal expression coin- in selected spinal motor neuron subtypes after the onset of cided with Runx1 , detected by using a previously de- differentiation and is both necessary and sufficient to suppress scribed anti-Runx1 antibody (17), in the second branchial arch, interneuron-specific developmental programs and promote main- where Runx1 mRNA expression has been demonstrated (18) tenance of motor neuron characteristics. These findings show an [supporting information (SI) Fig. S1 A–C]. The specificity of the important role for Runx1 during the consolidation of selected spinal motor neuron identities. Moreover, they suggest a require- anti-Runx1 antibody was demonstrated by the absence of im- ment for a persistent suppression of interneuron within munoreactivity in Runx1-deficient embryos (Fig. S1 E and F). lacZ/ϩ ␤ maturing motor neurons. In the spinal cord of Runx1 embryos, -gal expression began at ϷE9.5 in a small number of ventrolateral cells at levels ϩ lateral motor column ͉ median motor column ͉ runt ͉ spinal cord C1–C4 (Fig. 1A, arrowheads). ␤-gal cells did not express spinal accessory column neuronal progenitor markers such as Pax6 or Nkx2.2, nor the general cell proliferation marker Ki67 (Fig. 1 A–C). Moreover, we observed no detectable overlap between the expression of pecific transcription factor codes within exclusive ventral ␤ Sprogenitor domains regulate motor neuron and interneuron -gal and the interneuron markers (V0 interneurons), En1 differentiation in the developing spinal cord (1, 2). Some deter- (V1), and Chx10 (V2) (Fig. 1 D–F), although some occasional

minants of both lineages are coexpressed in mitotic progenitors overlap with Chx10 was observed at later stages (see Fig. 3 NEUROSCIENCE ϩ (3), raising the questions of what molecules control the diver- below). Instead, ␤-gal cells expressed the postmitotic motor gence and maintenance of motor neuron and interneuron dif- neuron markers Isl1 and choline acetyltransferase (ChAT) (3, 4, ferentiation programs. Genetic studies suggest that the Hb9 20) (Fig. 1 G and H, arrows). Only a subset of the Isl1ϩ cells is required to suppress interneuron programs actively in matur- expressed ␤-gal, suggesting that Runx1 is expressed in a restricted ing motor neurons (3, 4), but other effectors of the mechanisms number of spinal motor neurons. The expression of ␤-gal that promote the divergence of motor and interneuron fates faithfully recapitulated the Runx1 expression pattern in the remain to be determined. spinal cord (Fig. S1 H–J). These results indicate that Runx1 In both invertebrates and vertebrates, the runt/Runx in the spinal cord is first activated in a subpopulation family encodes DNA-binding transcription factors that mediate of postmitotic motor neurons, but not their progenitors, at transactivation or repression depending on specific contexts (5). cervical levels C1–C4. Members of this transcription factor family regulate neuron subtype specification and axon target connectivity in (6–8), chick (9, 10), and mice (11–16). The runt/Runx family Runx1 Expression Is Activated in Ventrally Exiting Motor Neurons member AML1/Runx1 (Runx1) is expressed in selected popula- After the Onset of Differentiation. To determine the identity of the cervical Isl1ϩ neurons in which ␤-gal is first expressed in tions of motor neurons in the murine and avian spinal cord, ϩ suggesting that it is involved in motor neuron development (13, Runx1lacZ/ embryos, we compared the expression of ␤-gal with 17). Here, we show that mouse Runx1 is expressed in restricted that of the Hb9 protein, which is expressed in virtually all motor groups of ventrally exiting cervical motor neurons during their neurons whose axons exit via the ventral root and innervate postmitotic development. Loss of Runx1 function does not affect skeletal muscles (3, 4). No visible overlap of ␤-gal and Hb9 the survival of those motor neurons but results in a loss of expression was observed at E9.5 (data not shown) and E10.5, the expression of motor neuron-specific genes and a concomitant peak of motor neuron generation (Fig. 1 J–L). At E9.5, we also activation of expression of interneuron-specific genes. Con- versely, ectopic expression of Runx1 in the spinal cord of developing chick embryos suppresses interneuron gene expres- Author contributions: N.S. and A.R.O.F. contributed equally to this work; N.S., A.K., and S.S. sion and promotes motor neuron differentiation programs. designed research; N.S., A.R.O.F., and A.L. performed research; A.L. and A.M. contributed These results identify a role for Runx1 in the establishment of new reagents/analytic tools; N.S., A.R.O.F., A.K., and S.S. analyzed data; and N.S., A.K., and selected motor neuron identities and suggest that maturing S.S. wrote the paper. motor neurons must continually suppress interneuron-specific The authors declare no conflict of interest. developmental programs. This article is a PNAS Direct Submission. ʈTo whom correspondence should be addressed at: Montreal Neurological Institute, 3801 Results rue University, Montreal, QC H3A 2B4, Canada. E-mail: [email protected]. Runx1 Is Expressed in Specific Postmitotic Motor Neurons in the Mouse This article contains supporting information online at www.pnas.org/cgi/content/full/ Cervical Spinal Cord. To determine the role of Runx1 in spinal 0711299105/DCSupplemental. motor neuron development, we first characterized its expression © 2008 by The National Academy of Sciences of the USA

www.pnas.org͞cgi͞doi͞10.1073͞pnas.0711299105 PNAS ͉ April 29, 2008 ͉ vol. 105 ͉ no. 17 ͉ 6451–6456 Downloaded by guest on October 1, 2021 Fig. 1. Runx1 expression in spinal motor neurons. (A–H) Expression of ␤-gal and the indicated in the spinal cord of E9.5 (A–C and G) or E10.5 (D–F and H) Runx1lacZ/ϩ embryos. ␤-gal expression does not overlap with proliferation (A–C) or ventral interneuron (D–F) markers, but it overlaps with the motor neuron markers ChAT and Isl1 (G and H, arrows). Arrowheads in (A–C) point to ␤-galϩ cells. (I) Summary of Runx1 expression (green) in the cervical spinal cord (SC) at E9.5; the location of the progenitor domain, ventral interneuron 0, 1, and 2 domains, and vMN domain are indicated. (J–O) Expression of ␤-gal and Hb9 in the cervical spinal cord of Runx1lacZ/ϩ embryos. No overlap is observed at E10.5 (J–L, arrowheads), whereas ␤-galϩ/Hb9ϩ cells are visible in a ventromedial domain at E11.5 (M–O, arrows; shown at higher magnification in the Inset). (P–R) Coexpression of ␤-gal and Phox2b in the cervical spinal cord of E9.5 Runx1lacZ/ϩ embryos. Virtually all ␤-galϩ cells coexpress Phox2b (arrows). (S) Analysis of axonal projections in E15.5 Runx1lacZ/ϩ embryos by using retrograde labeling by rhodamine- dextran (Rho) placed in the anterior portion of the trapezius muscle. Arrows point to examples of ␤-galϩ cells that were retrogradely labeled. (T) Schematic displaying the location of the ventrolateral domain of the spinal cord analyzed in S.(U) Summary of Runx1 expression in the cervical SC at E11.5; the lateral domain containing Runx1ϩ cells that express Isl1 and Phox2b and project to the anterior trapezius muscle is shown in yellow, and the ventromedial domain containing Runx1ϩ/Hb9ϩ cells is shown in green; LEP, lateral exit point. When shown, dotted lines depict outline of the spinal cord. (D–S) Dorsal is to the Top and lateral to the Right.

failed to detect an overlap between the expression of ␤-gal and and Hb9 expression (E9.5–E10.5). These results suggest that that of Lhx3, a protein whose expression at this stage transiently Runx1 becomes activated in specific vMNs after their initial marks the majority of ventrally exiting motor neurons (vMNs) as differentiation and during their developmental maturation. well as their precursors (20) (data not shown). Coexpression of ␤-gal and Hb9 was first observed in the ventromedial spinal cord Runx1 Is Expressed in Selected Types of Ventrally Exiting Motor at ϷE11 (Fig. 1 M–O, arrows, and data not shown). At this stage, Neurons and Is Not Required for Their Generation or Survival. In Runx1 expression expands ventrally and can be detected at more E13.5 Runx1lacZ/ϩ embryos, when distinct motor columns are caudal levels of the spinal cord (C5–T1) (13). This analysis of discernable, two groups of ␤-galϩ motor neurons were observed Runx1lacZ/ϩ embryos suggests that the onset of Runx1 expression at the forelimb level (C5–C8) (Fig. 2 A–L). One group was in vMNs follows the initial activation of Hb9 expression. composed of motor neurons of the medial component of the In E9.5-E10.5 Runx1lacZ/ϩ embryos, most, if not all, of the axial muscle-innervating median motor column (MMCm) (20), ␤-galϩ cells coexpressed the homeodomain protein Phox2b (Fig. based on their ventromedial location (Fig. 2A, vertical arrows) 1 P–R), which is selectively expressed in spinal accessory (SAC) and the expression of ChAT, Lhx3 (Fig. 2 A–F), and Isl1 (Fig. 2 motor (nXI) neurons (21). SAC neurons are present at spinal J–L). The second group consisted of motor neurons of the lateral levels C1–C4, express Isl1 but not Hb9, and send their axons out motor column (LMC), based on their ventrolateral location and of the spinal cord through lateral exit points located midway expression of retinaldehyde dehydrogenase 2 (RALDH2) (Fig. along the dorsoventral axis of the spinal cord. The exiting axons 2 G–I, horizontal arrows). ␤-gal expression was found in sub- assemble into the spinal accessory nerve, which innervates populations of both medial LMC (LMCm) motor neurons, which branchial arch-derived muscles in the neck (22). Consistent with express Isl1 but not Lim1, and project to ventral limb muscles, these observations, at E15.5 a number of ␤-galϩ cells at levels and lateral LMC (LMCl) motor neurons, which express Lim1 C1–C4 were retrogradely labeled from the anterior trapezius and innervate dorsal limb muscles (17, 23) (Fig. 2 G–O). muscle, a lateral cervical muscle of the neck innervated by the Consistent with these results, a group of ␤-galϩ neurons was spinal accessory nerve (Fig. 1S). Together, these results show retrogradely labeled from the deltoideus muscle (Fig. S2 A–C). that Runx1 expression is activated in two distinct spatiotemporal This result was specific because ␤-galϩ neurons were not retro- patterns. It is first expressed in dorsally exiting SAC motor gradely labeled from the pectoralis muscle (Fig. S2 D–F). neurons starting at ϷE9.5 followed by a later expression in Together, these results argue that Runx1 is expressed in selected selected populations of vMNs after the peak of vMN generation populations of ventrally exiting MMC and LMC motor neurons.

6452 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0711299105 Stifani et al. Downloaded by guest on October 1, 2021 respectively (19, 24). Transheterozygous Runx1lacZ/rd embryos are Runx1-null and die at ϷE12.5 because of a lack of fetal liver-derived hematopoiesis (19). This embryonic lethality can be circumvented by examining Tie2-Cre;Runx1Flox:lacZ/Flox:lacZ em- bryos, in which the expression of Cre recombinase is under the control of the endothelial/hematopoietic-specific Tie2 promoter (25, 26), resulting in a selective reactivation of Runx1 in hema- topoietic cells but not in the nervous system. Concomitantly, this conditional reactivation abolishes ␤-gal expression in hemato- poietic but not neuronal cells (Fig. S3). The total number of ␤-galϩ cells in the spinal cord of both Runx1lacZ/rd and Tie2- Cre;Runx1Flox:lacZ/Flox:lacZ embryos was the same as in their control littermates at all stages examined (Fig. 3 A–C). This result shows that Runx1 is not essential for ␤-gal expression and maintenance nor for the generation and/or survival of the selected spinal SAC, MMC, and LMC neurons in which it is normally expressed.

Runx1 Is Important for Persistent Suppression of Interneuron Differ- entiation Programs and Sustained Expression of Motor Neuron Genes. We traced the fate of motor neurons that normally express Runx1 by following ␤-gal expression in Runx1 mutant embryos. The numbers of ␤-galϩ cells that coexpressed general vMN markers such as Isl1 and ChAT were decreased in mutant embryos compared with control littermates (Fig. 3 D and F). The expression of specific markers of MMC (i.e., Lhx3) or LMC (i.e., RALDH2) motor neurons was also decreased in Runx1-deficient spinal neurons (Fig. 3 D and F). In Runx1 mutant embryos of both genotypes, we also noted the presence of increased numbers of cells coexpressing ␤-gal and the exclusive spinal interneuron markers Pax2 and Chx10 (3, 27) (Fig. 3 E and G). Moreover, we found that Pax2, which is not normally expressed in Isl1ϩ spinal motor neurons (ref. 27; see also Fig. 3 H–J), was coexpressed with Isl1 and ␤-gal in Runx1 mutant motor neurons at E11.5 (Fig. 3 K–M). This situation was observed at several rostrocaudal positions, including level C2–C3 (Fig. S4 A–H) where Runx1- NEUROSCIENCE expressing cells correspond to SAC motor neurons (Fig. 1 P–R). We did not detect any significant change in the expression of Phox2b in Runx1 mutant SAC motor neurons at E11.5 (Fig. S4 I–O), suggesting that at least certain Phox2bϩ SAC neurons might coexpress Pax2 in Runx1-deficient embryos at this stage. Together, these results strongly suggest that Runx1 inactivation results in a derepression of interneuron-specific genes in post- mitotic motor neurons. We next tested whether Runx1 was sufficient to cause a suppression of interneuron-specific genes. To this end, we electroporated a GFP expression plasmid alone or together Fig. 2. Runx1 expression in MMC and LMC motor neuron subtypes. Expres- with a Runx1 expression plasmid into Hamburger and Ham- sion of ␤-gal and the indicated marker proteins in the ventral spinal cord of ilton (HH) (28) stage 14–16 chicken embryo neural tubes. E13.5 Runx1lacZ/ϩ embryos at spinal levels C5 (A–C), C6–C7 (D–L), or C7–C8 After incubation, the majority of GFPϩ cells coexpressed (M–O). In all pictures, dorsal is to the Top and lateral to the Right. ␤-gal Runx1 (Fig. S5). We determined the proportion of GFPϩ cells expression overlaps with ChAT, Lhx3, and Isl1 in MMCm motor neurons expressing motor neuron and interneuron markers at HH stage ␤ ϩ located in a ventromedial (medial; vertical arrows) domain. These -gal cells 27–28 (Fig. 4 A–E). Compared with control embryos express- decrease in number in a rostral-to-caudal direction (cf., A and J). ␤-gal expres- sion is also observed in ventrolateral (lateral; horizontal arrows) LMC motor ing only GFP, Runx1 caused a significant decrease in the neurons that express ChAT and RALDH2. Most of these cells express Isl1 at level expression of the interneuron-specific proteins Pax2 and C5-C6 (J–L), whereas an overlap with Lim1/2 expression is observed at level Chx10, but it did not affect the expression of Evx1, a V0 C7–C8 (M–O). (O) Dotted line outlines the lateral domain of Lim1/2 expression interneuron marker (Fig. 4F). In addition, ectopic Runx1 containing ␤-galϩ/Lim1/2ϩ cells; many Lim1/2ϩ cells found at more medial expression led to an increase in the expression of general positions, likely corresponding to V0/V1 interneurons, do not express ␤-gal; motor neuron markers such as Hb9 and Isl1, as well as the DRG, dorsal root ganglion. In all panels, arrows point to examples of double- motor- and interneuron markers Lim3 and Lim1/2 (Fig. 4G). labeled cells. Dotted lines, except in O, show outline of the spinal cord. (P) These effects were observed at both brachial level, where Summary of Runx1 expression at levels C5–C8 of the spinal cord (SC) of E13.5 Runx1 is expressed, and lumbar level, where Runx1 is not Runx1lacZ/ϩ embryos; Runx1ϩ cells mark a medial subdomain of the MMCm (light blue) and specific subdomains of both LMCm and LMCl (yellow). expressed (17), but occurred only in the ventral, and not dorsal, half of the spinal cord. Importantly, these effects were phe- nocopied by the oncogenic human fusion protein AML1/ETO To characterize the function of Runx1 in motor neuron (hereafter referred to as Runx1/ETO for consistency) (Fig. development, we examined two separate lines of Runx1-deficient S6). Runx1/ETO harbors the DNA-binding domain of Runx1 mice, termed Runx1lacZ/rd and Tie2-Cre;Runx1Flox:lacZ/Flox:lacZ, fused to the protein eight-twenty one, a potent transcriptional

Stifani et al. PNAS ͉ April 29, 2008 ͉ vol. 105 ͉ no. 17 ͉ 6453 Downloaded by guest on October 1, 2021 Fig. 4. Suppression of interneuron-specific genes and promotion of motor neuron programs by exogenous Runx1. (A–E) In ovo electroporations. Expression of GFP and either Hb9 (B and C) or Isl1 (D and E) in the ventral spinal cord of stage 27–28 chick embryos that were electroporated at stage 14–16 with GFP alone (B and D, control) or together with Runx1 (C and E, Runx1). (A) Schematic displaying the location of the ventral domain of the spinal cord of electroporated embryos analyzed in B–G.(F and G) Percentage of GFPϩ cells coexpressing the indicated markers in the ventral spinal cord of chick embryos electroporated with GFP alone (control) or together with Runx1 (Runx1). Error bars, S.E.M. *, P Յ 0.01; ***, P Յ 0.0001, Student’s t test. Ն5 sections were analyzed per embryo, n Ն 4 embryos per experimental condition.

Fig. 3. De-repression of interneuron differentiation programs in Runx1- ␤ ϩ deficient motor neurons. (A) Numbers of -gal cells in the ventral spinal cord cord. Together, the findings of this work strongly suggest that of E10.5 Runx1lacZ/ϩ (ϩ/Ϫ) and Runx1lacZ/rd (Ϫ/Ϫ) littermates. (B and C) Num- bers of ventromedial (medial) and ventrolateral (lateral) ␤-galϩ cells in the Runx1 suppresses interneuron differentiation programs within spinal cord of either E13.5 (B) or E18.5 (C) Tie2-Cre;Runx1Flox:lacZ/ϩ (ϩ/Ϫ) and developing motor neurons. Tie2-Cre;Runx1Flox:lacZ/Flox:lacZ (Ϫ/Ϫ) littermates. (D–G) Numbers of ␤-galϩ cells coexpressing the indicated marker proteins in the ventral spinal cord of either Discussion E13.5 (D and E) or E18.5 (F and G) Runx1-deficient or control littermates. This work demonstrates that mouse Runx1 is expressed in Ventromedial (MMC location) and ventrolateral (LMC location) ␤-gal expres- restricted populations of postmitotic motor neurons in the sion domains were counted separately. *, P Յ 0.01; **, P Յ 0.001; ***, P Յ cervical spinal cord. These cells include selected motor neu- 0.0001, Student’s t test. In all graphs, counts are represented as mean number rons of the SAC, MMCm, LMCm, or LMCl motor columns. of cells Ϯ SEM per section; Ն5 sections were analyzed per embryo, n Ն 4 embryos per genotype. (H–M) Expression of Isl1 and Pax2 in the ventral spinal The spinal Runx1 expression pattern in the mouse is more cord of E11.5 Runx1lacZ/ϩ (H–J, ϩ/Ϫ)orRunx1lacZ/rd (K–M, Ϫ/Ϫ) littermates. complex than in the chick, where it is restricted to a group of Merged images are shown in J and M.(K–M) Arrows point to examples of cells brachial LMCl motor neurons that innervate the dorsal fore- coexpressing Isl1 and Pax2; those cells also coexpressed ␤-gal (M Inset). Dorsal limb scapulohumeralis posterior muscle (17). Although it is Top and lateral is to the Right. remains to be determined whether chick Runx1 is also ex- pressed in SAC and/or MMCm motor neurons, its expression in a defined group of LMCl motor neurons that innervate a repressor that replaces the normal C-terminal transcription single muscle target suggests that Runx1ϩ cells in the chick activation and repression domains of Runx1 (5, 29). The brachial spinal cord define a particular LMC motor neuron ensuing fusion protein retains the DNA-binding specificity of pool (17). This observation raises the possibility that mouse Runx1 and the dedicated transcription repression activity of Runx1 expression might define individual motor neuron pools ETO. Thus, our results suggest that transcriptional repression (or subpopulations within pools) belonging to separate motor is important for the effect of exogenous Runx1 on motor columns. We have found that mouse Runx1 is expressed in only neuron and interneuron development in the ventral spinal subpopulations of Lhx3ϩ MMCm motor neurons or

6454 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0711299105 Stifani et al. Downloaded by guest on October 1, 2021 RALDH2ϩ LMC motor neurons. These observations raise the deficient neurons experience conflicting developmental programs possibility that Runx1 might be involved in mechanisms that because that situation is often associated with increased apoptosis. define specific motor neuron subpopulations in the cervical Moreover, we have observed no obvious perturbations of the cell spinal cord of mouse embryos. body position of Runx1-mutant neurons (N.S. and S.S., unpublished Runx1 is expressed in postmitotic motor neurons but not in observations). proliferating motor neuron precursors. Moreover, Runx1 is In summary, the present results implicate Runx1 in mechanisms expressed in selected vMNs only after E10.5, after the peak that suppress interneuron differentiation programs and consolidate period of vMN generation and the onset of Hb9 expression. the acquisition of specific motor neuron identities. Because Runx1 These findings suggest roles for Runx1 during postmitotic motor expression is not correlated with the initial separation of motor- and neuron development and not during the period of motor neuron interneuron lineages during the precursor-to-neuron transition, birth. In agreement with this possibility, Runx1 inactivation does they suggest further that the discrimination between those com- not perturb the generation of neurons in which it would be peting developmental programs is a prolonged process that initiates expressed, but it causes a decrease of both general (e.g., ChAT, at the level of precursor cells and continues not just after cell cycle Isl1) and specific (e.g., RALDH2) vMN markers. The loss of exit but also during further developmental maturation. These motor neuron markers in Runx1-deficient embryos is reminis- observations suggest that mechanisms that ensure a suppression of cent of the caused by Hb9 inactivation. Hb9 is interneuron differentiation programs persist during postmitotic dispensable for motor neuron generation but is required for motor neuron development. sustained expression of motor neuron markers (3, 4). The alteration of motor neuron gene expression observed in Runx1ϩ Materials and Methods vMNs when Runx1 is inactivated might be caused, at least in part, Mouse Lines. Runx1lacZ/ϩ and Runx1rd/ϩ mice were generated and genotyped as by a perturbation of the mechanisms that maintain a sustained described (19, 30). Runx1lacZ/ϩ males were crossed to Runx1rd/ϩ females to gen- expression of Hb9, with a consequent loss of other downstream erate double-heterozygous Runx1lacZ/rd embryos. Runx1lacZ/rd embryos lack Runx1 activity and die by E12.5 because of failed fetal liver-derived hematopoiesis, motor neuron markers. Because Runx1 is expressed in only a rd/rd Flox:lacZ/ϩ subpopulation of Hb9ϩ vMNs, other processes would promote similar to Runx1 embryos (19). Runx1 mice were generated and geno- typed as described in ref. 24. Runx1Flox:lacZ/ϩ females were crossed to Tie2- the maintenance of Hb9 expression in Runx1-negative motor Cre;Runx1Flox:lacZ/ϩ males, in which Cre recombinase expression is driven by the neurons. endothelial/hematopoietic-specific Tie2 promoter (25). For embryonic staging, The decreased expression of motor neuron markers caused the day of appearance of the vaginal plug was considered as E0.5. All animal by Runx1 inactivation is correlated with a converse increase in procedures were conducted in accordance with the guidelines of the Canadian the expression of ventral interneuron genes, including the V2 Council for Animal Care. interneuron marker Chx10. This effect is also similar to the phenotype observed in Hb9 mutant embryos, where cells Immunohistochemistry. Mouse embryos were collected and fixed, and cryostat destined to become motor neurons exhibit aberrant expression sections were prepared as described (13, 29). Sections were subjected to immu- of Chx10, suggesting that Hb9 is required for suppression of V2 nohistochemistry as described in ref. 13, with the following antibodies: mouse interneuron programs (3, 4). We observed Runx1-deficient monoclonals against En1 (clone 4G11; 1:10), Evx1 (clone 99.1–3A2; 1:100), Hb9 (clone 81.5C10; 1:25), Isl1 (clone 39.4D5; 1:50), Lim1/2 (clone 4F2; 1:5), Lim3 (clone motor neurons exhibiting a ‘‘hybrid’’ phenotype characterized 67.4E12; 1:25), Nkx2.2 (clone 74.5A5; 1;100), Pax6 (clone Pax6; 1:10), and ␤-gal NEUROSCIENCE by the coexpression of motor- and interneuron genes. These (clone 40–1A) (obtained from the Developmental Studies Hybridoma Bank de- results suggest that when Runx1 is inactivated, postmitotic cells veloped under the auspices of the National Institute of Child Health and Human fated to develop as motor neurons undergo, in addition to a Development and maintained by the University of Iowa, Department of Biolog- reduced expression of motor neuron-specific genes, an acti- ical Sciences, Iowa City, IA); rabbit polyclonals against ␤-gal (1:5,000; Cappel), vation of interneuron gene expression. This possibility is Chx10 (1:4,000) and Runx1 (1:2,000) (gifts from T. Jessell, J. Dasen, and S. Brenner- consistent with our finding that forced expression of Runx1 in Morton, Columbia University, New York), En1 (1:1,000; a gift from A. Joyner, the ventral spinal cord of chick embryos results in a promotion Skirball Institute, New York), Lhx3/4 (1:3,000; a gift from S. Pfaff, The Salk Institute of general motor neuron characteristics (e.g., Hb9 and Isl1 for Biological Studies, La Jolla, CA), Pax2 (1:100; Zymed Laboratories), RALDH2 expression) and a decreased expression of interneuron genes. (1:1,500; a gift from P. McCaffery, University of Massachusetts Medical School, Waltham, MA), and Phox2b (1:1,000; a gift from J. F. Brunet, Ecole Normale Because our loss-of-function studies show that Runx1 is dis- Superieure, Paris, France); goat polyclonals against ␤-gal (1:5,000; Biogenesis) and pensable for motor neuron generation, it seems unlikely that ChAT (1:100; Chemicon); sheep polyclonal against Chx10 (1:500; Chemicon); and the decrease in interneuron-specific gene expression caused by guinea pig polyclonal against Isl1/2 (1:5,000; a gift from S. Pfaff). All images were exogenous Runx1 was the result of an interference created by captured by using a DVC black and white camera mounted on an Axioskop the enhanced activation of motor neuron programs. fluorescence microscope (Zeiss). Runx1 is a dual-function transcription factor that can acti- vate or repress transcription in a context-dependent manner In Ovo Electroporations. For details, see SI Materials and Methods. (5). It is therefore possible that Runx1 represses the expression ϩ of specific interneuron genes within developing motor neu- Retrograde Axonal Labeling. E15.5 Runx1lacZ/ embryos were collected into rons. This possibility is in agreement with our demonstration ice-cold PBS, and several muscles, including the anterior trapezius, deltoideus, that Runx1/ETO, a dedicated transcriptional repressor, has the and pectoralis, were injected with a solution of rhodamine-conjugated dex- tran (molecular weight, 3,000; Molecular Probes), as described (3, 23). Embryos same phenotypic effect as full-length Runx1 when ectopically were incubated for5hat30°C in oxygenated PBS, followed by fixation, expressed in the neural tube of chick embryos. Taken together, embedding in OCT compound, cryosectioning, and immunohistochemistry. these findings suggest that Runx1 is involved in mechanisms that suppress interneuron differentiation programs and con- ACKNOWLEDGMENTS. We thank E. Stamateris, Z. Dong, L. Liu, Y. Tang, T. tribute to the stabilization of selected motor neuron fates. Basmacioglu, and M. Bouchard-Levasseur for invaluable assistance; Drs. N. Speck We have thus far found no evidence that the altered gene (Dartmouth Medical School) and M. Yanagisawa (University of Texas Southwest- ern Medical Center) for mouse lines; and Drs. T. Jessell, P. McCaffery, and S. Pfaff expression profile associated with Runx1 inactivation in developing for antibodies. This research was funded by Canadian Institutes of Health Re- motor neurons causes a complete conversion of motor neurons to search (CIHR) Neuromuscular Research Partnership Grants MOP-42479 (to S.S.) an interneuron fate. Throughout embryogenesis, Runx1 mutant and MOP-775556/IG-74068 (CIHR) (to A.K.) and grants from the Medical Research embryos do not exhibit any loss of the spinal neurons where Runx1 Council (U.K.) and European Union Framework Programme VI integrated project EuroStemCell (to A.M.). A.F. and N.S. were supported by a Montreal Neurological would be expressed had it not been inactivated, suggesting that Institute J. Timmins Costello Fellowship and a Government of Canada Student- those cells can receive appropriate trophic support. The lack of ship, respectively. A.K. is an EJLB Scholar, and S.S. is a Chercheur National of the increased cell death also suggests that it is unlikely that Runx1- Fonds de la Recherche en Sante du Quebec.

Stifani et al. PNAS ͉ April 29, 2008 ͉ vol. 105 ͉ no. 17 ͉ 6455 Downloaded by guest on October 1, 2021 1. Jessell TM (2000) Neuronal specification in the spinal cord: Inductive signals and 17. Dasen JS, Tice BC, Brenner-Morton S, Jessell TM (2005) A Hox regulatory network transcriptional codes. Nat Rev Genet 1:20–29. establishes motor neuron pool identity and target-muscle connectivity. Cell 123:477– 2. Shirasaki R, Pfaff SL (2002) Transcriptional codes and the control of neuronal identity. 491. Annu Rev Neurosci 25:251–281. 18. Simeone A, Daga A, Calabi F (1995) Expression of runt in the mouse embryo. Dev Dyn 3. Thaler J, et al. (1999) Active suppression of interneuron programs within developing 203:61–70. motor neurons revealed by analysis of homedomain factor HB9. Neuron 23:675–687. 19. North T, et al. (1999) Cbfa2 is required for the formation of intra-aortic hematopoietic 4. Arber S, et al. (1999) Requirement for the gene Hb9 in the consolidation of clusters. Development 126:2563–2575. motor neuron identity. Neuron 23:659–674. 20. Sharma K, et al. (1998) LIM homeodomain factors Lhx3 and Lhx4 assign subtype 5. Cameron ER, Neil JC (2004) The Runx genes: Lineage-specific oncogenes and tumor identities for motor neurons. Cell 95:817–828. suppressors. Oncogene 24:4308–4314. 21. Hirsch MR, Glover JC, Dufour HD, Brunet JF, Goridis C (2007) Forced expression of 6. Duffy JB, Kania MA, Gergen JP (1991) Expression and function of the Drosophila gene Phox2b homeodomain transcription factor induces a branchio-visceromotor axonal runt in early stages of neural development. Development 113:1223–1230. phenotype. Dev Biol 303:687–702. 7. Dormand EL, Brand AH (1998) Runt determines cell fate in the Drosophila embryonic 22. Snider WD, Palavali V (1990) Early axon and dendritic outgrowth of spinal accessory CNS. Development 125:1659–1667. motor neurons studied with DiI in fixed tissues. J Comp Neurol 297:227–238. 8. Kaminker JS, Canon J, Salecker I, Banerjee U (2002) Control of photoreceptor axon 23. Kania A, Jessell TM (2003) Topographic motor projections in the limb imposed by LIM target choice by transcriptional repression of Runt. Nat Neurosci 5:746–750. homeodomain protein regulation of ephrin-A:EphA interactions. Neuron 38:581–596. 9. Chen AI, de Nooij J, Jessell TM (2006) Graded activity of transcription factor Runx3 24. Samokhvalov IM, et al. (2006) Multifunctional reversible knockout/reporter system specifies the laminar termination pattern of sensory axons in the developing spinal enabling fully functional reconstitution of the AML1/Runx1 and rescue of he- cord. Neuron 49:395–408. matopoiesis. Genesis 44:115–121. 10. Marmigere F, et al. (2006) The Runx1/AML1 transcription factor selectively regulates 25. Kisanuki YY, et al. (2001) Tie2-Cre transgenic mice: A new model for endothelial development and survival of TrkA nociceptive sensory neurons. Nat Neurosci 9:180– cell-lineage analysis in vivo. Dev Biol 230:230–242. 187. 26. Ward NL, Putoczki T, Mearow K, Ivanco TL, Dumont DJ (2005) Vascular-specific growth 11. Inoue K, et al. (2002) Runx3 controls the axonal projection of proprioceptive dorsal root factor angiopietin 1 is involved in the organization of neuronal processes. J Comp ganglion neurons. Nat Neurosci 5:946–954. Neurol 482:244–256. 12. Levanon D, et al. (2002) The Runx3 transcription factor regulates development and 27. Burrill JD, Moran L, Goulding MD, Sauressig H (1997) PAX2 is expressed in multiple survival of TrkC dorsal root ganglia neurons. EMBO J 21:3454–3463. spinal cord interneurons, including a population of EN1ϩ interneurons that require 13. Theriault FM, Roy P, Stifani S (2004) AML1/RUNX1 is important for the development of PAX6 for their development. Development 124:4493–4503. hindbrain cholinergic branchiovisceral motor neurons and selected cranial sensory 28. Hamburger V, Hamilton HL (1951) A series of normal stages in the development of the neurons. Proc Natl Acad Sci USA 101:10343–10348. chick embryo. J Morphol 88:49–92. 14. Chen CL, et al. (2006) Runx1 determines nociceptive sensory neuron phenotype and is 29. Theriault FM, et al. (2005) Role for Runx1 in the proliferation and neuronal differen- required for thermal and neuropathic pain. Neuron 49:365–377. tiation of selected progenitor cells in the mammalian nervous system. J Neurosci 15. Kramer I, et al. (2006) A role for Runx1 transcription factor signaling in dorsal root 25:2050–2061. ganglion sensory neuron diversification. Neuron 49:379–393. 30. Wang Q, et al. (1996) Disruption of the Cbfa2 gene causes necrosis and hemorrhaging 16. Yoshikawa M, et al. (2007) Runx1 selectively regulates cell fate specification and axonal in the central nervous system and blocks definitive hematopoiesis. Proc Natl Acad Sci projections of dorsal root ganglion neurons. Dev Biol 303:663–674. USA 93:3444–3449.

6456 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0711299105 Stifani et al. Downloaded by guest on October 1, 2021