- Email: [email protected]
Expression of the helix-loop-helix factor, Hes3, during embryo development suggests a role in early midbrain-hindbrain patterning
Mechanisms of Development 62 (1997) 227–237
Expression of the helix-loop-helix factor, Hes3, during embryo development suggests a role in early midbrain-hindbrain patterning Corrinne G. Lobe* Institute for Molecular Biology, LSB 425, McMaster University, 1280 Main Street W., Hamilton, ON, L8S 4K1, Canada Received 8 July 1996; accepted 23 January 1997
Abstract The Hes gene family members are mammalian homologues of the Drosophila hairy and Enhancer of split genes. hairy and Enhancer of split function in both segmentation and in the Notch neurogenic pathway during Drosophila embryo development. Previous expression data suggested a conserved role for the Hes genes in the Notch signalling pathway, but not in segmentation. Here, Hes3 expression during mouse embryogenesis is described. During early development of the central nervous system, Hes3 is expressed specifically in the region of the midbrain/hindbrain boundary, and in rhombomeres 2, 4, 6 and 7. This pattern suggests that Hes3 may have a conserved role as a segmentation gene. Later in development, Hes3 is co-expressed with other neurogenic gene homologues in the developing central nervous system and epithelial cells undergoing mesenchyme induction. 1997 Elsevier Science Ireland Ltd. Keywords: Neurogenic genes; Segmentation; Midbrain/hindbrain; Rhombomere
1. Introduction Drosophila hairy is a member of the pair-rule class of genes and functions both in early segmentation and in the neurogenic pathway. In segmentation, hairy (h) represses transcription of another pair-rule gene, fushi tarazu (ftz) (Chenn and McConnell, 1995; Kuhn et al., 1995; Liem et al., 1995). This, in part, establishes the expression patterns of successive pair-rule and segment polarity genes that are required for the division of the embryo into the characteristic 14 segments along the antero-posterior (A/P) axis. Hairy, together with the Enhancer of split (E(spl)) proteins, also functions in the neurogenic pathway by repressing transcription of two proneural genes, achaete and scute (Van Doren et al., 1994). The repression is accomplished through an N box DNA element located between the achaete and scute genes. This activity contributes to the patterning of proneural clusters (Skeath and Carroll, 1992) and the selection of cells within the proneural cluster to follow either a neural or epidermal fate (Jennings et al., 1994). * Corresponding author. Tel.: +1 905 5259140, ext. 27335; fax: +1 905 5212955; e-mail: [email protected]
The Hes genes are a family of mammalian genes that encode basic helix-loop-helix transcription factors (Sasai et al., 1992) that are homologues of the h and E(spl) genes of Drosophila (Klambt et al., 1989; Rushlow et al., 1989; Lecourtois and Schweisguth, 1995; Oka et al., 1995). Like their invertebrate counterparts, three of the Hes-encoded proteins (Hes1, Hes2 and Hes5) bind to N boxes, which have the consensus sequence CACNAG, and thereby repress transcription from a target promoter (Sasai et al., 1992; Ishibashi et al., 1993; Akazawa et al., 1992). Transcription inhibition through N box binding can occur by two different mechanisms: the first requires a conserved WRPW motif at the C-terminus. This motif is used for interaction with the Groucho protein, which acts as a transcriptional co-repressor (Paroush et al., 1994). The second mechanism is through the loosely conserved ‘Orange’ domain, which may confer specific interactions with Ebox binding basic helix-loop-helix proteins (Heitzler et al., 1996). Both of these mechanisms require a functional bHLH moiety for DNA binding. When assayed in vitro, the Hes proteins can also inhibit transcriptional activation directed from E boxes (CANNTG), independent of N box binding, even though only Hes2 appears capable of binding directly to the E box
0925-4773/97/$17.00 1997 Elsevier Science Ireland Ltd. All rights reserved PII S0925-4773 (97 )0 0665-5
228
C.G. Lobe / Mechanisms of Development 62 (1997) 227–237
element (Akazawa et al., 1992; Sasai et al., 1992; Ishibashi et al., 1993). Repression of transcription from E box elements by Hes proteins might occur through inhibition of certain other bHLH proteins that do bind to E boxes, namely E47, MyoD, Myf5 and Mash1 (Sasai et al., 1992). Hes proteins may form non-functional heterodimers with the E box-binding proteins, thus interfering with their ability to activate transcription (Bailey and Posakony, 1995). It is not yet clear whether this mechanism of transcriptional regulation is utilized in vivo (Van Doren et al., 1994). One of the Hes proteins, Hes3, lacks the basic domain required for DNA binding and does not bind to N or E boxes (Sasai et al., 1992). It does, however, possess the helix-loop-helix domain required for dimerization; thus, Hes3 is similar to the Id proteins (Bailey and Posakony, 1995) and to Drosophila Extramacrochaetae (Emc) (Ellis et al., 1990). Unlike Id and Emc, Hes3 possesses the Orange domain and the C-terminal WRPW motif required for interaction with the Groucho co-repressor. Hes3 is not able to inhibit transcription through N boxes, as the other Hes proteins can, but efficiently represses transcriptional activation from E boxes and inhibits the transcriptional activation by E47 through the E box in in vitro assays (Sasai et al., 1992). Therefore, Hes3, at least, seems able to repress transcription through E box elements indirectly, by dimerization with the E box-binding proteins. This also implies that Hes3 can effect only one part of the transcriptional repression that the other Hes proteins are capable of. In mice, gain- and loss-of-function experiments using Hes1 point to a role for the Hes genes in regulating neural differentiation, rather than determining an epidermal versus neural fate: Hes1 expression directed by a viral vector in cells of the central nervous system prevented neural and glial differentiation (Ishibashi et al., 1994). Conversely, a targeted gene inactivation of Hes1 lead to a severe phenotype in the central nervous system, apparently due to premature differentiation of neurons (Ishibashi et al., 1995). Therefore, Hes1 seems to be involved in delaying differentiation of neuronal cells, possibly through inhibition of Mash1, rather than in selection of a non-neuronal cell fate. The only suggestion to date of a conserved role for Hes genes in segmentation comes from expression of Hes1 during somite formation (Sasai et al., 1992). Other expression reported for the Hes genes does not indicate a function in segmental patterning of the developing embryo: Hes1 is expressed in many tissues, mainly epithelial tissues undergoing mesenchymal induction (Sasai et al., 1992); Hes5 is expressed throughout the ventricular zone of the central nervous system (Takebayashi et al., 1995); Hes3 was detected in cells of adult cerebellum, but not in embryos (Sasai et al., 1992). While examining the expression of mouse homologues of the neurogenic genes (mNGs), I found that Hes3 is also expressed during embryogenesis. Its initial expression in the developing central nervous system suggests a role in
early midbrain and hindbrain patterning along the anteroposterior axis. Later expression occurs in epithelial tissues and overlaps the other mNGs, pointing to a later role in regulating cell differentiation.
2. Results 2.1. Hes3 expression in primitive ectoderm and the midhindbrain region during neurulation (7–8.5 dpc) Gene expression in embryos from 7 to 10 dpc was analyzed by whole mount in situ hybridization (Wilkinson and Nieto, 1993). Hes3 transcripts were initially detected in 7– 7.5 dpc embryos in the ectoderm adjacent to the primitive streak, up to and slightly ahead of the node (Fig. 1A; arrowheads denote the position of the node). As the headfold formed, the anterior boundary of this signal became more sharply defined in the region of the prospective hindbrain. Expression was observed in the neural plate ahead of the node and in ectoderm lateral to the primitive streak (Fig. 1B). By comparison, Hes1 at this stage was expressed throughout the neural plate rostral to the node and in a line of cells forming a semi-circle around the caudal side of the node (Fig. 1C,D). By 8.5 dpc, strong signals for the Hes3 transcript were localized to the region of the mesencephalic/metencephalic boundary, two bands in the hindbrain and throughout the open neural plate except for the floor plate (Fig. 1E,F). Caudal to the open neural plate, the ectoderm on either side of the primitive streak also continued to express Hes3 (Fig. 1E,F, small arrows). In FVB/N inbred mice, the two bands of expression in the hindbrain were distinct at the 4–5-somite stage and persisted until approximately the 7-somite stage (Theiler, 1989). However, in the outbred CD1 strain, we found that Hes3 expression in the hindbrain continued past the 7somite stage and spinal cord expression was initiated earlier, at 8.5 dpc compared to 9.5 dpc, than in FVB/N mice. We utilized the persistent hindbrain expression in CD1 mice to compare expression to Krox20. Krox20 is expressed in rhombomere (r) 3 beginning at the 0–4somite stage, and r5 at 4–7 somites (Wilkinson et al., 1989). Its expression there is required for the formation of r3 and r5 (Swiatek and Gridley, 1993). CD1 embryos were stained either for Hes3 (Fig. 2A,D,G,J), Krox20 (Fig. 2C,F,I,K) or both transcripts (Fig. 2B,E,H). This clearly demonstrated that Hes3 is expressed strongest in r1, weakly in r2, r4, r6 and r7, and is off in r3 and r5 at the same time that Krox20 is expressed only in r3 and r5. The alternating pattern of Hes3 and Krox20 expression in the hindbrain seemed to develop simultaneously (Fig. 2A–C, arrowhead). It was therefore not possible to determine if the products of one of the genes may be regulating the other. The expression in the prospective mid/hindbrain region
C.G. Lobe / Mechanisms of Development 62 (1997) 227–237
229
Fig. 1. Expression of Hes3 and Hes1 in 7.0–8.5 dpc embryos, analyzed by whole mount in situ hybridization with riboprobes. (A) Expression of Hes3 is initially detected in the embryonic ectoderm, just ahead of the node (arrowhead) and more caudally flanking the primitive streak. (B) After headfold formation, Hes3 signal continues in the ectoderm but extends further rostrally into the base of the headfold. (C) Hes1 expression in headfold stage embryos occurs in the neural plate (np) and the node region (arrowhead). (D) Ventral view of the embryo in (C) to show Hes1 signal in a semicircle around the caudal side of the node region (arrowhead). (E) At 8.5 dpc, Hes3 transcripts are detected dorsally in the mesencephalon and metencephalon (mes/met) and in two bands of the hindbrain neurectoderm which correspond to r4, r6 and r7. Prominent expression also continues in the neural plate and ectoderm adjacent to the primitive streak (ps; arrow). fb, forebrain; ht, heart. (F) Dorsal view of the embryo in (E), showing expression in the mesencephalon, metencephalon and r4, r6 and r7, in the neural plate (arrowhead) and ectoderm flanking the primitive streak (small arrows). Numbers designate rhombomere segments.
is also intriguing, particularly with respect to other mNGs at the same stage: one of the groucho-related genes, Grg3, is expressed in an overlapping domain, in the mid- and hindbrain, and therefore could potentially interact with Hes3 there (Leon and Lobe, 1996). Grg4, on the other hand, is expressed in two diffuse bands, in the rostral mesencephalon and directly behind the mesencephalic/ metencephalic boundary, just anterior and posterior to the region of prominent Hes3 expression (Fig. 3A,B).
Hes1 transcripts are located with Grg4 in the rostral mesencephalon, just anterior to Hes3 (Fig. 3D). These patterns further suggest the mNGs may play a role in the early patterning of the brain. A gene which was demonstrated to be required for patterning of the mid-hindbrain region is En1. Its expression begins at around or just before the 1-somite stage in the presumptive mesencephalon and metencephalon (Davis and Joyner, 1988; Rowitch and McMahon, 1995) overlapping Hes3 expression. However,
230
C.G. Lobe / Mechanisms of Development 62 (1997) 227–237
Fig. 2. Comparison of Hes3 and Krox20 expression in CD1 embryos. Embryos were hybridized either with only Hes3 (A,D,G,J), with Krox20 (C,F,I,K) or with both Hes3 and Krox20 (B,E,H). (A–C) Embryos at 8.0 dpc (0–4 somite) display a gap in Hes3 (A) hindbrain expression as Krox20 (C) is first expressed in r3. The embryo in (A) has turned due to handling during hybridization. (D–I) In 8.5 dpc embryos (5–8 somites) Hes3 and Krox20 expression are reciprocal in the hindbrain, with Hes3 expression in r1, r2, r4, r6 and r7 and Krox20 in r3 and r5. (D–F) Dorsal views; (G–I) lateral view. In (D), the small arrow points to ectoderm flanking the primitive streak and the arrowhead points to expression in the open neural plate. (J,K) At 9.0 dpc, a gap in Hes3 expression persists in r3 and r5 (J) at the same time that Krox20 is specifically expressed there (K). In all panels, small arrowheads mark the position of r3 and r5.
En1 expression extends to the ventral-most part of the neural fold, whereas Hes3 is restricted to the dorsal part (Fig. 3A,C). 2.2. Expression during early brain development (9–10.5 dpc) Following embryo turning at around 9.0 dpc, Hes3 signal was initially still localized to the dorsal part of the midhindbrain boundary and the open neural tube in the tail region in FVB/N embryos (Fig. 4A). However, in slightly older embryos, the signal extended caudally through the entire length of the neural tube and rostrally in scattered cells in the dorsal midbrain (Fig. 4B) and later in the forebrain (Fig. 4C). Sections of embryos after whole mount in situ hybridization revealed the Hes3 signal was localized to a dorso-lateral region of the neural tube (Fig. 4G,H, arrows and arrowheads). Mash1 expression was first detected in embryos after turning and was generally found in adjacent parts of the neural tube to Hes3. For instance, Mash1 transcripts were initially present in the dorsal part of the mesencephalon (Fig. 4D) (Guillemot and Joyner, 1993). Later expression was observed in the mesencephalon and spinal cord in a domain more dorsal than Hes3 (Fig. 4E,F,I), as well as in cells flanking the floor plate (Fig. 4I) (Lo et al., 1991; Guillemot and Joyner, 1993). Mash1 was not initially present in the cerebello-mesencephalic flexure, where Hes3 was strongest (Fig. 4B,E, arrows) (Guillemot and Joyner, 1993). By 10.5 dpc, the signal for Hes3 transcripts in the cen-
tral nervous system became restricted to a band of cells just rostral of the isthmus, in the region of the prospective inferior colliculus (Fig. 4J,K) and the open neural plate in the tail. By contrast, a strong En1 signal was detected in cells on both sides of the flexure that marks the boundary between the midbrain and hindbrain (Fig. 4L). 2.3. Hes3 is expressed more extensively in the nervous system during organogenesis (11–16.5 dpc) Embryos from 11.5 dpc to 16.5 dpc were sectioned and then subjected to hybridization with digoxygenin-labeled RNA probes to analyze gene expression. At 11.5 and 12.5 dpc, Hes3 expression was detected throughout the length of the central nervous system and in the peripheral nervous system (Fig. 5A–C). In the brain of 11.5 dpc embryos, transcripts were localized to the roof of the telencephalon (Fig. 5A, te) and the lateral walls and roof of the diencephalon. This expression overlaps that of the reported Mash1 expression in the brain (Lo et al., 1991; Guillemot and Joyner, 1993). There was also a weak Hes3 signal in the floor of the diencephalon. At the mesencephelon, strongest expression was in the lateral walls, just above the sulcus limitans, in the ventricular zone next to the lumen. Expression was also observed in the roof of the caudal region of the midbrain and isthmus. A strong signal in a few cells of the floorplate was observed in the hindbrain and spinal cord (Fig. 5A, fp). This signal is complementary to that of Mash1 in cells on either side of the floor plate (Lo et al., 1991; Guillemot and Joyner, 1993).
C.G. Lobe / Mechanisms of Development 62 (1997) 227–237
231
Fig. 3. Expression in 8.5 dpc (7–8 somite stage) embryos. (A) Hes3 expression persists in the dorsal region of the mesencephalon/metencephalon (mes/ met) boundary. A strong signal in the ectoderm adjacent to the primitive streak (ps) and open neural plate (np) is also present, but the expression in r4, r6 and r7 is quite low at this stage in FVB/N mice. fb, forebrain; ht, heart. (B) Grg4, a groucho homologue is expressed in the mesencephalon and metencephalon (mes, met), with highest levels flanking the Hes3 expression domain (arrows). Grg4 is also expressed in the forebrain (fb), somites, and presomitic mesoderm (ps, primitive streak). (C) En1 expression also occurs in the region of the mesencephalon/metencephalon boundary (mes/met) but extends ventrally to the floor plate. fb, forebrain; mes, mesencephalon; met, metencephalon; np, neural plate; ps, primitive streak; ht, heart. (D) Hes1 is expressed in the mesencephalon just rostral to Hes3 expression (arrow), and in the forebrain (fb), somites and a diffuse band in the primitive streak region (ps).
Expression in the ventricular zone next to the lumen, dorsal to the sulcus limitans, also occurred at the level of the otic vesicle and continued down through the caudal part of the hindbrain and the spinal cord (arrowhead in Fig. 5A,B). In the spinal cord, cells immediately lining the lumen in the ventricular zone and in the ventral horn of the marginal zone also stained for Hes3 transcripts at 11.5 dpc. Comparison of Hes3 to Mash1 expression in the spinal cord at 12.5 dpc showed that Hes3 signal was in the dorsal-most part of the spinal cord and the floor plate where Mash1 was absent (Fig. 5B,C; fp, floorplate). Expression of the two genes did overlap in the dorsal region of the ventricular zone. However, the Hes3 signal was confined to cells adjacent to the lumen, whereas Mash1 signal extended out to where the ventricular zone meets the marginal zone (Fig. 5C). Later in central nervous system development, at 14.5 and 16.5 dpc, Hes3 transcripts were located in the ventricular zone of the cerebral cortex (Fig. 5D, cc; Fig. 6A, vz). In the midbrain, Hes3 signal was prominent dorsally in the tectal neuroepithelium (strongest in the inferior colliculus;
Fig. 5D and 6C, ic) and ventrally in the post tegmental neuroepithelium (Fig. 5D, pn). Expression of Hes3 in the external granule layer of the cerebellum was first apparent as it formed at 14.5 dpc (Fig. 5D, eg), and continued through 16.5 dpc (Fig. 6C, eg). Expression also persisted in the medulla and spinal cord in the ventricular zone adjacent to the lumen (arrowhead in Fig. 5D; Fig. 5F, dm). Other CNS expression was detected in the lens and the pigmented epithelium layer of the developing eye (Fig. 5A, ey) and the neuroepithelium of the otic vesicle and cochlea (Fig. 5D, co). Expression in the nasosensory epithelium was first apparent at 11.5 dpc (Fig. 5A) and continued at later stages (Fig. 5B and 6A; ne). In the peripheral nervous system, cells of the cranial and dorsal root ganglia expressed Hes3 (cn, tg, af, and ov in Fig. 5A,D; drg in Fig. 5G). These tissues are also positive for Notch1, Grg3, Grg4, and Hes1 expression (Weinmaster et al., 1991; Franco del Amo et al., 1992; Reaume et al., 1992; Sasai et al., 1992; Weinmaster et al., 1992; Koop et al., 1996; Leon and Lobe, 1996). Apart from the nervous system, signal was detected in
232
C.G. Lobe / Mechanisms of Development 62 (1997) 227–237
Fig. 4. Expression analysis in 9.0–10.0 dpc embryos. (A) At 9.0 dpc, Hes3 is weakly expressed in the mesencephalon and rostral part of the metencephalon (arrowhead), and in the dorsal part of the open neural tube in the tail region (arrow). (B) By 9.5 dpc, Hes3 expression in the neural tube is also seen more rostrally in the mesencephalon (*; ms, mesencephalon; mt, metencephalon) with the strongest expression in the cerebello-mesencephalic flexure (arrow). Expression is detected in the spinal cord (sc), appearing in a rostral to caudal gradient. (C) A slightly older embryo than that shown in (B) displays stronger expression throughout the brain. (D) Mash1 expression occurs in the dorsal region of the mesencephalon, at the same stage as Hes3, shown in (A). (E) Mash1 expression at 9.5 dpc is found throughout the dorsal part of the midbrain, hindbrain and spinal cord, except for the cerebellomesencephalic flexure (arrow), where Hes3 expression is strongest (see B). (F) Expression of Mash1 slightly later than the embryo shown in (E). Mash1 signal is stronger throughout the CNS and appears similar to Hes3 expression at the same stage (shown in C; arrow, cerebello-mesencephalic flexure). (G,H) Cross-sections of the embryo shown in (C) demonstrate that the Hes3 signal is in a dorsal-lateral part of the brain (arrow) and spinal cord (arrowhead; di, diencephalon; met, metencephalon; my, myelencephalon; np, neural plate; ht, heart). (I) Cross section of the embryo shown in (F) reveals Mash1 expression is more dorsal than Hes3 in the central nervous system (arrow; my, myelencephalon; ht, heart). (J) Between 9.5 and 10.5 dpc, the Hes3 signal becomes restricted to the mesencephalon, appearing as two bands, one just rostral of the mes/met boundary and one slightly more anterior. (K) In a 10.0 dpc embryo stained for Hes3 expression the strongest signal occurs rostral to the isthmus, in the inferior colliculus (ic; mes, mesencephalon; met, metencephalon). (L) At 10.0 dpc, a strong En1 signal is present in both the inferior colliculus (ic) and metencephalon (met) neighboring the mesencephalic isthmus. (mes, mesencephalon).
epithelial structures, in a similar pattern to that reported for the Notch1 and Grg genes (Weinmaster et al., 1991; Dehni et al., 1995; Koop et al., 1996; Leon and Lobe, 1996). These included the basal layer of the skin (bs in Fig. 5A and 6B), the hair and whisker follicles (Fig. 5D and 6A,B; wh), the adrenal medulla and cortex of the kidney (Fig. 5G, ag, ki), endoderm lining of the gut (Fig. 5G, gu), salivary gland, lung (Fig. 5G, lu), thymic and tooth bud epithelia (Fig. 5D and 6A, tb). The roof of Rathke’s pocket, adjacent to the infundibulum from the diencephalon, was also positive for Hes3 signal (Fig. 5E, pi). This part of Rathke’s pocket will form the epithelial part of the pituitary, the pars intermedia, whereas the more ventral mesenchymal portion will form the glandular part of the pituitary, the pars distalis (Rugh, 1990). We also hybridized the Hes3 RNA probe with P21 brain sections to compare our results at that stage to the previous report of Hes3 expression in the cerebellum (Sasai et al., 1992). We observed expression restricted to the Purkinje cells of the cerebellum (data not shown) as described previously (Sasai et al., 1992).
3. Discussion Hes3 expression was previously reported for the adult cerebellum, but was not detected during embryo development (Sasai et al., 1992). The analysis here reveals that Hes3 is expressed during mouse embryogenesis, from 7 dpc onward, in the developing nervous system and epithelial tissues. Comparison to the expression patterns of previously characterized developmental control genes raises the possibility that Hes3 is involved both in early patterning along the A/P axis and in a signalling pathway with the other mNGs. The early expression of Hes3 and Hes1 around the primitive streak and node borders the area where cells delaminate from the ectoderm and migrate through the streak. This is reminiscent of the pattern of gene expression in the morphogenetic furrow of the Drosophila eye, in which the HLH factor, Emc, is expressed in all cells ahead of the progressing furrow and hairy is expressed in a line of cells just preceding the furrow (Brown et al., 1995; Heberlein and Moses, 1995; Heberlein et al., 1995). Expression
C.G. Lobe / Mechanisms of Development 62 (1997) 227–237
233
Fig. 5. Hes3 expression detected in sections of 11.5–14.5 dpc embryos. (A) Cross-section of 11.5 dpc embryo at the level of the forebrain and hindbrain. A schematic of the orientation of the section is shown to the left. Signals for Hes3 expression are in the nasosensory epithelia (ne), lining the olfactory pit (op), the dorsal ventricular zone of the telencephalon (te), pigment layer of the retina (ey), semilunar ganglia of the trigeminal nerve (tg), accousticofacialis ganglia VII and VIII (af), a few cells lining the otic vesicle (ov), cranial nerves (cn), floor plate (fp) and ventricular zone above the sulcus limitans (arrowhead) in the myelencephalon (my), and the basal layer of the skin (bs). BA, branchial arch. (B,C) Serial cross sections through the spinal cord of 12.5 dpc embryo hybridized with Hes3 (B) or Mash1 (C). Hes3 signal is strongest in the dorsal-most part of the spinal cord in cells of the ventricular zone lining the lumen (arrowhead) and a few cells of the floor plate (fp). Signal is also present in the dorsal root ganglia (drg). Mash1 signal is strongest throughout the dorsal ventricular zone of the spinal cord (bracketed). (D) Parasaggital section of the head, showing Hes3 signal in the whisker follicles (wh), nasosensory epithelia (ne), toothbud progenitor (tb), trigeminal ganglia (tg), cells lining the cochlea (co), the ventricular zone of the cerebral cortex (cc), inferior colliculus (ic), post-tegmental neuroepithelium (pn) and brainstem (arrowhead), and external granular layer (eg) of the prospective cerebellum (cb). tm, tegmentum; sc, superior colliculus. (E) A higher magnification showing Hes3 expression in the pars intermedia (pi) of the developing pituitary. pd, pars distalis. (F) Higher magnification of the region of the cerebellum from (B). Signal is present in the external granule layer (eg) of the cerebellum (cb), the choroid plexus (cp) and the dorsal medulla (dm). (G) Sagittal section of the trunk region. Signals corresponding to Hes3 expression occur in the lung epithelia (lu), liver (li), endodermal lining of the gut (gu), kidney (ki), adrenal gland (ag), and dorsal root ganglia (drg). ht, heart.
234
C.G. Lobe / Mechanisms of Development 62 (1997) 227–237
Fig. 6. Expression of Hes3 in 16.5 dpc embryo sections. (A) Sagittal section of the head. Hes3 signals correspond to whisker follicles (wh), epithelia of the toothbud progenitors (tb), nasosensory epithelia (ne), the ventricular zone (vz) of the cerebral cortex (cc), sensory cells lining the cochlea (co), and the external granule layer of the cerebellum (cb). mb, midbrain; to, tongue. (B) A higher magnification of the nose region, showing expression of Hes3 in the basal layer of the skin (bs) and the inner root sheath of the whisker follicles (wh). (C) Parasaggital section of the mesencephalon/metencephalon boundary. Expression of Hes3 occurs in the ventricular zone of the inferior colliculus, the external granule layer (eg) of the cerebellum (cb) and cells in the choroid plexus (cp).
of Hes3 and Hes1 around the node and primitive streak may represent part of a conserved regulatory pathway to organize morphogenetic changes and cell differentiation (Brown et al., 1995; Heberlein and Moses, 1995). The expression pattern of Hes3 during embryonic brain development suggests a role in early patterning of the midbrain and hindbrain region. Hes3 appears to mark cells at the mesencephalic/metencephalic border. Its expression is bracketed by Grg4, a groucho homologue, and Hes1. Other genes expressed around the mid/hindbrain boundary are Pax2, Wnt1, En1 and En2 (Wilkinson et al., 1987; Davis and Joyner, 1988; Rowitch and McMahon, 1995). Gene targeting of Wnt1 and En1 resulted in
loss of the midbrain and adjacent cerebellar component of the metencephalon, demonstrating their important role in patterning this part of the brain (McMahon and Bradley, 1990; Wurst et al., 1994). It was further demonstrated that the Wnt1 phenotype corresponded to loss of the En1expressing cells (McMahon et al., 1992). This is comparable to the interaction for the Wnt (wg) and En genes during Drosophila segmentation, in which maintenance of En expression requires Wnt (DiNardo et al., 1988; Martinez-Arias et al., 1988; Bejsovec and Martinez-Arias, 1991; Heemskerk et al., 1991). However, both in Drosophila and in mouse, Wnt1 expression was not required for the initiation of En1 expression. In Drosophila, initiation
C.G. Lobe / Mechanisms of Development 62 (1997) 227–237
of En and Wg expression patterns is determined by ftz, Eve, and prd, which in turn are regulated by hairy. Promoter analysis of the En2 gene revealed Pax binding sites (Song et al., 1996), and Pax2 expression in the mid/hindbrain precedes expression of Wnt1 and En1 (Rowitch and McMahon, 1995), indicating that the regulatory hierarchy between Pax, Wnt and En has also been conserved. Thus, it is possible that in the mouse, the Hes genes also have a conserved regulatory role upstream of Pax, En and Wnt genes. The early expression of Hes3 in the mid/hindbrain region is in accord with this speculation. A targeted gene inactivation of Hes3 may reveal whether it plays a role in early mid/hindbrain patterning through regulation of homologues of the Drosophila segmentation genes. The early expression of Hes3 in the hindbrain also suggests a role in segmentation. It is first expressed throughout the neural plate but its expression is lost, first in r3 and then in r5. This pattern occurs at approximately the same time that Krox20 expression appears in r3 and r5 and precedes the morphological appearance of rhombomeres. Inactivation of the Krox20 gene results in a loss of r3 and r5 confirming that Krox20 is required for formation of those two rhombomeres. The specific complementary expression of Hes3 in regions of the hindbrain where Krox20 is not expressed suggests that both Hes3 and Krox20 are required for correct rhombomere formation. It was not possible to discern if one of Hes3 or Krox20 appeared first and therefore whether one of them could be regulating expression of the other. In Drosophila, h and E(spl) gene products negatively regulate achaete and scute (Ohsako et al., 1994; Van Doren et al., 1994). Interestingly, the strongest Hes3 expression in the neural tube lies in the cerebello-mesencephalic flexure, which corresponds to the region of the neural tube where Mash1 is not initially present (Guillemot and Joyner, 1993). Likewise, Hes3 expression in the spinal cord lies adjacent to Mash1 initially in the dorsolateral spinal cord and floorplate, and later is in the ventricular zone within the domain of Mash1 expression. This is again suggestive of a conserved regulatory role for Hes3, in this case in the negative regulation of the Achaete-scute homologue, Mash1. Hes3 expression at mid- to late-embryogenesis occurs in many types of epithelial cells undergoing mesenchymal induction, a common theme for the neurogenic gene homologues, Notch1, Grg1, Grg3, Grg4 and Hes1 (Weinmaster et al., 1991; Dehni et al., 1995; Koop et al., 1996; Leon and Lobe, 1996). Therefore, Hes3 may act with the other mNGs to regulate cell proliferation and differentiation in developing epithelial tissues. The Drosophila neurogenic genes are also expressed during epithelial/ mesenchymal interactions, where they seem to designate epithelial cell fates (Tepass and Hartenstein, 1995). Hes3 is unique among the Hes proteins in that it has the helix-loop-helix, Orange, and WRPW domains but lacks the basic domain and does not bind to DNA (Sasai et al.,
235
1992; Dawson et al., 1995). Therefore, it probably functions similarly, but not identically, to the other Hes proteins. Its expression pattern is also distinct from the other Hes genes, in that it displays a definitive pattern along the antero-posterior axis. These results suggest that Hes3 has a role not only in regulating cell differentiation with the other Hes, Grg and Notch genes, but also an early role in patterning the midbrain and hindbrain.
4. Experimental procedures 4.1. RNA probes cDNA clones for rat Hes3 and Hes1 in the pSV2-CMV expression vector were kindly provided by R. Kageyama (Sasai et al., 1992). The EcoRI inserts were subcloned into pBluescript-KS (Stratagene). The Hes3 clone was linearized with BamHI and transcribed with T3 RNA polymerase to generate a 985 bp antisense RNA probe. The Hes1 template was generated by linearizing with BstEII and transcribed with T7 RNA polymerase. The Grg4 sequence used for an RNA template was a 459 bp sequence that lies within the WD40 domain (Koop et al., 1996). The En1 and Mash1 cDNAs were kindly provided by Dr. A. Joyner, and the Krox20 cDNA by Dr. J. Rossant. RNA synthesis was directed using T7, T3 or SP6 RNA polymerase, including digoxygenin-labeled nucleotide mix as recommended by the supplier (Boehringer-Mannheim). Products were analyzed on a 1% agarose gel and approximately 1 mg/ml of the digoxygenin-labeled antisense RNA was used for hybridization of embryos and slides. 4.2. Whole-mount in situ hybridization of 7.5–10.5 dpc embryos FVB/N or CD1 mice, as indicated in the figures, were mated and the day of the plug was taken as 0.5 dpc. Embryos of 7.5–10.5 dpc were dissected into PBS and fixed overnight at 4°C in 4% paraformaldehyde in PBS. The following day they were passed through a graded methanol series and then stored in methanol at −20°C. The protocol for whole mount in situ hybridization of embryos was as described (Wilkinson and Nieto, 1993). After overnight hybridization, embryos were washed twice for 30 min at 70°C in solution 1 (50% formamide, 5 × SSC pH 4.5, 1% SDS), 10 min at 70°C in a 1:1 mix of solution 1 and solution 2 (0.5 M NaCl, 10 mM Tris–HCl pH 7.5, 0.1% Tween-20), three 5 min washes in solution 2, two 30 min incubations at 37°C in solution 2, a 5 min wash in solution 2, 5 min in solution 3 (50% formamide, 2 × SSC pH 4.5), two 30 min washes in solution 3 at 65°C, three 5 min washes in TBST (137 mM NaCl, 25 mM Tris– HCl pH 7.6, 3 mM KCl, 0.1% Tween-20) and a 90 min block in 10% serum/TBST. Embryos were then incubated
236
C.G. Lobe / Mechanisms of Development 62 (1997) 227–237
overnight at 4°C with alkaline phosphatase-coupled antidigoxygenin antibody (Boehringer-Mannheim). The antibody was pre-blocked for 90 min at 4°C with 13 dpc embryo powder and 1% serum in TBST. After overnight incubation with antibody, embryos were washed three times for 5 min and four times for 1 h in TBST, and then washed overnight in TBST. The following day embryos were washed three times for 10 min in NTMT (100 mM NaCl, 100 mM Tris–HCl pH 9.5, 50 mM MgCl2, 0.1% Tween-20, 2 mM levamisole) and then stained in 337.5 mg/ml NBT/175 mg/ml BCIP in NTMT. Photographs were taken using illuminated light on a Leica Wild M3C dissecting microscope. For sectioning after whole mount hybridization, samples were cryoprotected in 20% sucrose/PBS, frozen and sectioned at 10 mm on a Leitz cryostat at −20°C. Slides were directly mounted in 7% gelatin/50% glycerine and photographed on a Zeiss Axiophot microscope with DIC optics. 4.3. In situ hybridization of embryo and brain sections Embryos of 11.5–16.5 dpc were dissected into PBS and fixed overnight in 4% PFA/PBS at 4°C. Three week old post-natal brains were fixed by perfusion and stored in 4% PFA/PBS at 4°C overnight. After fixation, tissues were rinsed through a graded ethanol series, xylene, and embedded in paraffin. Tissues were sectioned at 20 mm. The slides were then hybridized and washed exactly as described (Wilkinson and Nieto, 1993) but without RNase treatment in the washes. The alkaline phosphatase staining reaction was allowed to proceed for 6 h to overnight. Following a rinse in PBS and water, slides were airdried, mounted in Permount and photographed on a Zeiss Axiophot with DIC optics.
Acknowledgements I thank Drs. J.R. Jacobs, A. Nagy and W. Wurst for useful discussions, R. Kageyama for the Hes3 cDNA, A. Joyner for the En1 and Mash1 probes and J. Rossant for the Krox20 probe. This work was supported by a grant from the Medical Research Council of Canada. CL is supported by a Scholarship from the Medical Research Council of Canada.
References Akazawa, C., Sasai, Y., Nakanishi, S. and Kageyama, R. (1992) J. Biol. Chem. 267, 21879–21885. Bailey, A.M. and Posakony, J.W. (1995) Genes Dev. 9, 2609–2622. Bejsovec, A. and Martinez-Arias, A. (1991) Development 113, 471–485. Brown, N.L., Sattler, C.A., Paddock, S.W. and Carroll, S.B. (1995) Cell 80, 879–887. Chenn, A. and McConnell, S.K. (1995) Cell 82, 631–641.
Davis, C.A. and Joyner, A.L. (1988) Genes Dev. 2, 1736–1744. Dawson, S.R., Turner, D.L., Weintraub, H. and Parkhurst, S.M. (1995) Mol. Cell. Biol. 15, 6923–6931. Dehni, G., Liu, Y., Husain, J. and Stifani, S. (1995) Mech. Dev. 53, 369– 381. DiNardo, S., Sher, E., Heemskerk-Jongens, J., Kassis, J.A. and O’Farrell, P.H. (1988) Nature 332, 604–609. Ellis, H.M., Spann, D.R. and Polakony, J.W. (1990) Cell 61, 27–38. Franco del Amo, F., Smith, D.E., Swiatek, P.J., Gendron-Maguire, M., Greenspan, R.J., McMahon, A.P. and Gridley, T. (1992) Development 115, 737–744. Guillemot, F. and Joyner, A.L. (1993) Mech. Dev. 42, 171–185. Heberlein, U. and Moses, K. (1995) Cell 81, 987–990. Heberlein, U., Singh, C.M., Luk, A.Y. and Donohoe, T.J. (1995) Nature 373, 709–711. Heemskerk, J., Dinardo, S., Kostriken, R. and O’Farrell, P.H. (1991) Nature 352, 404–410. Heitzler, P., Bourouis, M., Ruel, L., Carteret, C. and Simpson, P. (1996) Development 122, 161–171. Ishibashi, M., Sasai, Y., Nakanishi, S. and Kageyama, R. (1993) Eur. J. Biochem. 215, 645–652. Ishibashi, M., Moriyoshi, K., Sasai, Y., Shiota, K., Nakanishi, S. and Kageyama, R. (1994) EMBO J. 13, 1799–1805. Ishibashi, M., Ang, S.-L., Shiota, K., Nakanishi, S., Kageyama, R. and Guillemot, F. (1995) Genes Dev. 9, 3136–3148. Jennings, B., Preiss, A., Delidakis, C. and Bray, S. (1994) Development 120, 3537–3548. Klambt, C., Knust, E., Tietze, K. and Campos-Ortega, J.A. (1989) EMBO J. 8, 203–210. Koop, K.E., MacDonald, L.M. and Lobe, C.G. (1996) Mech. Dev. 59, 73–87. Kuhn, R., Schwenk, F., Aguet, M. and Rajewsky, K. (1995) Science 269, 1427–1429. Lecourtois, M. and Schweisguth, F. (1995) Genes Dev. 9, 2598–2608. Leon, C. and Lobe, C.G. (1996) Dev. Dyn. 208, 11–24. Liem, K.F., Tremml, G., Roelink, H. and Jessell, T.M. (1995) Cell 82, 969–979. Lo, L.-C., Johnson, J.E., Wuenschell, C.W., Saito, T. and Anderson, D.J. (1991) Genes Dev. 5, 1524–1537. Martinez-Arias, A., Baker, N.E. and Ingham, P.W. (1988) Development 103, 157–170. McMahon, A.P. and Bradley, A. (1990) Cell 62, 1073–1085. McMahon, A.P., Joyner, A.L., Bradley, A. and McMahon, J.A. (1992) Cell 69, 581–595. Ohsako, S., Hyer, J., Panganiban, G., Oliver, I. and Caudy, M. (1994) Genes Dev. 8, 2743–2755. Oka, C., Nakano, T., Wakeham, A., de la Pompa, J.L., Mori, C., Sakai, T., Okazaki, S., Kawaichi, M., Shiota, K., Mak, T.W. and Honjo, T. (1995) Development 121, 3291–3301. Paroush, Z., Finley, R.L.J., Kidd, T., Wainwright, S.M., Ingham, P.W., Brent, R. and Ish-Horowicz, D. (1994) Cell 79, 805–815. Reaume, A.G., Conlon, R.A., Zirngibl, R., Yamaguchi, T.P. and Rossant, J. (1992) Dev. Biol. 154, 377–387. Rowitch, D.H. and McMahon, A.P. (1995) Mech. Dev. 52, 3–8. Rugh, R. (1990) The Mouse. Its Reproduction and Development. Oxford University Press, Oxford. Rushlow, C.A., Hogan, A., Pinchin, S.M., Howe, K.R., Lardelli, M.T. and Ish-Horowicz, D. (1989) EMBO J. 8, 3095–3103. Sasai, Y., Kageyama, R., Tagawa, Y., Shigemoto, R. and Nakanishi, S. (1992) Genes Dev. 6, 2620–2634. Skeath, J.B. and Carroll, S.B. (1992) Development 114, 939–946. Swiatek, P.J. and Gridley, T. (1993) Genes Dev. 7, 2071–2083. Takebayashi, K., Akazawa, C., Nakanishi, S. and Kageyama, R. (1995) J. Biol. Chem. 270, 1342–1349. Tepass, U. and Hartenstein, V. (1995) Development 121, 393–405. Theiler, K. (ed.) (1989) The House Mouse. Development and Normal Stages from Fertilisation to 4 Weeks of Age. Springer, Berlin.
C.G. Lobe / Mechanisms of Development 62 (1997) 227–237 Van Doren, M., Bailey, A.M., Esnayra, J., Ede, K. and Posakony, J.W. (1994) Genes Dev. 8, 2729–2742. Weinmaster, G., Roberts, V.J. and Lemke, G. (1991) Development 113, 199–205. Weinmaster, G., Roberts, V.J. and Lemke, G. (1992) Development 116, 931–941. Wilkinson, D.G. and Nieto, M.A. (1993) Guide to techniques in mouse development. In Wasserman, P.M. and DePamphalis, M.L. (eds.),
237
Methods in Enzymology, Academic Press, San Diego, CA, Vol. 225, pp. 361–373. Wilkinson, D.G., Bailes, J.A. and McMahon, A.P. (1987) Cell 50, 79– 88. Wilkinson, D.G., Bhatt, S., Chavrier, P., Bravo, R. and Charnay, P. (1989) Nature 337, 461–464. Wurst, W., Auerbach, A. and Joyner, A. (1994) Development 120, 2065–2075.
Expression of the helix-loop-helix factor, Hes3, during embryo development suggests a role in early midbrain-hindbrain patterning Corrinne G. Lobe* Institute for Molecular Biology, LSB 425, McMaster University, 1280 Main Street W., Hamilton, ON, L8S 4K1, Canada Received 8 July 1996; accepted 23 January 1997
Abstract The Hes gene family members are mammalian homologues of the Drosophila hairy and Enhancer of split genes. hairy and Enhancer of split function in both segmentation and in the Notch neurogenic pathway during Drosophila embryo development. Previous expression data suggested a conserved role for the Hes genes in the Notch signalling pathway, but not in segmentation. Here, Hes3 expression during mouse embryogenesis is described. During early development of the central nervous system, Hes3 is expressed specifically in the region of the midbrain/hindbrain boundary, and in rhombomeres 2, 4, 6 and 7. This pattern suggests that Hes3 may have a conserved role as a segmentation gene. Later in development, Hes3 is co-expressed with other neurogenic gene homologues in the developing central nervous system and epithelial cells undergoing mesenchyme induction. 1997 Elsevier Science Ireland Ltd. Keywords: Neurogenic genes; Segmentation; Midbrain/hindbrain; Rhombomere
1. Introduction Drosophila hairy is a member of the pair-rule class of genes and functions both in early segmentation and in the neurogenic pathway. In segmentation, hairy (h) represses transcription of another pair-rule gene, fushi tarazu (ftz) (Chenn and McConnell, 1995; Kuhn et al., 1995; Liem et al., 1995). This, in part, establishes the expression patterns of successive pair-rule and segment polarity genes that are required for the division of the embryo into the characteristic 14 segments along the antero-posterior (A/P) axis. Hairy, together with the Enhancer of split (E(spl)) proteins, also functions in the neurogenic pathway by repressing transcription of two proneural genes, achaete and scute (Van Doren et al., 1994). The repression is accomplished through an N box DNA element located between the achaete and scute genes. This activity contributes to the patterning of proneural clusters (Skeath and Carroll, 1992) and the selection of cells within the proneural cluster to follow either a neural or epidermal fate (Jennings et al., 1994). * Corresponding author. Tel.: +1 905 5259140, ext. 27335; fax: +1 905 5212955; e-mail: [email protected]
The Hes genes are a family of mammalian genes that encode basic helix-loop-helix transcription factors (Sasai et al., 1992) that are homologues of the h and E(spl) genes of Drosophila (Klambt et al., 1989; Rushlow et al., 1989; Lecourtois and Schweisguth, 1995; Oka et al., 1995). Like their invertebrate counterparts, three of the Hes-encoded proteins (Hes1, Hes2 and Hes5) bind to N boxes, which have the consensus sequence CACNAG, and thereby repress transcription from a target promoter (Sasai et al., 1992; Ishibashi et al., 1993; Akazawa et al., 1992). Transcription inhibition through N box binding can occur by two different mechanisms: the first requires a conserved WRPW motif at the C-terminus. This motif is used for interaction with the Groucho protein, which acts as a transcriptional co-repressor (Paroush et al., 1994). The second mechanism is through the loosely conserved ‘Orange’ domain, which may confer specific interactions with Ebox binding basic helix-loop-helix proteins (Heitzler et al., 1996). Both of these mechanisms require a functional bHLH moiety for DNA binding. When assayed in vitro, the Hes proteins can also inhibit transcriptional activation directed from E boxes (CANNTG), independent of N box binding, even though only Hes2 appears capable of binding directly to the E box
0925-4773/97/$17.00 1997 Elsevier Science Ireland Ltd. All rights reserved PII S0925-4773 (97 )0 0665-5
228
C.G. Lobe / Mechanisms of Development 62 (1997) 227–237
element (Akazawa et al., 1992; Sasai et al., 1992; Ishibashi et al., 1993). Repression of transcription from E box elements by Hes proteins might occur through inhibition of certain other bHLH proteins that do bind to E boxes, namely E47, MyoD, Myf5 and Mash1 (Sasai et al., 1992). Hes proteins may form non-functional heterodimers with the E box-binding proteins, thus interfering with their ability to activate transcription (Bailey and Posakony, 1995). It is not yet clear whether this mechanism of transcriptional regulation is utilized in vivo (Van Doren et al., 1994). One of the Hes proteins, Hes3, lacks the basic domain required for DNA binding and does not bind to N or E boxes (Sasai et al., 1992). It does, however, possess the helix-loop-helix domain required for dimerization; thus, Hes3 is similar to the Id proteins (Bailey and Posakony, 1995) and to Drosophila Extramacrochaetae (Emc) (Ellis et al., 1990). Unlike Id and Emc, Hes3 possesses the Orange domain and the C-terminal WRPW motif required for interaction with the Groucho co-repressor. Hes3 is not able to inhibit transcription through N boxes, as the other Hes proteins can, but efficiently represses transcriptional activation from E boxes and inhibits the transcriptional activation by E47 through the E box in in vitro assays (Sasai et al., 1992). Therefore, Hes3, at least, seems able to repress transcription through E box elements indirectly, by dimerization with the E box-binding proteins. This also implies that Hes3 can effect only one part of the transcriptional repression that the other Hes proteins are capable of. In mice, gain- and loss-of-function experiments using Hes1 point to a role for the Hes genes in regulating neural differentiation, rather than determining an epidermal versus neural fate: Hes1 expression directed by a viral vector in cells of the central nervous system prevented neural and glial differentiation (Ishibashi et al., 1994). Conversely, a targeted gene inactivation of Hes1 lead to a severe phenotype in the central nervous system, apparently due to premature differentiation of neurons (Ishibashi et al., 1995). Therefore, Hes1 seems to be involved in delaying differentiation of neuronal cells, possibly through inhibition of Mash1, rather than in selection of a non-neuronal cell fate. The only suggestion to date of a conserved role for Hes genes in segmentation comes from expression of Hes1 during somite formation (Sasai et al., 1992). Other expression reported for the Hes genes does not indicate a function in segmental patterning of the developing embryo: Hes1 is expressed in many tissues, mainly epithelial tissues undergoing mesenchymal induction (Sasai et al., 1992); Hes5 is expressed throughout the ventricular zone of the central nervous system (Takebayashi et al., 1995); Hes3 was detected in cells of adult cerebellum, but not in embryos (Sasai et al., 1992). While examining the expression of mouse homologues of the neurogenic genes (mNGs), I found that Hes3 is also expressed during embryogenesis. Its initial expression in the developing central nervous system suggests a role in
early midbrain and hindbrain patterning along the anteroposterior axis. Later expression occurs in epithelial tissues and overlaps the other mNGs, pointing to a later role in regulating cell differentiation.
2. Results 2.1. Hes3 expression in primitive ectoderm and the midhindbrain region during neurulation (7–8.5 dpc) Gene expression in embryos from 7 to 10 dpc was analyzed by whole mount in situ hybridization (Wilkinson and Nieto, 1993). Hes3 transcripts were initially detected in 7– 7.5 dpc embryos in the ectoderm adjacent to the primitive streak, up to and slightly ahead of the node (Fig. 1A; arrowheads denote the position of the node). As the headfold formed, the anterior boundary of this signal became more sharply defined in the region of the prospective hindbrain. Expression was observed in the neural plate ahead of the node and in ectoderm lateral to the primitive streak (Fig. 1B). By comparison, Hes1 at this stage was expressed throughout the neural plate rostral to the node and in a line of cells forming a semi-circle around the caudal side of the node (Fig. 1C,D). By 8.5 dpc, strong signals for the Hes3 transcript were localized to the region of the mesencephalic/metencephalic boundary, two bands in the hindbrain and throughout the open neural plate except for the floor plate (Fig. 1E,F). Caudal to the open neural plate, the ectoderm on either side of the primitive streak also continued to express Hes3 (Fig. 1E,F, small arrows). In FVB/N inbred mice, the two bands of expression in the hindbrain were distinct at the 4–5-somite stage and persisted until approximately the 7-somite stage (Theiler, 1989). However, in the outbred CD1 strain, we found that Hes3 expression in the hindbrain continued past the 7somite stage and spinal cord expression was initiated earlier, at 8.5 dpc compared to 9.5 dpc, than in FVB/N mice. We utilized the persistent hindbrain expression in CD1 mice to compare expression to Krox20. Krox20 is expressed in rhombomere (r) 3 beginning at the 0–4somite stage, and r5 at 4–7 somites (Wilkinson et al., 1989). Its expression there is required for the formation of r3 and r5 (Swiatek and Gridley, 1993). CD1 embryos were stained either for Hes3 (Fig. 2A,D,G,J), Krox20 (Fig. 2C,F,I,K) or both transcripts (Fig. 2B,E,H). This clearly demonstrated that Hes3 is expressed strongest in r1, weakly in r2, r4, r6 and r7, and is off in r3 and r5 at the same time that Krox20 is expressed only in r3 and r5. The alternating pattern of Hes3 and Krox20 expression in the hindbrain seemed to develop simultaneously (Fig. 2A–C, arrowhead). It was therefore not possible to determine if the products of one of the genes may be regulating the other. The expression in the prospective mid/hindbrain region
C.G. Lobe / Mechanisms of Development 62 (1997) 227–237
229
Fig. 1. Expression of Hes3 and Hes1 in 7.0–8.5 dpc embryos, analyzed by whole mount in situ hybridization with riboprobes. (A) Expression of Hes3 is initially detected in the embryonic ectoderm, just ahead of the node (arrowhead) and more caudally flanking the primitive streak. (B) After headfold formation, Hes3 signal continues in the ectoderm but extends further rostrally into the base of the headfold. (C) Hes1 expression in headfold stage embryos occurs in the neural plate (np) and the node region (arrowhead). (D) Ventral view of the embryo in (C) to show Hes1 signal in a semicircle around the caudal side of the node region (arrowhead). (E) At 8.5 dpc, Hes3 transcripts are detected dorsally in the mesencephalon and metencephalon (mes/met) and in two bands of the hindbrain neurectoderm which correspond to r4, r6 and r7. Prominent expression also continues in the neural plate and ectoderm adjacent to the primitive streak (ps; arrow). fb, forebrain; ht, heart. (F) Dorsal view of the embryo in (E), showing expression in the mesencephalon, metencephalon and r4, r6 and r7, in the neural plate (arrowhead) and ectoderm flanking the primitive streak (small arrows). Numbers designate rhombomere segments.
is also intriguing, particularly with respect to other mNGs at the same stage: one of the groucho-related genes, Grg3, is expressed in an overlapping domain, in the mid- and hindbrain, and therefore could potentially interact with Hes3 there (Leon and Lobe, 1996). Grg4, on the other hand, is expressed in two diffuse bands, in the rostral mesencephalon and directly behind the mesencephalic/ metencephalic boundary, just anterior and posterior to the region of prominent Hes3 expression (Fig. 3A,B).
Hes1 transcripts are located with Grg4 in the rostral mesencephalon, just anterior to Hes3 (Fig. 3D). These patterns further suggest the mNGs may play a role in the early patterning of the brain. A gene which was demonstrated to be required for patterning of the mid-hindbrain region is En1. Its expression begins at around or just before the 1-somite stage in the presumptive mesencephalon and metencephalon (Davis and Joyner, 1988; Rowitch and McMahon, 1995) overlapping Hes3 expression. However,
230
C.G. Lobe / Mechanisms of Development 62 (1997) 227–237
Fig. 2. Comparison of Hes3 and Krox20 expression in CD1 embryos. Embryos were hybridized either with only Hes3 (A,D,G,J), with Krox20 (C,F,I,K) or with both Hes3 and Krox20 (B,E,H). (A–C) Embryos at 8.0 dpc (0–4 somite) display a gap in Hes3 (A) hindbrain expression as Krox20 (C) is first expressed in r3. The embryo in (A) has turned due to handling during hybridization. (D–I) In 8.5 dpc embryos (5–8 somites) Hes3 and Krox20 expression are reciprocal in the hindbrain, with Hes3 expression in r1, r2, r4, r6 and r7 and Krox20 in r3 and r5. (D–F) Dorsal views; (G–I) lateral view. In (D), the small arrow points to ectoderm flanking the primitive streak and the arrowhead points to expression in the open neural plate. (J,K) At 9.0 dpc, a gap in Hes3 expression persists in r3 and r5 (J) at the same time that Krox20 is specifically expressed there (K). In all panels, small arrowheads mark the position of r3 and r5.
En1 expression extends to the ventral-most part of the neural fold, whereas Hes3 is restricted to the dorsal part (Fig. 3A,C). 2.2. Expression during early brain development (9–10.5 dpc) Following embryo turning at around 9.0 dpc, Hes3 signal was initially still localized to the dorsal part of the midhindbrain boundary and the open neural tube in the tail region in FVB/N embryos (Fig. 4A). However, in slightly older embryos, the signal extended caudally through the entire length of the neural tube and rostrally in scattered cells in the dorsal midbrain (Fig. 4B) and later in the forebrain (Fig. 4C). Sections of embryos after whole mount in situ hybridization revealed the Hes3 signal was localized to a dorso-lateral region of the neural tube (Fig. 4G,H, arrows and arrowheads). Mash1 expression was first detected in embryos after turning and was generally found in adjacent parts of the neural tube to Hes3. For instance, Mash1 transcripts were initially present in the dorsal part of the mesencephalon (Fig. 4D) (Guillemot and Joyner, 1993). Later expression was observed in the mesencephalon and spinal cord in a domain more dorsal than Hes3 (Fig. 4E,F,I), as well as in cells flanking the floor plate (Fig. 4I) (Lo et al., 1991; Guillemot and Joyner, 1993). Mash1 was not initially present in the cerebello-mesencephalic flexure, where Hes3 was strongest (Fig. 4B,E, arrows) (Guillemot and Joyner, 1993). By 10.5 dpc, the signal for Hes3 transcripts in the cen-
tral nervous system became restricted to a band of cells just rostral of the isthmus, in the region of the prospective inferior colliculus (Fig. 4J,K) and the open neural plate in the tail. By contrast, a strong En1 signal was detected in cells on both sides of the flexure that marks the boundary between the midbrain and hindbrain (Fig. 4L). 2.3. Hes3 is expressed more extensively in the nervous system during organogenesis (11–16.5 dpc) Embryos from 11.5 dpc to 16.5 dpc were sectioned and then subjected to hybridization with digoxygenin-labeled RNA probes to analyze gene expression. At 11.5 and 12.5 dpc, Hes3 expression was detected throughout the length of the central nervous system and in the peripheral nervous system (Fig. 5A–C). In the brain of 11.5 dpc embryos, transcripts were localized to the roof of the telencephalon (Fig. 5A, te) and the lateral walls and roof of the diencephalon. This expression overlaps that of the reported Mash1 expression in the brain (Lo et al., 1991; Guillemot and Joyner, 1993). There was also a weak Hes3 signal in the floor of the diencephalon. At the mesencephelon, strongest expression was in the lateral walls, just above the sulcus limitans, in the ventricular zone next to the lumen. Expression was also observed in the roof of the caudal region of the midbrain and isthmus. A strong signal in a few cells of the floorplate was observed in the hindbrain and spinal cord (Fig. 5A, fp). This signal is complementary to that of Mash1 in cells on either side of the floor plate (Lo et al., 1991; Guillemot and Joyner, 1993).
C.G. Lobe / Mechanisms of Development 62 (1997) 227–237
231
Fig. 3. Expression in 8.5 dpc (7–8 somite stage) embryos. (A) Hes3 expression persists in the dorsal region of the mesencephalon/metencephalon (mes/ met) boundary. A strong signal in the ectoderm adjacent to the primitive streak (ps) and open neural plate (np) is also present, but the expression in r4, r6 and r7 is quite low at this stage in FVB/N mice. fb, forebrain; ht, heart. (B) Grg4, a groucho homologue is expressed in the mesencephalon and metencephalon (mes, met), with highest levels flanking the Hes3 expression domain (arrows). Grg4 is also expressed in the forebrain (fb), somites, and presomitic mesoderm (ps, primitive streak). (C) En1 expression also occurs in the region of the mesencephalon/metencephalon boundary (mes/met) but extends ventrally to the floor plate. fb, forebrain; mes, mesencephalon; met, metencephalon; np, neural plate; ps, primitive streak; ht, heart. (D) Hes1 is expressed in the mesencephalon just rostral to Hes3 expression (arrow), and in the forebrain (fb), somites and a diffuse band in the primitive streak region (ps).
Expression in the ventricular zone next to the lumen, dorsal to the sulcus limitans, also occurred at the level of the otic vesicle and continued down through the caudal part of the hindbrain and the spinal cord (arrowhead in Fig. 5A,B). In the spinal cord, cells immediately lining the lumen in the ventricular zone and in the ventral horn of the marginal zone also stained for Hes3 transcripts at 11.5 dpc. Comparison of Hes3 to Mash1 expression in the spinal cord at 12.5 dpc showed that Hes3 signal was in the dorsal-most part of the spinal cord and the floor plate where Mash1 was absent (Fig. 5B,C; fp, floorplate). Expression of the two genes did overlap in the dorsal region of the ventricular zone. However, the Hes3 signal was confined to cells adjacent to the lumen, whereas Mash1 signal extended out to where the ventricular zone meets the marginal zone (Fig. 5C). Later in central nervous system development, at 14.5 and 16.5 dpc, Hes3 transcripts were located in the ventricular zone of the cerebral cortex (Fig. 5D, cc; Fig. 6A, vz). In the midbrain, Hes3 signal was prominent dorsally in the tectal neuroepithelium (strongest in the inferior colliculus;
Fig. 5D and 6C, ic) and ventrally in the post tegmental neuroepithelium (Fig. 5D, pn). Expression of Hes3 in the external granule layer of the cerebellum was first apparent as it formed at 14.5 dpc (Fig. 5D, eg), and continued through 16.5 dpc (Fig. 6C, eg). Expression also persisted in the medulla and spinal cord in the ventricular zone adjacent to the lumen (arrowhead in Fig. 5D; Fig. 5F, dm). Other CNS expression was detected in the lens and the pigmented epithelium layer of the developing eye (Fig. 5A, ey) and the neuroepithelium of the otic vesicle and cochlea (Fig. 5D, co). Expression in the nasosensory epithelium was first apparent at 11.5 dpc (Fig. 5A) and continued at later stages (Fig. 5B and 6A; ne). In the peripheral nervous system, cells of the cranial and dorsal root ganglia expressed Hes3 (cn, tg, af, and ov in Fig. 5A,D; drg in Fig. 5G). These tissues are also positive for Notch1, Grg3, Grg4, and Hes1 expression (Weinmaster et al., 1991; Franco del Amo et al., 1992; Reaume et al., 1992; Sasai et al., 1992; Weinmaster et al., 1992; Koop et al., 1996; Leon and Lobe, 1996). Apart from the nervous system, signal was detected in
232
C.G. Lobe / Mechanisms of Development 62 (1997) 227–237
Fig. 4. Expression analysis in 9.0–10.0 dpc embryos. (A) At 9.0 dpc, Hes3 is weakly expressed in the mesencephalon and rostral part of the metencephalon (arrowhead), and in the dorsal part of the open neural tube in the tail region (arrow). (B) By 9.5 dpc, Hes3 expression in the neural tube is also seen more rostrally in the mesencephalon (*; ms, mesencephalon; mt, metencephalon) with the strongest expression in the cerebello-mesencephalic flexure (arrow). Expression is detected in the spinal cord (sc), appearing in a rostral to caudal gradient. (C) A slightly older embryo than that shown in (B) displays stronger expression throughout the brain. (D) Mash1 expression occurs in the dorsal region of the mesencephalon, at the same stage as Hes3, shown in (A). (E) Mash1 expression at 9.5 dpc is found throughout the dorsal part of the midbrain, hindbrain and spinal cord, except for the cerebellomesencephalic flexure (arrow), where Hes3 expression is strongest (see B). (F) Expression of Mash1 slightly later than the embryo shown in (E). Mash1 signal is stronger throughout the CNS and appears similar to Hes3 expression at the same stage (shown in C; arrow, cerebello-mesencephalic flexure). (G,H) Cross-sections of the embryo shown in (C) demonstrate that the Hes3 signal is in a dorsal-lateral part of the brain (arrow) and spinal cord (arrowhead; di, diencephalon; met, metencephalon; my, myelencephalon; np, neural plate; ht, heart). (I) Cross section of the embryo shown in (F) reveals Mash1 expression is more dorsal than Hes3 in the central nervous system (arrow; my, myelencephalon; ht, heart). (J) Between 9.5 and 10.5 dpc, the Hes3 signal becomes restricted to the mesencephalon, appearing as two bands, one just rostral of the mes/met boundary and one slightly more anterior. (K) In a 10.0 dpc embryo stained for Hes3 expression the strongest signal occurs rostral to the isthmus, in the inferior colliculus (ic; mes, mesencephalon; met, metencephalon). (L) At 10.0 dpc, a strong En1 signal is present in both the inferior colliculus (ic) and metencephalon (met) neighboring the mesencephalic isthmus. (mes, mesencephalon).
epithelial structures, in a similar pattern to that reported for the Notch1 and Grg genes (Weinmaster et al., 1991; Dehni et al., 1995; Koop et al., 1996; Leon and Lobe, 1996). These included the basal layer of the skin (bs in Fig. 5A and 6B), the hair and whisker follicles (Fig. 5D and 6A,B; wh), the adrenal medulla and cortex of the kidney (Fig. 5G, ag, ki), endoderm lining of the gut (Fig. 5G, gu), salivary gland, lung (Fig. 5G, lu), thymic and tooth bud epithelia (Fig. 5D and 6A, tb). The roof of Rathke’s pocket, adjacent to the infundibulum from the diencephalon, was also positive for Hes3 signal (Fig. 5E, pi). This part of Rathke’s pocket will form the epithelial part of the pituitary, the pars intermedia, whereas the more ventral mesenchymal portion will form the glandular part of the pituitary, the pars distalis (Rugh, 1990). We also hybridized the Hes3 RNA probe with P21 brain sections to compare our results at that stage to the previous report of Hes3 expression in the cerebellum (Sasai et al., 1992). We observed expression restricted to the Purkinje cells of the cerebellum (data not shown) as described previously (Sasai et al., 1992).
3. Discussion Hes3 expression was previously reported for the adult cerebellum, but was not detected during embryo development (Sasai et al., 1992). The analysis here reveals that Hes3 is expressed during mouse embryogenesis, from 7 dpc onward, in the developing nervous system and epithelial tissues. Comparison to the expression patterns of previously characterized developmental control genes raises the possibility that Hes3 is involved both in early patterning along the A/P axis and in a signalling pathway with the other mNGs. The early expression of Hes3 and Hes1 around the primitive streak and node borders the area where cells delaminate from the ectoderm and migrate through the streak. This is reminiscent of the pattern of gene expression in the morphogenetic furrow of the Drosophila eye, in which the HLH factor, Emc, is expressed in all cells ahead of the progressing furrow and hairy is expressed in a line of cells just preceding the furrow (Brown et al., 1995; Heberlein and Moses, 1995; Heberlein et al., 1995). Expression
C.G. Lobe / Mechanisms of Development 62 (1997) 227–237
233
Fig. 5. Hes3 expression detected in sections of 11.5–14.5 dpc embryos. (A) Cross-section of 11.5 dpc embryo at the level of the forebrain and hindbrain. A schematic of the orientation of the section is shown to the left. Signals for Hes3 expression are in the nasosensory epithelia (ne), lining the olfactory pit (op), the dorsal ventricular zone of the telencephalon (te), pigment layer of the retina (ey), semilunar ganglia of the trigeminal nerve (tg), accousticofacialis ganglia VII and VIII (af), a few cells lining the otic vesicle (ov), cranial nerves (cn), floor plate (fp) and ventricular zone above the sulcus limitans (arrowhead) in the myelencephalon (my), and the basal layer of the skin (bs). BA, branchial arch. (B,C) Serial cross sections through the spinal cord of 12.5 dpc embryo hybridized with Hes3 (B) or Mash1 (C). Hes3 signal is strongest in the dorsal-most part of the spinal cord in cells of the ventricular zone lining the lumen (arrowhead) and a few cells of the floor plate (fp). Signal is also present in the dorsal root ganglia (drg). Mash1 signal is strongest throughout the dorsal ventricular zone of the spinal cord (bracketed). (D) Parasaggital section of the head, showing Hes3 signal in the whisker follicles (wh), nasosensory epithelia (ne), toothbud progenitor (tb), trigeminal ganglia (tg), cells lining the cochlea (co), the ventricular zone of the cerebral cortex (cc), inferior colliculus (ic), post-tegmental neuroepithelium (pn) and brainstem (arrowhead), and external granular layer (eg) of the prospective cerebellum (cb). tm, tegmentum; sc, superior colliculus. (E) A higher magnification showing Hes3 expression in the pars intermedia (pi) of the developing pituitary. pd, pars distalis. (F) Higher magnification of the region of the cerebellum from (B). Signal is present in the external granule layer (eg) of the cerebellum (cb), the choroid plexus (cp) and the dorsal medulla (dm). (G) Sagittal section of the trunk region. Signals corresponding to Hes3 expression occur in the lung epithelia (lu), liver (li), endodermal lining of the gut (gu), kidney (ki), adrenal gland (ag), and dorsal root ganglia (drg). ht, heart.
234
C.G. Lobe / Mechanisms of Development 62 (1997) 227–237
Fig. 6. Expression of Hes3 in 16.5 dpc embryo sections. (A) Sagittal section of the head. Hes3 signals correspond to whisker follicles (wh), epithelia of the toothbud progenitors (tb), nasosensory epithelia (ne), the ventricular zone (vz) of the cerebral cortex (cc), sensory cells lining the cochlea (co), and the external granule layer of the cerebellum (cb). mb, midbrain; to, tongue. (B) A higher magnification of the nose region, showing expression of Hes3 in the basal layer of the skin (bs) and the inner root sheath of the whisker follicles (wh). (C) Parasaggital section of the mesencephalon/metencephalon boundary. Expression of Hes3 occurs in the ventricular zone of the inferior colliculus, the external granule layer (eg) of the cerebellum (cb) and cells in the choroid plexus (cp).
of Hes3 and Hes1 around the node and primitive streak may represent part of a conserved regulatory pathway to organize morphogenetic changes and cell differentiation (Brown et al., 1995; Heberlein and Moses, 1995). The expression pattern of Hes3 during embryonic brain development suggests a role in early patterning of the midbrain and hindbrain region. Hes3 appears to mark cells at the mesencephalic/metencephalic border. Its expression is bracketed by Grg4, a groucho homologue, and Hes1. Other genes expressed around the mid/hindbrain boundary are Pax2, Wnt1, En1 and En2 (Wilkinson et al., 1987; Davis and Joyner, 1988; Rowitch and McMahon, 1995). Gene targeting of Wnt1 and En1 resulted in
loss of the midbrain and adjacent cerebellar component of the metencephalon, demonstrating their important role in patterning this part of the brain (McMahon and Bradley, 1990; Wurst et al., 1994). It was further demonstrated that the Wnt1 phenotype corresponded to loss of the En1expressing cells (McMahon et al., 1992). This is comparable to the interaction for the Wnt (wg) and En genes during Drosophila segmentation, in which maintenance of En expression requires Wnt (DiNardo et al., 1988; Martinez-Arias et al., 1988; Bejsovec and Martinez-Arias, 1991; Heemskerk et al., 1991). However, both in Drosophila and in mouse, Wnt1 expression was not required for the initiation of En1 expression. In Drosophila, initiation
C.G. Lobe / Mechanisms of Development 62 (1997) 227–237
of En and Wg expression patterns is determined by ftz, Eve, and prd, which in turn are regulated by hairy. Promoter analysis of the En2 gene revealed Pax binding sites (Song et al., 1996), and Pax2 expression in the mid/hindbrain precedes expression of Wnt1 and En1 (Rowitch and McMahon, 1995), indicating that the regulatory hierarchy between Pax, Wnt and En has also been conserved. Thus, it is possible that in the mouse, the Hes genes also have a conserved regulatory role upstream of Pax, En and Wnt genes. The early expression of Hes3 in the mid/hindbrain region is in accord with this speculation. A targeted gene inactivation of Hes3 may reveal whether it plays a role in early mid/hindbrain patterning through regulation of homologues of the Drosophila segmentation genes. The early expression of Hes3 in the hindbrain also suggests a role in segmentation. It is first expressed throughout the neural plate but its expression is lost, first in r3 and then in r5. This pattern occurs at approximately the same time that Krox20 expression appears in r3 and r5 and precedes the morphological appearance of rhombomeres. Inactivation of the Krox20 gene results in a loss of r3 and r5 confirming that Krox20 is required for formation of those two rhombomeres. The specific complementary expression of Hes3 in regions of the hindbrain where Krox20 is not expressed suggests that both Hes3 and Krox20 are required for correct rhombomere formation. It was not possible to discern if one of Hes3 or Krox20 appeared first and therefore whether one of them could be regulating expression of the other. In Drosophila, h and E(spl) gene products negatively regulate achaete and scute (Ohsako et al., 1994; Van Doren et al., 1994). Interestingly, the strongest Hes3 expression in the neural tube lies in the cerebello-mesencephalic flexure, which corresponds to the region of the neural tube where Mash1 is not initially present (Guillemot and Joyner, 1993). Likewise, Hes3 expression in the spinal cord lies adjacent to Mash1 initially in the dorsolateral spinal cord and floorplate, and later is in the ventricular zone within the domain of Mash1 expression. This is again suggestive of a conserved regulatory role for Hes3, in this case in the negative regulation of the Achaete-scute homologue, Mash1. Hes3 expression at mid- to late-embryogenesis occurs in many types of epithelial cells undergoing mesenchymal induction, a common theme for the neurogenic gene homologues, Notch1, Grg1, Grg3, Grg4 and Hes1 (Weinmaster et al., 1991; Dehni et al., 1995; Koop et al., 1996; Leon and Lobe, 1996). Therefore, Hes3 may act with the other mNGs to regulate cell proliferation and differentiation in developing epithelial tissues. The Drosophila neurogenic genes are also expressed during epithelial/ mesenchymal interactions, where they seem to designate epithelial cell fates (Tepass and Hartenstein, 1995). Hes3 is unique among the Hes proteins in that it has the helix-loop-helix, Orange, and WRPW domains but lacks the basic domain and does not bind to DNA (Sasai et al.,
235
1992; Dawson et al., 1995). Therefore, it probably functions similarly, but not identically, to the other Hes proteins. Its expression pattern is also distinct from the other Hes genes, in that it displays a definitive pattern along the antero-posterior axis. These results suggest that Hes3 has a role not only in regulating cell differentiation with the other Hes, Grg and Notch genes, but also an early role in patterning the midbrain and hindbrain.
4. Experimental procedures 4.1. RNA probes cDNA clones for rat Hes3 and Hes1 in the pSV2-CMV expression vector were kindly provided by R. Kageyama (Sasai et al., 1992). The EcoRI inserts were subcloned into pBluescript-KS (Stratagene). The Hes3 clone was linearized with BamHI and transcribed with T3 RNA polymerase to generate a 985 bp antisense RNA probe. The Hes1 template was generated by linearizing with BstEII and transcribed with T7 RNA polymerase. The Grg4 sequence used for an RNA template was a 459 bp sequence that lies within the WD40 domain (Koop et al., 1996). The En1 and Mash1 cDNAs were kindly provided by Dr. A. Joyner, and the Krox20 cDNA by Dr. J. Rossant. RNA synthesis was directed using T7, T3 or SP6 RNA polymerase, including digoxygenin-labeled nucleotide mix as recommended by the supplier (Boehringer-Mannheim). Products were analyzed on a 1% agarose gel and approximately 1 mg/ml of the digoxygenin-labeled antisense RNA was used for hybridization of embryos and slides. 4.2. Whole-mount in situ hybridization of 7.5–10.5 dpc embryos FVB/N or CD1 mice, as indicated in the figures, were mated and the day of the plug was taken as 0.5 dpc. Embryos of 7.5–10.5 dpc were dissected into PBS and fixed overnight at 4°C in 4% paraformaldehyde in PBS. The following day they were passed through a graded methanol series and then stored in methanol at −20°C. The protocol for whole mount in situ hybridization of embryos was as described (Wilkinson and Nieto, 1993). After overnight hybridization, embryos were washed twice for 30 min at 70°C in solution 1 (50% formamide, 5 × SSC pH 4.5, 1% SDS), 10 min at 70°C in a 1:1 mix of solution 1 and solution 2 (0.5 M NaCl, 10 mM Tris–HCl pH 7.5, 0.1% Tween-20), three 5 min washes in solution 2, two 30 min incubations at 37°C in solution 2, a 5 min wash in solution 2, 5 min in solution 3 (50% formamide, 2 × SSC pH 4.5), two 30 min washes in solution 3 at 65°C, three 5 min washes in TBST (137 mM NaCl, 25 mM Tris– HCl pH 7.6, 3 mM KCl, 0.1% Tween-20) and a 90 min block in 10% serum/TBST. Embryos were then incubated
236
C.G. Lobe / Mechanisms of Development 62 (1997) 227–237
overnight at 4°C with alkaline phosphatase-coupled antidigoxygenin antibody (Boehringer-Mannheim). The antibody was pre-blocked for 90 min at 4°C with 13 dpc embryo powder and 1% serum in TBST. After overnight incubation with antibody, embryos were washed three times for 5 min and four times for 1 h in TBST, and then washed overnight in TBST. The following day embryos were washed three times for 10 min in NTMT (100 mM NaCl, 100 mM Tris–HCl pH 9.5, 50 mM MgCl2, 0.1% Tween-20, 2 mM levamisole) and then stained in 337.5 mg/ml NBT/175 mg/ml BCIP in NTMT. Photographs were taken using illuminated light on a Leica Wild M3C dissecting microscope. For sectioning after whole mount hybridization, samples were cryoprotected in 20% sucrose/PBS, frozen and sectioned at 10 mm on a Leitz cryostat at −20°C. Slides were directly mounted in 7% gelatin/50% glycerine and photographed on a Zeiss Axiophot microscope with DIC optics. 4.3. In situ hybridization of embryo and brain sections Embryos of 11.5–16.5 dpc were dissected into PBS and fixed overnight in 4% PFA/PBS at 4°C. Three week old post-natal brains were fixed by perfusion and stored in 4% PFA/PBS at 4°C overnight. After fixation, tissues were rinsed through a graded ethanol series, xylene, and embedded in paraffin. Tissues were sectioned at 20 mm. The slides were then hybridized and washed exactly as described (Wilkinson and Nieto, 1993) but without RNase treatment in the washes. The alkaline phosphatase staining reaction was allowed to proceed for 6 h to overnight. Following a rinse in PBS and water, slides were airdried, mounted in Permount and photographed on a Zeiss Axiophot with DIC optics.
Acknowledgements I thank Drs. J.R. Jacobs, A. Nagy and W. Wurst for useful discussions, R. Kageyama for the Hes3 cDNA, A. Joyner for the En1 and Mash1 probes and J. Rossant for the Krox20 probe. This work was supported by a grant from the Medical Research Council of Canada. CL is supported by a Scholarship from the Medical Research Council of Canada.
References Akazawa, C., Sasai, Y., Nakanishi, S. and Kageyama, R. (1992) J. Biol. Chem. 267, 21879–21885. Bailey, A.M. and Posakony, J.W. (1995) Genes Dev. 9, 2609–2622. Bejsovec, A. and Martinez-Arias, A. (1991) Development 113, 471–485. Brown, N.L., Sattler, C.A., Paddock, S.W. and Carroll, S.B. (1995) Cell 80, 879–887. Chenn, A. and McConnell, S.K. (1995) Cell 82, 631–641.
Davis, C.A. and Joyner, A.L. (1988) Genes Dev. 2, 1736–1744. Dawson, S.R., Turner, D.L., Weintraub, H. and Parkhurst, S.M. (1995) Mol. Cell. Biol. 15, 6923–6931. Dehni, G., Liu, Y., Husain, J. and Stifani, S. (1995) Mech. Dev. 53, 369– 381. DiNardo, S., Sher, E., Heemskerk-Jongens, J., Kassis, J.A. and O’Farrell, P.H. (1988) Nature 332, 604–609. Ellis, H.M., Spann, D.R. and Polakony, J.W. (1990) Cell 61, 27–38. Franco del Amo, F., Smith, D.E., Swiatek, P.J., Gendron-Maguire, M., Greenspan, R.J., McMahon, A.P. and Gridley, T. (1992) Development 115, 737–744. Guillemot, F. and Joyner, A.L. (1993) Mech. Dev. 42, 171–185. Heberlein, U. and Moses, K. (1995) Cell 81, 987–990. Heberlein, U., Singh, C.M., Luk, A.Y. and Donohoe, T.J. (1995) Nature 373, 709–711. Heemskerk, J., Dinardo, S., Kostriken, R. and O’Farrell, P.H. (1991) Nature 352, 404–410. Heitzler, P., Bourouis, M., Ruel, L., Carteret, C. and Simpson, P. (1996) Development 122, 161–171. Ishibashi, M., Sasai, Y., Nakanishi, S. and Kageyama, R. (1993) Eur. J. Biochem. 215, 645–652. Ishibashi, M., Moriyoshi, K., Sasai, Y., Shiota, K., Nakanishi, S. and Kageyama, R. (1994) EMBO J. 13, 1799–1805. Ishibashi, M., Ang, S.-L., Shiota, K., Nakanishi, S., Kageyama, R. and Guillemot, F. (1995) Genes Dev. 9, 3136–3148. Jennings, B., Preiss, A., Delidakis, C. and Bray, S. (1994) Development 120, 3537–3548. Klambt, C., Knust, E., Tietze, K. and Campos-Ortega, J.A. (1989) EMBO J. 8, 203–210. Koop, K.E., MacDonald, L.M. and Lobe, C.G. (1996) Mech. Dev. 59, 73–87. Kuhn, R., Schwenk, F., Aguet, M. and Rajewsky, K. (1995) Science 269, 1427–1429. Lecourtois, M. and Schweisguth, F. (1995) Genes Dev. 9, 2598–2608. Leon, C. and Lobe, C.G. (1996) Dev. Dyn. 208, 11–24. Liem, K.F., Tremml, G., Roelink, H. and Jessell, T.M. (1995) Cell 82, 969–979. Lo, L.-C., Johnson, J.E., Wuenschell, C.W., Saito, T. and Anderson, D.J. (1991) Genes Dev. 5, 1524–1537. Martinez-Arias, A., Baker, N.E. and Ingham, P.W. (1988) Development 103, 157–170. McMahon, A.P. and Bradley, A. (1990) Cell 62, 1073–1085. McMahon, A.P., Joyner, A.L., Bradley, A. and McMahon, J.A. (1992) Cell 69, 581–595. Ohsako, S., Hyer, J., Panganiban, G., Oliver, I. and Caudy, M. (1994) Genes Dev. 8, 2743–2755. Oka, C., Nakano, T., Wakeham, A., de la Pompa, J.L., Mori, C., Sakai, T., Okazaki, S., Kawaichi, M., Shiota, K., Mak, T.W. and Honjo, T. (1995) Development 121, 3291–3301. Paroush, Z., Finley, R.L.J., Kidd, T., Wainwright, S.M., Ingham, P.W., Brent, R. and Ish-Horowicz, D. (1994) Cell 79, 805–815. Reaume, A.G., Conlon, R.A., Zirngibl, R., Yamaguchi, T.P. and Rossant, J. (1992) Dev. Biol. 154, 377–387. Rowitch, D.H. and McMahon, A.P. (1995) Mech. Dev. 52, 3–8. Rugh, R. (1990) The Mouse. Its Reproduction and Development. Oxford University Press, Oxford. Rushlow, C.A., Hogan, A., Pinchin, S.M., Howe, K.R., Lardelli, M.T. and Ish-Horowicz, D. (1989) EMBO J. 8, 3095–3103. Sasai, Y., Kageyama, R., Tagawa, Y., Shigemoto, R. and Nakanishi, S. (1992) Genes Dev. 6, 2620–2634. Skeath, J.B. and Carroll, S.B. (1992) Development 114, 939–946. Swiatek, P.J. and Gridley, T. (1993) Genes Dev. 7, 2071–2083. Takebayashi, K., Akazawa, C., Nakanishi, S. and Kageyama, R. (1995) J. Biol. Chem. 270, 1342–1349. Tepass, U. and Hartenstein, V. (1995) Development 121, 393–405. Theiler, K. (ed.) (1989) The House Mouse. Development and Normal Stages from Fertilisation to 4 Weeks of Age. Springer, Berlin.
C.G. Lobe / Mechanisms of Development 62 (1997) 227–237 Van Doren, M., Bailey, A.M., Esnayra, J., Ede, K. and Posakony, J.W. (1994) Genes Dev. 8, 2729–2742. Weinmaster, G., Roberts, V.J. and Lemke, G. (1991) Development 113, 199–205. Weinmaster, G., Roberts, V.J. and Lemke, G. (1992) Development 116, 931–941. Wilkinson, D.G. and Nieto, M.A. (1993) Guide to techniques in mouse development. In Wasserman, P.M. and DePamphalis, M.L. (eds.),
237
Methods in Enzymology, Academic Press, San Diego, CA, Vol. 225, pp. 361–373. Wilkinson, D.G., Bailes, J.A. and McMahon, A.P. (1987) Cell 50, 79– 88. Wilkinson, D.G., Bhatt, S., Chavrier, P., Bravo, R. and Charnay, P. (1989) Nature 337, 461–464. Wurst, W., Auerbach, A. and Joyner, A. (1994) Development 120, 2065–2075.