Distinct temporal expression of mouse Nkx-5.1 and Nkx-5.2 homeo☐ genes during brain and ear development

m

D

ELSEVIER

Mechanisms of Development 52 (1995) 37 1-38 I

Distinct temporal expression of mouse Nkx-5.1 and Nkx-5.2 homeobox genes during brain and ear development Silke Rinkwitz-Brand&

Matthias Justus, Ira Oldenettel, Hans-Henning

Arnold, Eva Bober*

Department r!f‘Moleculur und Cellulur Biology, Technical University Braunschweig, Spielmannstrasse 7, 38106 Braunschweig, Germuny

Received 31 March 1995; revision received 24 April 1995; accepted 15 May 1995

Abstract

The mouse N&-5.1 and Nkx-5.2 genes have been identified by sequence homology to Drosophila NK genes within the homeobox domain. Here, we report the isolation of the Nkx-5.2 cDNA and a detailed comparative analysis of the spatio-temporal expression patterns for Nkx-5.1 and Nkr-5.2 genes. Nkx-5.2 transcripts are first detected in E13.5 embryos where they colocalize with Nkx-5.1 mRNA in the developing central nervous system and the inner ear. However, the onset of Nkx-5.1 transcription begins much earlier in 10 somite stage embryos (E8.5) in the otic placode and the branchial region. Nkx-5.1 expression in the ear persists until birth, whereas in branchial arches it is transient between E8.5 to El 1.5. Transcript distribution appears regionalized in the otic vesicle concentrating at the anterior and posterior margin and later at the dorsal side of the otocyst. These domains are distinct from regions expressing Pax-2 and sek, two other early markers for otic development. From E11.5 to birth several Nkx-5.1 expression domains appear in the brain between the ventral diencephalon and the myelencephalon. The same expression domains also exist for Nkr-5.2 beginning at E13.5. The regionally restricted expression pattern of both Nkx-5 genes during mouse development suggests their involvement in cell type specification of neuronal cells. Keywords:

Homeobox;

Nkx-genes; Neural development; Inner ear

1. Introduction Homeobox containing genes play an important role in various developmental processes. In vertebrates, Hox genes have been shown to convey positional specification along the antero-posterior axis of the body and the proximo-distal axis of the limbs (Duboule, 1992; Hunt and Krumlauf, 1992; Krumlauf, 1993). More recently, new homeobox containing genes have been identified which are expressed in more rostra1 regions than the Hox genes. Members of this group include emx-I, emx-2, dlx-1, dlx-2 and gbx-2 (Simeone et al., 1992a; Bulfone et al., 1993). Based on their expression patterns a role in controlling specification of head structures including development of the central nervous system (CNS) and sensory organs has been proposed. Interestingly, many members of this novel class of homeobox genes exert a biphasic expression with an early phase possibly involved in positional specifica* Corresponding 3918178.

author, Tel.: +49 531 3915721; Fax: +49 531

tion of segmental body units and a later phase involved in cell type specification (Heuer et al., 1992). This dual role is in agreement with similar dual functions of segmentation and gap genes in Drosophila. For example, eve and ftz are important for establishing segmental identities and later specifying neuronal cell types (Doe et al., 1987, 1988). The mouse otx gene, the homologue of Drosophila otd, appears to convey positional identity to specific regions of the developing brain (neuromeres) and is later expressed in a subset of deep-layer neurons of the cerebral cortex (Simeone et al., 1992b; Figdor and Stern, 1993). Based on sequence homology to four distinct Drosophila NK genes (Kim and Nierenberg, 1989), homologous homeobox genes have been identified in the mouse. Members of this new family include several Nkx-2 genes with defined expression patterns in specific regions of the embryo (Price et al., 1992; Lints et al., 1993). Recently, we described the partial gene sequences of Nkx-5.1 and Nkx-5.2, two novel mouse genes, containing homeoboxes that are slightly diverged from the clas-

0925.4773/95/$09.50 0 1995 Elsevier Science Ireland Ltd. All rights reserved 0925-4773(95)00414-V

SSDI

S. Rinkwitz-Brandt et al. I Mechanisms of Development S.? (1995) 371-381

312

sical NK genes (Bober et al., 1994a). In situ hybridisation with Nkr-5.1 probes revealed activity in central and peripheral nervous structures of midgestation mouse embryos. Here, we present a detailed comparison of the expression of both mouse Nkx-5 genes during early embryonic and fetal development. In whole mount hybridisations Nkx-5.1 transcripts first appear in branchial arches and the otic placode at E8.5, approximately two days earlier than previously described. Comparison with other markers of otic development such as Pax-2 (Dressler et al., 1990) and sek (Nieto et al., 1992) demonstrates that Nkx-5.1 is one of a group of regulatory molecules which label different populations of cells in the early otic vesicle. In contrast, Nkx-5.2 expression is undetectable prior to E13.5. Subsequently, both Nkx-5.1 and Nkx-5.2 genes are active in the same regions of the brain and inner ear. These spatio-temporal expression patterns suggest that both genes may play distinctive but overlapping roles in the development of specific neuronal structures. 2. Results 2.1. Nucleotide sequence of Nkr-5.2 cDNA and its expression in brain and the inner ear during prenatal mouse development A small genomic fragment encoding the Nkn-5.2 homeobox has been described previously (Bober et al., 1994a). This fragment failed to yield interpretable signals when used as hybridisation probe on tissue sections. In order to generate a better probe, we isolated the corresponding Nkx-5.2 cDNA from an El 1.5 mouse embryo. The longest available sequence of 1470 nucleotides contains an open reading frame for 286 amino acids and the complete 3’-untranslated region including the poly(A) addition signal sequence (Fig. 1). The Nkx-5.2 homeobox and two short conserved sequences immediately downstream are more than 90% identical to the same regions of the Nkx-5.1 gene. No significant sequence homology between both genes and any other known NK genes exists outside of these domains. A 600 bp SmaI fragment located downstream of the homeobox was used to generate a Nkx-5.2-specific hybridisation probe (Fig. 1B). In whole-mounted embryo preparations and on tissue sections of E7.5 to E12.5 embryos no transcripts were detected (Fig. 2C,F). However, at E13.5 Nkx-5.2 expression was observed in the brain and in the inner ear (Figs. 3-6). In particular, the ventral diencephalon, the mesencephalon, the pons and the medulla oblongata, as well as the inner ear, showed strong Nkx-5.2 expression domains. Significantly, these regions of the brain do also express Nku-5. J at this developmental

Fig.

I. CA) Nkr-5.2

cDNA

and the

corresponding protein sequence from the

Schematic representation of the Nkr-5.2

cDNA,

situ hybridisation probe; S, Sma 1; X, Xho 1.

period (compare Fig. 2G,H,J and Fig. 3C,F,H for Nkx-5.i and Nkx-5.2 hybridizations, respectively). 2.2. Detailed analysis of Nkx-5.2 expression domains in specific forebrain and hindbrain regions: coexpression of Nkx-5.1 and Nkx-5.2 To provide a detailed spatial representation of Nkr-5.2 expressing brain areas during fetal development, in situ hybridizations were performed on sections taken through the brain of E14.5 to E17.5 embryos in three cutting planes (transversal, frontal and sagittal). Figs. 3-6 show representative hybridizations. At E14.5 the brain already undergoes progressed neuronal and cytoarchitectural differentiation. Therefore, the expression domains of Nkx-5.2 at this stage can be correlated to individual anatomical brain structures. Seven discrete expression domains were observed: two Nkx-5.2 domains are located in the ventral diencephalon, one in the mesencephalon and four in the hindbrain (see the schematic representation in Fig. 7D). Of the hindbrain domains, one resides in the metencephalon and three in the medulla oblongata. The most prominent expression is seen in two separate non-overlapping rostra1 and caudal domains of the diencephalon. Both domains are confined to the hypothalamus with similar dorsal boundaries (Fig. 3A,D,E). Nku-5.2 is not expressed in the thalamic region (Fig. 3B,C). The rostra1 hybridisation signal is located in the preoptic area and the caudal one in the mammillary area (Fig. 3D-G, see also Fig. 4). The latter domain extends into the ventral part of the hypothalamus which is not clearly identifiable at this stage (Fig. 4C,D). Nkr-5.2 expression domains in the mesencephalon and hindbrain are organized in two parallel bundels running from dorsal to ventral on both sides of the midline. These bundels were recognized as symetrical round or elliptic areas in the tegmentum (Fig. 3C), the pons (Fig. 3D,F) and the medulla oblongata (Fig. 3G-K). In addition, a single unpaired Nkx-5.2 expression domain was found just around the fourth ventricle (Fig. 3G). The later stages of fetal development are characterized by increasing complexity of the differentiating brain structures. At this developmental stage proliferation and migration of neuronal precursor cells cease and the formation of fibers becomes clearly visible. The distribution of Nkx-5.2 transcripts at E17.5 essentially remains the same as in brains of E14.5 embryos. The rostra1 diencephalic expression domain is now confined to the periventricular preoptic nucleus and the mediolateral preoptic nucleus (Fig. 5A). The caudal signals appear within the dorsomedial hypothalamic nucleus, the dorsal part of the premammillary nucleus and the arcuate nucleus (Figs.

longest open reading frame. the homeodomain is indicated by the box. (B)

tilled bar indicates the homeobox, open bm below represents the region used for the synthesis of the in

S. Rinkwitz-Brand?

A

et nl. I Mechanisms

ofDevelopment

371

52 (1995) 371-381

l

CTTGGACGTTCTTGGTTTCCTCCGATTTCTTATGAACCCAATG

61

TGGGGAAGGGATGTCCGGCGGCCGGTGGCGTCTCCAGCTTCACCATCCAGTCCATCCTGG

121

GCGGGGGCCCTCCGAGGCACCGCGGGAGCCCGCCGGGTGGCCAGCCAGG~CGCAGCCT

181

GTCTGTGTCCTCGGAGGAGGAGGAGCCGGAGAAGGTTGGAGGCTCCGGCTTGCTTCTGC

241

CCAGATCCGCACGGCCCCAAGGAGCCAAGCCCTAAGCCCT~GCACCATACCCCCATCCCTTTCCCT

301

TGCCTGGGTACCCCCAAGGGCAGCGGAGGCGCAGGGCCTGCGGCCTCGGAGCGCACGCCA

L

G

R

PRMAARKM

SWFPPISYE P

WGRDVRR

V

A

S

P

A

S

P

Q

AGALRGTAGARRVAS

C

D

P

H

s

E

G

PKEPSPKHHTPI E

GGAGPAAS

LGTPKGS

361

TTCCTTTCTCCTTCTCACCCGGACTTTAAGGAAGAGAAGGAGAGGCTTTTGCCGGCGGGC

421

TCGCCTTCTCCGGGCCCGGAGCGGCCACGGGACGGCGGTGCGGAGCGCC~CCGGAGCG

481

GCCAAGAAGAAGACGCGCACCGTCTTCTCCCGCAGCCAAGTGTACCAGCTCGAGTCTACC

P

FLSPSH S

P

s

P

P

s

w

T

Q

FJ

KAPACFC

VCVLGGGGAGEGW P

s

G

P

E

D R

F

K

E

P

F

P

R

T

P

T

G

A

EKERLLPAG

PRDGGAERQ SRSQVYQLEST

AKKKTRTVF

TTCGACATGAAACGCTACCTGAGCAGCTCGGAGCGCGCTTGCCTCGCCTCCAGCCTGCAG

541

L

FDMKRY

s

s

E

s

R

A

C

LASSLQ

CTCACCGAGACCCAGGTTAAGACTTGGTTCCAGAATCGCCGC~C~GTGG~GCGGCAG

601

NRRNKWKRQ

LTETQVHTWFQ

661

CTCTCAGCCGAGCTTGAGGCAGCCAACATGGCATGGCACACGCGTCGGCGCAGACTCTCGTGGGC

721

ATGCCGCTCGTGTTTCGGGACAGTTCGCTGCTGCGAGTGCCGGTGCCGCGCTCTCTGGCC

781

TTCCCGGCTCCGCTTTATTACCCGAGCAGCAATCTCTCGGCCTG

ASAQTLVG

LSAELEAANMAH M F

P

L

V

F

R

PAPLYYPS

D

S

SLLRVPV SNLSAL

P

R

S

L

A

P

L

Y

N

L

841

TACAACAAGCTTGACTACTGACATGCTCGCCCGCCGGCTCCCGGGACGCCCCTGGTCGCC

901

TGCAGCTGCCTGCATAGCCCGGCGCGGTCGGGATCCTTCGCAGAGCTCCGGAGCAGACGC

961

GCGCGTGTCCAGAAATATTAAGAAATACACCATGTGTATTCGTTATGTCTTATTTATGGC

1021

CTCTTTCCTACCTTTTGTTTTGTTTTGTTTTATTTTGTTTTGTTTTGTTTTGAGTATTTA

1081

TCTGCATTCCTCACGCCAGAGCACCTGCTTTCTACACAAACTCGCGTTTGA

1141

AACTATTTTGGAAGGGGAGGCTCTCGGGTGGTGGTGGGAAGTGGGGTTTTGATTAGGTCG

1201

GTTTTGACCAGTAGCAAGAAA

YNKLDY.

AAAAAAAAAAACATCAGAAAAACCACCAACAGAACCAGG

1261 1321

GAGCCGAACCAACTAGGGCTTTGGTCGTTGAACTELATCACTTCTCTTTATTCCCTTCATC

1381 1441

AAAAATAAAACGTTTTAAACATGGTTGCTAAAAAAAAAAAAAA

B 5’

;,i,

3’

I

I

S. Rinkwitz-Brandt et al. I Mechanisms of Development 52 (1995) 371-381

374

hv

Fig. 2. Expression of Nkx-5.2 in prenatal mouse development: comparison with Nkx-5.1. The first row (A-C) shows transverse sections at E10.5. At this stage high level of the Nkx-5.1 expression is detectable in the otic vesicle (B); Nkw-5.2 hybridisation reveals no signals (C). In the second row (DF) coronal sections at El 1.5 at the level of the otic vesicle also show Nkr-5.1 (E) but no Nkx-5.2 (F) expression. The third row (G-J) shows examples of N/a-5.1 expression on transverse sections of an E14.5 embryo. Nkr-5.1 is expressed in the same structures as Nkr-5.2: mesencephalon (G), preoptic area, mammillary area and pons (H), inner ear and medulla oblongata (J): compare with Fig. 3C, F and .I respectively.

5B,C and 6B,C). The characteristically elongated expression domains in midbrain and hindbrain may correspond to the nuclei of afferent nerve fibers. In the medulla Nkx5.2 expression appears in the solitary tract nucleus (Figs. 5D and 6C). The detailed expression domains of Nkx-5.2 in fetal brain and inner ear were also found when Nkx-5.1 was

used as probe suggesting that both genes are subject to identical or similar spatial activation processes. For a better concept of the three-dimensional Nkx-5 expression domains whole brain preparations of an E15.5 embryo and a newborn mouse were hybridised. In addition, the inner ear of the E15.5 embryo was dissected and hybridised. As shown in Fig. 7A, the ventral aspect of the E15.5

S. Rinkwitz-Brundt

et (11./ Mechunisms

of Development 52

(1995) 371-381

315

Fig. 3. N/W5.2 expression domains in the E14.5 brain. Serial transversal sections through the brain from dorsal (the top of the head in B) to more ventral areas. Planes of sections are given in panel L. Panels A, E and I show Nissl stained sections corresponding approximately to the dark field images in B-D, F-H and J-K, respectively. The most dorsal section (B) shows no hybridisation. Clear Nkr-5.2 signals are seen in the mesencephalon (CL mes), and two hypothalamic regions designated as mammillary (ma) and preoptic (poa) areas in panel D. At the level of section F the broadest signal distribution is observed in the hypothalamus (ma, poa) with a clear restriction to the preoptic area at the more ventral level (G). There is also hybridisation in the pons (F; pn) and the medulla oblongata (G, H; me). Nkr-5.2 transcripts are also detected at the more ventral level of the medulla (J, K) and in the semicircular canals (SC) of the inner ear (H. J).

brain reveals V-shaped Nkx-5.1 positive bundels in the preoptic area and two symetric stripes of expression in the mammillary region and in the medulla oblongata. In contrast, the same areas of the newborn brain exhibit only

barely detectable Nkx-5.1 expression (Fig. 7C). Expression domains located more dorsally cannot be recognized easily by this technique and are shown on the scheme (Fig. 7D). Whole mount hybridisation of the inner ear

316

S. Rinkwitz-Brandt

et al. I Mechunisms

of Development 52

(1995) 371-381

Fig. 4. The hypothalamic distribution of the Nkx-5.2 transcripts on mid-sagittal sections from E14.5 brain (A) shows Nissl stained section, the hybridising structures are designated. (B-D) Dark field images of sections hybridised with Nkx-5.2. Section B is slightly lateral to C and D, which are medial. N/U-5.2 expression is localized in the preoptic area (port), the mammillary area (ma), and the ventral hypothalamus (ht).

reveals strong N~J-5. I expression in all three semicircular canals (Fig. 7B). Similar results as shown here for Nkx5.1 were also obtained for Nkx-5.2 (data not shown). 2.3. Nkx-5. I gene expression

begins in the otic placode

and in the first and second branchial

arch at E8.5

In a previous study, the first Nkx-5.1 transcripts were detected by in situ hybridisation on tissue sections of E10.5 embryos. Reinvestigation on whole-mounted embryos now revealed that Nkx-5.1 expression sets in at E8.5 with one domain in the otic placode or vesicle and another one in the branchial arches (Fig. 8). At the lo12 somite stage, the first Nkr-5.1 transcripts appear in the invaginating otic placode which is already clearly discernible at this stage (Fig. 8B). Nkx-5.1 positive cells seem confined to the single cell layer of columnar epithelium within the placode. To the best of our knowledge, Nkx-5.1 constitutes the first molecular marker specific for this early stage of otic development. From E8.5 to E9.5 the invagination of the otic placode continues and leads to formation of the otic vesicle. During this period Nkx-5.1 distribution undergoes a reorganization. While Nkx-5.1 transcripts are initially distributed uniformly throughout the epithelium of the otic placode, they later appear to concentrate at the anterior and posterior margins, and at the dorsal side of the otic vesicle, whereas the ventral side is essentially free of signals (Fig. 8D,E; see also Fig. 9A).

The other expression domain of Nkx-5.1 in the branchial arches appears approximately at the same time as in the otic placode. At 10 somite stage Nkx-5. I hybridisation is detected at the ventral side of the embryo (Fig. lB, arrow) with increasing signal intensity at later stages (Fig. 8C). At E9.0 (16 somites) the signal is V-shaped in the cleft formed by the posterior edge of the first and the anterior edge of the second branchial arch (Fig. 8C, small arrows). The relative intensity of the Nkx-5.1 hybridisation signals in the otic vesicle and the branchial region varies during early developmental stages. While Nkx-5.1 expression levels appear initially similar in both structures (Fig. 8B), in 16 somite embryos expression in branchial arches increases relative to the otic vesicle (Fig. 8C). At later stages, beyond the 20 somite stage, the signal in the branchial arches drastically decreases relative to the otic vesicle and is visualized only after prolonged staining (Fig. 8E,F). 2.4. Nkx-5.1 is expressed in the early otic vesicle in regions different from Pax-2 and sek genes

It has been demonstrated

that different parts of the otolater into different structural components of the inner ear (Li et al., 1978). We compared Nkx-5.1 transcript distribution with the expression of Pax-2 (Dressler et al., 1990) and sek (Nieto et al., 1992), two other mouse genes expressed during cyst acquire specific fates and develop

S. Rinkwitz-Brundt et ul. I Mechanisms ofDevelopment 52 (1995) 371-381

311

Fig. 5. Nkx-5.2 expression domains in the E17.5 brain. Coronal sections through the head in four different planes from rostra1 to caudal were hybridised with iVkx-5.2probe. Each panel (A-D) shows a Nissl stained section to the left and corresponding dark field images to the right. Planes of sections (A-D) are indicated schematically in (E). (mlpo) medial preoptic nucleus, (pvpof periventricular preoptic nucleus, (dmh) dorsomedial hypothalamic nucleus, (prm) premammillary nucleus, (arc) arcuate nucleus, (me) medulla oblongata. otic development. Already at the 20 somite stage transcripts for Nkx-5.1, Pax-2, and sek are located in different regions of the otocyst (Fig. 9A-C respectively). Both Nkx-5.1 and sek are expressed in the dorsal part of the otocyst, while Pax-2 expression is confined to its ventral wall. Nkx-5.1 transcripts are also present at the anterior and posterior poles of the otocyst. Thus, Pax-2 and Nku-5.1 are expressed on opposite sites in a complementary pattern with a possible slight overlap at the common border. sek seems expressed in a subset of dorsal Nkx-5.1 expressing cells but does not overlap with Pax-2. This distinct transcript distribution for different regulatory genes suggests that patterning processes in the otic vesicle early

may start well before recognizable entiation.

morphological

differ-

3. Discussion This work describes a detailed expression of Nkx-5.1 and Nkr-5.2, two novel homeobox genes of the NKrelated gene family. Using in situ hybridisation on whole-mounted embryo preparations we determined the onset of Nkx-5.1 gene expression at E8.5, the 10-12 somite stage. Until E10.5, transcripts for the Nkx-5.1 gene are restricted to two small regions: the otic placode and parts of the first and second

378

Fig. 6. Nkx-5.2 hybridisation on midsagittal sections through the head of an E17.5 embryo. Nkx-5.2 expression is seen in the preoptic area (poa), the dorsomedial hypothalamic nucleus (dmh), the dorsal part of the premammillary nucleus (prm) and in the medulla oblongata (me). (A) presents a Nissl stained section with Nkx-5 expressing structures marked; (B) and (C) are two parallel sections showing the hybridising structures in dark field illumination; (D) planes of sections.

branchial arches. Later, distinct CNS structures also express this gene. Interestingly, the Nkx-5.2 gene which is closely linked to Nkx-5.1 on mouse chromosome 7 (Bober et al., 1994a) is first activated at E13.5 as estimated by in situ hybridisation techniques. 3.1. Distinct domains in the developing

brain express

both Nkx-5 genes

Transcript distribution of several homeobox genes, such as dlx, otx, emx, gbx, Nkx-2 and genes of the wnt gene family correlates with neuromeric boundaries and thus supports a concept of segmental origin of the forebrain (Figdor and Stern, 1993; Puelles and Rubenstein, 1993; Rubenstein et al., 1994; Guthrie, 1995). The diencephalic expression of the Nkx-5 genes also seems to respect these boundaries confining them to neuromere Dl according to the designation of Figdor and Stern (1993) or to the ventral parts of neuromeres P4 and P5 according to Rubenstein et al. (1994). In this region Nkx-5 transcripts may overlap with transcripts for Dlx and Nkx-2.1 (Puelles and Rubenstein, 1993). However, the onset of Nkr-5 expression clearly occurs 2-4 days later. Therefore, it seems unlikely that Nkx-5 genes play a role in the establishment of neuromeric identities as was suggested for these earlier expressed regulatory genes. Significantly, the onset of Nkx-5. I transcription correlates with first cellular differentiation processes in the hypothalamus (Altman and Bayer, 1988). Therefore, as already suggested (Bober

et al., 1994a), the Nkx-5 genes may play a role in neuronal cell specification. Despite structural similarity and the regional overlap of Nkx-5.1 and Nkx-5.2 expression, both genes may fulfill different functions as the onset of their expression is quite different. This is reminiscent of the Drosophila genes tinman (NK-4) and bag pipe (NK3), another pair of closely linked NK genes. These genes are also activated consecutively during Drosophila development and play a role in the specification of mesodermal cell lineages (Bodmer et al., 1990; Dohrmann et al., 1990; Azpiazu et al., 1993). 3.2. Nkx-5. I represents an early marker for otic development

The first morphologically identifiable structure of the prospective ear is the otic plate at the 2-3 somite stage (E8.0). The columnar epithelium of the otic placode becomes discernible approximately 12 h later at the 8-10 somite stage. After an additional 12 h (13 to 20 somites) rapid invagination of the otic placode can be observed eventually resulting in the formation of the otic vesicle at the 20-24 somite stage (E9.5) (Sher, 1971; Anniko and WillstrGm, 1984). Transcripts for Nlur-5.1 are already detectable by in situ hybridisation in the forming otic placode (10 somites). Therefore, Nkx-5.1 represents the earliest known molecular marker for otic development described so far. It remains to be established whether Nkx5.1 fulfills any critical function(s) during the initial steps

S. Rinkwitz-Brand?

et al. I Mechanisms

qf Development 52 (1995) 371-381

379

Fig. 7. Nkr-5.1 expression in whole mount brain preparations of an E15.5 embryo (A) and a l-day neonate (C) (ventral view) and in the inner ear prepared from E15.5 (B). Arrows and arrow head in (C) point at hybridising structures which correspond to those indicated in (A). (D) shows schematically all Nkx-5 expression domains in fetal brain, (poa) preoptic area, (ma) mammillaty area, (me) medulla, (psc) posterior semicircular canal, (asc) anterior semicircular canal, (co) cochlea, (pn) pons, (mes) mesencephalon, (ht) hypothalamus, (dt) dorsal thalamus, (vt) ventral thalamus, (III) and (IV) brain ventricles.

of inner ear development. Functional importance of the Nkx-5.1 gene for otic development can be tested in mice lacking the gene product. Knock-out mice generated for other members of the NK-gene family have demonstrated significant roles for cell and organ specification. For example, the Nkx-5 related gene Hox-II is essential for spleen development (Roberts et al., 1994). The nucleotide sequence of the Hox-II homeobox shows 64% identity with the corresponding Nkx-5.1 domain with sequence homology extending several amino acids downstream of the homeobox. Another example is the Nkx-2.5 gene which is important for proper heart development (Harvey, 1994). Therefore, one can expect that Nkx-5.1 might also be essential for the formation of the organs in which it is predominantly expressed. 3.3. Nlcx-5.1 expression in branchial arches The expression in the first branchial cleft does not correlate to any morphologically or functionally defined regions. The epibranchial placodes which are also located at

the cleft between first and second branchial arch seem to be more proximal than the Nkx-5.1 expression domain (Webb and Noden, 1993). Interestingly, based on the Hox-a2 -I- phenotype in mice, it has been postulated that zones of polarizing activity should exist at the edges of the first two branchial arches (Gendron-Maguire et al., 1993; Rijli et al., 1993). However, no gene products have been identified so far to be responsible for this signalling. It remains to be established whether the Nkx-5.1 protein is involved in these patterning processes. The expression of the gooscoid gene also correlates with this putative polarizing zone (Gaunt et al., 1993), albeit gooscoid is activat ed in this region almost two days later than Nkx-5.1. Hence, if both genes play a role in the putative signal cascade, Nkx-5.1 would act upstream of gooscoid and could be involved in the generation of the polarizing activity with gooscoid being one of the responding genes involved in determination of specific regional structures,

380

S. Rinkwitz-Brandt et (11.I Mechanismsqf Development52 (1995) 371-381

Fig. 8. Expression of Nkx-5.1 in mouse embryos during E8.0 to El 1.O. In situ hybridizations with Nkx-5.1 digoxygenin-labelled probe on wholemounted embryo preparations. (A) E8.0 (4-6 somites) shows no signal; (B) ES.5 (8-10 somites) shows hybridisation in the otic placode and the branchial arch region (arrow); (C) E9.0 (16 somites) reveals strong signal in the cleft between first and second branchial arches (small arrows) and weak hybridisation at the anterior tip of the otic vesicle (arrow) (D) E9.5 (22 somites) shows strong hybridisation in the otic vesicle detected after short staining reaction; (E and F) El 1.0 (34 somites) reveal hybridisation in the otic vesicle after short staining (E) and additional hybridisation in the branchial arches (small arrows) after prolonged staining reaction (F).

4. Material and methods 4. I. cONA cloning and sequencing The Nkx-5.2 cDNA was isolated from a lambda gtll cDNA library prepared from an El 1.5 mouse embryo (Clontech). A previously characterized genomic Nkx-5.2 fragment containing the homeobox was used as probe (Bober et al., 1994a). The longest cDNA fragment (1470 bp) obtained in this screen was subcloned into KS Bluescript vector (Stratagene) to prepare riboprobes. Dideoxy-sequencing was performed in Ml3 vectors. 4.2. In situ hybridisation on tissue sections In situ hybridizations were performed on 1Opm thick paraffin tissue sections as described previously (Bober et al., 1991, 1994a). The Nkx-5. I antisense RNA probe was prepared as described (Bober et al., 1994a); Nkx-5.2 hy-

bridizations were performed with the antisense RNA transcript corresponding to 600 bp of the extreme 3’sequences of the Nkx-5.2 cDNA (Fig. 8B). 4.3. In situ hybridisation on whole-mounted

embryos and

brain preparations

Embryos and brain preparations from NMRI mice were processed for whole mount in situ hybridisation as described previously (Bober et al., 1994b). Probes for whole mount hybridizations were synthesized using digoxygenin-labelled nucleotides (Boehringer Mannheim) on cDNA templates as described above. Acknowlegments We are grateful to P. Gruss (Gottingen) and D. Wilkinson (London) for the Pax-2 and sek probes, respectively. We thank T. Braun for fruitful discussions and critical

S. Rinkwitz-Brandt et al. /Mechanisms

ofDevelopmenr 52 (1995) 371-381

Fig. 9. Nkx-5.1, Pax-2 and sek are expressed in different regions of the otic vesicle. Whole-mounted E9.5 (24 somites) embryos were hybridised Nkx-5.1 (A), Pax-2 (B) and sek (C) probes. Anterior (a), posterior (p), dorsal (d) and ventral (v) sides of the otic vesicle as well as the mandibular (md) are marked.

reading of the manuscript. The excellent technical assistance of S. Hoffmann is gratefully acknowleged. We also thank T. Hadrys for technical help. This work was supported by Deutsche Forschungsgemeinschaft, grant No. SFB 271, Teilprojekt Bl. S. R.-B. was supported by the Boehringer Ingelheim Fonds. References Altman, J. and Bayer, S.A. (1988) J. Comp. Neurol. 275, 346-377. Anniko, M. and Willstrom, S.-O. (1984) Am. J. Otolaryngol. 5, 373381. Azpiazu, N. and Frasch, M. (1993) Genes Dev. 7, 1325-1340. Bober, E., Lyons, G.E., Braun, T., Cossu, G., Buckingham, M. and Arnold, H.H. (1991) J. Cell Biol. 113, 1255-1265. Bober, E., Baum, C., Braun, T. and Arnold, H.H. (1994a) Dev. Biol. 162,288-303. Bober, E., Franz, T., Arnold, H.H., Gruss, P. and Tremblay, P. (1994b) Development 120,603-612. Bodmer, R., Jan, L.Y. and Jan, Y.N. (1990) Development 110, 661669. Bulfone, A. (1993) J. Neurosci. 13, 3155-3172. Doe, Q.C., Hiromi, Y., Gehring, W.J. and Goodman, C.S. (1987) Science 239, 170-175. Doe, Q.C., Smouse, D. and Goodman, C.S. (1988) Nature 333. 376378. Dohrmann, C., Azpiazu, N. and Frasch, M. (1990) Genes Dev. 4,20982111. Dressier, G.R., Deutsch, U., Chowdury, K., Nomes, H.O. and Gruss, P. (1990) Development 109,787-795. Duboule, D. (1992) BioEssays 14,375-384. Figdor, M.C. and Stem, C.D. (1993) Nature 363,630-634.

with arch

Gaunt, S.J., Blum, M. and De Robertis. E.M. (1993) Development 117, 769-778. Gendron-Maguire, M., Mallo, M., Zhang, M. and Gridley, T. (1993) Cell 75.1317-1331. Guthrie, S. (1995) Trends Neurosci. 18.74-79. Harvey, R.P., Lyons, I., Li, R., Parsons, L.M., Hartley, L., Andrews, J. and Smith, M. (1994) J. Cell. Biochem. (suppl. 18D). Wll3. Heuer, J.G. and Kaufman, T.C. (1992) Development 115.3547. Hunt, P. and Krumlauf, R. (1992) Annu. Rev. Cell Biol. 8, 227-256. Kim, Y. and Nirenberg, M. (1989) Proc. Natl. Acad. Sci. USA 86, 7716-7720. Krumlauf, R. (1993) Curr. Opin. Genet. Dev. 3.621625. Li, C.W., Van de Water, T.R. and Ruben, R.J. (1978) J. Morphol. 157, 249-268. Lints, T.J., Parsons, L.M., Hartley, L.. Lyons. I. and Harvey, R.P. (1993) Development 119,419-43 1. Nieto, M.A., Gilardi-Hebenstreit, P., Chamay, P. and Wilkinson, D.G. (1992) Development 116, 1137-1150. Price, M., Lazzaro, D., Pohl, T., Mattei, M.-G., Ruther, U., Olivio, J.C., Duboule, D. and Di Lauro, R. (1992) Neuron 8,24l-255. Puelles, L. and Rubenstein, J.L.R. (1993) Trends Neurosci. 16, 472479. Rijli, F.M., Mark, M., Lakkaraju, S., Dierich, A., Dolle, P. and Chambon, P. (1993) Cell 75, 1333-1349. Roberts, C.W., Shutter, J.R. and Korsmeyer, S.J. (1994) Nature 368, 747-749. Rubenstein, J.L.R., Martinez, S., Shimamura. K. and Puelles. L. (1994) Science 266.578-580. Sher, A.E. (197 1) Acta Otolaryngol. Suppl. 285, l-77. Simeone, A., Acampora, D., Gulisano, M., Stornaiuolo, A. and Boncinelli, E. (1992a) Nature 358, 687-690. Simeone, A., Gulisano, M., Acampora, D., Stornaiuolo, A., Rambaldi, M. and Boncinelli, E. (1992b) EMBO J. 11.2541-2550. Webb, J.F. and Noden, D.M. (1993). Am. Zool. 33,434-447.