(en)The invention relates to novel neurogenin proteins, nucleic acids and antibodies.

1.ApplicationNumber: US-94193610-A
1.PublishNumber: US-2011287480-A1
2.Date Publish: 20111124
3.Inventor: ANDERSON DAVID J.
MA QIUFA
SOMMER LUKAS

4.Inventor Harmonized: ANDERSON DAVID J(US)
MA QIUFA(US)
SOMMER LUKAS(US)

5.Country: US
6.Claims:
(en)The invention relates to novel neurogenin proteins, nucleic acids and antibodies.

7.Description:
(en)RELATED APPLICATIONS
This application is a continuation application of U.S. patent application Ser. No. 12/024,558, which is a divisional application of U.S. patent application Ser. No. 10/425,259, filed Apr. 29, 2003 which is a continuation of U.S. patent application Ser. No. 08/932,411, filed on Sep. 17, 1997, which is a continuation of U.S. patent application Ser. No. 08/772,009, filed on Dec. 19, 1996; which is a continuation of U.S. patent application Ser. No. 08/722,570, filed Sep. 27, 1996. The contents of these applications are incorporated herein by reference in their entirety.


FIELD OF THE INVENTION
The invention relates to novel neurogenin proteins, nucleic acids and antibodies.
BACKGROUND OF THE INVENTION
Transcription factors in the basic-helix-loop-helix (bHLH) family have been shown to play a central role in cell type determination, in several tissues and organisms. For example, MyoD and myf5 are necessary and sufficient for mammalian myogenesis, while the proneural genes achaete-scute and atonal perform similar functions during Drosophila neurogenesis (for reviews, see (Weintraub, 1993, Cell. 75, 1241-1244; Jan and Jan, (1994) Ann. Rev. of Genet. 28, 373-393.). In both vertebrate myogenesis and fly neurogenesis, multiple functionally-interchangeable bHLH proteins act in networks and/or cascades within their respective lineages (Jan and Jan, 1993). For example, at least four different bHLH proteins are sequentially expressed during murine muscle development: MyoD/myf5; myogenin and MRF4 (Olson and Klein, 1994). Similarly in Drosophila expression of achaete-scute is followed by that of asense during peripheral neurogenesis (Brand et al., 1993; Dominguez and Campuzano, 1993; Jarman et al., 1993a). The function of such cascades is not yet clear, although it has been suggested that the later-acting genes function in differentiation rather than in determination (Weintraub, 1993; Lee et al., 1995). Although numerous bHLH proteins expressed during vertebrate neurogenesis have been identified (Johnson et al., 1990; Akazawa et al., 1992; Ferreiro et al., 1992; Sasai et al., 1992; Ishibashi et al., 1993; Turner and Weintraub, 1994; Akazawa et al., 1995; Lee et al., 1995; Shimizu et al., 1995), none so far examined exhibits the functional properties expected of a neuronal determination factor ((Guillemot et al., 1993; Sommer et al., 1995); see below).
One feature that characterizes the proneural genes in Drosophila is their interaction with the genetic circuitry underlying lateral inhibition. Lateral inhibitory interactions between neuroectodermal cells, mediated by the products of the neurogenic genes Notch and Delta, result in the selection of a single sensory organ precursor (SOP) cell from a group of developmentally equivalent undetermined cells called a “proneural cluster” (Ghysen et al., 1993). All cells in the proneural cluster initially express achaete-scute, but during the selection process proneural gene expression becomes restricted at high levels to the SOP (Cubas et al., 1991). This restriction occurs because the proneural genes promote expression of Delta (Hinz et al., 1994; Kunisch et al., 1994), and their expression and function are in turn inhibited by signalling through Notch (for review, see (Ghysen et al., 1993)). Thus, cells which express sufficient achaete-scute, and hence Delta, to inhibit proneural activity in their neighbors adopt an SOP fate (for discussion, see (Chitnis, 1995)). In this way, the proneural genes both promote a neural fate cell-autonomously, and inhibit this fate non-autonomously.
Lateral inhibition mediated by vertebrate homologs of Notch and Delta has recently been demonstrated to regulate primary neurogenesis in Xenopus (Chitnis et al., 1995). Primary neurons differentiate in three parallel rows within the neural plate; between these rows undifferentiated neural plate cells are set aside for later waves of neurogenesis. Expression of X-Delta-1 defines three broader longitudinal domains that prefigure these territories of primary neurogenesis. Ectopic expression of a dominant negative form of X-Delta-1 (X-Delta-1 Stu ) increases the density of neurons that differentiate within each territory, but does not increase the width of each territory or the overall area of the neural plate (Chitnis et al., 1995). Conversely, expression of constitutively active forms of X-Notch-1 suppresses primary neurogenesis (Coffman et al., 1990; Coffman et al., 1993). These data suggest that the three territories of primary neurogenesis in Xenopus (medial, intermediate and lateral) are analogous to proneural clusters in Drosophila . This in turn implies the existence of one or more bHLH proteins whose expression defines these prospective neurogenic territories.
Several bHLH proteins expressed during Xenopus neurogenesis have been identified. One such protein, NeuroD, can exert a neuronal determination function when ectopically expressed, but the endogenous XNeuroD gene is not expressed early enough to play a proneural role (Lee et al., 1995). Several Xenopus homologs of achaete-scute have also been identified
(Ferreiro et al., 1992; Zimmerman et al., 1993; Turner and Weintraub, 1994). Ectopic expression of one of these, XASH-3, can induce neural plate expansion (Ferreiro et al., 1994; Turner and Weintraub, 1994) or ectopic neurogenesis within the neural plate (Chitnis and Kintner, 1996), depending on the dose of injected RNA. Unlike NeuroD, however, XASH-3 is incapable of converting epidermal cells to neurons. Moreover, XASH-3 is expressed in a very restricted region of the neural plate, corresponding to the future sulcus limitans (Zimmerman et al., 1993). Thus, there must be other bHLH genes whose expression pattern and function are more consistent with proneural activity.
Accordingly, it is an object of the invention to provide such bHLH genes such as the neurogenin family. Thus, the invention provides recombinant neurogenin proteins and variants thereof, and to produce useful quantities of these neurogenin proteins using recombinant DNA techniques.
It is a further object of the invention to provide recombinant nucleic acids encoding neurogenin proteins, and expression vectors and host cells containing the nucleic acid encoding the neurogenin protein.
An additional object of the invention is to provide polyclonal and monoclonal antibodies directed against neurogenin proteins.
A further object of the invention is to provide methods for producing the neurogenin proteins.
SUMMARY OF THE INVENTION
In accordance with the objects outlined above, the present invention provides recombinant nucleic acids encoding neurogenin proteins.
In a further aspect, the invention provides expression vectors comprising transcriptional and translational regulatory DNA operably linked to DNA encoding a neurogenin protein, and host cells containing the expression vectors.
In an additional aspect, the invention provides methods for producing neurogenin proteins comprising the steps of culturing a host cell transformed with an expressing vector comprising a nucleic acid encoding a neurogenin protein and expressing the nucleic acid to produce a neurogenin protein.
In a further aspect, the invention provides recombinant neurogenin proteins.
In a further aspect, the invention provides polyclonal or monoclonal antibodies to neurogenin proteins.



BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 depicts the alignment of the amino acid sequences of rat neurogenin-1 (top) (SEQ ID NO:1) and Xenopus (bottom; labelled X-NGNR-1.a) neurogenin-1 (SEQ ID NO:2). The bHLH region is marked in bold type. Solid lines indicated amino acid identity, the dots are conservative substitutions. Selection of the initiator methionine was according to Kozak's rule (Kozak, 1984), and identification of in-frame upstream termination codons (data not shown).
FIG. 2 depicts the alignment of the rat neurogenin-1 bHLH domain (SEQ ID NO:3) with other bHLH domains (SEQ ID NOS:4-11). Identity is shown in bold type. References for the compared sequences are: NeuroD (Lee et al., 1995)/BETA2 (Naya et al., 1995) (SEQ ID NO:5), MATH-2/Nex-1 (Bartholoma and Nave, 1994; Shimizu et al., 1995) (SEQ ID NO:6), MATH-1 (Akazawa et al., 1995) (SEQ ID NO:8), KW8 (Kume et al., 1996) (SEQ ID NO:7), Drosophila atonal (Jarman et al., 1993b)(SEQ ID NO:9), MASH1 (Johnson et al., 1990) (SEQ ID NO:10), AS-C T5 (Villares and Cabrera, 1987) (SEQ ID NO:11).
FIG. 3 depicts the nucleic acid sequence of rat neurogenin 1 (SEQ ID. NO:12); the protein sequence is shown in SEQ ID NO:1.
FIG. 4 depicts the nucleic acid sequence of mouse neurogenin 1 (SEQ ID. NO:13); the elucidated protein sequence is shown in SEQ ID NO:14.
FIG. 5 depicts the nucleic acid sequence of one of the Xenopus neurogenin 1 (SEQ ID NO:15) isoforms; the elucidated protein sequence is shown in SEQ ID NO:2.
FIG. 6 depicts the nucleic acid sequence of the other Xenopus neurogenin (SEQ ID NO:16) isoform; the protein sequence may be easily determined.
FIGS. 7A , 7 B and 7 C depict the nucleic acid sequence of mouse neurogenin 2 (SEQ ID. NO:17).
FIG. 8 depicts the amino acid sequence of mouse neurogenin 2 (SEQ ID NO:18). The bHLH domain is boxed.
FIGS. 9A and 9B depict the nucleic acid sequence of mouse neurogenin 3 (SEQ ID NO:19).
FIG. 10 depicts the amino acid sequence of mouse neurogenin 3 (SEQ ID NO:20). The bHLH domain is boxed.
FIG. 11 depicts the alignment of neurogenin 2 (SEQ ID NO:18) with Xenopus (bottom; labelled X-NGNR-1.a) (SEQ ID NO:2). The bHLH region is marked in bold type. Solid lines indicated amino acid identity, the dots are conservative substitutions. The two proteins exhibit 54% overall sequence identity and 68% overall sequence similarity to each other. Their bHLH domains (boxed) are 87% identical. Neurogenin 1 and neurogenin 3 exhibit similar extents of identity to X-NGNR-1.a within the bHLH domain, but exhibit 41 and 39% overall sequence identity with the Xenopus protein, respectively.



DETAILED DESCRIPTION OF THE INVENTION
The present invention provides novel neurogenin proteins. In a preferred embodiment, the neurogenin proteins are from vertebrates, more preferably from mammals, and in the preferred embodiment, from rats, mice or humans. However, using the techniques outlined below, neurogenin proteins from other organisms may also be obtained.
There are a number of closely related neurogenin proteins and nucleic acids that are all considered neurogenin proteins. As outlined below, there are at least three separate yet related neurogenin proteins in the mouse, neurogenin 1, neurogenin 2 and neurogenin 3. Neurogenin 1 has also been found in rat and Xenopus , and it is expected that at least all three neurogenin proteins are found in a number of organisms.
A neurogenin protein of the present invention may be identified in several ways. “Neurogenin” includes neurogenin 1, 2 or 3. A neurogenin nucleic acid or neurogenin protein is initially identified by substantial nucleic acid and/or amino acid sequence homology to the sequences shown in the Figures, including homology to neurogenin 1, 2 or 3. Such homology can be based upon the overall nucleic acid or amino acid sequence.
As used herein, a protein is a “neurogenin protein” if the overall homology of the protein sequence to the amino acid sequences of the neurogenins depicted herein is preferably greater than about 35% to 40%, more preferably greater than about 60% and most preferably greater than 80%. In some embodiments the homology will be as high as about 95 or 98%. This homology will be determined using standard techniques known in the art, such as the Best Fit sequence program described by Devereux et al., Nucl. Acid Res. 12:387-395 (1984) or the BLASTX program (Altschul et al., J. Mol. Biol. 215, 403-410). The alignment may include the introduction of gaps in the sequences to be aligned. In addition, for sequences which contain either more or fewer amino acids than the proteins disclosed herein, it is understood that the percentage of homology will be determined based on the number of homologous amino acids in relation to the total number of amino acids. Thus, for example, homology of sequences shorter than that shown in the Figures, as discussed below, will be determined using the number of amino acids in the shorter sequence.
Neurogenin proteins may be identified in one aspect by significant homology to the areas other than the bHLH domain, i.e. the N- and C-terminal portions of the sequences depicted in the Figures. This homology is preferably greater than about 40%, with greater than about 50 or 60% being particularly preferred and greater than about 80% being especially preferred. In some cases the homology will be greater than about 90 to 95 or 98%.
In addition, a neurogenin protein preferably also has significant homology to the neurogenin bHLH domain as described herein. This homology is preferably greater than about 75%, with greater than about 80% being particularly preferred and greater than about 85% being especially preferred. In some cases the homology will be greater than about 90 to 95 or 98%.
Neurogenin proteins of the present invention may be shorter or longer than the amino acid sequences shown in the Figures. Thus, in a preferred embodiment, included within the definition of neurogenin proteins are portions or fragments of the sequences depicted herein. Fragments of neurogenin proteins are considered neurogenin proteins if a) they share at least one antigenic epitope; b) have at least the indicated homology; c) and preferably have neurogenin biological activity.
Neurogenin proteins may also be identified as being encoded by neurogenin nucleic acids. Thus, neurogenin proteins are encoded by nucleic acids that will hybridize to the sequences depicted in FIGS. 3 (SEQ ID NO:12), 4 (SEQ ID NO:13), 5 (SEQ ID NO: 15), or 6 (SEQ ID NO: 16), as outlined herein.
In a preferred embodiment, when the neurogenin protein is to be used to generate antibodies, the neurogenin protein must share at least one epitope or determinant with one or more of the full length proteins depicted herein. By “epitope” or “determinant” herein is meant a portion of a protein which will generate and/or bind an antibody. Thus, in most instances, antibodies made to a smaller neurogenin protein will be able to bind to the full length protein. In a preferred embodiment, the epitope is unique; that is, antibodies generated to a unique epitope show little or no cross-reactivity, for example, with other bHLH domains. The neurogenin antibodies of the invention specifically bind to neurogenin proteins. By “specifically bind” herein is meant that the antibodies bind to the protein with a binding constant in the range of at least 10 4 -10 6 M −1 , with a preferred range being 10 4 -10 9 M −1 .
In the case of the nucleic acid, the overall homology of the nucleic acid sequence is commensurate with amino acid homology but takes into account the degeneracy in the genetic code and codon bias of different organisms. Accordingly, the nucleic acid sequence homology may be either lower or higher than that of the protein sequence. Thus the homology of the nucleic acid sequence as compared to the nucleic acid sequences of FIGS. 3 (SEQ ID NO:12) 4 (SEQ ID NO:13), 5 (SEQ ID NO:15), and 6 (SEQ ID NO:16) is preferably greater than 40%, more preferably greater than about 50%, particularly greater than about 60% and most preferably greater than 75%. In some embodiments the homology will be as high as about 80 to 90 to 95 or 98%.
In a preferred embodiment, a neurogenin nucleic acid encodes a neurogenin protein. As will be appreciated by those in the art, due to the degeneracy of the genetic code, an extremely large number of nucleic acids may be made, all of which encode the neurogenin proteins of the present invention. Thus, having identified a particular amino acid sequence, those skilled in the art could make any number of different nucleic acids, by simply modifying the sequence of one or more codons in a way which does not change the amino acid sequence of the neurogenin.
In one embodiment, the nucleic acid homology is determined through hybridization studies. Thus, for example, nucleic acids which hybridize under high stringency to the nucleic acid sequences shown in the Figures or their complements are considered neurogenin genes. High stringency conditions are known in the art; see for example Maniatis et al., Molecular Cloning: A Laboratory Manual, 2d Edition, 1989, and Short Protocols in Molecular Biology, ed. Ausubel, et al., both of which are hereby incorporated by reference.
In another embodiment, less stringent hybridization conditions are used; for example, moderate or low stringency conditions may be used, as are known in the art; see Maniatis and Ausubel, supra.
The neurogenin proteins and nucleic acids of the present invention are preferably recombinant. As used herein, “nucleic acid” may refer to either DNA or RNA, or molecules which contain both deoxy- and ribonucleotides. The nucleic acids include genomic DNA, cDNA and oligonucleotides including sense and anti-sense nucleic acids. Such nucleic acids may also contain modifications in the ribose-phosphate backbone to increase stability and half life of such molecules in physiological environments.
The nucleic acid may be double stranded, single stranded, or contain portions of both double stranded or single stranded sequence. By the term “recombinant nucleic acid” herein is meant nucleic acid, originally formed in vitro, in general, by the manipulation of nucleic acid by endonucleases, in a form not normally found in nature. Thus an isolated neurogenin nucleic acid, in a linear form, or an expression vector formed in vitro by ligating DNA molecules that are not normally joined, are both considered recombinant for the purposes of this invention. It is understood that once a recombinant nucleic acid is made and reintroduced into a host cell or organism, it will replicate non-recombinantly, i.e. using the in vivo cellular machinery of the host cell rather than in vitro manipulations; however, such nucleic acids, once produced recombinantly, although subsequently replicated non-recombinantly, are still considered recombinant for the purposes of the invention.
Similarly, a “recombinant protein” is a protein made using recombinant techniques, i.e. through the expression of a recombinant nucleic acid as depicted above. A recombinant protein is distinguished from naturally occurring protein by at least one or more characteristics. For example, the protein may be isolated or purified away from some or all of the proteins and compounds with which it is normally associated in its wild type host, and thus may be substantially pure. For example, an isolated protein is unaccompanied by at least some of the material with which it is normally associated in its natural state, preferably constituting at least about 0.5%, more preferably at least about 5% by weight of the total protein in a given sample. A substantially pure protein comprises at least about 75% by weight of the total protein, with at least about 80% being preferred, and at least about 90% being particularly preferred. The definition includes the production of a neurogenin protein from one organism in a different organism or host cell. Alternatively, the protein may be made at a significantly higher concentration than is normally seen, through the use of a inducible promoter or high expression promoter, such that the protein is made at increased concentration levels. Alternatively, the protein may be in a form not normally found in nature, as in the addition of an epitope tag or amino acid substitutions, insertions and deletions, as discussed below.
Also included with the definition of neurogenin protein are other neurogenin proteins of the neurogenin family, and neurogenin proteins from other organisms, which are cloned and expressed as outlined below. Thus, probe or degenerate polymerase chain reaction (PCR) primer sequences may be used to find other related neurogenin proteins from humans or other organisms. As will be appreciated by those in the art, particularly useful probe and/or PCR primer sequences include the unique areas of the neurogenin nucleic acid sequence. Thus, useful probe or primer sequences may be designed to the bHLH domain or the N- and C-terminal portions of the sequences. As shown in the examples, using the bHLH domain to generate probes can result in the cloning of both neurogenin proteins and related but distinct proteins, and thus if these sequences are used, further screening and/or sequencing may be required to identify neurogenin proteins. As is generally known in the art, preferred PCR primers are from about 15 to about 35 nucleotides in length, with from about 20 to about 30 being preferred, and may contain inosine as needed. The conditions for the PCR reaction are well known in the art.
Once the neurogenin nucleic acid is identified, it can be cloned and, if necessary, its constituent parts recombined to form the entire neurogenin protein nucleic acid. Once isolated from its natural source, e.g., contained within a plasmid or other vector or excised therefrom as a linear nucleic acid segment, the recombinant neurogenin nucleic acid can be further used as a probe to identify and isolate other neurogenin nucleic acids. It can also be used as a “precursor” nucleic acid to make modified or variant neurogenin nucleic acids and proteins.
Using the nucleic acids of the present invention which encode a neurogenin protein, a variety of expression vectors are made. The expression vectors may be either self-replicating extrachromosomal vectors or vectors which integrate into a host genome. Generally, these expression vectors include transcriptional and translational regulatory nucleic acid operably linked to the nucleic acid encoding the neurogenin protein. “Operably linked” in this context means that the transcriptional and translational regulatory DNA is positioned relative to the coding sequence of the neurogenin protein in such a manner that transcription is initiated. Generally, this will mean that the promoter and transcriptional initiation or start sequences are positioned 5′ to the neurogenin protein coding region. The transcriptional and translational regulatory nucleic acid will generally be appropriate to the host cell used to express the neurogenin protein; for example, transcriptional and translational regulatory nucleic acid sequences from Bacillus are preferably used to express the neurogenin protein in Bacillus . Numerous types of appropriate expression vectors, and suitable regulatory sequences are known in the art for a variety of host cells.
In general, the transcriptional and translational regulatory sequences may include, but are not limited to, promoter sequences, ribosomal binding sites, transcriptional start and stop sequences, translational start and stop sequences, and enhancer or activator sequences. In a preferred embodiment, the regulatory sequences include a promoter and transcriptional start and stop sequences.
Promoter sequences encode either constitutive or inducible promoters. The promoters may be either naturally occurring promoters or hybrid promoters. Hybrid promoters, which combine elements of more than one promoter, are also known in the art, and are useful in the present invention.
In addition, the expression vector may comprise additional elements. For example, the expression vector may have two replication systems, thus allowing it to be maintained in two organisms, for example in mammalian or insect cells for expression and in a procaryotic host for cloning and amplification. Furthermore, for integrating expression vectors, the expression vector contains at least one sequence homologous to the host cell genome, and preferably two homologous sequences which flank the expression construct. The integrating vector may be directed to a specific locus in the host cell by selecting the appropriate homologous sequence for inclusion in the vector. Constructs for integrating vectors are well known in the art.
In addition, in a preferred embodiment, the expression vector contains a selectable marker gene to allow the selection of transformed host cells. Selection genes are well known in the art and will vary with the host cell used.
The neurogenin proteins of the present invention are produced by culturing a host cell transformed with an expression vector containing nucleic acid encoding a neurogenin protein, under the appropriate conditions to induce or cause expression of the neurogenin protein. The conditions appropriate for neurogenin protein expression will vary with the choice of the expression vector and the host cell, and will be easily ascertained by one skilled in the art through routine experimentation. For example, the use of constitutive promoters in the expression vector will require optimizing the growth and proliferation of the host cell, while the use of an inducible promoter requires the appropriate growth conditions for induction. In addition, in some embodiments, the timing of the harvest is important. For example, the baculoviral systems used in insect cell expression are lytic viruses, and thus harvest time selection can be crucial for product yield.
Appropriate host cells include yeast, bacteria, archebacteria, fungi, and insect and animal cells, including mammalian cells. Of particular interest are Drosophila melanogaster cells, Saccharomyces cerevisiae and other yeasts, E. coli, Bacillus subtilis , SF9 cells, C129 cells, 293 cells, Neurospora , BHK, CHO, COS, and HeLa cells.
In a preferred embodiment, the neurogenin proteins are expressed in mammalian cells. Mammalian expression systems are also known in the art. A mammalian promoter is any DNA sequence capable of binding mammalian RNA polymerase and initiating the downstream (3′) transcription of a coding sequence for neurogenin protein into mRNA. A promoter will have a transcription initiating region, which is usually placed proximal to the 5′ end of the coding sequence, and a TATA box, using a located 25-30 base pairs upstream of the transcription initiation site. The TATA box is thought to direct RNA polymerase II to begin RNA synthesis at the correct site. A mammalian promoter will also contain an upstream promoter element (enhancer element), typically located within 100 to 200 base pairs upstream of the TATA box. An upstream promoter element determines the rate at which transcription is initiated and can act in either orientation. Of particular use as mammalian promoters are the promoters from mammalian viral genes, since the viral genes are often highly expressed and have a broad host range. Examples include the SV40 early promoter, mouse mammary tumor virus LTR promoter, adenovirus major late promoter, herpes simplex virus promoter, and the CMV promoter.
Typically, transcription termination and polyadenylation sequences recognized by mammalian cells are regulatory regions located 3′ to the translation stop codon and thus, together with the promoter elements, flank the coding sequence. The 3′ terminus of the mature mRNA is formed by site-specific post-translational cleavage and polyadenylation. Examples of transcription terminator and polyadenylation signals include those derived from SV40.
The methods of introducing exogenous nucleic acid into mammalian hosts, as well as other hosts, is well known in the art, and will vary with the host cell used. Techniques include the use of viruses such as retroviruses and adenoviruses, dextran-mediated transfection, calcium phosphate precipitation, polybrene mediated transfection, protoplast fusion, electroporation, encapsulation of the polynucleotide(s) in liposomes, and direct microinjection of the DNA into nuclei.
In a preferred embodiment, neurogenin proteins are expressed in bacterial systems. Bacterial expression systems are well known in the art.
A suitable bacterial promoter is any nucleic acid sequence capable of binding bacterial RNA polymerase and initiating the downstream (3′) transcription of the coding sequence of neurogenin protein into mRNA. A bacterial promoter has a transcription initiation region which is usually placed proximal to the 5′ end of the coding sequence. This transcription initiation region typically includes an RNA polymerase binding site and a transcription initiation site. Sequences encoding metabolic pathway enzymes provide particularly useful promoter sequences. Examples include promoter sequences derived from sugar metabolizing enzymes, such as galactose, lactose and maltose, and sequences derived from biosynthetic enzymes such as tryptophan. Promoters from bacteriophage may also be used and are known in the art. In addition, synthetic promoters and hybrid promoters are also useful; for example, the tac promoter is a hybrid of the trp and lac promoter sequences. Furthermore, a bacterial promoter can include naturally occurring promoters of non-bacterial origin that have the ability to bind bacterial RNA polymerase and initiate transcription.
In addition to a functioning promoter sequence, an efficient ribosome binding site is desirable. In E. coli , the ribosome binding site is called the Shine-Delgarno (SD) sequence and includes an initiation codon and a sequence 3-9 nucleotides in length located 3-11 nucleotides upstream of the initiation codon.
The expression vector may also include a signal peptide sequence that provides for secretion of the neurogenin protein in bacteria. The signal sequence typically encodes a signal peptide comprised of hydrophobic amino acids which direct the secretion of the protein from the cell, as is well known in the art. The protein is either secreted into the growth media (gram-positive bacteria) or into the periplasmic space, located between the inner and outer membrane of the cell (gram-negative bacteria).
The bacterial expression vector may also include a selectable marker gene to allow for the selection of bacterial strains that have been transformed. Suitable selection genes include genes which render the bacteria resistant to drugs such as ampicillin, chloramphenicol, erythromycin, kanamycin, neomycin and tetracycline. Selectable markers also include biosynthetic genes, such as those in the histidine, tryptophan and leucine biosynthetic pathways.
These components are assembled into expression vectors. Expression vectors for bacteria are well known in the art, and include vectors for Bacillus subtilis, E. coli, Streptococcus cremoris , and Streptococcus lividans , among others.
The bacterial expression vectors are transformed into bacterial host cells using techniques well known in the art, such as calcium chloride treatment, electroporation, and others.
In one embodiment, neurogenin proteins are produced in insect cells. Expression vectors for the transformation of insect cells, and in particular, baculovirus-based expression vectors, are well known in the art.
In a preferred embodiment, neurogenin protein is produced in yeast cells. Yeast expression systems are well known in the art, and include expression vectors for Saccharomyces cerevisiae, Candida albicans and C. maltosa, Hansenula polymorpha, Kluyveromyces fragilis and K lactis, Pichia guillerimondii and P. pastoris, Schizosaccharomyces pombe , and Yarrowia lipolytica . Preferred promoter sequences for expression in yeast include the inducible GAL1,10 promoter, the promoters from alcohol dehydrogenase, enolase, glucokinase, glucose-6-phosphate isomerase, glyceraldehyde-3-phosphate-dehydrogenase, hexokinase, phosphofructokinase, 3-phosphoglycerate mutase, pyruvate kinase, and the acid phosphatase gene. Yeast selectable markers include ADE2, HIS4, LEU2, TRP1, and ALG7, which confers resistance to tunicamycin; the neomycin phosphotransferase gene, which confers resistance to G418; and the CUP1 gene, which allows yeast to grow in the presence of copper ions.
The neurogenin protein may also be made as a fusion protein, using techniques well known in the art. Thus, for example, for the creation of monoclonal antibodies, if the desired epitope is small, the neurogenin protein may be fused to a carrier protein to form an immunogen. Alternatively, the neurogenin protein may be made as a fusion protein to increase expression, or for other reasons.
Also included within the definition of neurogenin proteins of the present invention are amino acid sequence variants. These variants fall into one or more of three classes: substitutional, insertional or deletional variants. These variants ordinarily are prepared by site specific mutagenesis of nucleotides in the DNA encoding the neurogenin protein, using cassette or PCR mutagenesis or other techniques well known in the art, to produce DNA encoding the variant, and thereafter expressing the DNA in recombinant cell culture as outlined above. However, variant neurogenin protein fragments having up to about 100-150 residues may be prepared by in vitro synthesis using established techniques. Amino acid sequence variants are characterized by the predetermined nature of the variation, a feature that sets them apart from naturally occurring allelic or interspecies variation of the neurogenin protein amino acid sequence. The variants typically exhibit the same qualitative biological activity as the naturally occurring analogue, although variants can also be selected which have modified characteristics as will be more fully outlined below.
While the site or region for introducing an amino acid sequence variation is predetermined, the mutation per se need not be predetermined. For example, in order to optimize the performance of a mutation at a given site, random mutagenesis may be conducted at the target codon or region and the expressed neurogenin variants screened for the optimal combination of desired activity. Techniques for making substitution mutations at predetermined sites in DNA having a known sequence are well known, for example, M13 primer mutagenesis and PCR mutagenesis. Screening of the mutants is done using assays of neurogenin protein activities.
Amino acid substitutions are typically of single residues; insertions usually will be on the order of from about 1 to 20 amino acids, although considerably larger insertions may be tolerated. Deletions range from about 1 to about 20 residues, although in some cases deletions may be much larger. It should be noted that some of the neurogenin proteins differ substantially in length; thus, for example, mouse neurogenin 2 is 263 amino acids in length and mouse neurogenin 3 is 211 amino acids.
Substitutions, deletions, insertions or any combination thereof may be used to arrive at a final derivative. Generally these changes are done on a few amino acids to minimize the alteration of the molecule. However, larger changes may be tolerated in certain circumstances. When small alterations in the characteristics of the neurogenin protein are desired, substitutions are generally made in accordance with the following chart:











Original Residue
Exemplary Substitutions





Ala
Ser


Arg
Lys


Asn
Gln, His


Asp
Glu


Cys
Ser


Gln
Asn


Glu
Asp


Gly
Pro


His
Asn, Gln


Ile
Leu, Val


Leu
Ile, Val


Lys
Arg, Gln, Glu


Met
Leu, Ile


Phe
Met, Leu, Tyr


Ser
Thr


Thr
Ser


Trp
Tyr


Tyr
Trp, Phe


Val
Ile, Leu



































Substantial changes in function or immunological identity are made by selecting substitutions that are less conservative than those shown in Chart I. For example, substitutions may be made which more significantly affect: the structure of the polypeptide backbone in the area of the alteration, for example the alpha-helical or beta-sheet structure; the charge or hydrophobicity of the molecule at the target site; or the bulk of the side chain. The substitutions which in general are expected to produce the greatest changes in the polypeptide's properties are those in which (a) a hydrophilic residue, e.g. seryl or threonyl, is substituted for (or by) a hydrophobic residue, e.g. leucyl, isoleucyl, phenylalanyl, valyl or alanyl; (b) a cysteine or proline is substituted for (or by) any other residue; (c) a residue having an electropositive side chain, e.g. lysyl, arginyl, or histidyl, is substituted for (or by) an electronegative residue, e.g. glutamyl or aspartyl; or (d) a residue having a bulky side chain, e.g. phenylalanine, is substituted for (or by) one not having a side chain, e.g. glycine.
The variants typically exhibit the same qualitative biological activity and will elicit the same immune response as the naturally-occurring analogue, although variants also are selected to modify the characteristics of the neurogenin proteins as needed. Alternatively, the variant may be designed such that the biological activity of the neurogenin protein is altered.
In one embodiment, bHLH variants are made. In one embodiment, the bHLH domain may be eliminated entirely. Alternatively, any or all of the amino acids of a bHLH domain may be altered or deleted. In a preferred embodiment, one or more of the amino acids of the domain are substituted by other amino acids. Thus, amino acids corresponding to the neurogenin bHLH domain residues may be altered.
In one embodiment, the neurogenin nucleic acids, proteins and antibodies of the invention are labelled. By “labelled” herein is meant that a compound has at least one element, isotope or chemical compound attached to enable the detection of the compound. In general, labels fall into three classes: a) isotopic labels, which may be radioactive or heavy isotopes; b) immune labels, which may be antibodies or antigens; and c) colored or fluorescent dyes. The labels may be incorporated into the compound at any position.
In a preferred embodiment, the neurogenin protein is purified or isolated after expression. Neurogenin proteins may be isolated or purified in a variety of ways known to those skilled in the art depending on what other components are present in the sample. Standard purification methods include electrophoretic, molecular, immunological and chromatographic techniques, including ion exchange, hydrophobic, affinity, and reverse-phase HPLC chromatography, and chromatofocusing. For example, the neurogenin protein may be purified using a standard anti-neurogenin antibody column. Ultrafiltration and diafiltration techniques, in conjunction with protein concentration, are also useful. For general guidance in suitable purification techniques, see Scopes, R., Protein Purification, Springer-Verlag, NY (1982). The degree of purification necessary will vary depending on the use of the neurogenin protein. In some instances no purification will be necessary.
Once expressed and purified if necessary, the neurogenin proteins are useful in a number of applications as will be apparent to those in the art. For example, the proteins may be used to generate antibodies, which are then useful to purify the protein as outlined above. The antibodies are useful in diagnostic assays to detect neurogenin proteins. The proteins are also useful in neurogenesis.
The following examples serve to more fully describe the manner of using the above-described invention, as well as to set forth the best modes contemplated for carrying out various aspects of the invention. It is understood that these examples in no way serve to limit the true scope of this invention, but rather are presented for illustrative purposes. All references cited herein are incorporated by reference.
EXAMPLES
Example 1
Cloning of Rat and Xenopus Neurogenin 1 Genes
Previous work has identified MASH1 as a bHLH protein expressed in autonomic but not sensory ganglia of the mammalian PNS (Johnson et al., 1990; Lo et al., 1991; Guillemot and Joyner, 1993; Guillemot et al., 1993). We sought to isolate cDNAs encoding bHLH proteins expressed, conversely, in sensory but not autonomic ganglia. Degenerate RT-PCR was performed using cDNA prepared from embryonic day 13.5 (E13.5) rat dorsal root ganglia (DRG), using oligonucleotide primers derived from conserved regions of bHLH subfamilies including MASH1 and NeuroD. Random-primed cDNA template prepared from E13.5 rat embryonic DRGs was subjected to 40 cycles of PCR (1′ @ 94° C.; 2 ′ @ 45 ° C.; 2′ @ 65° C.). The 5′ primers were: 5′CGCGGATCC(A/C)GNAA(C/T)GA(A/G)(A/C)G(G/C/T)GA(A/G)(A/C)3′ (SEQ ID NO:21) and 5′CGCGGATCCGCNAA(C/T)GC(A/C/T)(A/C)G(G/CT)GA(A/G)(A/C)G (SEQ ID NO:22), which were derived from RNERER (SEQ ID NO:29) and ANARER (SEQ ID NO:30), respectively, and contain a BamH1 site at the end. The 3′ primers were 5′CCGGAATTCGT(T/C)TC(A/G/C)A(T/C)(T/C)TT(A/G)CT(A/C/G)A(A/G/T)(T/C)TT3′ (SEQ ID NO:23) and 5′CCGGAATTCGT(T/C)TC(A/G/C)A(T/C)(T/C)TT(A/G/T)GA(A/C/G)A(A/G/T)(T/C)TT3′ (SEQ ID NO:24), both of which are the reverse translation of K(L/M)SK(V/I)ET (SEQ ID NO:25) and contain a EcoR1 site at the end. The 130 bp-PCR product was purified from a polyacrylamide gel, cloned into M13 mp 19 (New England Bio-labs), and sequenced. The ngn1 PCR product was then used to screen a lambda ZAP cDNA library prepared from rat E13.5 DRG (Saito et al., 1995). The isolated 1.2 kb and 1.7 kb cDNA clones were sequenced on both strands by Caltech sequencing core facility. The 1.7 kb cDNA fragment, encoding a predicted protein of 244 amino acid residues, was then used to screen a mouse lambda-2 129 genomic library (a gift of Z. Chen). From a 17-kb positive clone, a 2.0 kb fragment hybridizing to rat ngn1 cDNA was isolated and sequenced, which contains an open reading frame (ORF) differing from the rat one at six positions. The 2.2 kb-X-ngnr-1.a cDNA was isolated by screening a Xenopus St 17 cDNA library (Kintner and Melton, 1987) at low stringency using the rat cDNA as probe. An isoform of X-ngnr-1.a called X-ngnr 1.b was also isolated from this screen, and is depicted in the figures. X-ngnr-1.a and X-ngnr-1.b show the same expression patterns and phenotypes in mRNA injection (data not shown), and are referred to collectively as X-NGNR-1 in the text. Nucleotide sequences of the rat, mouse and two isoforms of the Xenopus neurogenin 1 have been deposited in GeneBank database.
Preliminary experiments indicated that mouse ngn1 mRNA caused ectopic neurogenesis when microinjected into Xenopus embryos (not shown), a phenotype similar to that obtained by over-expression of NeuroD (Lee et al., 1995). To determine whether this phenotype reflected the existence of a Xenopus gene with similar functional characteristics, the murine ngn1 cDNA probe was used to screen a stage 17 (St. 17) Xenopus cDNA library at low-stringency. Several ngn1-related cDNAs were obtained. This cDNA, which we have named Xenopus ngn-related-1(X-ngnr-1), encodes a polypeptide of 215 amino acids displaying 82% sequence identity to mouse neurogenin within the bHLH domain ( FIG. 2 ) (SEQ ID NO:4).
Example 2
Expression of Neurogenin 1 During Neurogenesis
Non-radioactive in situ hybridization to frozen sections of mouse embryos was performed as previously described (Birren et al., 1993). Whole-mount in situ hybridization was performed essentially as described (Chitnis et al., 1995) using digoxigenin-labelled antisense probes for X-NGNR-1, XNeuroD (Lee et al., 1995), X-Delta-1 and N-tubulin (Chitnis et al., 1995), with the following modification. At the step of developing the alkaline phosphatase reaction using NBT/BCIP substrates, 0.45 μl rather than 4.5 μl of NBT stock solution (75 mg/ml in 70% dimethyl formamide) was added to each 1 ml of staining buffer. This change reduced background staining and improved visualization of low-abundance mRNAs.
A preliminary analysis of ngn1 mRNA expression in mouse embryos by in situ hybridization revealed that expression of this gene is apparently restricted to the nervous system (data not shown). Within the nervous system, the expression of neurogenin 1 is spatially or lineally restricted; for example it is expressed in the ventral half of the spinal cord, except for a narrow domain just below the roofplate. In the peripheral nervous system, ngn1 mRNA is expressed in developing sensory but not in autonomic ganglia. Interestingly, a comparison of the expression of ngn1 and NeuroD expression on adjacent serial sections revealed that the two genes appear to be sequentially expressed in overlapping regions or neural cell populations. In the ventral spinal cord, for example, ngn1 mRNA is expressed throughout the ventricular zone, in regions where uncommitted progenitors are located, while NeuroD transcripts are expressed at the lateral border of the ventricular zone. However both genes show a similar dorso-ventral restriction in their domains of expression within the spinal cord (except that NeuroD is not expressed below the roof plate). A similar spatial segregation is seen in the mesencephalic-diencephalic region. These data suggest that neurogenin 1 and NeuroD may function in similar regions of the murine nervous system but at sequential stages of neurogenesis.
A similar spatial overlap but temporal displacement was found for the expression of X-ngnr-1 and XNeuroD in Xenopus . For example, at St. 12, X-ngnr-1 expression is observed in three broad patches within the neural plate, that demarcate the medial, intermediate and lateral territories where primary motorneurons, interneurons and sensory neurons, respectively, will later differentiate. In contrast, no XNeuroD expression is yet detected at this stage, nor have any primary neurons yet differentiated. XNeuroD mRNA can be detected at St. 13.5, in narrow rows of cells located within the three domains of primary neurogenesis which is already underway at this time. At this stage, X-ngnr-1 is expressed in a similar region of the neural plate but in many more cells than XNeuroD. A similar sequential expression of X-NGNR-1 and XNeuroD is seen in the trigeminal placode. Thus in Xenopus and in mouse, expression of neurogenin1/X-NGNR-1 precedes but spatially overlaps that of NeuroD/XNeuroD.
Example 3
Inducement of Neurogenesis
The observation that expression of X-ngnr-1 temporally and spatially prefigures the expression of XNeuroD, taken together with the sequence homology between the two genes, led us to test whether expression of X-NGNR-1 like that of XNeuroD is sufficient to induce premature and/or ectopic primary neurogenesis. We therefore injected X-ngnr-1 mRNA into one blastomere of two-cell stage embryos, and examined the pattern of neurogenesis after further development by whole mount in situ hybridization using a probe for N-tubulin, a neuron-specific marker in Xenopus (Chitnis et al., 1995). The distribution of β-galactosidase activity translated from a co-injected lacZ mRNA, as determined by counterstaining with XGal, was used to assess the overall distribution of the injected mRNAs in each individual embryo. Two types of negative controls were used: the uninjected side of the same embryo, and separate embryos injected only with lacZ mRNA.
The X-NGNR-1 open reading frame was cloned in-frame into the EcoR1 site of the vector pMT-CS2 (Turner and Weintraub, 1994). Capped X-NGNR-1 mRNA was transcribed using SP6 RNA polymerase as described (Kintner and Melton, 1987). X-NGNR-1 mRNA was co-injected with lacZ mRNA (as a marker) into one blastomeres of two-celled embryos (Coffman et al., 1993). Notch ICD and X-Delta-1 Stu RNAs were prepared as described (Chitnis et al., 1995). Injection of LacZ mRNA alone was done as a control. Histochemical staining for β-galactosidase was performed to visualize the distribution of injected mRNAs. Embryos were collected at the neural plate stage (St13-14) or tail-bud stage (St 26) and subjected to in situ hybridization using probes as indicated in the figure legends. Animal cap assays were performed as described previously (Ferreiro et al., 1994).
Over-expression of X-ngnr-1 mRNA caused extensive ectopic neurogenesis within the neural plate (100% of embryos examined; ≧100 embryos injected).
At St. 13.5, for example, nearly every cell on the injected side of the embryo appeared to express N-tubulin, whereas on the uninjected side the three stripes of primary neurogenesis were clearly distinguishable (data not shown); moreover within these stripes the distribution of neurons was more scattered than on the injected side. In addition to the increased extent of neurogenesis, the timing of neuronal differentiation was accelerated on the injected side, so that N-tubulin + cells were seen on the injected side at St. 12.5, a time at which no expression of this marker was detected on the contralateral control side (not shown). Moreover, X-ngnr-1 also caused ectopic neuronal differentiation in regions of non-neurogenic ectoderm that flank the neural plate. The consequence of this was most easily observed at St. 24-26 (tail bud stage), where supernumerary neurons were observed within the epidermis. Extensive ectopic neurogenesis was also observed in the most anterior part of the embryo, where for example the eye was missing and replaced by amorphous neural tissue (data not shown). In these respects, the phenotype of X-ngnr-1 mRNA-injected embryos appeared similar or identical to that previously reported for XNeuroD (Lee et al., 1995). In the latter case, it has been confirmed by injections at the 32-cell stage that the ectopic neurons in the skin indeed reflect a conversion of non-neurogenic ectoderm (and not, for example, simply conversion of migrating trunk neural crest cells) (Lee et al., 1995), and we infer by extension that the same holds true for X-NGNR-1-injected embryos.
The fact that over-expression of X-ngnr-1 yielded an XNeuroD-like phenotype, coupled with the fact that endogenous X-ngnr-1 expression temporally precedes and overlaps that of XNeuroD, suggested that the latter might be a target of transcriptional activation by the former. To test this, embryos injected with X-ngnr-1 mRNA were hybridized with an XNeuroD probe. A massive, ectopic induction of endogenous XNeuroD mRNA was observed in X-ngnr-1 mRNA-injected embryos (100% of embryos examined; >50 embryos injected). The extent of ectopic expression was similar to that observed with an N-tubulin probe. By contrast, injection of XNeuroD mRNA did not increase the expression of endogenous X-ngnr-1 mRNA, although it did induce ectopic neurogenesis as previously reported (Lee et al., 1995). These data suggest that the neurogenic phenotype of X-ngnr-1 mRNA-injected embryos may reflect an induction of endogenous XNeuroD, and suggest that the sequential expression of these two genes during Xenopus neurogenesis reflects a unidirectional cascade in which the former induces transcription of the latter.
To ensure that the ectopic neurogenesis promoted by injection of X-NGNR-1 RNA was not a secondary consequence of induction of mesodermal tissue, we performed animal cap experiments. Animal caps from embryos injected in both blastomeres with various RNAs at the 2-cell stage were dissected and allowed to develop in vitro, after which time they were assayed for expression of various marker mRNAs by RNase protection (Ferreiro et al., 1994). As expected from the whole mount in situ analysis, injection of X-NGNR-1 RNA caused induction of expression of N-tubulin mRNA. No mesodermal induction was detected under these conditions, by criteria of expression of muscle-specific actin mRNA (Ferreiro et al., 1994). This control experiment indicates that the promotion of neurogenesis by X-NGNR-1 is a direct effect on naive ectoderm, and not an indirect result of mesoderm induction.
The effect of X-NGNR-1 was also compared to that of noggin (Smith and Harland, 1992) in the animal cap experiments. Noggin, which promotes neural induction (Lamb et al., 1993) by antagonizing the epidermalizing effect of BMP-4 (Sasai et al., 1995), induced expression of NCAM, a marker of undifferentiated neural tissue (Kintner and Melton, 1987), but not of the neuronal differentiation marker N-tubulin (Chitnis et al., 1995). X-NGNR-1, by contrast, induced expression of both NCAM and N-tubulin mRNAs, although the level of the latter transcript was higher. These data are consistent with the idea that noggin promotes neuralization but is insufficient for neuronal differentiation (Ferreiro et al., 1994), while X-NGNR-1 promotes both neuralization and consequent neuronal differentiation.
X-NGNR-1 Expression Precedes, and Can Activate Expression of, X-Delta-1
The foregoing data indicated that X-ngnr-1 is expressed earlier than XNeuroD, and is capable of inducing expression of XNeuroD as well as of promoting ectopic neurogenesis. Thus, like NeuroD, X-NGNR-1 can exert a neuronal determination function when over-expressed. But can X-NGNR-1 normally play this role in vivo? To address this question we examined the timing of X-NGNR-1 expression relative to that of X-Delta-1. In Xenopus as in Drosophila , X-Delta-1 encodes a lateral inhibitory ligand that controls a choice between neuronal and non-neuronal fates (Chitnis et al., 1995). By definition, therefore, at the time Delta is first expressed this choice has not yet been made.
During early gastrulation (St. 10.5), X-ngnr-1 mRNA can be detected at the lateral margins of the prospective neural plate. At this stage, X-Delta-1 mRNA is not yet expressed in this region, although it is detected in an area adjacent to the blastopore. By midgastrulation (St. 11.5), both X-ngnr-1 and X-Delta-1 mRNAs can be detected in three distinct patches within the neural plate, prefiguring the regions where primary neurogenesis will occur. Within these regions the domain of X-ngnr-1 expression appears to encompass that of X-Delta-1. At the same stage, X-ngnr-1 expression can be observed in the presumptive trigeminal placode, where X-Delta-1mRNA is not yet detectable. At neither of these stages is expression of XNeuroD detected (not shown). These data indicate that expression of X-ngnr-1 precedes that of X-Delta-1 in both the CNS (neural plate) and the PNS (trigeminal placode), whereas XNeuroD is not expressed until after X-Delta-1.
In Drosophila , the proneural genes (achaete-scute) activate expression of Delta (Hinz et al., 1994; Kunisch et al., 1994). The fact that expression of X-ngnr-1 precedes but spatially overlaps that of X-Delta-1 suggested, therefore, that the former might be capable of activating expression of the latter. In support of this idea, injection of synthetic X-NGNR-1 RNA induced ectopic expression of endogenous X-Delta-1 mRNA (100% of embryos examined; embryos injected), whereas control injections of lacZ mRNA had no such effect. Thus, like the proneural genes in Drosophila , X-ngnr-1 can activate expression of a lateral inhibitory ligand that controls a choice between neuronal and non-neuronal fates, within a group of developmentally equivalent cells.
Xnotch1 ICD Inhibits Both the Expression and Function of X-NGNR-1
Within the three domains of the neural plate, X-ngnr-1 mRNA-expressing cells appear scattered, rather than contiguous. In Drosophila , the expression of achaete-scute is restricted to sensory organ precursor cells by lateral inhibitory interactions mediated by Notch and Delta (Ghysen et al., 1993). This suggested by analogy that the expression of X-ngnr-1 might similarly be restricted to subsets of neural precursors by lateral inhibition. Three different experiments support this idea. First, injection of a dominant-active form of Notch (Struhl et al., 1993) (the intracellular domain, or ICD), which inhibits primary neurogenesis, also repressed the expression of endogenous X-ngnr-1 mRNA (18/18 embryos tested); in contrast control injections of lacZ mRNA had no such effect. Conversely, blocking lateral inhibition by injection of a dominant-negative form of X-Delta-1 (X-Delta-1 Stu ) (Chitnis et al., 1995), caused an apparent increase in the density of strongly X-ngnr-1-positive cells, as well as a slight expansion of the X-ngnr-1-positive domain in 60% of injected embryos (31/50 embryos tested). In contrast such an effect was not seen in control lacZ-injected embryos (except in one isolated case out of 39 embryos examined). This second result suggested that the density of X-ngnr-1-expressing cells within each domain of primary neurogenesis is normally limited by lateral inhibition. The fact that exogenous Notch ICD is, moreover, able to strongly suppress endogenous X-ngnr-1 expression supports the idea that this lateral inhibition is mediated, at least in part, by endogenous X-Notch genes.
To determine whether Notch-mediated signalling can inhibit the function as well as the expression of X-ngnr-1, exogenous X-ngnr-1 mRNA was co-injected with either lacZ mRNA, or lacZ mRNA plus Notch ICD mRNA. An inhibition of X-ngnr-1-promoted ectopic neurogenesis was observed with high penetrance (27/29 embryos tested), indicating that the function as well as the expression of X-NGNR-1 is sensitive to inhibition by Notch ICD . However, within the injected side of the experimental embryos, the inhibition of neurogenesis showed variable expressivity, and appeared most complete in those regions which received the highest amount of the co-injected mRNAs (as determined by XGal staining). Assuming that all three co-injected mRNAs are similarly distributed, this result could indicate that a certain threshold amount of Notch ICD mRNA is necessary to overcome neurogenesis promoted by exogenous X-ngnr-1 mRNA, and/or that there is a non-linear relationship between the amount of Notch ICD mRNA injected and the amount of inhibitory activity produced.
The initial expression of X-ngnr-1 occurs in three territories (medial, intermediate and lateral) which demarcate the domains in which primary neurogenesis will eventually occur (Chitnis et al., 1995). Our data suggest that within these territories, lateral inhibition restricts X-NGNR-1 expression to a limited number of neuronal precursor cells. During this process, X-NGNR-1 positively regulates X-Delta-1, and is in turn negatively regulated (in adjacent cells) by signalling through X-Notch-1, a receptor for X-Delta-1. Thus cells expressing higher levels of X-NGNR-1 will inhibit expression of X-NGNR-1 in their neighbors and thereby suffer less inhibition, leading to yet higher levels of X-NGNR-1 expression, and so on. As X-NGNR-1 expression becomes restricted to presumptive neuronal precursors, it leads (directly or indirectly) to expression of XNeuroD. Once the cells express sufficiently high levels of XNeuroD, they undergo neuronal differentiation. In this view, lateral inhibition is part-and-parcel of the cell fate decision process itself, although it may also be engaged to prevent further differentiation after the decision has been made.
This model postulates a role for X-NGNR-1 that is analogous to that deduced for the Drosophila proneural genes, such as achaete-scute and atonal, from both loss-of-function and gain-of-function genetic experiments (Ghysen et al., 1993). While Xenopus is advantageous for gain-of-function perturbations, loss-of-function perturbations are less readily achieved in this system. However our conclusions are not based solely on gain-of-function phenotypes, but also on the timing and place of X-ngnr-1 expression in relation to that of other regulatory genes, as well as on the regulation of endogenous X-ngnr-1 mRNA expression by manipulation of the lateral inhibition machinery.
For example, our conclusion that X-Delta-1 is positively regulated by X-NGNR-1 is consistent with the fact that the initial expression of X-ngnr-1 precedes and spatially overlaps that of X-Delta-1, as well as with the fact that injection of X-ngnr-1 mRNA induces ectopic expression of endogenous X-Delta-1 mRNA. Similarly, in Drosophila , the proneural genes positively regulate expression of Delta (Hinz et al., 1994; Kunisch et al., 1994). Likewise, our conclusion that expression of X-ngnr-1 is restricted to subsets of cells by lateral inhibition is consistent with the observation that this gene exhibits a scattered pattern of expression within each of the three territories of primary neurogenesis, and that the density of X-NGNR-1-expressing cells can be increased within these territories by injection of a dominant-negative form of X-Delta-1. Finally, there is a good correlation between the ability of injected X-Delta-1 Stu mRNA to increase the density of X-NGNR-1 expressing cells, and also to increase the density of N-tubulin-expressing cells which subsequently differentiate. This correlation suggests that the number of neuronal precursors may normally be determined by the number of cells that express X-NGNR-1 above a given threshold level.
The observation that X-ngnr-1 both activates, and is inhibited by, the lateral inhibitory circuitry raises the paradox of how a neurogenic phenotype can nevertheless be obtained by over-expressing this gene. The simplest answer is that the injected X-NGNR-1 RNA bypasses X-Notch-mediated transcriptional repression of the endogenous X-ngnr-1 gene. However, our data suggest that Notch is also able inhibit the function of X-NGNR-1 translated from exogenous RNA, either by a post-transcriptional mechanism or by inhibiting expression of X-NGNR-1 target genes. Nevertheless, this inhibition appears to require high levels of X-Notch ICD expression, being strongest in those regions which contain the highest level of co-injected RNAs. In the situation where X-NGNR-1 RNA alone is injected, the level of endogenous X-Notch signalling may simply be insufficient to override the large amounts of exogenous X-NGNR-1 protein. Moreover, the possibility that Notch ICD inhibits X-NGNR-1 function artefactually, because it is injected in a form that normally does not exist in vivo (see (Nye et al., 1994) for further discussion), cannot be excluded.
Relationship of XASH-3 and X-NGNR-1
Xash3 is the only other neural bHLH gene which is known to be expressed as early as X-ngr-1 in the neural plate (Zimmerman et al., 1993). The available evidence, however, more clearly identifies X-ngnr-1 as a vertebrate analog of the Drosophila proneural genes during primary neurogenesis. Firstly, while the expression of X-ngnr-1 correlates extremely well with the three domains of the neural plate where primary neurons form, Xash-3 is expressed in an “intermediate” zone of the neural plate which may in fact correspond to the sulcus limitans (Zimmerman et al., 1993). Secondly, the activity of Xash3 in ectopic expression studies appears to be different from that of X-NGNR-1. For instance, ectopic expression of XASH-3 at high levels causes an expansion of neural tissue (Ferreiro et al., 1994; Turner and Weintraub, 1994), a phenotype never observed with X-NGNR-1. Ectopic expression of XASH-3 can also induce ectopic neuronal differentiation, but only effectively when lateral inhibition is also blocked using the dominant-negative X-Delta-1, and only then within the posterior neural plate (Chitnis and Kintner, 1996). X-NGNR-1 does not have similar restrictions in its activity, and can promote neurogenesis anteriorly, and outside the neural plate. Finally, although exogenous XASH-3 is sensitive to lateral inhibition mediated by X-Notch-1 and X-Delta-1 (Chitnis and Kintner, 1996), there is no evidence that endogenous Xash-3 expression is normally regulated by such inhibition. Thus, X-ngnr-1 fulfills more of the criteria expected for a gene whose activity defines the “proneural” domains wherein primary neurons arise in the neural plate.
The function of Xash-3 remains enigmatic. The fact that ectopic expression of this gene expands undifferentiated neural tissue suggests that it could normally act to prevent or delay overt neuronal differentiation within the restricted domain of the neural plate where it is expressed. Consistent with this idea, injection of Notch ICD which expands the neural plate also expands the domain of Xash-3 mRNA expression (unpublished observations). This is exactly the opposite of what is observed in the case of X-ngnr-1, whose expression is repressed under these conditions. An insensitivity of Xash-3 to transcriptional inhibition by X-Notch-1 signalling would allow the continued expression of this gene to maintain the uncommitted neural character of neural plate cells, while the high sensitivity of XASH-3 to functional inhibition by Notch (Chitnis and Kintner, 1996) would prevent these cells from undergoing overt neuronal differentiation. Whether XASH-3 also contributes to primary or secondary neurogenesis, but in different cells or at a different step in the pathway as X-NGNR-1, remains to be determined.
X-NGNR-1 Performs Two Distinct and Temporally Separated Functions
In our experiments, injection of X-ngnr-1 mRNA results in the induction of both X-Delta-1 and XNeuroD. During normal development, however, expression of XNeuroD is delayed relative to that of X-Delta-1. How is the sequential expression of these two putative target genes of X-NGNR-1 normally achieved? One possibility is that X-Delta-1 requires a lower threshold of X-NGNR-1 activity than XNeuroD to be activated, and that it takes time for X-NGNR-1 to accumulate to levels sufficient to induce XNeuroD. Another explanation is that a co-factor is required together with X-NGNR-1 to activate XNeuroD, and that expression of this co-factor is delayed. Precedent for such a temporal separation of transcription factor functions is found during mother-daughter segregation in yeast, where the Swi5 protein acts first to activate ASH1 expression (which in turn blocks Swi5p function in daughter cells (Bobola et al., 1996; Sil and Herskowitz, 1996)), and later to activate HO expression in mother cells. The delay in HO activation by Swi5p reflects an induction of the necessary co-activators Swi4p and Swi6p (Amon, 1996). Interestingly, in both Xenopus neurogenesis and yeast the temporal separation provides a time window for these determinative factors (Swi5p or X-NGNR-1) to provide an inhibitory signal to neighboring or daughter cells, while allowing them to later promote an alternative fate cell-autonomously.
Determination versus Differentiation Genes
We have documented a sequential expression of ngn1 and NeuroD mRNAs during both murine and Xenopus neurogenesis, and in the latter system have further demonstrated a unidirectional functional cascade for these genes. Thus in Xenopus neurogenesis, as in mammalian skeletal myogenesis and Drosophila neurogenesis, structurally-related bHLH proteins function in cascades (Jan and Jan, 1993). The timing and location of NeuroD expression have previously been suggested to reflect a function for this gene in neuronal differentiation (Lee et al., 1995), perhaps analogous to that of myogenin during muscle development (for review, see (Weintraub, 1993)). By extension, the upstream gene X-ngnr-1 would function in determination, analogous to the roles of MyoD and myf5 during myogenesis. Such a conclusion would be consistent with our demonstration that the regulation and function of X-ngnr-1 are similar to that of the proneural genes, which control the determination of neural precursors in Drosophila (Campuzano and Modolell, 1992).
What is different about determination and differentiation bHLH factors? The similar actions of these proteins in gain-of-function experiments suggest that they may differ only in the time and place of their expression, or in the downstream genes they regulate (Jan and Jan, 1993). On the other hand, they may possess intrinsic functional differences that have so far escaped detection. For example, it has been proposed that muscle differentiation factors are less sensitive to inhibitors than are determination factors (Weintraub, 1993). However, such a differential sensitivity to inhibitors has not been demonstrated in myogenesis, although MyoD function can be inhibited by Notch ICD (Kopan et al., 1994).
Previously, a differential sensitivity of XASH-3 and XNeuroD to lateral inhibition was demonstrated (Chitnis and Kintner, 1996). Although Xash-3 is expressed earlier than XNeuroD, there is no evidence that the two genes function in a cascade; indeed the latter gene is expressed in many regions where it is not preceded by the former. While X-NGNR-1 and XNeuroD do appear to function in a cascade, they do not appear differentially sensitive to inhibition by co-injected X-Notch ICD , in side-by-side comparisons (Ma et al., unpublished data). It is possible that these genes are differentially sensitive to direct inhibition by Notch signalling at the transcriptional level, but this is currently difficult to test since inhibition of X-NGNR-1 expression by Notch ICD indirectly prevents expression of XNeuroD.
The determination function proposed for X-ngnr-1 may, therefore, primarily reflect the developmental context in which this gene is expressed. Expression of X-ngnr-1 in the neurectoderm generates a group of competent cells, from which a subset is later selected for overt neuronal differentiation. The state of competence imposed by X-ngnr-1 expression is insufficient to allow differentiation, precisely because the expression and function of this gene are sensitive to lateral inhibition. However, through its ability to upregulate X-Delta-1, X-NGNR-1 confers the capacity to engage in a winner-take-all competition between alternative cell states: neuronal or non-neuronal. While increased expression of X-NGNR-1 imposes a bias towards the neuronal state, this state is unstable until subsequent events render the cell insensitive to further inhibition. The nature of the events that stabilize the neuronal state and commit the cell irreversibly to neuronal differentiation remain to be established.
Example 4
Cloning of Neurogenin 2 and Neurogenin 3 from Mouse
In an extension of the above approach, additional genes were cloned, using Monc-1 cells, an immortalized neural crest cell line which can be grown under differentiated or undifferentiated conditions (Sommer, et al., (1995). Neuron. 15, 1245-1258.).
RT-PCR and Isolation of bHLH Transcription Factor cDNA Clones
Total cellular RNA was extracted (Chomczynski and Sacchi, (1987) Anal. Biochem. 162, 156-159) from Monc-1 cells (Sommer et al., 1995) that had been differentiated for two days. Oligo-dT primed reverse transcription of total RNA was carried out for 1 hr at 42° C. using Superscript II (Gibco). First strand cDNA served as template for PCR amplification using Taq polymerase (Fisher) and the following degenerate primers: Asc10 (5′ primer; cgggatccAAT/C GA/CI C/AGI GAA/G C/AGI AAT/C C/AGI A/GT) (SEQ ID NO:26); and Asc11 (3′ primer; cggaattcAG/AI GTT/C TCI AT/CT/C TTI G/CA/TI AT/G/AT/C TT) (SEQ ID NO:27). The Asc10 primer encodes NE/ARERNRV/M (SEQ ID NO:28) and contains a BamHI site (small letters) at the 5′ end. The Asc11 primer encodes the reverse translation of KM/LSKV/IETL (SEQ ID NO:31) and includes an EcoR1 site (small letters) at the 5′ end.
PCR was carried out in a Perkin Elmer DNA Thermal Cycler for 8 cycles with a denaturation step at 94° C. for 1 min, annealing at 43° C. for 1 min and extension at 72° C. for 1 min, followed by an additional 40 cycles with the annealing temperature raised to 55° C. The Taq polymerase was added at 94° C. during the first cycle using Amplivax (Perkin Elmer). The 130 by PCR product was purified from a NuSieve agarose gel (FMC Bioproducts) by Qiaquick (Qiagen), subject to a second round of PCR (40 cycles of 1′ at 90° C., 1′ at 55° C., 1′ at 72° C.), digested with EcoR1 and BamHI, and cloned into the Bluescript SK-vector (pBS) (Stratagene Cloning Systems). Altogether, 70 clones were characterized, and identified according to the sequences of helix 1 and the loop.
They comprised MASH1 (Johnson, et al., (1990) Nature. 346, 858-861) (8 clones), SCLERAXIS (Cserjesi et al., (1995) Development 121, 1099-1110) (5 clones), as well as the eight ATONAL-related factors. Among those, NGN1 was represented by 14 clones; NGN2, 4 clones; NGN3, 16 clones; NeuroD, 1 clone; MATH2/Nex1, 2 clones; NDRF/KW8NeuroD2, 2 clones; NeuroD4, 5 clones; and NeuroD5, 1 clone; the sequences of the rest of the 70 clones did not represent bHLH factors.
Isolation of ngn2 and ngn3 Genomic Clones
The PCR products encoding the novel bHLH domains were used individually to screen a mouse lambda-2 129 genomic library (a gift from Z. Chen). From positive genomic clones containing the genes ngn2 and ngn3, respectively, a 2.0 kb (ngn2) and a 1.8 kb (ngn3) restriction fragment hybridizing to the respective PCR products were isolated, cloned into pBS, and then sequenced on both strands by the Caltech sequencing core facility.
The PCR fragments encoding the helix 1-loop domain of ngn2 and ngn3 were then used individually as hybridization probes to isolate clones containing additional coding sequences from a mouse genomic library. Hybridizing restriction fragments from these genomic clones were subcloned and sequenced. Sequencing of these subclones revealed that the open reading frames of both ngn2 and ngn3 are each contained within a single exon, a situation similar to that of other bHLH genes previously characterized, such as Mash1 (Guillemot et al., 1993). The deduced polypeptide sequences of NGN2 and NGN3 consist of 263 and 214 amino acids, respectively ( FIG. 1A ), and display a 41-47% overall sequence identity to each other and to NGN1. Of the three members of the NEUROGENIN-subfamily known so far in mouse, NGN2 shows the highest homology to Xenopus NGN-RELATED FACTOR-1a (X-NGNR-1a), a protein suggested to control the choice of neuronal fate by uncommitted neural plate cells in Xenopus (Ma et al., 1996). (54% overall sequence identity of NGN2 to X-NGNR-1a, vs. 41% for NGN1 and 39% for NGN3; see FIG. 1 (SEQ ID NOS:1 and 2) and (Ma et al., 1996)).
Alignment of the bHLH domains of NGN2 and NGN3 with those of other murine ATONAL-like sequences identified in this study revealed that these bHLH domains define two distinct subfamilies. The first subfamily consists of NGN1, NGN2 and NGN3, while the members of the second subfamily are very similar to NeuroD (Lee et al., 1995; Naya et al., 1995) and comprise MATH2Nex1 (Bartholoma and Nave, 1994; Shimizu et al., 1995) and NDRF/KW8/NeuroD2 (Kume et al., 1996; McCormick et al., 1996; Yasunami et al., 1996). In the NEUROGENIN subfamily, the junction region between the basic domain and helix 1 (“jnc.”) as well as many residues of the loop and helix 2 are highly conserved. Within this region, moreover, several amino acids are conserved among all members of the NEUROGENIN subfamily but are divergent in members of the NeuroD subfamily.
Members of the NeuroD subfamily, on the other hand, are characterized by sequence identity within the loop region and helix 2 (in contrast to the NEUROGENIN subfamily members). Furthermore the sequence of the loop is specific for members of the NeuroD subfamily, as are several amino acids in helix 2 ( FIG. 2 , dotted residues in NeuroD consensus). On this basis, we have tentatively assigned to this subfamily two additional novel ATONAL-like sequences isolated in our screen, called NeuroD4 and NeuroD5. Confirmation of this assignment will await the isolation of full-length clones for these two genes. The sequence of MATH 1, the first mammalian ATONAL homolog to be isolated (Akazawa et al., 1995), was not detected in our RT-PCR screen. It was not included in the comparison presented here because it does not fit well in either subfamily, and may represent yet a third subfamily of ATONAL-like genes.
In Situ Hybridization
Non-radioactive in situ hybridization with digoxigenin labeled riboprobes was carried out on frozen sections of paraformaldehyde-fixed mouse embryos according to Birren et al. (Birren et al., (1993) Development 119, 597-610). Non-radioactive whole mount in situ hybridization was performed using a modification of the procedure described by Wilkinson (Wilkinson, (1992) In Situ Hybridization: A Practical Approach, ed. D. G. Wilkinson, IRL Press, Oxford). Antisense riboprobes used were as follows: ngn2 and ngn3 (isolated in this study); ngn1 (Ma et al., (1996) Cell 87, in press); neuroD (Lee, et al., (1995). Science. 268, 836-844.); and rat erbB3 (a gift from Hai Wang).
ngn2 Expression Precedes that of ngn1 in Both the Spinal Cord and the Neural Crest
The expression of the ngn's in the trunk region was investigated, by performing in situ hybridization to transverse sections of developing mouse embryos. Between E8.5 and E10.5, both ngn1 and ngn2 become strongly expressed in the ventricular zone (VZ) of the ventral neural tube, as well as in a limited area of the dorsal neural tube near the roof plate. In the dorsal neural tube, however, the domain of ngn2 expression was broader than that of ngn1. This difference was also observed at E12, when ngn1 expression became restricted to the basal half of the neural tube and a narrow stripe adjacent to the roof plate, while ngn2 expression persisted throughout the dorso-ventral axis of the spinal cord (data not shown; see also Ma et al., 1996). Interestingly, ngn1 was expressed in a more lateral part of the VZ while ngn2 was detectable in both the medial and lateral parts of the VZ. In contrast to ngn1 and ngn2, ngn3 mRNA was most strongly expressed in a very restricted region of the spinal cord, just dorsal to the floor plate, which was detectable as early as E9 and persisted until E14 (data not shown) when a few ngn3-expressing cells were still detectable in this region. At E12, however, a second more dorsally-located domain of weaker ngn3 expression was detected. This second domain appeared to overlap with the expression of both ngn1 and ngn2, however ngn3 was expressed closer to the ependymal canal than was ngn1, in this region.
In the peripheral nervous system (PNS), ngn2 expression clearly preceded that of ngn1 and was detectable in cells dorso-lateral to the neural tube as early as E8.5. On adjacent sections, ngn1 mRNA was not yet detectable in this region. The timing and location of ngn2 expression in this position strongly suggests that this gene is expressed by migrating neural crest cells, a conclusion supported by whole mount in situ hybridization experiments. At E10.5, persistent but weak ngn2 expression was detected in subsets of cells in the trunk dorsal root sensory ganglia (DRG), which also contained more numerous ngn1-expressing cells as previously shown (Ma et al., 1996). The position and morphology of the ngn2-expressing cells, which are located at the dorso-lateral aspect of the DRG, suggests that they are non-neuronal cells, either neuronal precursors and/or satellite cells. Neither ngn2 nor ngn1 were detectable in sympathetic or enteric ganglia (not shown), which instead express the bHLH factor Mash1 (Lo, et al., (1991). Genes & Dev. 5, 1524-1537). ngn2 expression in DRG was rapidly extinguished, so that by E12 no signal was detectable. In contrast, ngn1 expression in trunk DRG persisted through E12 and was not extinguished until E14 (data not shown). Expression of ngn3 was not detected in either trunk neural crest or its derivatives at any of the stages examined.
Apparent Axial Differences in Neural Crest and Its Derivatives Defined by ngn2 Expression
To obtain further evidence that ngn2 is expressed by early-migrating trunk neural crest cells, the expression of this gene was examined at E8.5-E9.5 by whole mount in situ hybridization. At E9.25, ngn2 was expressed in two longitudinal stripes, that define its dorsal and ventral domains of expression in the spinal cord. A third longitudinal stripe of ngn2 expression was also detected ventro-lateral to the dorsal stripe. The position of this stripe was consistent with the idea that it represents expression of ngn2 in early emigrating crest cells. This interpretation is supported by the similar location of cells expressing erbB3, an independent marker of migrating neural crest (Meyer and Birchmeier, (1995) Nature 378, 386-390). Surprisingly, the domain of strong ngn2-expression in neural crest seemed not to extend as far anteriorly as did the domains of its expression in the spinal cord, but rather declined sharply within the cervical region. Analysis of embryos from earlier stages confirmed that ngn2 expression in neural crest of cervical regions was low or undetectable at E8.5 as well (data not shown). In contrast migrating crest cells expressing erbB3 were detectable in the cervical region as well as in the trunk. However, we cannot presently distinguish whether the relatively lower level of ngn2 expression in cranial crest reflects a true difference in its expression by neural crest cells, or rather fewer crest cells in this region at the stages we examined.
Nevertheless, an apparent axial difference in levels of ngn2 expression by neural crest cells was supported by the analysis of transverse sections through the thoracolumbar and cervical regions of E8.5 embryos. Moreover, similar axial differences were observed in forming DRG at later developmental stages. This prediction was confirmed by comparison of transverse sections through different axial levels of later stage embryos. At both E9.5 and E10.5, ngn2 transcripts were detectable in DRG at thoracolumbar but not cervical levels. By contrast, ganglia at both axial levels express ngn1 mRNA as detected on adjacent serial sections. Taken together, these data suggest that cells expressing ngn2 contribute mainly to developing DRGs of the thoracolumbar but not the cervical level. This in turn provides molecular evidence that trunk neural crest might be distinct from cervical neural crest from the earliest stages of migration.
In contrast to the low or undetectable levels of ngn2 expression in migrating cervical neural crest cells, the transcript was abundantly expressed in the epibranchial placodes. This difference appears to correlate with the fact that these placodes contribute neurons to most of the cranial sensory ganglia, while the non-neuronal cells derive from the neural crest in this region (for review, see (Noden, (1993) J. Neurobiol. 24, 248-261). Consistent with this correlation, ngn2 mRNA was later detected in several (but not all) of the cranial sensory ganglia, including the VII th and IX th ganglia (see below). This correlation supports, but does not prove, the idea that many of the ngn2 expressing cells are precursors to neurons, an idea consistent with the function of the ngn2 homolog X-ngnr-1 in Xenopus neurogenesis (Ma, et al., 1996).
Neurogenins Exhibit Partially-overlapping Domains of Expression in the Developing CNS
Previous studies of ngn1 (Ma et al., 1996) as well as our whole mount in situ hybridization data indicated that members of the ngn subfamily are also expressed in the developing brain. To further characterize this expression we hybridized adjacent transverse sections through E12 embryos with probes for the three ngn's. Both ngn1 and ngn2 were expressed in common but restricted regions of the CNS, in the VZ. For example, high levels of ngn2 mRNA were detected in cortical neuroepithelium, which also expressed ngn1 albeit at much lower levels. Similarly, a portion of the diencephalon expressed high levels of ngn2 and lower levels of ngn1. Interestingly, the expression domains of the ngn's collectively excluded large areas, such as the striatum, which instead express Mash1 (Lo et al., (1991) Genes & Dev. 5, 1524-1537). In other studies we have shown by double-label in situ hybridization that the domains of ngn1 and Mash1 expression in the developing forebrain are complementary and non-overlapping (Ma et al., in preparation). Thus the expression domains of the ngn's and Mash1 appear collectively to account for most or all of the neuroepithelium of the CNS and the ganglia of the PNS.
Although at first sight, the domains of ngn1 and ngn2 expression in the developing brain appeared co-extensive, closer inspection revealed that in some areas the boundaries of ngn2 expression were distinct from those of ngn1. For instance, at E10.5 the ventral boundary of ngn2 mRNA expression in developing cortex appeared to extend slightly below that of ngn1 (data not shown). Similarly, in sagittal sections of E12 hindbrain, the domain of ngn2 expression extended posterior to that of ngn1. Such cases of partial overlap were observed in other areas of the brain as well. For example at E12 in parasagittal sections through the lateral part of the dorsal thalamus, ngn2 but not ngn1 was detected, while both transcripts were detected in mid-sagittal sections passing through more medial regions of this structure (not shown). Similarly the posterior tegmental epithelium at E12 expressed ngn2 but not ngn1. Finally, ngn3 expression was even more restricted than that of the other genes, being mostly limited to the hypothalamic region at E12 (data not shown). Expression of ngn3 in this region was mutually exclusive with that of the other two ngn's.
Interestingly, we did not observe major brain regions that express high levels of ngn1 but not ngn2 mRNAs. However, this was indeed observed in the periphery: for example, the trigeminal ganglion and olfactory epithelium expressed ngn1 but not ngn2. Conversely, the retina and IX th cranial ganglion expressed ngn2 but not ngn1. Within the facio-accoustic ganglion complex, ngn1 was expressed in the epithelium of the otic vesicle and in the accoustic component of the facio-accoustic complex (the future vestibule-cochlear ganglion (VIII th g)), whereas ngn2 was expressed by the facial component of the facio-accoustic complex (the future geniculate ganglion (VII th g)). These examples of complementarity, taken together with their sequence similarity, suggest that ngn1, 2 and 3 may play similar roles in different regions, lineages or sensory organs of the CNS and PNS.
NeuroD Expression Follows that of the Neurogenins in Many Regions and Lineages
In contrast to the more restricted expression domains of the ngn's, neuroD expression encompasses essentially all of the brain areas and peripheral ganglia and sense organs that express ngn1, 2 or 3. Thus, for example, in the hindbrain neuroD expression spans the region of neuroepithelium that expresses ngn1 and ngn2. In the hypothalamus, neuroD was expressed within the region that expresses ngn3. Similarly, in the periphery all of the structures examined express neuroD irrespective of whether they expressed ngn1 or ngn2.
As reported previously for ngn1 (Ma et al., 1996), within many of the regions or structures where their expression overlaps, neuroD was expressed at a later step in neuronal differentiation than were the neurogenins. This was seen most clearly in cortex, where neuroD mRNA was displaced laterally towards the pial surface, which contained newly-differentiated post mitotic neuroblasts, while ngn1 and ngn2 were expressed in the ventricular and subventricular zones which contain proliferating neural precursors. Other examples of such sequential expression were seen in the hindbrain and the hypothalamus. Apparent exceptions to this rule were seen in the olfactory epithelium and retina, where the expression of both ngn's was detected in laminae that also expressed neuroD; however this apparent lack of laminar separation may simply reflect the relative thinness of these developing sensory epithelia. In the PNS, the lack of a laminar structure makes it impossible to infer sequential expression from spatial segregation. However, analysis of expression at different ages indicated that ngn1 and ngn2 were detected in several cranial sensory ganglia (e.g., IX th , VII th ) at stages when neuroD mRNA was not yet detectable (data not shown). The sequential expression of various neurogenins and neuroD in the mammalian nervous system is consistent with the fact that in Xenopus , X-ngnr-1 functions upstream of, and activates expression of, XneuroD (Ma et al., 1996).
ngn3 is Expressed in Pancreatic Islet Cell Progenitors
As neuroD is also expressed in the pancreas where it is thought to activate expression of islet cell-specific genes (Naya, et al., (1995). Genes & Dev. 9, 1009-1019), we sought to determine whether by analogy to the nervous system there might be members of the ngn subfamily expressed earlier than neuroD in this peripheral endocrine tissue. At E12, when only relatively weak expression of neuroD was detected, strong expression of ngn3 was observed in cells at a similar position on adjacent serial sections. Expression of ngn1 and ngn2 mRNAs was not detected. Expression of ngn3 in the pancreas persisted through at least E14, the latest stage examined, when it identified clusters of apparent islet cells also expressing neuroD. While our data do not directly demonstrate co-expression of these two genes in the same cells, by analogy to the nervous system it seems likely that ngn3 functions upstream of neuroD in a cascade controlling islet cell development.
Sequential and Regionally-restricted Expression of Different bHLH Factors during Mammalian Neurogenesis
Diversity among bHLH factors is thought to subserve different developmental events in different lineages, which occur both in series and in parallel (Jan and Jan, (1993). Cell. 75, 827-830; Weintraub, H. (1993). Cell. 75, 1241-1244). The data presented here are entirely consistent with this idea: different members of the ngn subfamily are expressed in different regions or lineages of the nervous system, while neuroD is expressed downstream of the ngn's in most of these regions. This apparent cascade relationship extends to the endocrine pancreas, where ngn3 seems to be expressed upstream of neuroD. While it is currently difficult to directly demonstrate sequential expression of individual ngn's and neuroD within the same cell lineage, it seems likely by analogy to Xenopus (Ma et al., 1996) that this is indeed the case. Not only is X-ngnr-1 able to activate XneuroD, but mouse ngn1 has been demonstrated to do so in Xenopus , as well (Ma et al., 1996). As ngn2 is even more closely related to X-ngnr-1 than is ngn1, it is likely that NGN2 is also an upstream activator of neuroD; by extension, the same may hold true for ngn3.
Although we have emphasized the striking convergence of different ngn's onto neuroD, additional neuroD-like genes have been identified in this and in other studies (Kume, et al., (1996). Biochem. Biophys. Res. Commun. 219, 526-530; Yasunami et al., (1996) Bicochem. Biophys. Res. Comm. 220, 754-758). As targeted mutations in the neuroD gene in mice interfere with pancreatic but not neuronal development (Lee et al., personal communication), it seems likely that at least some of these neuroD-related genes are functionally-redundant with neuroD itself and therefore that their expression must partially overlap. In that case, different ngn's may activate multiple neuroD-like genes in at least some regions.
In Drosophila , five positive-acting bHLH factors have been identified which promote neurogenesis (Jan and Jan, 1993), and a similar number of positive bHLH factors have been implicated in mammalian (Olson and Klein, (1994) Genes & Dev. 8, 1-8). In contrast, the number of mammalian neural-specific, putatively positive-acting bHLH factors is at least three times as large (for review, see (Guillemot, 1995)). (A growing subfamily of negative-acting bHLH factors related to Drosophila hairy and enhancer of split (Sasai, et al., (1992). Genes & Dev. 6, 2620-2634) is not included in this calculation). Why are so many bHLH genes necessary in the mammalian nervous system? Evidence from Drosophila suggests that different proneural genes specify not only the neural fate per se (Campuzano and Modolell, (1992) Trends Genet. 8, 202-208.), but also different neural precursor identities (Jarman, et al., (1993b). Cell. 73, 1307-1321; Skeath and Doe, (1996) Curr. Biol. 6, 1146-1152). Thus the relatively large number of atonal-related bHLH genes may reflect the great diversity of cell types in the mammalian nervous system. It is important to emphasize that the domains of ngn expression exclude large regions which instead express Mash1 (Ma, et al., in preparation). In this respect, it is curious that a similar extent of diversity has not yet been observed for the Mash-related subfamily, only one member of which is specifically expressed in the mammalian nervous system (Guillemot et al., (1994) Nature 371, 333-336; Johnson et al., (1990) Nature 346, 858-861). Whether this reflects technical or rather biological factors remains to be determined.
As in muscle (for review, see (Weintraub, 1993)), the multiplicity of neural bHLH genes also leads to partial functional redundancy. Thus, mutations in neuroD have no apparent neural phenotype (J. Lee, personal communication), and a targeted disruption of Mash1 affects the PNS but not the CNS (Guillemot, et al., (1993). Cell. 75, 463-476), despite the fact that this latter gene is widely expressed in the brain and spinal cord (Guillemot and Joyner, (1993) Mech. Devel. 42, 171-185; Lo et al., 1991). This apparent functional redundancy may arise from cross-regulatory interactions between related bHLH genes, as demonstrated for achaete-scute in Drosophila (Martinez and Modolell, (1991) Science 251, 1485-1487) and the myogenic bHLH genes in mammals (for review see (Weintraub, 1993)). The overlapping expression of ngn1 and ngn2 in some brain regions (e.g., cortex) may similarly reflect such cross-regulation, and suggests that these two genes may exhibit partial functional redundancy as well. Single and double mutants in these genes, currently under construction, should resolve this issue.
Overlapping Expression of Neurogenins in the PNS and CNS
Several regions of the CNS and PNS, such as the cortex and DRG, show at least partially-overlapping expression of ngn1 and ngn2. The sensitivity and resolution of double-label in situ hybridization have been insufficient, at least in our hands, to resolve the issue of whether these two genes are expressed in the same cells, or in distinct but intermingled cell populations. As mentioned above, the precedent for cross-regulatory interactions between related bHLH genes argues for the former interpretation. However, there are also several reasons to consider the latter possibility as well. First, expression of the two genes is not obligatorily coupled: we have demonstrated many regions which express one gene but not the other. Second, in those places where overlap does occur, it is often incomplete: thus in the cortex and hindbrain the boundaries of ngn1 and ngn2 expression are distinct. Third, all of the regions or structures where ngn1 and ngn2 expression overlaps are known to contain phenotypically distinct but intermingled neuronal cell types (Ramon y Cajal, (1995) Histology of the nervous system of man and vertebrates, Oxford University Press).
A final reason is the precedent provided by recent revisions in our understanding of the roles that diverse bHLH factors play in myogenesis (Braun et al., (1994) Development 120, 3083-3092; Patapoutian et al., (1995) Development 121, 3347-3358; Smith et al., (1994) J. Cell Biol. 127, 95-105). Initially, myf5, myoD and myogenin were all thought to be sequentially expressed in a common myogenic cell population within the somites (for review, see (Lyons, (1994) Am. Zool. 34, 305-312)). Subsequent work has revealed, however, that this is not the case: rather, myf5 and myoD are expressed by two distinct myoblast populations (Braun and Arnold, 1996), both of which subsequently express myogenin (Smith et al., 1994). Moreover, expression of myf5 and myoD in these two lineages initiates at different times, so that myogenin has already been expressed in the myf5-dependent lineage before myoD expression has been initiated in the other lineage. In the same way, the sequential but overlapping expression of ngn1 and ngn2 in developing sensory ganglia could reflect their expression in distinct sublineages on different schedules. If so, then the expression of these two genes would provide the first evidence that the lineages that generate different sensory neuron subtypes (Snider, (1994) Cell 77, 627-638) are segregated from a very early stage.
The idea that ngn1 and ngn2 are involved in the specification of different sensory neuron subtypes is further suggested by their pattern of expression in cranial sensory ganglia. We have shown that several of these ganglia express ngn2 but not ngn1, e.g., the VII th and IX th ganglia. Interestingly, the neurons in these ganglia are primarily derived from the epibranchial placodes (Noden, 1993), which also express high levels of ngn2 (but not ngn1). Taken together, these facts are consistent with the idea that the ngn2-expressing cells in these cranial ganglia represent placode-derived neuronal precursors. Other cranial ganglia, however, such as the V th (trigeminal), express ngn1 but not ngn2. Interestingly, these other ganglia contain crest-derived as well as placode-derived neurons (Le Douarin, (1982) The Neural Crest, Cambridge University Press), and it is intriguing to consider that in such dual-origin ganglia ngn1 may be expressed exclusively by crest-derived neuronal precursors. Alternatively, ngn1 and ngn2 might define different sensory sublineages irrespective of their placodal vs. crest origin. Whatever the case, expression of ngn1 and 2 in different cranial sensory sublineages would be consistent with the idea that these genes mark distinct sublineages in trunk sensory ganglia as well.
Cryptic Cellular Heterogeneity Revealed by Expression of the Neurogenins
In addition to their potential importance in understanding the molecular control of neurogenesis, the ngn's may provide useful markers of specific neural progenitor subpopulations for cellular studies of neurogenesis. This is seen in the developing spinal cord where, for example, ngn1, 2 and 3 each define distinct groups of cells within the VZ, at different positions along the dorso-ventral axis. The developmental significance of these cell groups is not clear, but they indicate that the VZ of the spinal cord is patterned from very early developmental stages. The ngn's may thus provide early markers of responses to inductive signals which are thought to pattern this axis of the spinal cord (Basler et al., (1993) Cell in press; Liem et al., (1995) Cell 82, 969-979; Yamada et al., (1993) Cell 73, 673-686).
Similarly, in the neural crest ngn2 mRNA abundance appears to differ between thoracolumbar and cranial regions. This suggests that there may be positional differences in early migrating neural crest populations, even caudal to the hindbrain crest which is known to be heterogeneous (Hunt and Krumlauf, (1991) Cell 66, 1075-1078). In the trunk region, moreover, ngn2 is expressed in migrating crest and early sensory ganglia but is never detected in autonomic derivatives (which instead express Mash1 (Lo et al., 1991)). This is consistent with the idea (Le Douarin, (1986) Science 231, 1515-1522) that the sensory and autonomic lineages of the crest segregate early in migration. ngn2 may therefore provide a useful marker to understand the control of this early lineage segregation event. In this way, the identification of the ngns may provide a new window into the early origins of neuronal diversity.
Sequential and Regionally-restricted Expression of Different bHLH Factors during Mammalian Neurogenesis
Diversity among bHLH factors is thought to subserve different developmental events in different lineages, which occur both in series and in parallel (Jan and Jan, 1993; Weintraub, 1993). The data presented here are entirely consistent with this idea: different members of the ngn subfamily are expressed in different regions or lineages of the nervous system, while neuroD is expressed downstream of the ngn's in most of these regions. This apparent cascade relationship extends to the endocrine pancreas, where ngn3 seems to be expressed upstream of neuroD. While it is currently difficult to directly demonstrate sequential expression of individual ngn's and neuroD within the same cell lineage, it seems likely by analogy to Xenopus (Ma et al., 1996) that this is indeed the case. Not only is X-ngnr-1 able to activate XneuroD, but mouse ngn1 has been demonstrated to do so in Xenopus , as well (Ma et al., 1996). As ngn2 is even more closely related to X-ngnr-1 than is ngn1, it is likely that NGN2 is also an upstream activator of neuroD; by extension, the same may hold true for ngn3.
Although we have emphasized the striking convergence of different ngn's onto neuroD, additional neuroD-like genes have been identified in this and in other studies (Kume et al., 1996; McCormick et al., 1996; Yasunami et al., (1996) Biochem. Biophys. Res. Comm. 220, 754-758). As targeted mutations in the neuroD gene in mice interfere with pancreatic but not neuronal development (Lee et al., personal communication), it seems likely that at least some of these neuroD-related genes are functionally-redundant with neuroD itself and therefore that their expression must partially overlap. In that case, different ngn's may activate multiple neuroD-like genes in at least some regions.
In Drosophila , five positive-acting bHLH factors have been identified which promote neurogenesis (Jan and Jan, 1993), and a similar number of positive bHLH factors have been implicated in mammalian myogenesis (Olson and Klein, 1994). In contrast, the number of mammalian neural-specific, putatively positive-acting bHLH factors is at least three times as large (for review, see (Guillemot, (1995) Biol. Cell. 84, 3-6). (A growing subfamily of negative-acting bHLH factors related to Drosophila hairy and enhancer of split (Sasai et al., (1992) Genes & Dev. 6, 2620-2634) is not included in this calculation). Why are so many bHLH genes necessary in the mammalian nervous system? Evidence from Drosophila suggests that different proneural genes specify not only the neural fate per se (Campuzano and Modolell, 1992), but also different neural precursor identities (Jarman et al., 1993b; Skeath and Doe, 1996). Thus the relatively large number of atonal-related bHLH genes may reflect the great diversity of cell types in the mammalian nervous system. It is important to emphasize that the domains of ngn expression exclude large regions which instead express Mash1 (Ma, et al., in preparation). In this respect, it is curious that a similar extent of diversity has not yet been observed for the Mash-related subfamily, only one member of which is specifically expressed in the mammalian nervous system (Guillemot et al., 1994; Johnson et al., 1990). Whether this reflects technical or rather biological factors remains to be determined.
As in muscle (for review, see (Weintraub, 1993)), the multiplicity of neural bHLH genes also leads to partial functional redundancy. Thus, mutations in neuroD have no apparent neural phenotype (J. Lee, personal communication), and a targeted disruption of Mash1 affects the PNS but not the CNS (Guillemot et al., 1993), despite the fact that this latter gene is widely expressed in the brain and spinal cord (Guillemot and Joyner, 1993; Lo et al., 1991). This apparent functional redundancy may arise from cross-regulatory interactions between related bHLH genes, as demonstrated for achaete-scute in Drosophila (Martinez and Modolell, 1991) and the myogenic bHLH genes in mammals (for review see (Weintraub, 1993)). The overlapping expression of ngn1 and ngn2 in some brain regions (e.g., cortex) may similarly reflect such cross-regulation, and suggests that these two genes may exhibit partial functional redundancy as well. Single and double mutants in these genes, currently under construction, should resolve this issue.
Overlapping Expression of Neurogenins in the PNS and CNS
Several regions of the CNS and PNS, such as the cortex and DRG, show at least partially-overlapping expression of ngn1 and ngn2. The sensitivity and resolution of double-label in situ hybridization have been insufficient, at least in our hands, to resolve the issue of whether these two genes are expressed in the same cells, or in distinct but intermingled cell populations. As mentioned above, the precedent for cross-regulatory interactions between related bHLH genes argues for the former interpretation. However, there are also several reasons to consider the latter possibility as well. First, expression of the two genes is not obligatorily coupled: we have demonstrated many regions which express one gene but not the other. Second, in those places where overlap does occur, it is often incomplete: thus in the cortex and hindbrain the boundaries of ngn1 and ngn2 expression are distinct. Third, all of the regions or structures where ngn1 and ngn2 expression overlaps are known to contain phenotypically distinct but intermingled neuronal cell types (Ramon y Cajal, 1995).
A final reason is the precedent provided by recent revisions in our understanding of the roles that diverse bHLH factors play in myogenesis (Braun and Arnold, 1996; Braun et al., 1994; Patapoutian et al., 1995; Smith et al., 1994). Initially, myf5, myoD and myogenin were all thought to be sequentially expressed in a common myogenic cell population within the somites (for review, see (Lyons, 1994)). Subsequent work has revealed, however, that this is not the case: rather, myf5 and myoD are expressed by two distinct myoblast populations (Braun and Arnold, 1996), both of which subsequently express myogenin (Smith et al., 1994). Moreover, expression of myf5 and myoD in these two lineages initiates at different times, so that myogenin has already been expressed in the myf5-dependent lineage before myoD expression has been initiated in the other lineage. In the same way, the sequential but overlapping expression of ngn1 and ngn2 in developing sensory ganglia could reflect their expression in distinct sublineages on different schedules. If so, then the expression of these two genes would provide the first evidence that the lineages that generate different sensory neuron subtypes (Snider, 1994) are segregated from a very early stage.
The idea that ngn1 and ngn2 are involved in the specification of different sensory neuron subtypes is further suggested by their pattern of expression in cranial sensory ganglia. We have shown that several of these ganglia express ngn2 but not ngn1, e.g., the VII th and IX th ganglia. Interestingly, the neurons in these ganglia are primarily derived from the epibranchial placodes (Noden, 1993), which also express high levels of ngn2 (but not ngn1). Taken together, these facts are consistent with the idea that the ngn2-expressing cells in these cranial ganglia represent placode-derived neuronal precursors. Other cranial ganglia, however, such as the V th (trigeminal), express ngn1 but not ngn2. Interestingly, these other ganglia contain crest-derived as well as placode-derived neurons (Le Douarin, 1982), and it is intriguing to consider that in such dual-origin ganglia ngn1 may be expressed exclusively by crest-derived neuronal precursors. Alternatively, ngn1 and ngn2 might define different sensory sublineages irrespective of their placodal vs. crest origin. Whatever the case, expression of ngn1 and 2 in different cranial sensory sublineages would be consistent with the idea that these genes mark distinct sublineages in trunk sensory ganglia as well.
Cryptic Cellular Heterogeneity Revealed by Expression of the Neurogenins
In addition to their potential importance in understanding the molecular control of neurogenesis, the ngn's may provide useful markers of specific neural progenitor subpopulations for cellular studies of neurogenesis. This is seen in the developing spinal cord where, for example, ngn1, 2 and 3 each define distinct groups of cells within the VZ, at different positions along the dorso-ventral axis. The developmental significance of these cell groups is not clear, but they indicate that the VZ of the spinal cord is patterned from very early developmental stages. The ngn's may thus provide early markers of responses to inductive signals which are thought to pattern this axis of the spinal cord (Basler et al., 1993; Liem et al., 1995; Yamada et al., 1993).
Similarly, in the neural crest ngn2 mRNA abundance appears to differ between thoracolumbar and cranial regions. This suggests that there may be positional differences in early migrating neural crest populations, even caudal to the hindbrain crest which is known to be heterogeneous (Hunt and Krumlauf, 1991). In the trunk region, moreover, ngn2 is expressed in migrating crest and early sensory ganglia but is never detected in autonomic derivatives (which instead express Mash1 (Lo et al., 1991)). This is consistent with the idea (Le Douarin, 1986) that the sensory and autonomic lineages of the crest segregate early in migration. ngn2 may therefore provide a useful marker to understand the control of this early lineage segregation event. In this way, the identification of the ngns may provide a new window into the early origins of neuronal diversity.
Sequential and Regionally-restricted Expression of Different bHLH Factors during Mammalian Neurogenesis
Diversity among bHLH factors is thought to subserve different developmental events in different lineages, which occur both in series and in parallel (Jan and Jan, 1993; Weintraub, 1993). The data presented here are entirely consistent with this idea: different members of the ngn subfamily are expressed in different regions or lineages of the nervous system, while neuroD is expressed downstream of the ngn's in most of these regions. This apparent cascade relationship extends to the endocrine pancreas, where ngn3 seems to be expressed upstream of neuroD. While it is currently difficult to directly demonstrate sequential expression of individual ngn's and neuroD within the same cell lineage, it seems likely by analogy to Xenopus (Ma et al., 1996) that this is indeed the case. Not only is X-ngnr-1 able to activate XneuroD, but mouse ngn1 has been demonstrated to do so in Xenopus , as well (Ma et al., 1996). As ngn2 is even more closely related to X-ngnr-1 than is ngn1 ( FIG. 2 ), it is likely that NGN2 is also an upstream activator of neuroD; by extension, the same may hold true for ngn3.
Although we have emphasized the striking convergence of different ngn's onto neuroD, additional neuroD-like genes have been identified in this and in other studies (Kume et al., 1996; McCormick et al., 1996; Yasunami et al., 1996). As targeted mutations in the neuroD gene in mice interfere with pancreatic but not neuronal development (Lee et al., personal communication), it seems likely that at least some of these neuroD-related genes are functionally-redundant with neuroD itself and therefore that their expression must partially overlap. In that case, different ngn's may activate multiple neuroD-like genes in at least some regions.
In Drosophila , five positive-acting bHLH factors have been identified which promote neurogenesis (Jan and Jan, 1993), and a similar number of positive bHLH factors have been implicated in mammalian myogenesis (Olson and Klein, 1994). In contrast, the number of mammalian neural-specific, putatively positive-acting bHLH factors is at least three times as large (for review, see (Guillemot, 1995)). (A growing subfamily of negative-acting bHLH factors related to Drosophila hairy and enhancer of split (Sasai et al., 1992) is not included in this calculation). Why are so many bHLH genes necessary in the mammalian nervous system? Evidence from Drosophila suggests that different proneural genes specify not only the neural fate per se (Campuzano and Modolell, 1992), but also different neural precursor identities (Jarman et al., 1993b; Skeath and Doe, 1996). Thus the relatively large number of atonal-related bHLH genes may reflect the great diversity of cell types in the mammalian nervous system. It is important to emphasize that the domains of ngn expression exclude large regions which instead express Mash1 (Ma, et al., in preparation). In this respect, it is curious that a similar extent of diversity has not yet been observed for the Mash-related subfamily, only one member of which is specifically expressed in the mammalian nervous system (Guillemot et al., 1994; Johnson et al., 1990). Whether this reflects technical or rather biological factors remains to be determined.
As in muscle (for review, see (Weintraub, 1993)), the multiplicity of neural bHLH genes also leads to partial functional redundancy. Thus, mutations in neuroD have no apparent neural phenotype (J. Lee, personal communication), and a targeted disruption of Mash1 affects the PNS but not the CNS (Guillemot et al., 1993), despite the fact that this latter gene is widely expressed in the brain and spinal cord (Guillemot and Joyner, 1993; Lo et al., 1991). This apparent functional redundancy may arise from cross-regulatory interactions between related bHLH genes, as demonstrated for achaete-scute in Drosophila (Martinez and Modolell, 1991) and the myogenic bHLH genes in mammals (for review see (Weintraub, 1993)). The overlapping expression of ngn1 and ngn2 in some brain regions (e.g., cortex) may similarly reflect such cross-regulation, and suggests that these two genes may exhibit partial functional redundancy as well. Single and double mutants in these genes, currently under construction, should resolve this issue.
Overlapping Expression of Neurogenins in the PNS and CNS
Several regions of the CNS and PNS, such as the cortex and DRG, show at least partially-overlapping expression of ngn1 and ngn2. The sensitivity and resolution of double-label in situ hybridization have been insufficient, at least in our hands, to resolve the issue of whether these two genes are expressed in the same cells, or in distinct but intermingled cell populations. As mentioned above, the precedent for cross-regulatory interactions between related bHLH genes argues for the former interpretation. However, there are also several reasons to consider the latter possibility as well. First, expression of the two genes, is not obligatorily coupled: we have demonstrated many regions which express one gene but not the other. Second, in those places where overlap does occur, it is often incomplete: thus in the cortex and hindbrain the boundaries of ngn1 and ngn2 expression are distinct. Third, all of the regions or structures where ngn1 and ngn2 expression overlaps are known to contain phenotypically distinct but intermingled neuronal cell types (Ramon y Cajal, 1995).
A final reason is the precedent provided by recent revisions in our understanding of the roles that diverse bHLH factors play in myogenesis (Braun and Arnold, 1996; Braun et al., 1994; Patapoutian et al., 1995; Smith et al., 1994). Initially, myf5, myoD and myogenin were all thought to be sequentially expressed in a common myogenic cell population within the somites (for review, see (Lyons, (1994) Am. Zool. 34, 305-312). Subsequent work has revealed, however, that this is not the case: rather, myf5 and myoD are expressed by two distinct myoblast populations (Braun and Arnold, 1996), both of which subsequently express myogenin (Smith et al., 1994). Moreover, expression of myf5 and myoD in these two lineages initiates at different times, so that myogenin has already been expressed in the myf5-dependent lineage before myoD expression has been initiated in the other lineage. In the same way, the sequential but overlapping expression of ngn1 and ngn2 in developing sensory ganglia, for example, could reflect their expression in distinct sublineages on different schedules. If so, then the expression of these two genes would provide the first evidence that the lineages that generate different sensory neuron subtypes (Snider, 1994) are segregated from a very early stage.
The idea that ngn1 and ngn2 are involved in the specification of different sensory neuron subtypes is further suggested by their pattern of expression in cranial sensory ganglia. We have shown that several of these ganglia express ngn2 but not ngn1, e.g., the VII th and IX th ganglia. Interestingly, the neurons in these ganglia are primarily derived from the epibranchial placodes (Noden, 1993), which also express high levels of ngn2 (but not ngn1). Taken together, these facts are consistent with the idea that the ngn2-expressing cells in these cranial ganglia represent placode-derived neuronal precursors. Other cranial ganglia, however, such as the V th (trigeminal), express ngn1 but not ngn2. Interestingly, these other ganglia contain crest-derived as well as placode-derived neurons (Le Douarin, 1982), and it is intriguing to consider that in such dual-origin ganglia ngn1 may be expressed exclusively by crest-derived neuronal precursors. Alternatively, ngn1 and ngn2 might define different sensory sublineages irrespective of their placodal vs. crest origin. Whatever the case, expression of ngn1 and 2 in different cranial sensory sublineages would be consistent with the idea that these genes mark distinct sublineages in trunk sensory ganglia as well.
Cryptic Cellular Heterogeneity Revealed by Expression of the Neurogenins
In addition to their potential importance in understanding the molecular control of neurogenesis, the ngn's may provide useful markers of specific neural progenitor subpopulations for cellular studies of neurogenesis. This is seen in the developing spinal cord where, for example, ngn1, 2 and 3 each define distinct groups of cells within the VZ, at different positions along the dorso-ventral axis. The developmental significance of these cell groups is not clear, but they indicate that the VZ of the spinal cord is patterned from very early developmental stages. The ngn's may thus provide early markers of responses to inductive signals which are thought to pattern this axis of the spinal cord (Basler et al., 1993; Liem et (1995); Yamada et al., 1993).
Similarly, in the neural crest ngn2 mRNA abundance appears to differ between thoracolumbar and cranial regions. This suggests that there may be positional differences in early migrating neural crest populations, even caudal to the hindbrain crest which is known to be heterogeneous (Hunt and Krumlauf, 1991). In the trunk region, moreover, ngn2 is expressed in migrating crest and early sensory ganglia but is never detected in autonomic derivatives (which instead express Mash1 (Lo et al., 1991)). This is consistent with the idea (Le Douarin, 1986) that the sensory and autonomic lineages of the crest segregate early in migration. ngn2 may therefore provide a useful marker to understand the control of this early lineage segregation event. In this way, the identification of the ngns may provide a new window into the early origins of neuronal diversity.
BIBLIOGRAPHY


Akazawa, et al., (1995) J. Biol. Chem. 270, 8730-8738.
Akazawa, et al., (1992) J. Biol. Chem. 21879-21885.
Alonso et al., EMBO J. 1988 7:2585
Amon, A. (1996) Cell. 84, 651-654.
Artavanis-Tsakonas, et al., (1995) Science. 268, 225-232.
Bartholomaa, et al., (1994) Mech. Devel. 48, 217-228.
Basler et al., 1993 Cell 73:687
Birren et al., 1993 Development 119:597
Blaugrund et al., 1996 Development 122:309
Bobola, et al., (1996) Cell. 84, 699-710.
Brand, et al., (1993) Development. 119, 1-17.
Braun et al., 1996 EMBO J. 15:310
Braun et al., 1994 Development 120:3083
Campuzano, et al., (1992) Trends Genet. 8, 202-208.
Chitnis, et al., (1995) Nature. 375, 761-766.
Chitnis, et al., (1996) Development. in press,
Chitnis, A. B. (1995). Mol. Cell. Neurosci. 6, 311-321.
Chomczynski et al. (1987) Anal. Biochem. 162:156
Cserjesi et al., 1995 Development 121:1099
Coffman, et al., (1990) Science. 249, 1438-1441.
Coffman, et al., (1993) Cell. 73, 659-671.
Cubas, et al., (1991) Genes & Dev. 5, 996-1008.
Dominguez, et al., (1993) EMBO J. 12, 2049-2060.
Ferreiro, et al., (1994) Development. 120, 3649-3655.
Ferreiro, et al., (1992) Mech. Development. 40, 25-36.
Garrell et al., 1991 BioEssays 13:493
Ghysen, et al., (1993) Genes & Dev. 7, 723-733.
Guillemot 1995 Biol. Cell. 84:3-6
Guillemot, et al., (1993) Mech. Devel. 42, 171-185.
Guillemot, et al., (1993) Cell. 75, 463-476.
Hinz, et al., (1994) Cell. 76, 77-88.
Hunt et al., Cell 1991 66:1075
Ishibashi, et al., (1993) Eur. J. Biochem. 215, 645-652.
Jan, Y. N. and Jan, L. Y. (1993) Cell. 75, 827-830.
Jan, Y. N. and Jan, L. Y. (1994) Annu. Rev. Genet. 28:373-393
Jarman, et al., (1993a) Development. 119, 19-29.
Jarman, et al., (1993b) Cell. 73, 1307-1321.
Jarman et al., 1994 Nature 369:398
Johnson, et al., (1990) Nature. 346, 858-861.
Kintner, et al., (1987) Development. 99, 311-325.
Kopan, et al., (1994) Development. 120, 2385-2396.
Kozak, M. (1984) 12, 857-872.
Kume, et al., (1996) Biochem. Biophys. Res. Commun. 219, 526-530.
Kunisch, et al., (1994) Proc. Natl. Acad. Sci. U.S.A. 91, 10139-10143.
Lamb, et al., (1993) Science. 262, 713-718.
Le Dourin et al., 1982 The Neural Crest. Cambridge Univ. Press
Le Dourin 1986 Science 231:1515
Lee, et al., (1995) Science. 268, 836-844.
Liem et al., 1995 Cell 82:969
Lo, et al., (1991) Genes & Dev. 5, 1524-1537.
Lyons 1994 Am. Zool. 34:305
Ma et al., 1996 Cell 87:43
Martinez et al., 1991: Science 251:1485
McCormick et al., 1996: Mol. Cell. Biol. 16:5792
Meyer et al., 1995 Nature 378:386
Naya, et al., (1995) Genes & Dev. 9, 1009-1019.
Noden et al., 1993 J. Neurobiol. 24:248
Nye, et al., (1994) Development. 120, 2421-2430.
Olson, et al., (1994) Genes & Dev. 8, 1-8.
Patapoutian et al., 1995 Development 121:3347
Ramon y Cajal 1995: History of the Nervous System of Man and Vertebrates, Oxford University Press, New York.
Saito, et al., (1995) Mol. Cell. Neurosci. 6, 280-292.
Sasai, et al., (1992) Genes & Dev. 6, 2620-2634.
Sasai, et al., (1995) Nature. 376, 333-336.
Shimizu, et al., (1995) Eur. J. Biochem. 229, 239-248.
Sil, et al, (1996) Cell. 84, 711-722.
Skeath et al., 1996 Curr. Biol. 6:1146
Smith, et al., (1992) Cell. 70, 829-840.
Snider et al., 1994 J. Cell. Bio. 127:95
Sommer, et al., (1995) Neuron. 15, 1245-1258.
Struhl, et al., (1993) Cell. 74, 331-345.
Turner, et al, (1994) Genes Dev. 8, 1434-1447.
Villares, et al., (1987) Cell. 50, 415-424.
Weintraub, H. (1993). Cell. 75, 1241-1244.
Wilkinson 1992 in In Situ Hybridization: A Practical Approach, IRL Press, Oxford
Yamada et al., 1993 Cell 73:673
Yasunami et al., 1996 Biochem Biophys. Res. Comm. 220:754
Zimmerman, et al., (1993) Development. 119, 221-232.