The NK-2 Homeobox Gene and the Early Development of the Central Nervous System of Drosophila MARSHALL NIRENBERG,? KOHZO NAKAYAMA,’ NORIKO NAKAYAMA,’ YONGSOK KIM,’ DERVLA MELLERICK,? -LAN-HSIANG WANG,’ KEITH O. WEBBER,? AND RAJNIKANT LAD‘ 4 Laboratory of Biochemical Genetics National Heart, Lung, and Blood Institute National Institutes of Health Bethesda, Maryland 20892 ’ Department of Immunobiology Cancer Research Institute Kanazawa University Takaramachi 13-1 Kanazawa, Ishikawa 920, Japan © Laboratory of Molecular Cardtology National Heart, Lung, and Blood Institute National Institutes of Health Bethesda, Maryland 20892 4 Laboratory of Molecular Biology National Cancer Institute National Institutes of Health Bethesda, Maryland 20892 © Department of Psychiatry Hospital of the University of Pennsylvania Philadelphia, Pennsylvania My colleagues and I deciphered the genetic code gradually, over a period of five years, between 1961 and 1965. I then stopped working on the code and began working in the field of neurobiology. The logic that connects the ge- netic code to neurobiology is that information is processed in both genetic and neural systems. My interest in the NK-2 homeobox gene! of Drosophila stems from the observation that NK-2 is the earliest predominantly neural gene regulator that has been found thus far that is expressed in the ventrolateral neurogenic an- 224 NIRENBERG ¢t ai.: NK-2 HOMEOBOX GENE 225 FIGURE 1. Side view of stage 5 Drosophila embryo illustrates concentration gradients of proteins that regulate gene expression. The concentration of bicoid homeobox protein is high in the ante- rior (A) region and low in the posterior (P) region of the embryo. A concentration gradient of nanos protein is established in the posterior to anterior direction. A concentration gradient of dorsal protein is established in nuclei in the ventral (V) towards the dorsal (D) region of the em- bryo. lage, which gives rise to part of the central nervous system of the embryo. I will tell you what we know about the NK-2 gene and relate these findings to the early development of the Drosophila embryo and the central problem of understanding the principles that are used initially to construct part of the cen- tral nervous system of the embryo. The studies on NK-2 were performed by my colleagues, Yongsok Kim, Kohzo Nakayama, Noriko Nakayama, Dervla Mellerick, Lan-Hsiang Wang, Keith Webber, and Rajnikant Lad. The nucleus of a fertilized Drosophila embryo undergoes 13 rounds of nu- clear division in the first 130 minutes of embryonic development, resulting in an embryo that consists of a single cell with approximately 5,000 nuclei. Most of the nuclei move to the periphery during stage 4 (80-130 minutes after fertilization, nuclear divisions 10-133) and cell membranes form around each nucleus between 130 and 170 minutes after fertilization (stage 5). The anterior-posterior and ventral-dorsal axes of the embryo are established and different cell types are generated during stages 4 and 5 by the formation of concentration gradients of proteins that regulate gene expression (Fic. 1). A concentration gradient of bicoid homeobox protein is established in the anterior-posterior direction (high concentration in the anterior portion of the embryo, low concentration in the posterior portion**); and a concentration gradient of nanos protein is established in the posterior to anterior direc- tion.”-!0 Concomitantly, a concentration gradient of dorsal protein is estab- lished in nuclei, with the highest concentration of dorsal protein in nuclei in the ventral portion of the embryo and the lowest concentration in nuclei in the dorsal part of the embryo.!!-!4 These gene regulators and terminal gene 226 ANNALS NEW YORK ACADEMY OF SCIENCES regulators initiate the induction or repression of other genes that encode pro- teins that regulate gene expression and result in dynamically changing patterns of gene expression in different parts of the embryo, depending upon the con- centrations of gene regulators that the nuclei were exposed to. The anterior- posterior gradients of gene regulators result in the formation of vertical stripes of equivalent nuclei that were exposed to the same concentrations of gene reg- ulators, whereas the ventral to dorsal gradient of dorsal protein results in the formation of horizontal stripes of equivalent nuclei (for reviews see Refs. 15-18). In effect, the embryo is divided into a bilaterally symmetric checker- board of clusters of nuclei that express different combinations of genes for pro- teins that regulate genes. The position of a nucleus in the embryo therefore determines the initial developmental fate of the nucleus. In FIGURE 2 is shown the composite nucleotide sequence of NK-2 cDNA and genomic DNA and the deduced amino acid sequence of NK-2 protein.!° The NK-2 gene contains 3 exons and 2 introns; introns ] and 2 are approx- imately 1.68! and 3.1!° kb in length, respectively. 3 NK-2 protein contains two regions near the N-terminus that consist almost entirely of alternating acidic and basic amino acid residues or pairs of acidic and basic amino acid residues. The protein contains multiple Ala repeats, an Asn repeat, an acidic domain, followed by a homeodomain, which is not closely related to any other Drosophila homeodomain. The C-terminal region of NK-2 protein con- tains a 17—amino acid residue sequence of unknown function termed the NK- 2 box (amino acid residues 631-647) that has been highly conserved during evolution and a His-Ala repeat. The NK-2 gene was shown to reside on the sex chromosome at 1C]-5.1 The NK-2 homeodomain has been conserved during evolution. FiGURE 3A compares the amino acid sequence of the Drosophila NK-2 homeodo- main!!9 and NK-2-like homeodomains from Xenopus,?° mouse,?!-2? pla- naria,?3,24 leech,?> and tapeworm.© The similarity in amino acid sequence ranges from 95 to 67 percent. The mouse genome contains six copies of the NK-2 gene, presumably formed by gene duplication. The amino acid residues that comprise NK-2 homeodomain a-helices I, IH, and III were determined by NMR.’ Binding of a 77-amino acid residue protein that contains the NK-2 homeodomain to a high-affinity NK-2 binding site in DNA results in an increase in the length of a-helix III from 11 to 19 amino acid residues (from NK-2 homeodomain residues 42-52 to 42-60) and also increases the stability of the secondary structure of the homeodomain.?” Xenopus and mouse proteins with NK-2-like homeodomains also contain the highly conserved 17-amino acid residue NK-2 box after the homeodo- main (94-77% homology with the NK-2 box sequence of Drosophila NK-2 protein) (Fic. 3B). The Drosophila NK-3!:?° (bagpipe)?® homeodomain pro- tein also contains a sequence related to the NK-2 box (47% homology); how- NIRENBERG et al.: NK-2 HOMEOBOX GENE 227 CCAGTTGTTGATGACACTTCAGTTTGGAATGT SGCTCGTK SCGTGCAACT! A RACGACTCGCGACCCTTGAACTATCGAACTTTCCCGGTGATAA ACTT. ARACTTAAGOCT AAATCGAACTATGT CTOGAATT TAAGTTGCCCT ACCAGGA TACCOGCTTAA PI CATCOCAGIGCT ACCCAGTAACTACACATARACATATI (CTI PCOCGTOCT CCGAACACCGTAAAAGCTCCAA AACGCAAAAGTCCOCTC AATCAAACAGTOGTT A CA ur T CGTCGGOGTCCTTGGAGAGGACCCCCTCAAAGCGGGA TOGA GA TCGCGAGCGGGACAACAGCAGC GGT! CTGGGCAGCGCTGGCAGCTT GCCCGCATCGCCCCAAAGCGCTATCACGGTGA SAS LERTPSKEDRORERDNSSGLGSAGSLPASPQESEATTVS GTCCGTCCTCCCCAGCCACGCCGAAGGCCCOGCTSCCGACCT CAACEC CCT OBC: C SGCAAGGAGCGGCAGGAACGGC PSS PAT PKA PLPT ST PSLEREREBR BS DRE DR EDA KE RO EL BY ACGAGCGAGACAGAGATCACGAGAGATTCGCCGCAGT CTT CAGCACCGCTAGCACCACCUTGCCCACGAACACAAGTTCCAGT TCCGGATT GGCGCCCGAACAGCT 'CCGCATTCCGACGG ERDRDHERFAAVFSTASTTVPTNTS SSESGCLAPBEQLRIP TS CAGCGGA TTT CCGGGGCTCCACAGCATGAGTAGTCTTATGCTTCCATCGTCGGCGGCTST GECCGOCOCAGCA SEGSCCOCOT TPCT GCOS. 'GETCGCCCATCOTGC AAAPSGFPGLHSMSSLUMOEPS SRAVYVAAABRAA P FL PWS PILL y TECCECOSTGGAACCACGCCCTOCTACCAGCOGCCT TT TAT COGGCGGCCCT GCGAAACGCTTTGCCT OGCTTATTCGATOCAAAGGT GCOGTCOTCOCAACOCTCTGGCTICCATATAT PPWRNRHALULPAAFYPAALRNALPGLFOAKVP SSQREGPHI S COGACATCTTGAATTTGGAGGGCT.CTGAGCTGAAGAATOCAGCAGCTGCCGCCGCT GCTOCCACCCACCATGGCAGCGATTTGAGTCACCACT COGCCAGT GAGT CCACCAGTGGRCATC DILNLe#os &€&LKNAAAAAAAAAHHGSODLSHHSASESTSGHR GCOGCCAGGGATICT. CACACOTCGCCT TCGGCCGTETCGCCCA CGCCAGCGGGCGTTICCGCGGACGAGCACCACAATGGCA GCGGGACT GGGGGAGGA GCGBGAGANGCGGATCACCACA ¢qcSuHhTsPSAVSPTPAGVSADEHHNGSGTGGGAGEA D HHS QCACCACCGAGCATCACGCCCCTCCTAGCCAT AGCAACAGCACCC CCAGCA! STCACCICCTC! \GCAGCACCATC! SGGTGGCCCCAC TIEMHA PPSHE GOO PHY HOG PEEL LP Oo OO A VA P OL IRCOCTTTEGCGCACCATCAGAGOGGCGAAGCCCAAAGCCACGCCCATGCCAATOCAGCOOCCGCCCA TCT GCTGGCCAGOCACAATSCA GERI TACGICT OCT OT GCE GCCBECCAAT PLA HHOSGEAQSHAHANAAAAHLL ASH NAAAAA AY AA GO Y ATCT GCCCAACCTGCCCAAGAACTTCCCGGGGAGCTIT: GGCGACGAGATGTCCTOGTACCACCACATGGCOCAAACTA TGCT GCA GCACTCGGGCAGGAGT ‘GCGTGGATCAAGGAGAACG DPNOLPKNFPGSFGDEXSSYHHHRAQTMLAHS GRSEAWI RENE q AGCTATACGGT. ACCCAGCAGCCOGCCA GT. CCGGA TAGCACCT CCCCAGTT ACCTOGGAA GTGTCGTACACCTACATT GGT TCCAACTSCCAGACAT COCCT GOCCTTTCCGGCGATTACA LYGcG?TQqgQPASPDSTSPVTSEVSYTYIGSNCQTSPALSEDY K AGAGCTACAGCOGGTCGGCCGATAGCGATGCACTATCCOTGGGCGACGCCCTGCACACCCTICCATGGA TCCTCT GGT AAT GGAAGT GCCGGAGGOGCT CCGA CGGCCCA TOCCCTACACA 5 YS RSADSDALSVGDRDALHTLHGSSEGNGSAGGAPTAHALHAYD ACAACAATAATAATACGACAAACAACAATAACCACAGCCTGAAGGCL AC ‘OGT SCTCAACGA ‘CGACG ‘NNNNTOINNREBN HS LUKAEGINGAGSGHOD SL NE DG 1 ££ p12 p 120 240 360 4380 43 600 83 720 123 940 163 960 203 1080 243 1200 283 1320 323 1440 363 1560 403 1680 4a3 1800 483 1920 $23 ACGTGGACGACGCCGACGGCAGTGGCGUCGGGGAT GCAAAT GGA TCOGACGGT CTGCCAAA TARG bh AACGGAAGCGACGAGTCCTGTTCACCAAGGCGCAAACATATGAGCTGGAACGTC VRBDADGSGGGDANGSDGLPNWNKKRKERRVLF TKAQT Y EL ERA GGTTTOGA CAACAACGTT ACT T GAGTGOCCCGGAAGGCGA GCACCT GGCCAGTT TGA TCCGCCTGACGCCGACCCA GGTGAAGATCTGGTTTCAAAACTA TCGCT ACARGACGAAGC GGG rRQOQRYLSAPERREHUASLUIRLT PT QV RK IWF QN HW ORY K T OK RA 2040 563 2160 603 CGCAAAACGAGAAGGGCTACGAGGGTCATCCTGGTCT ACT GCACGGCCATGCCACCCATCCGTA TCACCOCAGTGCCCTCCATCSLCC CGN CGTTCCAGTTCTGGTGAGGA Q EKRKGYEGHuHPGtLLHGHA TH PH HPS AL P ACGGAAAGCCCTGCTTGGOCGATAGTT.CCARACTGGGAGCOGACTGCGT CT CCGTGTCATCAGCCACCGCCACCGCCA TGCAGAATOCCGCCGCOCATCACTTGGTTGCCCTAAA TOGA GERERPCLGDSSKLGADCVSVYVSSATATAMQNAAAH HL VALN GA CGGCCECCT AT CAACA TGCCGCTGCAGCGGCTOCCEGCOTTCACGCCCACGCCCATGCTCATACCCACOCCCACGGACACGGCCA TOCACACGCOCACGCCTAGA SESE C OCCT GST GST AA YQHAAAAAA G LHAHABHAHAHA HU GH GH P HAHA Q RA AWW P CCTAATATTOCTAGGAACTGGCATTCACGGGACTOGGGGCGGATAGGTOGA TAGTCAGGCCOGT GGAGCGATGAGCCAGAAGACCACGCCTCCAGATTCCCAGT GAAGCCAAGTAAAAGA : 2 TATACAATAATAATTGAAACACATACAAACCCCT ACT AGCGCTCAGT CTCCCGGACTTATCCTIGAGTCGTTTCTGTACATAGTTCATATATATARTCATATATATACCTAAATGACAT AAATTTATOCATTCCTOGATATCAGACCANTOCCCAT GACAAAAGAAAGCTACATTTCCTTCTABATAABAATCAATACTAGTGTCTCTTGTCAAGCOCATTT AAT CGCATAAAATTTA GAAGTAGATTCAT EET TPTTATETT TTT TTT TTAGCCCTAAGTCCAATCCCAAAGAT AAAATGTGTAAATATAATATAAAAATGAAGT AAA TCTGAAAAA TAGATTATTICAAAAGAGA AATAAATAATAGGAAATATGCAACTAAAAARAA FIGURE 2. The nucleotide sequence of NK-2 cDNA and the deduced amino acid sequence of NK-2 protein (from Ref. 19). The homeodomain is enclosed within a box. Other interesting amino acid sequences are underlined, such as an acidic domain before the homeodomain and the highly conserved NK-2 box sequence after the homeodomain (amino acid residues 631-647). An inverted triangle shows the position of an intron. 2280 643 2400 683 2520 723 2640 2760 2880 3000 3033 228 ANNALS NEW YORK ACADEMY OF SCIENCES Q-HELIX 3 A SPECIES Q-HELIX 1 QHELIX 2) Q=HELIX 3 sic a 1 11 22 28 38 42 52 60 HOMOL- . . : . . . : . OGY REF, NK-2 d KRKRRVLFTKAQTYELERRFRQQRYLSAPEREHLASLIRLTPTQVKIWFOQNHRYKTKRAQ 100 1, XeNK-2 x s M---R 95 20 Nkx-2,.2 m Ss M. R 95 21 Nkx-2.4 m SQ--V K--K M-H M--OA 83 21 Nkx-2.1 m R sQ--v K--K M-H M--OQA 82 21 Nkx-2.3 m R--P----SQ--VF K SLK~--$ R---C--QR 75 21, Nkx-2.5 m R=--P~---SQ--V------- Kroon en DQ---VLK--S-------~- R---C--OR 73 22 Nkx-2.6 m Q-~S----SQ--VLA---~- K T ALQ--S R---S-SQR 70 22 Dth-l =p ---=----- $-K-IL----H---KK N--G-S M---H 80 23, Dth-2 p R----I--SQ--I K--K N--N C--S- 82 23, Lox 10 1 R----I-~SQ--I K TF-G KSK 80 25 EgHbx-3 t QS----~-N-F-1ISQ--K---K----T-0--QE--HT~G----~-~~------ A--M--LF 67 26 B SPECIES % AAA HOMOLOGY REF. NK-2 d 26 SPRRVAVPVLVR-NGKPC 100 1, 19 XeNK-2 x ll ------------: D---- 94-20 Nkx-2.2 m 110 ------------: D---- 94.21 Nkx-2.3 m 140 Peewn-------: D---- 88S 21 Nkx-2.4 m 2300 terete nnn K'D---- 8821 Nkx-2.1 m 34 ----------- K*D---- 88 = 21 Nkx-2.5 m 13. PA--I-------- D---- 77 22 Nkx~-2.6 m 13. PA----~---- L-D---- 77 = 22 NK-3 (bagpipe) d 10 ASK--P-Q----ED-STT 47 1, 28-29 FIGURE 3. (A) Comparison of the amino acid sequence of the (d) Drosophila NK-2 homeo- domain with similar homeodomains from (x) Xenopus, (m) mouse, (p) planaria, (I) leech, and (t) tapeworm (from Ref. 19). The symbol (—) represents the same amino acid residue as NK-2. The amino acid residues of Drosophila NK-2 homeodomain a-helix 1, a-helix 2, and a-helix 3 were determined in NMR.2’ In the absence of DNA a-helix 3 extends from amino acid residue 42 through 52. However, binding of the NK-2 homeodomain to a high-affiniry NK-2 site in DNA increases the length of a-helix 3 (residues 42-60).2” (B) The NK-2 box is a highly con- served 17—-amino acid residue sequence that is found after the homeodomain in proteins related to NK-2. AAA represents the number of amino acid residues between the end of the homeo- domain and the beginning of the conserved NK-2 box sequence. The symbol (-) represents the same amino acid residue; (.) represents the absence of an amino acid residue. ever, a highly conserved NK-2 box was not detected in the planarian homeo- domain proteins, Dth-] or Dth-2. Northern analysis of poly A* RNA from Drosophila at various stages of de- velopment showed that NK-2 mRNA is present in highest concentration in 3-6-hr Drosophila embryos and then progressively decreases during further em- bryonic development.!? No NK-2 mRNA was detected in 0-3-hr Drosophila embryos; therefore, no maternal NK-2 mRNA was found. Larvae and pupae contain greatly reduced levels of NK-2 mRNA compared with that of 3-6-hr embryos; however, an increase in NK-2 mRNA was found in adult flies. FIGURE 4 shows the distribution of NK-2 mRNA in Drosophila embryos as a function of developmental age, determined by # situ hybridization. NIRENBERG et al.: NK-2 HOMEOBOX GENE 229 NK-2 gene expression is initiated during stage 4 in bilaterally symmetrical longitudinal stripes, one stripe on each side, in the ventral (i.c., medial) half of the ventrolateral neurogenic anlage (Fic. 4A). By stage 5, when the first cell membranes are forming around the nuclei, the stripes of nuclei that ex- press NK-2 extend from 0 to 90% of the embryo length and each stripe is 6 or 7 nuclei in width (Fic. 4B). NK-2 is also expressed in part of the proce- phalic neuroectodermal anlage, the endodermal anterior and posterior midgut anlagen, and the hindgut anlage. The ventral mesodermal primordium invag- inates during gastrulation, bringing the longitudinal NK-2 positive bands of neuroectodermal cells closer to the ventral midline, separated only by ventral midline mesectodermal cells, which do not contain NK-2 mRNA (Fics. 4C-F). At first the level of NK-2 mRNA in the band of NK-2-positive cells is fairly homogeneous; however, during early gastrulation clusters of cells with high levels of NK-2 mRNA appear that are separated by vertical stripes of cells, 1 or 2 cells in width, that contain lower levels of NK-2 mRNA, apparently due to repression of the NK-2 gene (Fics. 4E-F). Initially one cluster of cells with a high level of NK-2 mRNA is formed per hemisegment;, later two clusters of NK-2-positive cells appear per hemisegment (Fics. 4G—J). Germ band extension results in an increase in the length of the band of cells that synthesize NK-2 mRNA and a concomitant decrease in the width of the band to 2 to 3 cells per side (Fic. 41). Hence, the neuroectodermal cells that syn- thesize NK-2 mRNA give rise to medial and paramedial neuroblasts that con- tinue to synthesize NK-2 mRNA. Ganglion mother cells and neurons were found that express the NK-2 gene that perhaps are the progeny of neuroblasts that express the NK-2 gene. However, during later embryonic development, the abundance of NK-2 mRNA decreases in some neurons and is extinguished in others. Some neurons that express the NK-2 gene form commissures and others contribute to longitudinal connectives. FiGuRE 5 shows schematic diagrams of cross-sections of embryos at the cellular blastoderm stage (stage 5) before nuclei have been enclosed by cell membranes, and at the end of gastrulation (end of stage 7) after ventral me- sodermal primordium cells have invaginated. These cross-sections of embryos illustrate ventral-dorsal patterning during early development of the Drosophila embryo. The ventral to dorsal concentration gradient of dorsal protein!-!* ac- tivates the tvist3°-34 and snail3+3 genes in the most ventral nuclei, which correspond to the mesodermal anlage, and the NK-2 gene is activated in the ventral (medial) half of the ventrolateral neuroectodermal anlage. A gene reg- ulator specific for dorsal (lateral) neuroectoderm has not been identified thus far. Dorsal protein represses the decapentaplegic (dpp) gene,*’— which encodes a protein that is a homologue of TGF-8,°7#! and the zen-I and zen-2 homeo- box genes*2-6 in nuclei in the ventral and lateral parts of the embryo, but not in nuclei that become dorsolateral epidermoblasts or dorsal amnioserosa, 230 ANNALS NEW YORK ACADEMY OF SCIEN! CES PO aad cid NIRENBERG et al.: NK-2 HOMEOBOX GENE 231 respectively. Hence, the concentration gradient of dorsal protein establishes the ventral-dorsal axis of the embryo and divides the embryo into longitudinal bands of nuclei that have different developmental fates. After cell membranes form, the most ventral cells, the mesodermal anlage, invaginate, which brings the mesectodermal cells to the ventral midline. Neuroectodermal cells gradually segregate as neuroblasts between about 3.5 and 7.3 hours after fertilization*”~4° (Fic. 6). Eventually a monolayer of neuroblasts and glioblasts separated by ventral midline mesectodermal cells is formed above the epidermal cells. Doe*? has shown that thirty-one neuro- blasts or glioblasts delaminate per hemisegment, that each is a unique cell type, and thar the relative position of each neuroblast or glioblast in the set is determined. A Drosophila embryo contains 14 parasegments and additional segments in the head region. Approximately 800 ventrolateral neuroblasts or glioblasts segregate from the medial and lateral neuroectoderm per embryo and addi- tional neuroblasts and glioblasts are formed from mesectodermal cells.4? The pattern of neuroblasts is repeated in different segments possibly with some vari- ation. However, some genes that encode proteins that regulate gene expres- sion are known to be expressed only by cells in a single segment, or a few seg- ments. Although most of the proof is lacking, it is likely that many neuroblasts also express segment-specific gene regulators and that most of the neuroblasts per side eventually will be found to be unique cell types. Three longitudinal stripes of neuroblasts or glial precursors can be distin- guished on each side that are the precursors of neurons and glia of the ventral nerve cord: ventral midline mesectodermal cells that separate the right and left halves of the ventral nerve cord, medial neuroblasts or glial precursors that express the NK-2 homeobox gene, and lateral neuroblasts or glial precursors that have little or no NK-2 mRNA. The monolayer of neuroblasts that give rise to the ventral nerve cord is also divided along the anterior-posterior axis of the embryo into 14 paraseg- ments; most parasegments consist of posterior compartment neuroblasts that FIGURE 4. Distribution of NK-2 mRNA in Drosophila embryos as a function of developmental age (from Ref. 19). The RNA probe used for ## situ hybridization was from the 3’-untranslated region of NK-2 cDNA; the probe did nor contain the homeobox. (A) Expression of the NK-2 gene is initiated during stage 4, the syncytial blastoderm stage. (B) Stage 5-6, side view. (C) Ven- tral view, stage 6, early gastrulation. (D) Ventrolateral view in late stage 6 embryo. (E) Side view of embryo; gastrulation is almost completed. Late stage 7, about 185 minutes after fertilization. Notice the apparent segmentation of the NK-2-positive region. (F) Late stage 7 illustrating ap- parent segmentation of NK-2-positive region. (G) Side view stage 9 embryo 3.7—4.3 hours after fertilization. Two clusters of neuroectodermal cells and/or neuroblasts that contain NK-2 mRNA can be seen per hemisegment. (H) Side view of stage 9-10 embryo. (1) Ventral view of stage 10 embryo. Two clusters of medial neuroectodermal cells and/or neuroblasts that contain NK-2 mRNA are present per hemisegment. (J) Stage 9-10 embryo, side view. 232 ANNALS NEW YORK ACADEMY OF SCIENCES EMIX STAGE 7 140 MIEN. FN IE Obst atte hoody at ee ODER. DOWSAL Tabs ele Lari Prelster atin! avn rtrte i DT AUIEO1On KO ]R] ALB! a NEUROECTODERT : 7 ree om y VENTRAL “an \\\ NEUHOECTONER’ i air NESECTODERM MESODERM FIGURE 5. Schematic drawings of cross-sections of embryos before gastrulation (stage 5) and after gastrulation (end of stage 7) to show the ventral neuroectoderm nuclei or cells (yellow) that express the NK-2 gene. 31 KINDS OF NEUROBLASTS OR GLIOBLASTS NEUROBLAST SEGREGATION ames a ane at Kraee esis yA ene hey Lene y ANTE FIORE NEUROBLASTS lela] GLIOBLASTS aM pier eran en aes rh Se ef Sie) Sena hve} ABOUT $00 VENTROLATERAL NEUROBLASTS ARE FORMED. MOST NEUROBLASTS PROBABLY ARE UNIQUE CELL TYPES. FIGURE 6. Some of the neuroectodermal cells delaminate and form a monolayer of neuroblasts and/or glioblasts immediately above the epidermal cell layer. The figure on the deft is a schematic illustration of a cross-section of an embryo to show the monolayer of neuroblasts or glioblasts that delaminate from the neuroectodermal cell layer. Béve: ventral midline mesectodermal cells that express the sim gene. Yellow: medial neuroblasts that express the NK-2 gene. Red: lateral neuro- blasts and/or glioblasts. Upper right panel shows 31 kinds of neuroblasts or glioblasts that delam- inate per average hemisegment. Four of the delaminated cells have migrated to other positions; hence, only 27 delaminated cells are shown. Lower right panel: Ventral view of the monolayer of neuroblasts and/or glioblasts to illustrate ventral-dorsal and anterior-posterior patterns. The yellow medial neuroblasts and/or glioblasts express the NK-2 gene. NIRENBERG et al.: NK-2 HOMEOBOX GENE 233 express the engrailed (en) homeobox protein*?,5° and anterior compartment cells that do not express en. Hence the 31 neuroblasts or glial precursor cells that segregate per hemisegment are divided into four groups of cells, de- pending on the position of the cells and the expression of the NK-2 and en genes: medial anterior compartment cells that have high levels of NK-2 mRNA but no en mRNA, lateral anterior compartment cells that lack NK-2 and en mRNA, medial posterior compartment cells that have high levels of both NK-2 and en mRNA, and lateral posterior compartment cells that have en mRNA and low levels of NK-2 mRNA. Neuroblasts start to divide soon after they segregate from the neuroecto- dermal cell layer. Each neuroblast division gives rise to a small ganglion mother cell and a large neuroblast, which becomes smaller with each division (Fic. 7). Each ganglion mother cell divides only once and gives rise to two neurons. The first neuroblasts to originate divide about eight times, whereas the last neuroblasts divide five times.*”-*® Therefore, a single neuroblast may be the precursor of 10 to 16 neurons. The neuroblasts that express the NK-2 gene®! in a thoracic segment at the end of neuroblast segregation (late stage 11) are shown schematically in FicureE 8, It should be emphasized that the abundance of NK-2 mRNA changes dynamically during development. NK-2 is expressed by two longitu- dinal columns of medial neuroblasts on each side; however, the abundance of NK-2 mRNA usually is higher in the column of neuroblasts adjacent to the mesectodermal ventral midline cells than in the second column of neuro- blasts. All neuroblasts in the posterior compartment express the NK-2 gene; however, the lateral neuroblasts contain much less NK-2 mRNA than do the medial neuroblasts. Hence, a medial to lateral gradient of NK-2 mRNA is es- tablished in both the anterior and posterior compartments. The amount of NK-2 mRNA in neuroblasts 2-1], 2-3, 5-1, and 5-2 (that is, immediately after or before the posterior compartment) decreases during development, re- sulting in the formation of two clusters of neuroblasts that express the NK-2 gene per hemisegment, one cluster in the anterior compartment and the second consisting of posterior compartment neuroblasts. Some ganglion mother cells and neurons also express the NK-2 gene; however, the levels of NK-2 mRNA decrease markedly in some cells during later stages of embryonic development. These results show that about half of the ventrolateral neuro- blasts express the NK-2 gene and that medial neuroblasts contain higher levels of NK-2 mRNA than do intermediate or lateral neuroblasts. The pattern of expression of the NK-2 gene also was determined in various mutant lines of flies as a function of developmental age.5! The NK-2 gene was found to be activated initially by dorsal in the ventral half of the embryo. However, the NK-2 gene normally is not expressed in the mesodermal anlage because of repression by snail, in the mesectodermal anlage because of repres- sion by sim, or in part of the lateral neuroectodermal anlage or dorsal epi- 234 ANNALS NEW YORK ACADEMY OF SCIENCES FIGURE 7. POSTERIOR NN e=asiO) ; , rere Nal COMPARTMENT vTatsh zen (X) t dpp IN omanleiney GRADIENT L (X) 1 dorsal " sim snail twist Figure 8. FIGURE 9. NE URL Or=) od cero ra. ee LO lety: team =a Odea tuts) -DORSAL NEUROECTODERM -VENTRAL NEUROECTODERM MESECTODERM -MESODERM Ficure 10. NIRENBERG et al.: NK-2 HOMEOBOX GENE 235 dermal anlage because of repression mediated by dpp. Both dorsal and twist were found to be required to activate the NK-2 gene in the hindgut pri- mordium and the posterior midgut primordium. During stage 4 dorsal activates the twist, snail, and NK-2°! genes (Fic. 9). Dorsal represses dpp and the zen genes in the ventral and lateral portions of the embryo; however, the concentration of dorsal is too low for effective re- pression of dpp or zen genes in the dorsolateral and dorsal portions of the em- bryo, respectively (for reviews see Refs. 52-54). snail is expressed only in the most ventral nuclei, which comprise the mesodermal anlage.55-5° sim is acti- vated in both the mesectodermal and mesodermal anlage, but sim is expressed only in the mesectodermal anlage because of repression by snail.3457-6° The NK-2 gene is activated in the mesodermal, mesectodermal, ventral neuroec- todermal, and part of the dorsal neuroectodermal anlagen, but is repressed by snail in the mesodermal anlage, by sim in the mesectodermal anlage, and by a gene regulator that has not thus far been identified whose repression is mediated by dpp in the dorsal neuroectodermal anlage.®! mist is expressed in the mesodermal, mesectodermal, and the ventral portion of the ventral neuro- ectodermal anlagen.°!-62.35 fist protein activates the snad gene in the meso- dermal primordium*® and the NK-2 gene in the ventral portion of the neuro- ectoderm.5! snail represses the sim and NK-2 genes in the mesodermal anlage, 5-59.60 while sim represses the NK-2 gene in the mesectodermal an- lage.*! The hierarchical organization of gene regulation results in the appear- FIGURE 7. Neuroblast division. A neuroblast (green) divides 5-8 times. Each neuroblast di- vision is unequal and gives rise to a slightly smaller neuroblast and a much smaller ganglion mother cell (ed). Each ganglion mother cell divides once, giving rise to two neurons (yellow). FIGURE 8. A ventral view of neuroblasts in a thoracic segment at the end of stage 1] is shown using the neuroblast nomenclature of Doe.49 Neuroblasts shown in color express the NK-2 gene (from Ref. 51). The varying darkness of the neuroblast color from brown to mn represents the relative abundance of NK-2 mRNA; for example, in order of decreasing abundance of NK-2 mRNA, we see brown (neuroblast 4-1), orange (4-2), yellow (5-1), pale yellow (5-2), and tan (2-1). No NK-2 mRNA was detected in black neuroblasts. Medial neuroblasts closest to the ventral mid- line usually contain more NK-2 mRNA than do neuroblasts in a more lateral position in that vertical row. All posterior compartment neuroblasts express the NK-2 gene. A medial to lateral NK-2 mRNA gradient is present. FIGURE 9. Regulation of NK-2 gene expression deduced from the patterns of expression of the NK-2 gene in various mutant lines of flies (from Ref. 51). An arrowhead represents gene ac- tivation, while a terminal dar represents gene repression. (X) corresponds to an unidentified re- pressor mediated by dpp. FIGURE 10. Side view of an embryo showing the ventral-dorsal pattern of the anlagen indi- cated and the hierarchically organized regulation of NK-2 gene expression.5! An arrowhead cor- responds to gene activation, while a terminal dar represents repression. dpp indirectly mediates repression of the NK-2 gene in dorsal neuroectoderm, via (X), an unidentified repressor. The NK-2 gene is repressed by sim in the mesectodermal anlage and by snail in the mesodermal anlage. 236 ANNALS NEW YORK ACADEMY OF SCIENCES ance of six horizontal stripes which, from ventral to dorsal, comprise the me- sodermal, mesectodermal, ventral neuroectodermal, dorsal neuroectodermal, dorsoepidermal, and amnioserosa anlagen, as shown in Ficure 10. The ventral border of the horizontal stripe of nuclei that synthesize NK-2 mRNA is created by repression of the NK-2 gene by snail initially and then by sim, while the dorsal border of the stripe is created by a different, uniden- tified species of repressor mediated by dpp.5! Thus, the ventral and dorsal borders of the NK-2-positive stripe of nuclei are created independently by different species of repressors. The width of the NK-2-positive stripe of nuclei and the position of the stripe on the ventral-dorsal axis of the embryo are not fixed, but can be shifted by the combined effects of proteins that induce and repress the NK-2 gene. Rao, Vaessin, Jan, and Jan®? have shown previously that the position of the neuroectoderm in the Drosophila embryo is shifted in appropriate mutants. Levine and his colleagues®+ have shown that the leading and trailing edges of a vertical stripe of nuclei that express the eve gene (eve stripe 2) are formed independently by giant and Kriipple, respectively, which repress the eve gene. Although the gradient of inducer and the repres- sors are different, the formation of a horizontal stripe of nuclei that express the NK-2 gene resembles the formation of a vertical stripe of nuclei that ex- press the eve gene. Neuroectodermal cells develop at different rates and segregate as neuro- blasts at different times, depending upon their position in the embryo (for review see Ref. 67). Therefore, it is likely that the expression of a proneural gene is the rate-limiting step in the development of ventral neuroectodermal cells and segregation of medial neuroblasts. Our working hypothesis is that NK-2 is a proneural gene required for the formation of medial neuroblasts. Deletion of the NK-2 gene and some neighboring genes is a homozygous lethal deficiency and results in embryos with grossly defective ventral nerve cords that lack many neurons compared to wild-type embryos.® At the present time we are trying to obtain specific mutations of the NK-2 gene. Initially all nuclei in the ventrolateral neurogenic anlage are committed to the neuroblast pathway of development. However, only about 25% of the neuroectodermal cells segregate as neuroblasts; most of the remaining neuro- ectodermal cells become ventrolateral epidermoblasts. Campos-Ortega and others have identified a set of neurogenic genes, Notch, Delta, almondex, big brain, master mind, neuralized, and the Enhancer of split [E(spl)| complex of genes, whose expression is required to turn off the neuroblast pathway of de- velopment and/or turn on the epidermoblast pathway of development (for reviews see Refs. 67, 69, and 70). Delta and Notch encode cell membrane pro- teins that interact with one another and contain multiple EGF repeats”!~75 (Delta and Notch proteins are thought to function as a ligand and corte- sponding receptor). The E(sp/) complex of genes contains a cluster of related genes that encode similar basic helix-loop-helix DNA binding proteins (HLH- NIRENBERG ¢t al.: NK-2 HOMEOBOX GENE 237 m3, HLH-m5, HLH-m7, HLH-m8 [E(spl)], HLH-m8, HLH-my, and HLH-mé, which are thought to be required for epidermoblast develop- ment,’°79 although some redundancy of HLH proteins is likely. The avail- able information suggests that direct contact between a segregated neuroblast and neighboring neuroectodermal cells turns off the neural pathway of devel- opment and activates the epidermoblast pathway of development in the neuroectodermal cells, a process termed lateral inhibition. The switch from neuroectodermal to epidermoblast pathway of development is blocked by mu- tation of a neurogenic gene (or by deletion of the E(sp!) complex of genes re- sulting in overproduction of neuroblasts and underproduction of epidermo- blasts). Mellerick and Nirenberg®® have shown that a null mutation of the Delta gene or deletion of the E(spl) gene complex results in overproduction of neuroblasts that express the NK-2 gene. These results show that neuroec- todermal cells that express the NK-2 gene are sensitive to lateral inhibition and suggest that one or more of the E(spl) HLH proteins repress the NK-2 gene. Delta probably represses the NK-2 gene indirectly by signalling activation of E(spl) HLH genes. One possibility that remains to be explored is that repres- sion of the NK-2 gene by E(spl) HLH proteins may extinguish an NK-2- dependent pathway for medial neuroblast development and activate the epi- dermal pathway of development. The NK-2 homeodomain was expressed in E. coli and purified to essential homogeneity. Binding studies to oligodeoxynucleotides showed that the con- sensus nucleotide sequence for NK-2 homeodomain binding is TNAAGTGG, and that the Kp is approximately 2 x 10-10 M.8! Twenty high-affinity and ad- ditional lower-affinity NK-2 homeodomain binding sites were found in 2.2 kb of the 5’-upstream region of the NK-2 gene,*! which suggests that NK-2 protein may be required to maintain NK-2 gene expression. Little is known about the functions of the NK-2 homeodomain protein. Our working hypothesis is that NK-2 is a proneural gene that may be required for the development of a subset of neuroblasts. REFERENCES 1. Kim, Y. & M. NiRENBERG. 1989. Drosophila NK-homeobox genes. Proc. Natl. Acad. Sci. USA 86: 7716-7720. 2, Foe, V. E. & B. M. ALBerrs. 1983. Studies of nuclear and cytoplasmic behavior during the five mitotic cycles that precede gastrulation in Drosophila embryo- genesis. J. Cell Sci. 61: 31-70. 3. Campos-Orreca, J. A. & V. HARTENSTEIN. 1985. The embryonic development of Drosophila melanogaster. Springer-Verlag. Berlin-Heidelberg-New York. 4. FRONHOFER, H. G. & C. NUSSLEIN-VOLHARD. 1986. Organization of anterior pattern in the Drosophila embryo by the maternal gene bicoid. Nature 324: 120-125. 5. Drizver, W. & C. NUSSLEIN-VOLHARD. 1988. The bicoid protein determines 238 10. ll. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. ANNALS NEW YORK ACADEMY OF SCIENCES position in the Drosophila embryo in a concentration-dependent manner. Cell 54: 95-104. DrreEvER, W. & C. NUSSLEIN-VOLHARD. 1988. A gradient of bicoid protein in Drosophila embryos. Cell 54: 83-93. Inisu, V., R. LEHMANN & M. AKAM. 1989. The Drosophila posterior-group gene nanos functions by repressing hunchback activity. Nature 338: 646-648. Wana, C. & R. LEHMANN. 1991. Nanos is the localized posterior determinant in Drosophila. Cell 66: 637-647. Wuarton, RP. & G. STRUHL. 1991. RNA regulatory elements mediate contro] of Drosophila body pattern by the posterior morphogen nanos. Cell 67: 955-967. LEHMANN, R. & C. NUSSLEIN-VOLHARD. 1991. The maternal gene manos has a central role in posterior pattern formation of the Drosophila embryo. Develop- ment 112: 679-691. STEWARD, R., S. B. ZUSMAN, L. H. Huane & P. ScHEDL. 1988. The dorsal pro- tein is distributed in a gradient in early Drosophila embryos. Cell 55: 487-495. STEWARD, R. 1989. Relocalization of the dorsal protein from the cytoplasm to the nucleus correlates with its function. Cell 59: 1179-1188. Rotu, S., D. STEIN & C. NUSSLEIN-VOLHARD. 1989. A gradient of nuclear lo- calization of the dorsal protein determines dorsoventral pattern in the Dro- sophila embryos. Cell 59: 1189-1202. RusHiow, C. A., K. Han, J. L. MANLEY & M. Levine. 1989. The graded dis- tribution of the dorsal morphogen is initiated by selective nuclear transport in Drosophila. Cell 59: 1165-1177. IncHaM, P. W. 1987. The molecular genetics of embryonic pattern formation in Drosophila, Nature 335: 25-34. St. JoHNsTON, D. & C. NUSSLEIN-VOLHARD. 1992. The origin of pattern and polarity in the Drosophila embryo. Cell 68: 201-219. Hoc, M. & H. Jacke. 1993. Transcriptional regulation and spatial patterning in Drosophila. Curr. Opin. Genet. Dev. 3: 566-573. Kornsera, T. B. & T. Tasata. 1993. Segmentation of the Drosophila embryo. Curr. Opin. Genet. Dev. 3: 585-595. NaxkayaMA, K., N. NAKAYAMA, Y. S. Kim, K. O. WEBBER, R. LAD & M. NIREN- BERG. The NK-2 Homeobox Gene of Drosophila. Proc. Natl. Acad. Sci. USA. In press. Sau M. S., R. B. MicHer, K. M. GuLpInc & R. M. GRAINGER. 1993. A Xenopus homeobox gene defines dorsal-ventral domains in the developing brain. Development 118: 193-202. Pricer, M., D. Lazzaro, T. Poni, M-G. Maret, U. ROTHER, J. C. Otrvo, D. DuBOuLE & R. Dr Lauro. 1992. Regional expression of the homeobox gene Nkx-2.2 in the developing mammalian forebrain. Neuron 8: 241-255. Lrnts, T. J., L. M. Parsons, L. Harriey, I. Lyons & R. P. Harvey. 1993. Nkx-2.5: A novel murine homeobox gene expressed in early heart progenitor cells and their myogenic descendants. Development 119: 419-431. GARCIA-FERNANDEZ, J., J. BAGUNA & E. Saxo. 1993. Genomic organization and expression of the planarian homeobox genes Dth-1 and Dth-2. Development 118: 241-253. GARCIA-FERNANDEZ, J., J. BAGUNA & E. Saxo. 1991. Planarian homeobox genes: Cloning, sequence analysis, and expression. Proc. Natl. Acad. Sci. USA 88: 7338-7342. NARDELLI-HAEFLIGER, D. & M. SHANKLAND. 1993. Lax 10, a member of the NIRENBERG et al.: NK-2 HOMEOBOX GENE 239 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 41. 42. NK-2 homeobox gene class, is expressed in a segmental pattern in the endo- derm and in the cephalic nervous system of the leech Helobdella. Development 18: 877-892. OLIver, G., M. Vispo, A. MarLHos, C. MARTINEZ, B. Sosa-PINEDA, W. FIE- yitz & R. EHRLICH. 1992. Homeoboxes in flatworms. Gene 121; 337-342. Tsao, D. H. H., J. M. Gruscuus, L.-H. WANG, M. NIRENBERG & J. A. FerR- reTTI. 1994. Elongation of helix III of the NK-2 homeodomain upon binding to DNA: A secondary structure study by NMR. Biochemistry 33: 15053- 15060. Azprazu, N. & M. FrascH. 1993. tinman and bagpipe: two homeobox genes that determine cell fates in the dorsal mesoderm of Drosophila. Genes Dev. 7: 1325-1340. WesseER, K. O., Y. Kim, R. Lap, K. NAKAYAMA & M. NIRENBERG. Unpublished manuscript. Tussg, C., F. PERRIN-SCHMITT, C. STOEZEL & B. THIssE. 1991. Sequence specific transactivation of the Drosophila twist gene by the dorsal gene product. Cell 65: 1191-1201. JIANG, J., D. Kosman, Y. T. Ip & M. Levine. 1991. The dorsal morphogen gra- dient regulates the mesoderm determinant twist in early Drosophila embryos. Genes Dev. 5: 1881-1891. Pan, D., J. D. Huane & A. J. Coursey. 1991. Functional analysis of the Drv- sophila twist promoter reveals a dorsal-binding ventral activator region. Genes Dev. 5: 1892-1901. Pan, D. & A. J. Coury. 1992. The same dorsal binding sites mediate both ac- tivation and repression in a context dependent manner. EMBO J. U: 1837-1842. Leptin, M. 1991. twist and snail as positive and negative regulators during Dro- sopbila mesoderm development. Genes Dev. 5: 1568-1576. Kosman, D., Y. T. Ip, M. Levine & K. Arora. 1991. Establishment of the mesoderm-neuroectoderm boundary in the Drosophila embryo. Science 254: 118-122. Ip, Y. T., RE. Park, D. KosMAN, K. YAZDANBAKHSH & M. LEVINE. 1992. dorsal-twist interactions establish snail expression in the presumptive meso- derm of the Drosophila embryo. Genes Dev. 6: 1518-1530. St. JoHNSTON, R. D. & W. M. GELBart. 1987. decapentaplagic transcripts are local- ized along the dorsal-ventral axis of the Drosophila embryo. EMBO J. 6: 2785-2791. InIsH, V. F. & W. M. Gevpart. 1987. The decapentaplegic gene is required for dorsal-ventral patterning of the Drosophila embryo. Genes Dev. 1: 868-879. Ray, R. P., K. Aror, C. NUSSLEIN-VOLHARD & W. M. GELBarr. 1991. The con- trol of cell fate along the dorsal-ventral axis of the Drosophila embryo. Devel- opment 113: 35-54. Huana, J-D., D. H. Scrrwyrer, J. M. Surroxawa & A. J. Coury. 1993. The interplay between multiple enhancer and silencer elements defines the pattern of decapentaplegic expression. Genes Dev. 7: 694-704. Papcett, R. W., R. D. St. JounsTon & W. M. GELBart. 1987. A transcript from a Drosophila pattern gene predicts a protein homologous to the transforming growth factor 8 family. Nature 325: 81-84. Doyrs, H., K. Harpine, T. Hoey & M. Levine. 1986. Transcripts encoded by a homeobox gene are restricted to dorsal tissues of Drosophila embryos. Na- ture 323: 76-79. 43. 45. 47. 48. 49, 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. ANNALS NEW YORK ACADEMY OF SCIENCES Doy te, H. J., R. Kraut & M. Levine. 1989. Spatial regulation of zerknullt: A dorsal-ventral patterning gene in Drosophila. Genes Dev. 3: 1518-1533. Rusuiow, C., H. Doyie, T. Hory & M. Levine. 1987. Molecular characteri- zation of the zerknullt region of the Antennapedia gene complex in Drosophila. Genes Dev. 1: 1268-1279. Ip, Y. T., R. Kraut, M. Levine & C. A. RusHiLow. 1991. The dorsal morphogen is a sequence-specific DNA-binding protein that interacts with a long range re- pression element in Drosophila. Cell 64: 439-446. Jrane, J., C. A. Rustow, Q. ZHov, S. SMALL & M. Levine. 1992. Individual dorsal morphogen binding sites mediate activation and repression in the Drv- sophila embryo. EMBO J. 11: 3147-3154. HARTENSTEIN, V. & J. A. Campos-Orreca. 1987. Early neurogenesis in wild- type Drosophila melanogaster. Roux’s Arch. Dev. Biol. 196: 473-485. HARTENSTEIN, V., E. RUDLOFF & J. A. CAMPos-OrTEGA. 1984. The pattern of proliferation of the neuroblasts in the wild-type embryo of Drosophila melano- gaster. Roux’s Arch. Dev. Biol. 193: 308-325. Dog, C. Q. 1992. Molecular markers for identified neuroblasts and ganglion mother cells in the Drosophila central nervous system. Development 116: 855-863. PareL, N. H., B. SCHAFER, C. S. GOODMAN & R. HotmcGremM. 1989. The role of segment polarity genes during Drosophila neurogenesis. Genes Dev. 3: 890-904. MELLERICK, D. M. & M. NrreNBERG. Dorsal-ventral patterning genes restrict NK-2 homeobox gene expression to the ventral half of the central nervous system of Drosophila embryos. Dev. Biol. In press. Frrcuson, E. L. & K. V. ANDERSON. 199]. Dorsal-ventral pattern formation in the Drosophila embryo: The role of zygotically active genes. Curr. Top. Dev. Biol. 25: 17-43. Ip, Y. T. & M. Levine. 1992. The role of the dorsal morphogen gradient in Dro- sophila embryogenesis. Seminars Dev. Biol. 3: 15-23. STEWARD, R. & S$. Govinp. 1993. Dorsal-ventral polarity in the Drosophila em- bryo. Curr. Opin. Genet. Dev. 3: 556-561. Boutay, J. L., C. DENNEFELD & A. ALBERGA. 1987. The Drosophila develop- mental gene snail encodes a protein with nucleic acid binding fingers. Nature 330: 395-398. Aperca, A., J. L. Bouray, E. KEMPE, C. DENNEFELD & M. Haenuin. 1991. The snail gene required for mesoderm formation in Drosophila is expressed dy- namically in derivatives of all three germ layers. Development lll: 983-992. ‘Tomas, J. B., S. T. Crews & C. S, GOODMAN. 1988. Molecular genetics of the single-minded \ocus: A gene involved in the development of the Drosophila ner- vous system. Cell 52: 133-141. Crews, S. T., J. B. THomas & C. S. GOODMAN. 1988. The Drosophila single- minded gene encodes a nuclear protein with sequence similarity to the per gene product. Cell 52: 143-151. NamBwu, J. R, R. G. FRANKS, S. Hu & S. T. Crews. 1990. The single-minded gene of Drosophila is required for the expression of genes important for the de- velopment of CNS midline cells. Cell 63: 63-75. Kasat, Y., J. R. NAMBU, P. M. LIEBERMAN & S. T. Crews. 1992. Dorsal-ventral patterning in Drosophila: DNA binding of snail protein to the single-minded gene. Proc. Natl. Acad. Sci. USA 89: 3414-3418. Trussz, B., C. SrorrzeEL, M. EL MEssaL & F. PERRIN-SCHMITT. 1987. Genes of NIRENBERG et al.: NK-2 HOMEOBOX GENE 241 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74, 75. 76. 77. 78. the Drosophila maternal dorsal group control the specific expression of the zy- gotic gene twist in the presumptive mesodermal cells. Genes Dev. 1: 709-715. TuissE, B., C. STOETZEL, C. Gorostiza-IuIssE & F. PERRIN-SCHMITT. 1988. Se- quence of the fwist gene and nuclear localization of its protein in endomeso- dermal cells of early Drosophila embryos. EMBO J. 7: 2175-2183. Rao, Y., H. VAEssIN, L. Y. JAN & Y. N. JAN. 1991. Neuroectoderm in Drosophila embryos is dependent on the mesoderm for positioning but not for formation. Genes Dev. 5: 1577-1588. SMALL, S., R. Kraut, T. Hogy, R. Warrior & M. Levine. 1991. Transcriptional regulation of a pair-rule stripe in Drosophila. Genes Dev. 5: 827-839. STANOJEVIC, D., S. SMALL & M. Levine. 199]. Regulation of a segmentation stripe by overlapping activators and repressors in the Drvsophila embryo. Science 254: 1385-1387. SMALL, S., A. Brain & M. Levine. 1992. Regulation of even-skipped stripe 2 in the Drosophila embryo. EMBO J. 1: 4047-4057. Camros-OrtEGA, J. A., also GoopMaAN, C. S. 1993. In The Development of Dro- sopltila melanogaster. M, Bate & A. Martinez-Arias, Eds. Cold Spring Harbor Lab- oratory Press. Cold Spring Harbor, New York. MELLERICK, D. & M. NiRENBERG. Unpublished material. Caprera, C. V. 1992. The generation of cell diversity during early neurogenesis in Drosophila. Development 115: 893-901. Fortin, M. E. & S. ARTAVANIS-ISAKONAS. 1993. Notch: Neurogenesis is only part of the picture. Cell 75: 1245-1247. VaESSIN, H., K. A. BREMER, E. Knust & J. A. Campos-OrreGa. 1987. The neurogenic gene Dele of Drosophila melanogaster is expressed in neurogenic ter- ritories and encodes a putative transmembrane protein with EGF-like repeats. EMBO J. 6: 3433-3440. Wuarton, K. A., K. M. JOHANSEN, T. Xu & S. ARTAVANIS-ISAKONAS. 1985. Nucleotide sequence from the neurogenic locus Notch implies a gene product that shares homology with proteins containing EGF-like repeats. Cell 43: 567-581. Repay, I., R. J. FLEMING, R. G. FeHon, L. Cuersas, P. Cuersas & S. ARTAVANIS-ISAKONAS. 1991. Specific EGF repeats of Notch mediate interactions with Delta and Sermie: Implications for Not as a multifunctional receptor. Cell 67: 687-699. FeHON, R. G., P. J. Koon, I. Repay, C. L. REGAN, T. Xu, M. A. T. MusKavitcH & S. ARTAVANIS-Isakonas. 1990. Molecular interactions between the protein products of the neurogenic loci Noth and Delm, two EGF-homologous genes in Drosophila. Cell 61: 523-534. Branp, M. & J. A. Campos-Orreca. 1990. Second-site modifiers of the split mutation of Nowh define genes involved in neurogenesis in Drosophila mela-- nogaster. Roux’s Arch. Dev. Biol. 198: 275-285. Kiapr, C., E. Knust, K. Thetze & J. A. CaMpos-Orreca. 1989. Closely re- lated transcripts encoded by the neurogenic gene complex Enhancer of split of Drosophila melanogaster. EMBO J. 8: 203-210. DELIGAKIS, C. & S. ARTAVANIS“ISaKONAS. 1992. The Enhancer of split [E (spl)] locus of Drosophila encodes seven independent helix-loop-helix proteins. Proc. Natl. Acad. Sci. USA 89: 8731-8735. Kwnust, E., H. ScHRons, F. Graw & J. A. CAMPOS-ORTEGA. 1992. Seven genes of the Enhancer of split complex of Drosophila melanogaster encode helix-loop- helix proteins. Genetics 132: 505-518. 242 ANNALS NEW YORK ACADEMY OF SCIENCES 79. Scurons, H., E. Knust & J. A. CaMpos-Orteca. 1992. The Enhancer of split complex and adjacent genes in the 96F region of Drosophila melanogaster are re- quired for segregation of neural and epidermal progenitor cells. Genetics 132: 481-503. 80. Meverick, D. & M. NrrENBERG. Manuscript in preparation. 81. Wane, L.-H., R. CHMELIK & M. NirENBERG. Manuscript in preparation.