LETTERS TO NATURE translation start site for the KnJ protein as it is the first A TO codon, and is followed by one long open reading frame of 1,240 bp. Comparison of the encoded polypeptide sequence with pro- tein sequences in the Swiss Protein Database (release 14.0) using the F ASTP programS reveals limited but significant similarity to a variety of eukaryotic proteins that contain homeodomains9. This similarity is restricted to the amino-acid segments compos- ing the homeodomains. The homeodomain is a 61-amino-acid motif which is postulated to bind DNA through the helix-turn- helix motif9-11. An alignment of a portion of the Knl amino-acid sequence with homeodomain sequences from various other eukaryotes is presented in Fig. 2a. The Kn1 homeodomain is most similar to the homeodomains from the human Prl (pre-B cell leukaemia) 12.IJ and Schizosaccharomyces pombe MATpil4 proteins, showing -35% identity with each. The Kn1 homeo- domain contains identical residues or similar substitutions in six of the eight positions most conserved among homeodomains, and contains the four invariant residues present in the putative recognition helix (helix 3) of all non-yeast homeodomains9. Secondary structure algorithmsls predict a helices for Knl in the region of the helix-turn-helix motif and there are no helix- breaking residues in the predicted helical regions. In the KnJ gene, the homeobox is separated from amino-terminal coding sequences by the large (5 kb) third intron and is interrupted near the carboxy terminus of helix 1 (ref. 11) by intron four. CCG...C.ttc=ttCTTTCCCtt=TT ,. V CCCCAAACCCTTTTCCtt=GCCCCAAGAACTtt. tttt. tt ttC OGGC=CC=TTGCDC~ = GAC ..TGGGCG...AGcttG. ttGCTGC ACC ACTAC=ACTGGAGGTCCGCCCACttCC .TGGAG ttACCCAACACTTroGAG TT 2 , . M .., T O H r G V ,. GGCGC AAGC..CC ACGGCC .=C .CGGCC ..CACCACC. tt. tt ACCACC ACC .CCACCCGTGGGC .ttCttCC tt AGCGcCG=T. , .. G .S S H G H G H G O H M H H H H H H H p M .S S L S .V V ,. GCGCCGCTGCCGCCGC AACCGCC...CGC..=CCGCTGACCCTG..CACGGTGGCGGCCACTGGGMC AGCGGCGGT ..CGGCAAC , .. .p L p p 0 p p S .G L p L T L N T V. .T G N S G G S G N ,. CCGGTGCTGCAGCTTGCC..CGGTGGCGGCCTCC1CGACGC .TGCGtt AAGGC cccttG=tt=CCT ACGCAGGCGAC , .. p V L O L .N G G G L L ..C V K ...p S S S S p T .G .,.. . GttGAGGCCA tt...GCC TCA =GCACCCACACT ACT AC=CTCCTCACTGCCT ACCttGAGTGC AAC AAGGTGGGGGCACC. ., . V. .I ...I I S N p N r r S L L T .r L .C N .V G .p ". CCGG..GTGttGGCG ...CTGACGG... T.GCGC.GGAGGTGGAGGCGCGGCAGCGCACGG==TGGCC GC TGCOACGG.. ... p .V S .R L T .I .0 .V. .R O R T .L G G L ...T .,.. . CCG...CTGGACC ..= .TGGAGGCGT ACC ACG... TGCTGGTG ...=AGGG...AGCTGACGAGGCCGCTGC..GAGGC.. TGG.. , .. p .L .O r M ..r H .M L v. r R ..L T R p L O ..M .,.. Ttt. TGCGAAGGGTGG..ttGC.GC TGAAC=CA tt= ttGCTGCGCAAC .TCCTTTC .rc..!c tt =TG... ..., . r N R R V. S O L N S L S I S G R S L R M I L S S G S S ..22. G. tt GT AGCGGAGG.GAGACCGAOC=TGAAGTTG. TGCAC. TGGTGTGGACC CTGAAGCACC. TCTCCTGAAGAAA .,. .0 .G S G G .T .L p .v. .N G V. 0 .L .H H L L ..2.. T AC..TGGCT .ttTAAGCttGCttAAGC... ctt= AGGCTCGCCAGCAGCTCCTT , .2. r S G r L S S L .0 .L ...K K G K L p ...R O O L L 2.. . AGCTGG TGGG. tt ..C AC TAC ...TGGCCTT ACCCCtt = ..AAGGTGGC ACTGGCTGAG= .CCGGGCTTGACCTGAAGC .." , . S ...O H r ...r p S .T Q K V. L ..S T G L .L .0 ". .ttAAC AACTGG= .ttAACCAGCGGAAGCGGC.C-C. ...TGCACC ACCTGA TGA TGG. TGGG TACCACACCACC , 2 .. I N ..p I N O .K .N ..p S .M M M L M M .G r M T T ". ..TGCCTTCT .C. TGGACGGCCACTTCA ttAAC..CGGCGGGC TG T =GCT AOCCACCGGT. ttttGC=C. TTTCAC ACCC , 2 .. N .r r M .G M r I N .G G L r R L G .,.. CACGGCCT AOCT .T ACT.. TGGTTCC ...TG TCTGAAGT ACTGAAG.CAGGGGGGCT AOCT. ttT.. TG TTTGTGCCGC ACGCA TG , , .. .GC TG T AAGGAGGCC. TGCTT ..TT .TTCTGTTGCCGTTGCTACTCT. ttT .T. TGCGCCT. TGCCTCCGTGC .TG..CT .TGCTTTAGG , , .,. TGGTTGCTGCttC ACACTGTGGTGGTGTGCTTTTGCTTTTGTGTGGtt GT .TTGT .TGCG T AACCTGAC... TGG. ttCCTGA TTGCT AC , ... V. V .TGTTTGAAT..TTTGC.TG.ttT.GCT..='r..ttT..TGGTACGGCtt.TGTCTTGtt (., ..,." Received 16 November 1900; accepted 3 January 1991. L Gunning. B. E s. In 7IIe Cytoskeleton in Pfsnt Growth and Development (ed. Lloyd, C. W .) 229-292 (Academic, LonOOn,1982) 2. Euteneuer, U.. J8d<son, W T, & Mclntosh, J, R. ), Cell BIoi, 94, 644-653 (1982). 3. Hoffman-Ber1lng, H Blochim. biophy£ Acts 14, 182-194 (1954). 4 Masud8, H. & Cande, W Z. Cell 49, 193-202 (1987) 5 Mltchison, T. ). & Klrschner, M. W ), Cell BIoi. 101. 766-777 (1985). 6 Shpetner, H, s & Vallee, R, B Cell 59, 421-432 (1989). 7. Obar, R. A., Collins, C A., Hammarb8d{, ), A., shpetner, H, s. & Vallee. R. B. Nature 347, 256-261 (1990). 8 K...imoto. T. & Shlbad<a, H. Protoplssms (suppl 2) 95-103 (1988). 9. Welngarten, M. D., Lod<wood. A. H.. Hwo, s y & Kirschner. M. W. Pr"" nstn Acsd 5ci. U5.A 722, 1858-1662 (1975). 10 Keith, C. H.. FeramlSCQ,J. R, & shelanski, M. ). 1 Cell Bioi 88, 234-240 (1981). ACKNOWLED(i:I.£NTs This wor1<was supported by grants from the Ministry of Education. Science and Culture, an and the Mltsublshl Foundation. Taxol was supplied by the Drug Synthesis & Chemistry B , Division of Cancer Treatment, National Cancer Institute. The developmental gene Jnotted-l is a member of a maize homeobox gene family Erik Vollbrecht*, Bruce Veitt, Neelima Sinhat & Sarah Hake*t:l: /USDA-ARS Plant Gene Expression Centej)800 Buchanan Street, ~bany. Mornia194710. USA t Department of Plant Biology, University of California, Berkeley, California 94720, USA THE Knotted-l (Knl) locus is defined by several dominant gain-or- function mutations that alter leaf development. Foci or cells along the lateral veins do not differentiate properly, but continue to divide, forming outpocketings or knots. The ligule, a Cringe normally round at the junction or leaf blade and sheath, is often displaced and perpendicular to its normal positionl-3. The phenotype is manifested in all cell layers or the leaf blade, but is controlled by a subgroup or cells or the inner layer4. Mutations result from the insertion or transposable elementsS or a tandem duplication6. We show that the Knl gene encodes a homeodomain- containing protein, the first identified in the plant kingdom. Sequence comparisons strongly suggest that Knl acts as a tran- scription factor. Here we use the Knl homeobox to isolate other expressed homeobox genes in maize. The Knl homeobox may permit the isolation or genes that, like animal and fungal counter- parts7, regulate cell rate determination. We have isolated complementary DNAs of the Knl locus from maize seedling RNA using..genomic DNA that was pre- viously cloned by transposon tagging5. Reversion analysis had indicated that these sequences are closely associated with the Knl gene5.6. The sequence of the Knl transcript (Fig. 1) was identified from cDNA sequences and primer extension assays (data not shown). The predicted transcript size of 1,635 base pairs (bp) agrees closely with the 1.7 -kilobase (kb) messenger RNA identified in RNA blots (data not shown). The ATG at position 181 of the deduced transcription unit is probably the FIG. 1 Complementary DNA and predicted amino-acid sequence of maize Kn1. The first nucleotide in the sequence is the most 5' nucleotide identified by primer extension experiments (unpublished data); the open arrow at the 5' end indicates the 5' end of the longest cDNA; sequence upstream was derived from genomic clones. The amino-acid sequence of the open reading frame starts from the first A TG codon and is indicated beneath the nucleotide sequence. The histidine-rich or 'CAX' region, is doubly underlined and a candidate PEST sequence is singly underlined (see text). The homeodomain region is boxed. Filled arrows denote intron-exon junctions; the splice site donor and acceptor sequences were determined from genomic clones and agreed (not shown) with consensus sequences for plant introns32. The heterogeneity indicated at the 3' end of the transcript (open arrows) reflects different polyadenylation sites detected in the three sequenced cDNAs. Asterisk indicates the first in-frame stop codon. METHODS. Complementary DNA primed with oligo(dT). was synthesized using poly(A)+ RNA from 14-day-old wild-type seedlings and cloned into the EcoRI site of Agt10 (ref. 33). Six cDNA clones were isolated from a primary library of 500.000 recombinants using a probe derived from transposon-tagged genomic sequences5. The longest cDNAs were subcloned into Bluescript (Stratagene) plasmids. Double-stranded DNA was sequenced on both strands by the dideoxy-chain-termination method using T7 DNA polymerase and analysed with the PCGENE software package (Intelligenetics). 241 ; To whom correspondence shouldbe addressed NATURE. VOl350 .21 MARCH 1991