The work described in this thesis consists of two different types of genome characterizations of Arabidopsis thaliana. Photosynthesis is just one of the light reactions of plants that do not occur in animals. These include those that affect the waxy layer of the epidermal cells (cer mutants, 4) and the trichomes normally found on leaves and stems (dis and gl mutants, 5), as well as those that cause more easily visible effects, such as changes in floral morphology (ag, ap and pi mutants, 6) and growth habit (er and cp mutants, 7).
These results indicate that the Arabidopsis nuclear genome consists of predominantly unique sequences and that most of the nuclear repetitive DNA is ribosomal DNA. A second series of experiments (29) involved the cloning of the genes encoding the light-induced chlorophyll a/b binding protein from Arabidopsis. The three genes of the Arabidopsis chlorophyll a/b binding protein family contrast with the much larger number in the homologous families in other plants.
We would like to thank the other members of the Meyerowitz laboratory for comments on the manuscript. Therefore, the chloroplast genome constitutes the main component of the repetitive sequences found in Arabidopsis thaliana DNA made from whole plants.
Materials and Methods
One of the two libraries was amplified before use according to the method described by Maniatis et al. All restriction fragments in this set of clones show intense hybridization with the genomic DNA probe. In all clones, some of the fragments were present in greater molar amounts than others.
To determine the approximate number of copies of the rONA repeat unit in the haploid genome, a quantitative dot-blot procedure was used. These can be directly compared after correction for the specific activity of the different probes and the length of the segment being hybridized in the genomic DNA. Due to the similar apparent number of copies in the genomic DNA, it was considered likely that all of these represent chloroplast clones.
This filter was allowed to hybridize with 3 2P-labeled DNA of the 1.0 kb EcoRI restriction fragment, which had been purified on an agarose gel. A similar gel blotting experiment was performed to determine which region of clone ).bA t026 contained repetitive sequences. Analysis of these clones shows that Arabidopsis rDNA is typical of plant rDNA.
Variations in the copy number of such an element could result in heterogeneity of rONA repeats.
F=Eco RI
Xbal
Restriction maps of the regions containing the 125 storage protein genes of the two Arabidopsis strains Landsberg and Columbia. The arrows below the maps indicate the locations of the storage protein genes and the direction of transcription. The locations of the genes were determined by hybridization of an eDNA clone from Brassica napus to the cloned DNA and by DNA sequencing.
The locations of the fragments contained in the nAtl5l2, sAt2lOl, and sA t21 05 subclones are also indicated. Arrows represent the locations and directions of transcription of the 650 nucleotide RNA coding regions. Determining the times in seed development when each set of genes is expressed.
RNA was prepared from 2 seed pods of each of the ages shown at the top of the figure, denatured and separated on a formaldehyde-agarose gel. The protein sequence of the 148 amino acid open reading frame is also shown below the sequence. The sequence of the 5' end and flanking region of the right gene is also shown below the full gene sequence.
Diagrammatic representation of the structures of the l2S storage protein genes encoding the proteins shown in Figure 9. It is dear from the cryptic map of the 68(' gene cluster that thrt'f' miC\As do not dt·ri, ·e from a common precursor All the leftmost nurleotides iO of the sequence were determined in each of the complementary strands D~A.
That the ~ nurleotides in the transposablt' elt>rnt>nt "as dt>termined when comparing the S€'quency of a clone from our 6~('. This region contains three additional copies of the repeat element in 6 kb of conti~uous Ht"quen(·e. To analyze the functional relevance of the sequences found in the gum gene clusterRr.
The DXA sequence of the group III cDXA clone includes nucleotide positions 21f).t to 2498 in the genomic clone sequence>. The St'quenee of the end deri,·ed from the 3'-terminus of tht-' R:\A contained a poly(.-\) tract adjacent to genomic nucleotide position 5646.
II III IV
I loLouLou I log I yPhoSorAspLouAioLouG I y --.l---A -<:A ---A -A lA-<: I -I Metly ysLou Thr I leAio ThrAioLouAII oSor I I eloulou I log I ySorAioAsnVoiAioAenCys ---- .,l( -<1-<: lA-<:-( -<1 +--.--<: --T -T--( -(-lA--G 1--1UC T -T CysAspCysG 1 yCytPro lhr Thr Thr Thr ThrCysAlePro.t.rgThr lhrG lnProProCys lhr Thr Thr Thr Thr ThrThr Tt'lr Thr ThrCya.t.leProPro l GCGA T IG IGGAIGCCCCACUC Tar UC I AlI I G I&LGCCACG IACCACGCUCC TCCG TGCACUC T ACGACUCUCCACUC T lCTlG lGCGCClCCC ThrG lnG lnS rr Thr ThrG In nProProCyl Thr ThrSerL ysProThr ThrProL ysG In Thr Thr ThrG) nLeuProCys Thr ThrPro Thr Thr Thrl ytA 1 a lClCUCU TC llCCACGCUCC ICCAlGCACGlCA TC IUGCCCACCACACC T UGCAUCTACCACGCUCT lCCG TGCACUCACCCACCACCAC T UGGCC Th rlhrlhrLysProfhrThrThrLytAlelhrThrThr LysAlaThrThrThrLysProThrlhrlhrLytGinThrThrThrGinLeuProCyalhrlhrPro ACCACCACGUGCCCACCACCACT AUGCCACCACCACT UGGCCACCACCAC IUGCCCACCACCAC IUGCUACT ACCACGCUCT ICCGTGCACUCACCC ThrThrThrLysGtnThr ThrThrGlnleuProCysThrThrProThrThrThrLysProThrlhrfhrLysProThrThrThrLytProThrT~rThrLysPro ACCACCAC IU GCAUC IACCACGCUCT TCCG TGCACUCACCCACCACCAC T UGCCCACCACCACGUGCCCACCACGUGCCCACCACCACT UGCCC ThrThrThrLysProThrThrThrLysProThrThrThrlysProThrThrThrLysProThrThrThrLysProThrThrThrLysProTnrThrThrLysPro ACCACCACGAAGCCCACCACCACCAAGCCCACCACGAAGCCCCACCAC T AAGCCC ACCACACGAAGCCC ACCACCAC T UGCCCACCACCACGAAGCCC Thr Thr Thr !Pro Thr Thr ThrL ysProThr Thr Thrl ysProThr Thr ThrL ysPro Thr lhr Thrl ysPro Thr Thr Thrl ysPro Thr T~r ThrL ysPro ACCACCACGUGCCCACCACCAC IAAGCCCACCACAC ACGUGCCACCACf.AC T UGCCCACCACCACGU GCCCACC ACCACT UGCCC ACCACCACGUGCCC ThrThrThrLysProThrlhrThrlysProThrThrThrLysProThrThrThrLysProThrThrProlys ACCACCAC IUGCCCACCACCACGUGCCCACCACCACGUGCCCACCACT UGCCCACCACACCT UG CysL ysAspCysSorCysvotlloCysG I yProG I yG 1 yG I uProC ysProG I yCysSorA I oArgVa I Provot CysL ysAspLou I I eAsn leeuwel TGCUGGAIIG T ICA TGCGTGAII TGIGGACCTGG TGGCGAGCCGTGTCCTGGGTG I TCCGCACGGGT TCCCGTCTGCUAGA TCTGA TCUCAIIA IG G I yA I 1Cy1G luCyiG tnProCyiG I y ProG I yG I yL ysA leCyl ThrG I yCysProG tulysProG tnLouCysG I nG lnlou I IeSer Asp I It GG 1-<:C ---GAG -<:UCCG--1 A--G-<:--(A A -
In the region on the carboxy side of the proteins, eight c.'·steine residues and II other amino acid positions are identical in all respects. It is clear from the experiments described that the basis of coordinate control of the three 68C R~As does not lie in the processing of the R!\As of a common precursor. D~A sequencing of the 68C puff was initiated as a means of testing the hypothesis that three coordinately regulated genes of similar function achieve their coordination by possessing identical portions of regulatory sequences.
A final possible reason for our failure to find identical control sequences associated with each of the 68 genes (' is that all three genes are controlled by a single set of sequences' within or near the gene cluster. The DNA sequence that coding for each of the three 68C mRNAs has allowed us to predict the expected amino acid sequences of the 68 C protein products. The amino acid sequences indicate that the three 68C proteins are related to each other as a clustered gene family.
The members of this family show modular construction: each member has an amino-terminal part of 23 amino acids, consisting mainly of amino acids with hydrophobic side chains that are not present in the mature one. secreted form of the protein. The sgs-i and sgs-8 polypeptides have only these two modules. positionE-d bf'tw~>en the other two. with much of modulE'. I )oS.) rE sites) consisting of tandem repeats of the sequence Pro-Thr-Thr-Thr-Lys.
Amino acid sequence similarities between the three gene products of 68C!' they do not reflect similarity in the nucleotide sequences of the genes. The most striking feature of the evolution of this gene family is the appearance of the threonine-rich central module E' in sgs-3 (or its disappearance from sgs·7 and sgs-8). Studies of 68(' puff proteins in Drosophila species other than ~larwgaster may help answer some questions regarding the e\·evolutionary mechanisms raised by the modular evolution of adhesive polypeptides 68t').
