- Email: [email protected]
Sequence analysis of a Dictyostelium discoideum gene coding for an active dihydroorotate dehydrogenase in yeast
BIOCHIMIE, 1985, 67, 583-588
Mdmoires originaux
Sequence analysis of a Dictyostelium discoideum gene coding for an active dihydroorotate dehydrogenase in yeast. Michel J A C Q U E T °, Monique K A L E K I N E and Emmanuelle BOY-MARCOTI'E. Laboratoire de Biologie Expdrimentale, B6timent 400, Universitd Paris-Sud, 91405 Orsay Cedex (France). (Refu le 21-2-1985, accept~ le 26-14-1985).
R~sum~ - - Un fragment de DNA de Dictyostelium discoideum isold sur la base de sa capacitd de compldmenter la mutation ural de la levure, contient un gone qui code pour une activitd dihydroorotate dehydrogenase. La sdquence nucldotidique complOte de ce fragment de 1898 paires de bases a dtd ddtermhlde et r~vOle une phase de lecture ouverte capable de coder pour une chaine peptidique de masse mol~culaire 47 000. Ce gone pr~sente un usage prdfdrentiel des codons de faible force d'appariement. Onze codons potentiels ne sont pas utilisds, en particulier ceux ayant un G en 3~ position sont absents. Les rdgions en bordure de la phase de lecture ont une teneur anormalement dlev~e de nucldotides A et T (80 %). La rdgion amont du gone prdsente plusieurs particularitds intdressantes comme la presence de plusieurs sdquences r~p~tdes directes et inverses. Mots-cl~s : DNA recombinant / biosynth~se des pyrimidines / usage des codons / DNA espaceur.
S u m m a r y - - A Dictyostelium discoideum DNA fragment isolated on the basis o f its ability to complement the ura I mutation o f yeast, codes for a dihydroorotate dehydrogenase activity. Tile complete nucleotide sequence o f this 1898 bp fragment has been determined and reveals an open reading frame capable o f coding for a 369 amino acid polypeptide o f molecular mass 47.000. Tile gene shows preferential use o f codons with weak pairing forces. Eleven codons, mainly those with a G in the third position, are absent. Ti~eflanking sequences are unusually rich in A + T (80 %). Several direct and inverted repeats exist in the 5' flanking sequence. Key-words : recombinant DNA / pyrimidine biosynthesis / codon usage / spacer DNA.
Introduction
Several genes from Dictyostelium discoideum have been cloned and sequenced (for review see [11). Despite varying programs of gene expression, they all have similar and unusual features in their genomic organization and sequence. All are embeded within spacer sequences unusually 0 To whom all correspondence should be addressed.
rich in A + T [2] and seem to use a restricted set of codons. Since Dictyostelium discoideum belongs to one of the more primitive phyla of eukaryotes, as documented by recent evolutionary comparisons of ribosomal RNA sequences [3], it is interesting to know how its gene organization resembles to or differs from other organisms. We chose to extend this analysis to genes involved in basic
584
M. Jacquet and coll.
metabolism which could be compared to their homologs in other organisms as well as known Dictyostelium genes. The six enzymatic steps of the pyrimidine de novo biosynthetic pathway are similar in most organisms, they although differ by their integration at the enzymatic level or their regulation [4]. Although scarcely studied, the existence of this pathway in Dictyostelium is documented by the ability to grow amoebae on defined medium deprived of pyrimidine [5] and by the isolation o f auxotrophic mutants [6]. More recently, we took advantage of genetic engineering methods to study the genes of pyrimidine biosynthesis. Two genomic fragments from Dictyostelium were selected in yeast pyrimidine mutants on the basis o f their ability to restore uracil prototrophy. One of them is able to complement a mutation (ural) in the gene coding for the dihydroorotate dehydrogenase (DOD; EC 1-3-3-1), [7]. The other genomic fragment complem6nts two mutations of yeast : ura3 and ura5, and "codes for a bifunctional enzyme orotate pyrophosphorylase and orotate monophosphate (OMp) decarboxylase [8]. Higher eukaryotes also have a bifunctional enzyme called U M P synthase carrying out the last two steps of UMP biosynthesis [41. I n t h i s report, we demonstrate that the Dietyostelium DNA fragment which complements the ural mutation codes for a dihydroorotate dehydrogenase activity. The complete nucleotide sequence o f this fragment is given. This sequence contains an open reading frame allowing the synthesis of a 369 amino acid peptide chain of molecular mass 47.000. Preferential use o f codons is observed. This predicted sequence from Dictyostelium is the first reported for a dihydroorotate dehydrogenase and will be compared to others when available, such as the URAI gene of Saccharomyces cerevisiae or the PYRD gene from E. coli already cloned [9, 10]. The 552 nucleotides sequenced upstream o f the putative ATG initiation codon, like the other flanking regions sequenced in Dictyostelium genes, have an unusually high A + T content and several repeated segments.
Materials and methods
Biological material The Dictyostelium axenic strain AX3 (from the collection R.H. Kessin) was grown in HL/5 medium [6]. The yeast OL3 strain MATa leu2-2, Ieu2-112, his3-11,
his3-15, ural-21, ural-50) was used for transformation experiments as previously described [7]. Plasmids pDUI sub1, sub2, sub5 and sub6 [7] were grown in E. coli KI2 HB 101 and purified as described. Dihydroorotate dehydrogenase assays Yeast crude extracts were obtained by iysis of protoplasts generated by zymoliase treatment. Lysis was achieved in I0 mM Tris-Cl pH 8.0. Cell debris, nuclei and mitochondria were removed by centrifugation at 12000g for 15 minutes. Dictyostelium discoideum crude extracts were obtained by thawing frozen cells in 10 mM Tris-Cl, pH 8.0, with PMSF 1 [ag/ml and subsequent centrifugation at 12 000 g for 15 minutes. Dihydroorotate dehydrogenase measurements were performed according to the procedure described by Rawls et al. [11], except that the electron transport system was replaced by dichlorophenol indophenol (2,6 DCIP). Carboxy-~SC-dihydroorotic acid was prepared from carboxy-~4C-orotic acid 30-50 mCi/mM (New England Nuclear) as described by Smithers [12]. Purification of the products was performed by chromatography on DEAE cellulose paper as described by Rawls [13].
DNA sequencing Isolated fragments were sequenced by the Maxam and Gilbert procedure [14] after 5' labelling with kinase or 3' labelling using the Klenow fragment of DNA polymerase. Restriction enzymes and polymerases were purchased from BRL, New England Biolabs, and Boehringer, and used as described by the suppliers. ~2p ATP and tt3:P NTP were from Amersham.
Results and discussion
I) A DNA fragment coding f o r a dehydroorotate dehydrogenase activity As previously reported, a DNA fragment from
Dictyostelium discoideum was selected in a yeast ural mutant due to its capacity to restore uracil prototrophy [71. Plasmid DNA containing this fragment was transfered to and amplified in E. colistrain HBI01. Two different plasmids were found among the transformed bacterial clones. Both contained an insert capable ofoconfering uracil prototrophy on the yeast strain carrying the ural mutation. They differed only by a 450 bp segment present in pDUla and absent in pDUlb, which is not required for complementation. Both also contained a genomic fragment from Dictyostelium not required for complementation and not adjacent in genomic DNA, as determined from Southern blot analysis. This extra fragment
Dihydroorotate dehydrogenase gene of dictyostelium probably originated from coligation during in vitro recombination. Its presence in both plasmids suggests that they originate from the same construction and one (pDUIb)has been deleted during cloning either in yeast or in E. coli. The segment required for complementation has been subcloned in different plasmids. To insure that the DNA fragment was coding for the dihydroorotate dehydrogenase, the enzymatic activity was assayed in transformed yeast ural cells. The DNA fragment was subcloned from the original insert of pDUla in a different plasmid (pDUI sub2) which allows a higher growth rate of the transformed yeast in uracildeprived medium. Enzymatic activity was followed by conversion of labelled dihydroorotate rather than by reduction of a hydrogen carder in order to reduce the possibility of artifacts. Such an assay can be performed on crude extracts after removal of broken cell debris, membrane, nuclei, and mitochondria. As shown in Table I, no activity was detected in the yeast ural mutant. In contrast, in transformed cells, like in the wild-type strain, high activities are observed. This result establishes that complementation is due to the recovery of dihydroorotate dehydrogenase activity. TABLEI Dihydroorotate dehydrogenase activity in )'east and dictyostelium. Enzymatic activity nmole/min/mg protein Strain Yeast GRFI8 (URAI) Yeast OL 3 (ural) Yeast OL 3 (urAl +pDUI subl) Dictyostelium discoideum, AX3
6.5 < 0.1 4.0 0.76
The dihydroorotate dehydrogenase activity was measured as described in Methods in yeast and Dictyostelium extracts. The unit of activity corresponds to one nmole of orotate produced per minute per m g of proteins.
The enzymatic activity measured in extracts of Dictyostelium amobae was found to be soluble. The specific activity is lower than in yeast, like in the case for other enzymatic activities of the pyrimidine pathway [8]. The value found in transformed yeast cells is slightly lower than in ' wild-type yeast but higher than in Dictyostelium. It may be that the Dictyostelium enzyme has a
585
lower efficiency in yeast but that it is compensated by the higher copy number of the gene. It is interesting to note that Dictyostelium dihydroorotate dehydrogenase, like the yeast enzyme, appears to be soluble. This contrasts with their homologs in others eukaryotic organisms which are tighly bound to the inner mitochondrial membrane [4].
2) The coding sequence The two DNA fragments capable of complementing of the ural mutation, DUla and DUlb, were sequenced by the Maxam and Gilbert procedure [14] as outlined in Figure 1. Both fragments have identical sequences except for the region deleted in DUlb (Fig. 3). The complete nucleotide sequence is presented in Figure2. There is only one large open reading frame, starting with an ATG at position 552 and ending with a TAA at position 1659. This reading frame is flanked by near homopolymers of A and T, which are unlikely to be involved in coding. The putative AUG is within a sequence compatible with the eukaryotic canonical initiation sequence Pu XXAUGG [15]. Down-stream from the termination codon are two polyadenylation signals (AAUAAA) at 20 and 29 nucleotides from the UAA codon. The coding sequence appears to be free of introns. Dictyostelium introns have been shown to have a peculiar sequence compared to other eukaryotic introns [16] and this could have prevented the complementation of yeast auxotrophy. The amino acid sequence deduced from the open reading frame corresponds to a polypeptide of 369 amino acids of predicted molecular mass 47.000. The corresponding enzyme from E. coli encodedby the PYRD gene appears to be composed of a dimer of polypeptides of molecular masses close to 40.000 [10]. From comparison of preliminary DNA sequencing data in E. coli (Jensen personal communication) and yeast (Jund and Lacroute personal communication) no direct sequence homology can be detected among these genes neither at the nucleotide level nor at the amino acid level. However general amino acid composition, chain length and hydrophobicity pattern show some similarities. The lack of direct homologies between polypeptides from different species involved in the same enzymatic function, which was already suggested from the lack of cross hybridization between cloned genes (unpublished observation) leads to the conclusion that the enzymatic specificity should reside within the tertiary structure which could be established from various primary sequences.
M. Jacquet and coll.
586 8g
UUla
C
K
Ba
SH
S F
F
Ha
STD
rR
S R SR
T FAA
A O S
"
~t
I
~
m
+~"
~ ~tt
t~ P
~ ~ '
$
OUlb
FIG.
Sequencing strategy.
I. --
D U l a and DUIb fragments correspond to the region of p D U l a and p D U l b required for complementation as described earlier
[71. The restriction sites used for sequencing are Alu 1 (A), BamH I (Ba), Bgl II (Bg), Cla I (C), Dde I (D), Fok I (F), Hinf I (H), Hae I11 (Ha), Kpn i (K), Rsa 1 (R), Sau 3A (S) and Taq I (T). The arrows indicate the fragments sequenced by the Maxam and Gilbert procedure with 5' labelling (~) or 3' labelling ( - ) . Total lenght of DUIa is 1898 base pairs.
L;UIATdTAL ;ATC|GAAAT~GA|~CATTATAAC~AACCATAC~TAA~TAATTT~T~TTTTTTTTTTTTTTTTTTTTTTTTTTTTTT~AAAAAAAAAAA 1 ATATATT~GAAA~TA~TAAT~T~A~|TA~TAT~T~T~T~T~TTTTTTTTTcTTTAT~A~TTA~T~A~CAGTT5AA~AAACACATTCA ~t CC~GCAACAGA~CTTT~ACT~CAAAT~AGTA~CCACT~AA~[AGTAC~ATCCTCA4~AC~AAACAAC~T~T~TTTCTATTTAAAATA 181 |CCATTTTATTTTTATATATAT~TAT[AC|AAAAA~AA~AGAAAAGAAAGAAAAAAAAAAAAA~AAAAAAAAAAAAAAAAAAAAGGAG~ 271
AGA&~G~GGA~AAAAAGG;AGA~AAGAAAAAAAAAA~t;~AGAAAGAAAAAAAAAAAT~TT~3GAAATTAATAATC~GGAT~AAATTTTT 361 TAAT;AAA[[AAAA~TTGAAGATTA~CATACCTTT[~A~TTGGT~TTTTTTTTTT~TT~TCT~T~TTTCAAATT~GTTA~AATAGAA~A HET~LUTRPP~L[AS~LY~PkU~Li~YRA~P~LEAS~LY~$~RT~LUA~PASy`~U~LUAS~CLYPR~PHE~LEGLU AAAAAAAAAAAAT~GA~T~C~AATTAA|AAACCAATTTAT~ATA1[AATAAAA~|TCGGAA~ATA~TTTA~A~AATGGACCAT~CATTG ~L~TYkLY~LY~VALASPAR~AS~ASPAS~L~A~PP~!|S[kLY~[YR|LEASPP~tEL[U~LYG~P~LYSL~ALASE~PR~|LEGLYVAL AAC~A|ATAA~AAA~TA~ATA~AA~TGATAA~AAsCATCCA~TAAATATATC~ATTTCTT~GG~CAAAAA~TAGC~AGTCCAATTGGT~ 631 PR:~LAGLUPR~L[UL[UA~NSEk~LNTRPVALL~Pz~EALA~[UVALALA~LYPH[ASPL~UP~UTHR[~L~T~tR~LEARGSERH~S TACC~CCAGAACCATTAT[AAATAGTCAATGG~TTA~A~TTGCATTGGTAGCAGGATTT~ATTTACCAACA~T~AGACCATTAGAAGTC GLUH|SPhEG~¥~t~PR~VALP~NVA~[T[¥RL|UAS~L[UGLU~ERGLU~PLY$G~NPH~Tt~qL¥S~L~A~PSE~GL¥SERTHk AC~A~ATTTTGGTCATC~AGTACCAAATGTTATGTA~[TA~AT[~GCAAAGTGAGGA~AAACAATTTACAAAATCA~ATA~TGG|TCAA LiU~I$~LATHRGLNTItRILEPRUTM~TItkXL[ASPCL~LEUALAILETHRASNSFRPHEGLYUETPROSFR~[TGLYLYSGLUTYRLFU CAT~CATGCAACTCAA~C~AT~CCAACC~C~A~C~A~TAGCAATtACCAAT~CATT~TATGCCAAG1A~GGG~A~AG~GTA~C TY~LYSGLY]L~]H~L[UALA~t[SS[RTY~L~UGLYSFR~[Y~LNSER~ET|L1VAL~¢~LETH~GLYTH~ALA~ERS[R~LAH|SA~P TCTATAA~GTA~AA~A~T~GCACA~AG~TA~TTAb~TTCA~TCAATCAATGATCG[A~CA~ACA~GT~C~GC~A~T~GTGCACAT~ PHELEU~LNA~PL[ULEU~L~RGP~EV~L~Y~ETCY$ALAGLYALALY$M~T¢AL~LUV~LASNT~RSERC~PR~A~NVALVALTHR AC~T~CAAG~TTGTT~ATACG~TTC~AT~AT~T~T~T~G~GCAAAA~T~CT~A~TAaATTAT~TT~TCCAAATGTTG~C~ lOB| GLYGLU~LY~LN~LETY~H~5A$NPK~SP~L~V~LTV~LU|LE~ER~E~i~kL~UVALLY$~LULEU$~$~LYSA$NXL[P~jLFU CTGGCGAG~GTCAAA~CTATCATAA|~CA~TG~TCrATGkAAT~TCTTCAACTTTGG|CAAA~AGTTA~CGA~TAAGAATAT|CCAC 117L ~L~1LELYS~ALGLY~AL~TASPA~PL~UG~ULY~[T~LUGLYLEUL[U~LN~LNALA~LUApGALA~y~ALALAALA|L[A~AGLY ~CATTATTAAAG~TGG~GT~ATGGATGACTT~A~A~A~T~GAAGGA~TACTCCA~CAAGCT~A~GAGCT~GCCTT~CCGCAATTGCTG 12bl
~L~SNTh~LEUs~R~ETLY~ALTH~A~PL¥S|L~TH~LYGLUPRU$E~LEU~LYALA~L~U|ttk$~RGLYVAL~Y5~L¥ALA ~[AT~AA|ACrr[AT~AArGAAAGT~A~A[AAAAT~A~T~GTGAAC~ATCAT|~GT~C~TC~CGTT~AAC~GTG~TGT[[~T~GTG 1351
P~LEA~AL~ALAL~U~PT~PV~L~RTH~ALA~RSER~LEILEL~$L~S~L~ASN~E~LY~LEUL~LEUL~U~LY~Y~LY CTC~AAT~T~GCTCCT~TA~G~T~TCsAC~CT|CTT~AATCATTA~G~ACAA~ATAG~AAA[TAAAACTTTTGGGTTCTG GLVILEL~PLYSPK~GLUH|SPH~ ~SPASPP,~[L~UASNS£~GLYALAA~PILEALA~}TS~RAL~ILFGLYtEu~ETT~PASPP~OTY~ G~CTATT~AC~A~CCT~A~ATT|T~AT~AT~TTTTA~A~CTGGTGCT~ATATT~CTATGTCT~CAA~TGGTTTAATGTGGGATCC~f ILFALAUITLY~KRPIt|~A~NA~NA~LY~AS'~ A~A~T~T~GA~ATG~CA~AACA~TAATA~AA~TAATrAAACAATTGTAAAAAAAAAA~AAAAAAAATAAAAAAAAAATGTAAAAAA~ sA|ATATTA~AAAACA~TAAATAATTA1TTTAAATAGATA~TAA~TTTT~TATT~TATATATATTAAAATTAAA~TATTGG~TT~A 1711 ~C~GGA~C~[~T1TACTT~GAC^a~i~C~TTT~C~c~GT~T~TTC~TTGF~G~^GGTGG~C~GCC~Tii~GG~c~TTG~ 1~O1
TTTGGATC 10~1 FIG.
2.
--
Complete nzzcleotide sequence of DUIa and deduced amino acid sequence from the open reading frame.
Dihydroorotate dehydrogenase gene o f dictyostelium |a
Ib
587
A 6A TC TGAAA T TGATTCAT T A T A A C C A ~ T - ~ . A A A T A A T r T TTT TTTTTTTTTTTTTT T TT T T T T TT TT T TTT~TA~AAAAAAAAA :'TC TGIAAT rGATICATTATAACCA~CCATACC TI~,AAATAATTTTTT
~
F--P-
A
A
A I t" f ~ r T IC'AAAA l" I'A ~"TAA TI~[Tr'~ [ | ~ t T
C-.~-
ToToT+IoIoJ+T[ J T~J [ ~ o T T ~ T J . [ A T A A T T TACT, ACC AGTTG AAAAA^CAC AT TC A
G...~
B
C-CTGCAACAGATCI'TT CACT TCCAAATGAGTA~,CCAC TTAATTTAG TACC ATCC TCAAGTACTAAACAACCTTTTTTTCTATTTAAAATA
B A l C CA T T TT A T TI" T TATATAT AT TGTA T T~ C T AA~. A+~.GAA~.AGAAAAGAAAG~AAA AA AAAAAAA ~AAA AAA AIAAAAAA A A A ~ G G A G G
A E-t,,.-
D-t,..-
,,.~t~,AAAT
D""
T TGAAc;At t
H-,-
~ F- , -
A
t
B
A ...................
"13 C,-D,..-
H..-~..-
A T TaAAT TGGT TT r t T t t t r r t t t t T T~-CT r:rr t t t CA AAT TCG ~ ~ + , c ~ ~ ~. T. _~_r . I ~ ~ ~"
•
!r;.r2Z!J.~..~L~.JL_'J_c__~J.!.U_Tj.c.,~, B
-4-.G
•~AAAAAAAA AAATGGAA A .'~,~.~A A A A A A A A T GG A .~,
FIG. 3 . -
Special features of the 5"flanking sequence.
Nucleotide sequence upstream to the putative A T G in p D U l a and p D U l b . Overlined sequences A correspond to A or T homopolyiner longer t h a n 10 nucleotides. Underlined B sequences are T n C T m or their complements. C, D, E, F, G, H boxed sequences:are direct nr inverted repeats (see the arrow) of 8 or more nucleotides.
The codon usage pattern (Table II) for this polypeptide shows several unusual features. First, as expected from a species with a genome containing 78 % A + T , there is a preferential use o f A or T at the third position. Second, eleven codons are not used. Moreover, seven out of these are absent from other Dictyostelium genes of known sequence. This bias in codon usage seems to be a characteristic of Dictyostelium genes, affecting a family of codons with higher pairing forces [18], as if the translational machinery o f this organism had been selected for weak codon anticodon interactions.
3) The 5' f l a n k i n g s e q u e n c e Five hundred fifty-two nucleotides have been sequenced upstream from the putative ATG in p D U l a . This sequence presents several salient features (Fig. 3). The A + T content is unusually high (82% compared to 66% for the coding region). Several homopolymer stretches o f A or 1" are present within this region. Seven of them are longer than 10 nucleotides and 2 contain more than 30 nucleotides. Complementarity between these homopolymers could generate a large number of stems. It is noteworthy that a deletion has occurred during cloning within two T homopolymer (Fig. 3) to generate pDUlb. Therefore these structures are able to participate in recombination events. We do not know whether recombination took place in yeast or in E. coli. These homopo-
lymers are often contained in larger streches of purines or pyrimidines (B in Fig. 3) with either one G among the As or one C among the Ts. These repetitions are reminiscent of satellite DNA. The larger purine track contains 112 residues starting at position 303. High A + T content of intergenic sequences has been reported for all Dictyostelium genes so far sequenced [17]. They also seem to occur to a lesser extent in other organisms [19]. They might be involved in gene organization in chromatin. Nuclear non histone proteins specific for A + T polymers have been reported in Dictyostelium [20] and could play a role in defining domains for gene packaging a n d / o r gene activation. Figure 3 also shows the existence of repeated sequences. Only perfect homology of 8 or more nucleotides has been boxed; and labelled C to H. The meaning of these sequences is unknown. The C, G and F sequences could be related to TATA boxes. The yeast RNA polymerase is able to find both in pDUla and pDUlb transcriptional initiation signals since both of these plasmids complement yeast, although they differ by a large deletion. Preliminary S I mapping experiments performed in yeast transformed cells (data not shown) give the same capping site for both fragments at position 550 and additional sites for pDU1 at position 523, 516 and 507. The two plasmids share upstream the CAP site a T homopolymer and the F sequence which
M. Jacquet and coll.
588
TABLE II Codon usage table a
b
a
b
a
b
a
b
Phe Phe Leu Leu
UUU UUC UUA UUG
8 3 19 8
37 43 62 7
Ser Ser Ser Ser
UCU UCC UCA UCG
6 0 11 1
35 7 77 2
Tyr Tyr NS NS
UAU UAC UAA UAG
I1 1 1 --
23 30 I0 0
Cys Cys NS Trp
UGU UGC UGA UGG
4 1 -6
32 1 0 27
Leu Leu Leu Leu
CUU CUC CUA CUG
2 3 0 0
3 22 1 0
Pro Pro Pro Pro
CCU CCC CCA CCG
1 0 18 0
2 0 8t 0
His His Gin Gln
CAU CAC CAA CAG
8 I 11 0
17 15 72 1
Arg Arg Arg Arg
CGU CGC CGA CGG
2 0 0 1
44 3 3 0
lie Ile lie Met
AUU AUC AUA AUG
19 7 3 13
80 40 1 58
Thr Thr Thr Thr
ACU ACC ACA ACG
5 6 8 0
70 39 18 0
Asn Asn Lys Lys
AAU AAC AAA AAG
20 1 23 4
52 40 89 3
Ser Ser Arg Arg
AGU AGC AGA AGG
15 0 3 0
6 5 37 0
Vai Val Val Val
GUU GUC GUA GUG
9 4 7 0
106 30 13 1
Ala Ala Ala Ala
GCU GCC GCA GCG
12 1 13 0
100 23 16 0
Asp Asp Glu Glu
GAU GAC GAA GAG
20 3 12 6
137 17 74 0
Gly Gly Gly Gly
GGU GGC GGA GGG
22 2 5 0
122 2 8 0
a : Number of codons found in the open reading frame of DUla. b : Number of codons from other Dictyostelium genes from Kimmel and Firtel, 1983. Underlined values correspond to codons not used in any sequenced gene. c o u l d b e i n v o l v e d in R N A p o l y m e r a s e p o s i t i o ning. T h e E a n d F r e p e a t s are o n l y c o m p o s e d o f p u r i n e s . T h e i r r e p e t i t i o n c o u l d suggest s o m e r e c o g n i t i o n sites for proteins. F u r t h e r w o r k , u s i n g a D N A m e d i a t e d t r a n s f o r m a t i o n s y s t e m f o r Dictyostelium is r e q u i r e d to get m o r e insight on the p u t a t i v e r e g u l a t o r y role o f s u c h s e q u e n c e s in Dictyostelium.
Acknowledgments We thank R. Guilbaud for her skillful technical assistance. We are gratefull to M. Kaczoreck for introducing us to sequencing techniques and to R. Kessin for critical reading o f the manuscript and the English revision. This work has been supported by grants o f the Ddl~gation Gdndrale ~ la Recherche Scientifique and CNRS.
REFERENCES 1. Loomis, W.F. (1982) in : "Dictyostelium discoideum : A developmental System', published b~' Academic Press, NY. 2. Firtel, R.A., Timm, R., Kimmel, A.R. & Me Keown, M. (1979) Proc. Natl. Acad. Sci. USA, 76, 206-210.
3. Mc Carroll, R., Olsen, G.J., Stahl, Y.D., Woex, C.R. & Sogen, M.L. (1983) Biochemistry, 22, 858-868. 4. Jones, M. (1980) Ann. Rev. Biochem., 49, 253-279. 5. Franke, J. & Kessin, R. (1977) Proc. Natl. Acad. Sci. USA, 74, 2157-2161. 6. Franke, J. & Kessin, R. (1978) Nature, 272, 537-538. 7. Jacquet, M., Boy-Macotte, E., Rossier, C. & Kessin, R. (1982) J. Mol. andApplied Genetics, 1,513-525. 8. Boy-Marcotte, E., Vilaine, F., Camonis, J. & Jacquet, M. (1984) Mol. Gen. Genet., 193, 406-413. 9. Loison, G., Jund, R., Nguyen-Juilleret, M. & Lacroute, F. (1981) Curt. Genet., 3, 119-123. 10. Jensen, K.F., Larsen, J.N., Schack, L. & Sivertsen, A. (1984) Fur. J. Biochem., 140, 343-352. 11. Rawls, J., Carowls, Jr., Chambers, L. & Cohen, W.S. (I981) Biochem. Genet., 19, 115-127. 12. Smithers, G.W., Noero, A.M. & O'Sullivan, W.J. (1978) Annal. Biochem., 88, 93-103. 13. Rawls, J. (1979) Comp. Biochem. Physiol., 6213, 207-215. 14. Maxam, A. & Gilbert, W. (1977) Proc. Natl. Acad. Sci. USA, 74, 560-564. 15. Kosak, M. (1981) Nuc. Acid. Res., 9, 5223-5252. 16. Kimmel, A.R. & Firtel, R.A. (1980) Nuc. Acid. Res., 8, 5599-5610. 17. Kimmel, A.R. & Firtel, R.A. (1983) Nuc. Acid. Res., 11, 541-552. 18. Lagerkvist, U. (1981) Cell, 23, 305-306. 19. Moreau, J., Marcaud, L., Maschat, F., Kejzlarova Lepesant, J., Lepesant, J.A. & Scherrer, K. (198f) Nature, 295, 260-262. 20. Garreau, H. & Williams, J.G. (1983) Nuc. Acid. Res., 11, 8473-8484.
Mdmoires originaux
Sequence analysis of a Dictyostelium discoideum gene coding for an active dihydroorotate dehydrogenase in yeast. Michel J A C Q U E T °, Monique K A L E K I N E and Emmanuelle BOY-MARCOTI'E. Laboratoire de Biologie Expdrimentale, B6timent 400, Universitd Paris-Sud, 91405 Orsay Cedex (France). (Refu le 21-2-1985, accept~ le 26-14-1985).
R~sum~ - - Un fragment de DNA de Dictyostelium discoideum isold sur la base de sa capacitd de compldmenter la mutation ural de la levure, contient un gone qui code pour une activitd dihydroorotate dehydrogenase. La sdquence nucldotidique complOte de ce fragment de 1898 paires de bases a dtd ddtermhlde et r~vOle une phase de lecture ouverte capable de coder pour une chaine peptidique de masse mol~culaire 47 000. Ce gone pr~sente un usage prdfdrentiel des codons de faible force d'appariement. Onze codons potentiels ne sont pas utilisds, en particulier ceux ayant un G en 3~ position sont absents. Les rdgions en bordure de la phase de lecture ont une teneur anormalement dlev~e de nucldotides A et T (80 %). La rdgion amont du gone prdsente plusieurs particularitds intdressantes comme la presence de plusieurs sdquences r~p~tdes directes et inverses. Mots-cl~s : DNA recombinant / biosynth~se des pyrimidines / usage des codons / DNA espaceur.
S u m m a r y - - A Dictyostelium discoideum DNA fragment isolated on the basis o f its ability to complement the ura I mutation o f yeast, codes for a dihydroorotate dehydrogenase activity. Tile complete nucleotide sequence o f this 1898 bp fragment has been determined and reveals an open reading frame capable o f coding for a 369 amino acid polypeptide o f molecular mass 47.000. Tile gene shows preferential use o f codons with weak pairing forces. Eleven codons, mainly those with a G in the third position, are absent. Ti~eflanking sequences are unusually rich in A + T (80 %). Several direct and inverted repeats exist in the 5' flanking sequence. Key-words : recombinant DNA / pyrimidine biosynthesis / codon usage / spacer DNA.
Introduction
Several genes from Dictyostelium discoideum have been cloned and sequenced (for review see [11). Despite varying programs of gene expression, they all have similar and unusual features in their genomic organization and sequence. All are embeded within spacer sequences unusually 0 To whom all correspondence should be addressed.
rich in A + T [2] and seem to use a restricted set of codons. Since Dictyostelium discoideum belongs to one of the more primitive phyla of eukaryotes, as documented by recent evolutionary comparisons of ribosomal RNA sequences [3], it is interesting to know how its gene organization resembles to or differs from other organisms. We chose to extend this analysis to genes involved in basic
584
M. Jacquet and coll.
metabolism which could be compared to their homologs in other organisms as well as known Dictyostelium genes. The six enzymatic steps of the pyrimidine de novo biosynthetic pathway are similar in most organisms, they although differ by their integration at the enzymatic level or their regulation [4]. Although scarcely studied, the existence of this pathway in Dictyostelium is documented by the ability to grow amoebae on defined medium deprived of pyrimidine [5] and by the isolation o f auxotrophic mutants [6]. More recently, we took advantage of genetic engineering methods to study the genes of pyrimidine biosynthesis. Two genomic fragments from Dictyostelium were selected in yeast pyrimidine mutants on the basis o f their ability to restore uracil prototrophy. One of them is able to complement a mutation (ural) in the gene coding for the dihydroorotate dehydrogenase (DOD; EC 1-3-3-1), [7]. The other genomic fragment complem6nts two mutations of yeast : ura3 and ura5, and "codes for a bifunctional enzyme orotate pyrophosphorylase and orotate monophosphate (OMp) decarboxylase [8]. Higher eukaryotes also have a bifunctional enzyme called U M P synthase carrying out the last two steps of UMP biosynthesis [41. I n t h i s report, we demonstrate that the Dietyostelium DNA fragment which complements the ural mutation codes for a dihydroorotate dehydrogenase activity. The complete nucleotide sequence o f this fragment is given. This sequence contains an open reading frame allowing the synthesis of a 369 amino acid peptide chain of molecular mass 47.000. Preferential use o f codons is observed. This predicted sequence from Dictyostelium is the first reported for a dihydroorotate dehydrogenase and will be compared to others when available, such as the URAI gene of Saccharomyces cerevisiae or the PYRD gene from E. coli already cloned [9, 10]. The 552 nucleotides sequenced upstream o f the putative ATG initiation codon, like the other flanking regions sequenced in Dictyostelium genes, have an unusually high A + T content and several repeated segments.
Materials and methods
Biological material The Dictyostelium axenic strain AX3 (from the collection R.H. Kessin) was grown in HL/5 medium [6]. The yeast OL3 strain MATa leu2-2, Ieu2-112, his3-11,
his3-15, ural-21, ural-50) was used for transformation experiments as previously described [7]. Plasmids pDUI sub1, sub2, sub5 and sub6 [7] were grown in E. coli KI2 HB 101 and purified as described. Dihydroorotate dehydrogenase assays Yeast crude extracts were obtained by iysis of protoplasts generated by zymoliase treatment. Lysis was achieved in I0 mM Tris-Cl pH 8.0. Cell debris, nuclei and mitochondria were removed by centrifugation at 12000g for 15 minutes. Dictyostelium discoideum crude extracts were obtained by thawing frozen cells in 10 mM Tris-Cl, pH 8.0, with PMSF 1 [ag/ml and subsequent centrifugation at 12 000 g for 15 minutes. Dihydroorotate dehydrogenase measurements were performed according to the procedure described by Rawls et al. [11], except that the electron transport system was replaced by dichlorophenol indophenol (2,6 DCIP). Carboxy-~SC-dihydroorotic acid was prepared from carboxy-~4C-orotic acid 30-50 mCi/mM (New England Nuclear) as described by Smithers [12]. Purification of the products was performed by chromatography on DEAE cellulose paper as described by Rawls [13].
DNA sequencing Isolated fragments were sequenced by the Maxam and Gilbert procedure [14] after 5' labelling with kinase or 3' labelling using the Klenow fragment of DNA polymerase. Restriction enzymes and polymerases were purchased from BRL, New England Biolabs, and Boehringer, and used as described by the suppliers. ~2p ATP and tt3:P NTP were from Amersham.
Results and discussion
I) A DNA fragment coding f o r a dehydroorotate dehydrogenase activity As previously reported, a DNA fragment from
Dictyostelium discoideum was selected in a yeast ural mutant due to its capacity to restore uracil prototrophy [71. Plasmid DNA containing this fragment was transfered to and amplified in E. colistrain HBI01. Two different plasmids were found among the transformed bacterial clones. Both contained an insert capable ofoconfering uracil prototrophy on the yeast strain carrying the ural mutation. They differed only by a 450 bp segment present in pDUla and absent in pDUlb, which is not required for complementation. Both also contained a genomic fragment from Dictyostelium not required for complementation and not adjacent in genomic DNA, as determined from Southern blot analysis. This extra fragment
Dihydroorotate dehydrogenase gene of dictyostelium probably originated from coligation during in vitro recombination. Its presence in both plasmids suggests that they originate from the same construction and one (pDUIb)has been deleted during cloning either in yeast or in E. coli. The segment required for complementation has been subcloned in different plasmids. To insure that the DNA fragment was coding for the dihydroorotate dehydrogenase, the enzymatic activity was assayed in transformed yeast ural cells. The DNA fragment was subcloned from the original insert of pDUla in a different plasmid (pDUI sub2) which allows a higher growth rate of the transformed yeast in uracildeprived medium. Enzymatic activity was followed by conversion of labelled dihydroorotate rather than by reduction of a hydrogen carder in order to reduce the possibility of artifacts. Such an assay can be performed on crude extracts after removal of broken cell debris, membrane, nuclei, and mitochondria. As shown in Table I, no activity was detected in the yeast ural mutant. In contrast, in transformed cells, like in the wild-type strain, high activities are observed. This result establishes that complementation is due to the recovery of dihydroorotate dehydrogenase activity. TABLEI Dihydroorotate dehydrogenase activity in )'east and dictyostelium. Enzymatic activity nmole/min/mg protein Strain Yeast GRFI8 (URAI) Yeast OL 3 (ural) Yeast OL 3 (urAl +pDUI subl) Dictyostelium discoideum, AX3
6.5 < 0.1 4.0 0.76
The dihydroorotate dehydrogenase activity was measured as described in Methods in yeast and Dictyostelium extracts. The unit of activity corresponds to one nmole of orotate produced per minute per m g of proteins.
The enzymatic activity measured in extracts of Dictyostelium amobae was found to be soluble. The specific activity is lower than in yeast, like in the case for other enzymatic activities of the pyrimidine pathway [8]. The value found in transformed yeast cells is slightly lower than in ' wild-type yeast but higher than in Dictyostelium. It may be that the Dictyostelium enzyme has a
585
lower efficiency in yeast but that it is compensated by the higher copy number of the gene. It is interesting to note that Dictyostelium dihydroorotate dehydrogenase, like the yeast enzyme, appears to be soluble. This contrasts with their homologs in others eukaryotic organisms which are tighly bound to the inner mitochondrial membrane [4].
2) The coding sequence The two DNA fragments capable of complementing of the ural mutation, DUla and DUlb, were sequenced by the Maxam and Gilbert procedure [14] as outlined in Figure 1. Both fragments have identical sequences except for the region deleted in DUlb (Fig. 3). The complete nucleotide sequence is presented in Figure2. There is only one large open reading frame, starting with an ATG at position 552 and ending with a TAA at position 1659. This reading frame is flanked by near homopolymers of A and T, which are unlikely to be involved in coding. The putative AUG is within a sequence compatible with the eukaryotic canonical initiation sequence Pu XXAUGG [15]. Down-stream from the termination codon are two polyadenylation signals (AAUAAA) at 20 and 29 nucleotides from the UAA codon. The coding sequence appears to be free of introns. Dictyostelium introns have been shown to have a peculiar sequence compared to other eukaryotic introns [16] and this could have prevented the complementation of yeast auxotrophy. The amino acid sequence deduced from the open reading frame corresponds to a polypeptide of 369 amino acids of predicted molecular mass 47.000. The corresponding enzyme from E. coli encodedby the PYRD gene appears to be composed of a dimer of polypeptides of molecular masses close to 40.000 [10]. From comparison of preliminary DNA sequencing data in E. coli (Jensen personal communication) and yeast (Jund and Lacroute personal communication) no direct sequence homology can be detected among these genes neither at the nucleotide level nor at the amino acid level. However general amino acid composition, chain length and hydrophobicity pattern show some similarities. The lack of direct homologies between polypeptides from different species involved in the same enzymatic function, which was already suggested from the lack of cross hybridization between cloned genes (unpublished observation) leads to the conclusion that the enzymatic specificity should reside within the tertiary structure which could be established from various primary sequences.
M. Jacquet and coll.
586 8g
UUla
C
K
Ba
SH
S F
F
Ha
STD
rR
S R SR
T FAA
A O S
"
~t
I
~
m
+~"
~ ~tt
t~ P
~ ~ '
$
OUlb
FIG.
Sequencing strategy.
I. --
D U l a and DUIb fragments correspond to the region of p D U l a and p D U l b required for complementation as described earlier
[71. The restriction sites used for sequencing are Alu 1 (A), BamH I (Ba), Bgl II (Bg), Cla I (C), Dde I (D), Fok I (F), Hinf I (H), Hae I11 (Ha), Kpn i (K), Rsa 1 (R), Sau 3A (S) and Taq I (T). The arrows indicate the fragments sequenced by the Maxam and Gilbert procedure with 5' labelling (~) or 3' labelling ( - ) . Total lenght of DUIa is 1898 base pairs.
L;UIATdTAL ;ATC|GAAAT~GA|~CATTATAAC~AACCATAC~TAA~TAATTT~T~TTTTTTTTTTTTTTTTTTTTTTTTTTTTTT~AAAAAAAAAAA 1 ATATATT~GAAA~TA~TAAT~T~A~|TA~TAT~T~T~T~T~TTTTTTTTTcTTTAT~A~TTA~T~A~CAGTT5AA~AAACACATTCA ~t CC~GCAACAGA~CTTT~ACT~CAAAT~AGTA~CCACT~AA~[AGTAC~ATCCTCA4~AC~AAACAAC~T~T~TTTCTATTTAAAATA 181 |CCATTTTATTTTTATATATAT~TAT[AC|AAAAA~AA~AGAAAAGAAAGAAAAAAAAAAAAA~AAAAAAAAAAAAAAAAAAAAGGAG~ 271
AGA&~G~GGA~AAAAAGG;AGA~AAGAAAAAAAAAA~t;~AGAAAGAAAAAAAAAAAT~TT~3GAAATTAATAATC~GGAT~AAATTTTT 361 TAAT;AAA[[AAAA~TTGAAGATTA~CATACCTTT[~A~TTGGT~TTTTTTTTTT~TT~TCT~T~TTTCAAATT~GTTA~AATAGAA~A HET~LUTRPP~L[AS~LY~PkU~Li~YRA~P~LEAS~LY~$~RT~LUA~PASy`~U~LUAS~CLYPR~PHE~LEGLU AAAAAAAAAAAAT~GA~T~C~AATTAA|AAACCAATTTAT~ATA1[AATAAAA~|TCGGAA~ATA~TTTA~A~AATGGACCAT~CATTG ~L~TYkLY~LY~VALASPAR~AS~ASPAS~L~A~PP~!|S[kLY~[YR|LEASPP~tEL[U~LYG~P~LYSL~ALASE~PR~|LEGLYVAL AAC~A|ATAA~AAA~TA~ATA~AA~TGATAA~AAsCATCCA~TAAATATATC~ATTTCTT~GG~CAAAAA~TAGC~AGTCCAATTGGT~ 631 PR:~LAGLUPR~L[UL[UA~NSEk~LNTRPVALL~Pz~EALA~[UVALALA~LYPH[ASPL~UP~UTHR[~L~T~tR~LEARGSERH~S TACC~CCAGAACCATTAT[AAATAGTCAATGG~TTA~A~TTGCATTGGTAGCAGGATTT~ATTTACCAACA~T~AGACCATTAGAAGTC GLUH|SPhEG~¥~t~PR~VALP~NVA~[T[¥RL|UAS~L[UGLU~ERGLU~PLY$G~NPH~Tt~qL¥S~L~A~PSE~GL¥SERTHk AC~A~ATTTTGGTCATC~AGTACCAAATGTTATGTA~[TA~AT[~GCAAAGTGAGGA~AAACAATTTACAAAATCA~ATA~TGG|TCAA LiU~I$~LATHRGLNTItRILEPRUTM~TItkXL[ASPCL~LEUALAILETHRASNSFRPHEGLYUETPROSFR~[TGLYLYSGLUTYRLFU CAT~CATGCAACTCAA~C~AT~CCAACC~C~A~C~A~TAGCAATtACCAAT~CATT~TATGCCAAG1A~GGG~A~AG~GTA~C TY~LYSGLY]L~]H~L[UALA~t[SS[RTY~L~UGLYSFR~[Y~LNSER~ET|L1VAL~¢~LETH~GLYTH~ALA~ERS[R~LAH|SA~P TCTATAA~GTA~AA~A~T~GCACA~AG~TA~TTAb~TTCA~TCAATCAATGATCG[A~CA~ACA~GT~C~GC~A~T~GTGCACAT~ PHELEU~LNA~PL[ULEU~L~RGP~EV~L~Y~ETCY$ALAGLYALALY$M~T¢AL~LUV~LASNT~RSERC~PR~A~NVALVALTHR AC~T~CAAG~TTGTT~ATACG~TTC~AT~AT~T~T~T~G~GCAAAA~T~CT~A~TAaATTAT~TT~TCCAAATGTTG~C~ lOB| GLYGLU~LY~LN~LETY~H~5A$NPK~SP~L~V~LTV~LU|LE~ER~E~i~kL~UVALLY$~LULEU$~$~LYSA$NXL[P~jLFU CTGGCGAG~GTCAAA~CTATCATAA|~CA~TG~TCrATGkAAT~TCTTCAACTTTGG|CAAA~AGTTA~CGA~TAAGAATAT|CCAC 117L ~L~1LELYS~ALGLY~AL~TASPA~PL~UG~ULY~[T~LUGLYLEUL[U~LN~LNALA~LUApGALA~y~ALALAALA|L[A~AGLY ~CATTATTAAAG~TGG~GT~ATGGATGACTT~A~A~A~T~GAAGGA~TACTCCA~CAAGCT~A~GAGCT~GCCTT~CCGCAATTGCTG 12bl
~L~SNTh~LEUs~R~ETLY~ALTH~A~PL¥S|L~TH~LYGLUPRU$E~LEU~LYALA~L~U|ttk$~RGLYVAL~Y5~L¥ALA ~[AT~AA|ACrr[AT~AArGAAAGT~A~A[AAAAT~A~T~GTGAAC~ATCAT|~GT~C~TC~CGTT~AAC~GTG~TGT[[~T~GTG 1351
P~LEA~AL~ALAL~U~PT~PV~L~RTH~ALA~RSER~LEILEL~$L~S~L~ASN~E~LY~LEUL~LEUL~U~LY~Y~LY CTC~AAT~T~GCTCCT~TA~G~T~TCsAC~CT|CTT~AATCATTA~G~ACAA~ATAG~AAA[TAAAACTTTTGGGTTCTG GLVILEL~PLYSPK~GLUH|SPH~ ~SPASPP,~[L~UASNS£~GLYALAA~PILEALA~}TS~RAL~ILFGLYtEu~ETT~PASPP~OTY~ G~CTATT~AC~A~CCT~A~ATT|T~AT~AT~TTTTA~A~CTGGTGCT~ATATT~CTATGTCT~CAA~TGGTTTAATGTGGGATCC~f ILFALAUITLY~KRPIt|~A~NA~NA~LY~AS'~ A~A~T~T~GA~ATG~CA~AACA~TAATA~AA~TAATrAAACAATTGTAAAAAAAAAA~AAAAAAAATAAAAAAAAAATGTAAAAAA~ sA|ATATTA~AAAACA~TAAATAATTA1TTTAAATAGATA~TAA~TTTT~TATT~TATATATATTAAAATTAAA~TATTGG~TT~A 1711 ~C~GGA~C~[~T1TACTT~GAC^a~i~C~TTT~C~c~GT~T~TTC~TTGF~G~^GGTGG~C~GCC~Tii~GG~c~TTG~ 1~O1
TTTGGATC 10~1 FIG.
2.
--
Complete nzzcleotide sequence of DUIa and deduced amino acid sequence from the open reading frame.
Dihydroorotate dehydrogenase gene o f dictyostelium |a
Ib
587
A 6A TC TGAAA T TGATTCAT T A T A A C C A ~ T - ~ . A A A T A A T r T TTT TTTTTTTTTTTTTT T TT T T T T TT TT T TTT~TA~AAAAAAAAA :'TC TGIAAT rGATICATTATAACCA~CCATACC TI~,AAATAATTTTTT
~
F--P-
A
A
A I t" f ~ r T IC'AAAA l" I'A ~"TAA TI~[Tr'~ [ | ~ t T
C-.~-
ToToT+IoIoJ+T[ J T~J [ ~ o T T ~ T J . [ A T A A T T TACT, ACC AGTTG AAAAA^CAC AT TC A
G...~
B
C-CTGCAACAGATCI'TT CACT TCCAAATGAGTA~,CCAC TTAATTTAG TACC ATCC TCAAGTACTAAACAACCTTTTTTTCTATTTAAAATA
B A l C CA T T TT A T TI" T TATATAT AT TGTA T T~ C T AA~. A+~.GAA~.AGAAAAGAAAG~AAA AA AAAAAAA ~AAA AAA AIAAAAAA A A A ~ G G A G G
A E-t,,.-
D-t,..-
,,.~t~,AAAT
D""
T TGAAc;At t
H-,-
~ F- , -
A
t
B
A ...................
"13 C,-D,..-
H..-~..-
A T TaAAT TGGT TT r t T t t t r r t t t t T T~-CT r:rr t t t CA AAT TCG ~ ~ + , c ~ ~ ~. T. _~_r . I ~ ~ ~"
•
!r;.r2Z!J.~..~L~.JL_'J_c__~J.!.U_Tj.c.,~, B
-4-.G
•~AAAAAAAA AAATGGAA A .'~,~.~A A A A A A A A T GG A .~,
FIG. 3 . -
Special features of the 5"flanking sequence.
Nucleotide sequence upstream to the putative A T G in p D U l a and p D U l b . Overlined sequences A correspond to A or T homopolyiner longer t h a n 10 nucleotides. Underlined B sequences are T n C T m or their complements. C, D, E, F, G, H boxed sequences:are direct nr inverted repeats (see the arrow) of 8 or more nucleotides.
The codon usage pattern (Table II) for this polypeptide shows several unusual features. First, as expected from a species with a genome containing 78 % A + T , there is a preferential use o f A or T at the third position. Second, eleven codons are not used. Moreover, seven out of these are absent from other Dictyostelium genes of known sequence. This bias in codon usage seems to be a characteristic of Dictyostelium genes, affecting a family of codons with higher pairing forces [18], as if the translational machinery o f this organism had been selected for weak codon anticodon interactions.
3) The 5' f l a n k i n g s e q u e n c e Five hundred fifty-two nucleotides have been sequenced upstream from the putative ATG in p D U l a . This sequence presents several salient features (Fig. 3). The A + T content is unusually high (82% compared to 66% for the coding region). Several homopolymer stretches o f A or 1" are present within this region. Seven of them are longer than 10 nucleotides and 2 contain more than 30 nucleotides. Complementarity between these homopolymers could generate a large number of stems. It is noteworthy that a deletion has occurred during cloning within two T homopolymer (Fig. 3) to generate pDUlb. Therefore these structures are able to participate in recombination events. We do not know whether recombination took place in yeast or in E. coli. These homopo-
lymers are often contained in larger streches of purines or pyrimidines (B in Fig. 3) with either one G among the As or one C among the Ts. These repetitions are reminiscent of satellite DNA. The larger purine track contains 112 residues starting at position 303. High A + T content of intergenic sequences has been reported for all Dictyostelium genes so far sequenced [17]. They also seem to occur to a lesser extent in other organisms [19]. They might be involved in gene organization in chromatin. Nuclear non histone proteins specific for A + T polymers have been reported in Dictyostelium [20] and could play a role in defining domains for gene packaging a n d / o r gene activation. Figure 3 also shows the existence of repeated sequences. Only perfect homology of 8 or more nucleotides has been boxed; and labelled C to H. The meaning of these sequences is unknown. The C, G and F sequences could be related to TATA boxes. The yeast RNA polymerase is able to find both in pDUla and pDUlb transcriptional initiation signals since both of these plasmids complement yeast, although they differ by a large deletion. Preliminary S I mapping experiments performed in yeast transformed cells (data not shown) give the same capping site for both fragments at position 550 and additional sites for pDU1 at position 523, 516 and 507. The two plasmids share upstream the CAP site a T homopolymer and the F sequence which
M. Jacquet and coll.
588
TABLE II Codon usage table a
b
a
b
a
b
a
b
Phe Phe Leu Leu
UUU UUC UUA UUG
8 3 19 8
37 43 62 7
Ser Ser Ser Ser
UCU UCC UCA UCG
6 0 11 1
35 7 77 2
Tyr Tyr NS NS
UAU UAC UAA UAG
I1 1 1 --
23 30 I0 0
Cys Cys NS Trp
UGU UGC UGA UGG
4 1 -6
32 1 0 27
Leu Leu Leu Leu
CUU CUC CUA CUG
2 3 0 0
3 22 1 0
Pro Pro Pro Pro
CCU CCC CCA CCG
1 0 18 0
2 0 8t 0
His His Gin Gln
CAU CAC CAA CAG
8 I 11 0
17 15 72 1
Arg Arg Arg Arg
CGU CGC CGA CGG
2 0 0 1
44 3 3 0
lie Ile lie Met
AUU AUC AUA AUG
19 7 3 13
80 40 1 58
Thr Thr Thr Thr
ACU ACC ACA ACG
5 6 8 0
70 39 18 0
Asn Asn Lys Lys
AAU AAC AAA AAG
20 1 23 4
52 40 89 3
Ser Ser Arg Arg
AGU AGC AGA AGG
15 0 3 0
6 5 37 0
Vai Val Val Val
GUU GUC GUA GUG
9 4 7 0
106 30 13 1
Ala Ala Ala Ala
GCU GCC GCA GCG
12 1 13 0
100 23 16 0
Asp Asp Glu Glu
GAU GAC GAA GAG
20 3 12 6
137 17 74 0
Gly Gly Gly Gly
GGU GGC GGA GGG
22 2 5 0
122 2 8 0
a : Number of codons found in the open reading frame of DUla. b : Number of codons from other Dictyostelium genes from Kimmel and Firtel, 1983. Underlined values correspond to codons not used in any sequenced gene. c o u l d b e i n v o l v e d in R N A p o l y m e r a s e p o s i t i o ning. T h e E a n d F r e p e a t s are o n l y c o m p o s e d o f p u r i n e s . T h e i r r e p e t i t i o n c o u l d suggest s o m e r e c o g n i t i o n sites for proteins. F u r t h e r w o r k , u s i n g a D N A m e d i a t e d t r a n s f o r m a t i o n s y s t e m f o r Dictyostelium is r e q u i r e d to get m o r e insight on the p u t a t i v e r e g u l a t o r y role o f s u c h s e q u e n c e s in Dictyostelium.
Acknowledgments We thank R. Guilbaud for her skillful technical assistance. We are gratefull to M. Kaczoreck for introducing us to sequencing techniques and to R. Kessin for critical reading o f the manuscript and the English revision. This work has been supported by grants o f the Ddl~gation Gdndrale ~ la Recherche Scientifique and CNRS.
REFERENCES 1. Loomis, W.F. (1982) in : "Dictyostelium discoideum : A developmental System', published b~' Academic Press, NY. 2. Firtel, R.A., Timm, R., Kimmel, A.R. & Me Keown, M. (1979) Proc. Natl. Acad. Sci. USA, 76, 206-210.
3. Mc Carroll, R., Olsen, G.J., Stahl, Y.D., Woex, C.R. & Sogen, M.L. (1983) Biochemistry, 22, 858-868. 4. Jones, M. (1980) Ann. Rev. Biochem., 49, 253-279. 5. Franke, J. & Kessin, R. (1977) Proc. Natl. Acad. Sci. USA, 74, 2157-2161. 6. Franke, J. & Kessin, R. (1978) Nature, 272, 537-538. 7. Jacquet, M., Boy-Macotte, E., Rossier, C. & Kessin, R. (1982) J. Mol. andApplied Genetics, 1,513-525. 8. Boy-Marcotte, E., Vilaine, F., Camonis, J. & Jacquet, M. (1984) Mol. Gen. Genet., 193, 406-413. 9. Loison, G., Jund, R., Nguyen-Juilleret, M. & Lacroute, F. (1981) Curt. Genet., 3, 119-123. 10. Jensen, K.F., Larsen, J.N., Schack, L. & Sivertsen, A. (1984) Fur. J. Biochem., 140, 343-352. 11. Rawls, J., Carowls, Jr., Chambers, L. & Cohen, W.S. (I981) Biochem. Genet., 19, 115-127. 12. Smithers, G.W., Noero, A.M. & O'Sullivan, W.J. (1978) Annal. Biochem., 88, 93-103. 13. Rawls, J. (1979) Comp. Biochem. Physiol., 6213, 207-215. 14. Maxam, A. & Gilbert, W. (1977) Proc. Natl. Acad. Sci. USA, 74, 560-564. 15. Kosak, M. (1981) Nuc. Acid. Res., 9, 5223-5252. 16. Kimmel, A.R. & Firtel, R.A. (1980) Nuc. Acid. Res., 8, 5599-5610. 17. Kimmel, A.R. & Firtel, R.A. (1983) Nuc. Acid. Res., 11, 541-552. 18. Lagerkvist, U. (1981) Cell, 23, 305-306. 19. Moreau, J., Marcaud, L., Maschat, F., Kejzlarova Lepesant, J., Lepesant, J.A. & Scherrer, K. (198f) Nature, 295, 260-262. 20. Garreau, H. & Williams, J.G. (1983) Nuc. Acid. Res., 11, 8473-8484.