Gene. 95 (1990) 123-127
123
Elsevier GENE 03739
Nitrogen regulation in Aspergillus: are two fingers better than one? (Gene areA; Aspergiilus nidulans; didactylous mutation; duplication; erythroid transcription factor; nitrogen metabolite repression; recombinant DNA; zinc finger)
Mark X. Caddick" and Herbert N. Arst Jr. b ° Department of Genetics and Microbiology, University of Liverpool. Liverpool 1,69 3BX (U.K.) Tel. (+ 44-5 i) 7943632, and b Departmem of Bacteriology. Royal Postgraduate Medical School, London WI2 0NN (U.K.) Received by R.W. Davies: 14 June 1990 Revised: 30 June 1990 Accepted: ! July 1990
SUMMARY
The areA gene, mediating nitrogen metabolite repression in AspergiUus nidulans, encodes a positive-acting regulatory protein with a single putative DNA-binding 'zinc finger' which is remarkably similar to the two 'zinc fingers' of the major regulatory protein of vertebrate erythroid cells (GF-I/Eryfl/NF-EI). The areA-300 mutation alters the specificity of gene activation in that it elevates expression of certain structural genes whilst reducing expression of certain others. It is an 'in-frame' tandem duplication of 417 bp including the entire DNA-binding region. The consequences of areA didactyly are further explored by construction of a double mutant having an altered loop residue in the N-terminal 'finger'.
INTRODUCTION
The positive-acting regulatory gene areA mediates nitrogen metabolite repression in the metabolically versatile fungus, A. nidulans (Arst and Cove, 1973; Wiame et al., 1985; Caddick etal., 1986). Nitrogen metabolite (or 'ammonium') repression, probably acting at the level of transcription (Arst and Scazzocchio, 1985; Wiame et al., 1985; L. Gorfinkiel, G. Diallinas and C. Scazzocchio as cited in Kudla et al., 1990), enables preferential utilisation
Correspondence to: Dr. H.N. Arst, Department of Bacteriology, Royal Postgraduate Medical School, Ducane Road, London WI2 0NN (U.K.) Tel. ( + 44-81)7403243; Fax ( + 44.81)7403169. Abbreviations: aa, amino acid(s); areA, regulatory gene mediating nitrogen metaboUte repression; ~-Ala, ~.alanine; bp, base pair(s); GF-1/Eryfl/NF-EI, major vertebrate erythroid transcription factor; nt, nucleotide(s); oligo,oligodeoxyribonucleotide(s);PCR, polymerase chain reaction; r repressed (null) phenotype; wt, wild type. 0378-1i 19/90/$03.50 © 1990Elsevier SciencePublishers B.V.(BiomedicalDivision)
of the favoured nitrogen sources ammonium and L.glutamine. The derived aa sequence of areA contains a single putative DNA-binding 'zinc finger' with considerable sequence similarity to the two 'zinc fingers' of the major regulatory protein of vertebrate erythroid cells, GF-I/Eryfl/NF-EI (Arst et al., 1989; Trainor et al., 1990; Kudla et al., 1990). Mutant area phenotypes are extraordinarily diverse, including alleles which elevate expression of certain structural genes whilst having a null phenotype for certain others (Arst and Cove, 1973; Hynes, 1975; A! Taho et al., 1984; Shaffer and Arst, 1984; Arst and Scazzocchio, 1985). We have previously described (Kudla et al., 1990) a pair of such specificityalleles with approximatelyreciprocal or 'mirror image' phenotypes, each changing a conserved (Kudla et al., 1990; Fu and Marzluf, 1990)central residue of the 'finger' loop. Here we report that another specificity mutation is an 'in-frame' duplication of the entire DNAbinding region and examine the consequences of areA didactyly.
124 TABLE 1 Partial phenotype of the areA-102 areA-300 double mutant allele with the areA-102 mutation located in the 5' copy of the duplication Growth a on media b containing:
Relevant genotype
L-His ~
2-Pyrrolidone c
p-Ala ©
Uric acid c
Chlorate + ammonium c
areA-102 areA-300 areA-1O2 areA-300
+ + + +
+ +
+ + + -
+ + +
+ + + -
wt(areA +)
-
++
+ + +
++
++
a Growth tests were performed as described by Arst et aL (1982) and scored a~er two to three days' incubation at 37°C. Growth scores range from + + + (maximal growth) > + + > + > + > - (little or no growth) and are not necessarily equivalent on different media. At least 50 different strains of each genotype were tested. u Solid minimal medium (Cove. 1966) containing I ~ (w/v) D-glucose as carbon source was used. ° Nitrogen sources were present at 10 mM [2-pyrrolidone and ammonium (as the (+)-tartrate)], 5 mM (L-His and p-Ala) or 595 #M (uric acid). Chlorate (as the potassium salt) was present at 100 raM. A minus score for L-His. 2-pyrrolidone or uric acid utHisation indicates nitrogen-starved growth (Arst and Scazzocehi:J, 1985), whereas minus scores on ~-Ala and chlorate media reflect even less growth as a consequence of toxicity (Arst and Scazzocchio, 1985). Strains carrying areA-102 ( + areA-300) are inhibited by ~-Ala even when an alternative nitrogen source is present, probably due to excessive uptake (H.N.A., unpublished). Supplement levels ( > 33/zM) of meso-inositol reverse such/~-Ala toxicity, indicating that the target of toxicity is inositoi biosynthesis or a closely related function. Sensitivity to chlorate toxicity in the presence of ammoninm indicates nitrogen metabolite derepression of nitrate reductase synthesis (Arst and Cove, 1969; Arst, 1982; Kudla et ai., 1990).
in nitrogen-metabolite derepression of an asparaginase and nitrate and nitrite reductases but has a null phenotype for expression of activities necessary for utilisation of 2-pyrrolidone, L-methionine and L-leucine as nitrogen sources (ShatTer and Arst, 1984; Shaffer et al., 1988; HN.A. and P.M. Shaffer, unpublished). Additionally and unexpectedly, areA-300 seems to relieve repression of thymine-7hydroxylase and asparaginase syntheses by atmospheric oxygen (Shaffer and Arst, 1984; Shaffer et al., 1988).
RESULTS AND DISCUSSION
(a) The areA.300 mutation is a duplication of the entire putative DNA-bluding region The 4-nitroquinoline. 1-oxide.induced mutation areA-300 was selected for its ability to elevate pyrimidine-salvage pathway activity levels (Shaffer and Arst, 1984), In addition to enhancing uptake of pyrimidine derivatives and strongly derepressing synthesis of thymine-7-hydroxylase, it results
wt A+D
mmA
fx
~
B+E
, i
i
& A-300 ~J-
A+D} B+E } C+D
~
~J
D I
~ ....... i
D
, I
I
I
Fig, I. Amplification of regions within area using PCR. Oligo primers (17-20 nt long) A to E were used to amplify regions ofareA from wt and areA-300 genomic DNA by PCR as described by Kudla et aL (1990), except that annealing temperatures of 55-60 ° C were used. The relative position and orientation of the primers, the combinations used and their respective products are indicated. For areA-300, the primer combinations A + D and B + E produced two distinct products (indicated by continuous and dashed lines) visualised by gel electrophoresis and Southern hybridisation. The ratio ofprodocts varied with PCR conditions, shorter extension times generally favouring the shorter (wt) product. Primers C and D are divergently oriented, with a single nt overlap, in the wt and thus failed to give a PCR product with wt genomic DNA. Consistent with the presence of a duplication in areA-300, PCR with C + D amplified a region of418 bp which was sequenced directly (Kudla et al., 1990), enabling identification of the internal junction of the duplication (indicated by nt numbers, 1917 followed by 1501). Additional PCR products of areA-300 were also sequenced to confirm the fidelity of the entire duplicated DNA. The region duplicated is indicated by open bars; nonduplicated regions are shown as blackened bars.
125 Two striking genetic observations suggested that areA300 might be a duplication of the DNA-binding region. (i) When selection of loss-of-function (areA') mutations was attempted in areA-300 strains (Arst, 1989) by exploiting their derepression of nitrate reductase synthesis and consequent chlorate hypersensitivity (,~rst and Cove, 1969; Table I), areA-300 reverted to areA + at high frequency. Of 125 spontaneous chlorate resistance mutations associated with events involving areA, 102 were area + reversions as compared to 23 areA" forward mutations. High revertability is a feature of duplication mutations (Kinsey and Fincham, 1979; Rambosek and Kinsey, 1984; Das oral., 1988). (ii) Whereas 15 of 23 mapped areA" mutations selected in wt or other areA specificity mutant strains are located in the putative DNA-binding region, all 13 selected in areA-300 strains which have been mapped lie upstream from it. This suggests a predominance of frameshift and polypeptide chain-termination alleles. Localisation of a high proportion of loss-of-function mutations in the DNA-binding region is 49~ A
~1 A
P
S
R
P
A~S
P
G
G
T
K
N
G
E
Q
N
G
P
GCTGCTCCATCGCGGCCCGCC~GTCCAGGCGGGACCA~GAACGGAGAGCAGAATGGTCCC 1~,8o ,ILo^1 V I'o'.' 51 D,
~
~
•
T T (C) T N (C) F T 0 T T P L W R R N P E G A C C A u ( : T ' ~ C ~ T ~ C T TTACGC~GAC CACACCGCTGTGGCGGCGTAAC CCTGAAGGT
15~0 53~
~ ~ ' ) G L F L K ~T~CAACGCCTGCGGTTTGTTTTTGAAGTTG
L
H G V V R P .L_ CACGGCGTGGTGCGCCCACTG
1600 5S/~
~ ~ ~ ~ ,, v
z K K a s
R N s A N s '. A v
TCGCTCAAGACCGATGTGATCAAGAAGCGTAACCGCAACAGCGCCAATAGCCTTGCCGTT
characteristic of transcriptional activator genes (e.g., Johnston and Dover, 1988). Functional redundancy of the DNA-binding region due to a dupfication in areA-300 could account for the paucity of mutations in this region. Physical evidence for a dupfication in areA-300 strains was obtained using PCR (Saiki et aL, 1985). Pairs of primers giving a single PCR product with wt DNA gave two products, one abnormally large, using areA-300 DNA (Fig. 1). From the size difference between the larger and smaller PCR products obtained using the same pair of primers and from Southern-blot analysis (not shown) a duplication of 400-450 bp was estimated. Additionally, a pair of divergent primers, giving no product with wt template, gave a product with areA-300 template (Fig. I). PCR products were used directly to sequence both copies of the duplicated region plus flanking regions, indicating a 417-bp 'in-frame' tandem duplication (Fig. 2). The derived 139-aa sequence of the duplicated region encompasses the 52-aa 'finger'-containing sequence of similarity to the DNAbinding regions of the major erythroid transcription factor GF-1/Eryfl/NF-El (Arst etal., 1989; Kudla etal., 1990). The N-terminal copy of the duplication is effectively truncated (by the presence of the C-terminal copy) 80 aa from the C terminus. Truncations of 55-124 aa from the C terminus result in nitrogen metabolite derepression of activities under area control (Arst, 1982; Kudla et al., 1990). Therefore, separation of the N-terminal DNAbinding domain from sequences near the C terminus might be responsible for nitrogen metabolite derepression associated with areA-300.
1660 57~ G S 8 R V S ~ K S A it K N S V 0 O v T P GGCTCTTCACGGGTATCCAAAAAGTCAGCCCGCAAGAACTCGGTTCAGCAGGTAACTCCG 1720 59~
T A P T $ S R A 0 g H T T S r. S P P A M ACTGCGCCTACCTCGAGCCGGGCCCAAAGCAATACGACATCCGAGTCTCCCCCAGCAATG 1780 614 P G S S G R G S G Y V P I A A A P P K S CCGGGTAGCTCCGGCCGAGGCTCTGGAGTCGTTCCTATTGCGGCTGCCCCGCCTAAATCT
18a10 63al
$ S A A T T~S P G T N N G AGCTCAGCGGCGACTAC~CTCCGGG CACAAACAACGGT
1gOO
1917_T
Fi& 2. Nucleotidesequenceof the regionduplicatedin areA-300strains. The completesequenceof areA was reportedby Kudiaetal. (1990)and is in the EMBL,GenBankand DDBJ NucleotideSequenceDatabases under the accessionnumberX52491. Numberingofnt is fromthe start codon. The derivedaa sequenceis shownin the single-lettercode and numberedfromthe initiatorMet. Numberedarrowsindicatethe limits ofthe duplicatedregion.The 52-aaregionof sequencesimilaritywiththe erythroid transcription factor GF-I/Eryfl/NF-EI (Arst etal., 1989; Kudlaetal., 1990)is linedbetweennt and aa sequences,the Cysresidues of the 'finger'are circledand the areA-102sequencechangeis indicated at nt 1576and aa 526. Possiblyrelevantto the originof the duplication is that 6 of the 9 nt at its beginningare identicalto the correspondingnt immediatelyfollowingit.
(b) A double mutant in which the N . t e r m i n a l 'finger' o f ueA-300
contains a mutant residue
Beyond assessment of the phenotype of areA-300, one approach to determining the consequences of the areA-300 duplication is to examine double mutants in which one copy of the duplicate¢i sequence is mutated. The specificity mutation areA-102 (Arst and Cove, 1973; Hynes, 1975; Wiame et al., 1985; Arst and Scazzocchio, 1985; Katz and Hynes, 1989) converts the central leucine residue in the 'finger' loop to valine (Kudla etal., 1990). By using strains carrying suitably positioned area r mutations, a double mutant having the areA-102 mutation in the 5' copy of the duplicated sequence was obtained (Fig. 3), recognised by its unique phenotype (Table I) and confirmed by sequencing of PCR-amplified DNA (Fig. 2). The presence of the areA-102 sequence change in the 5' copy of the duplication was predicted since it could arise by a single crossover (Fig. 3), whereas a triple crossover would be required to construct the corresponding 3' mutated copy recombinant (not shown). The phenotype of the areA-102 areA-300 double mutant (Table I) generally resembles that of either the areA-102 or
126
and regulate expression of any genes not normally under area control.
ACKNOWLEDGEMENTS
t
,,
~w
~,~
J
*,~i,~, i i
,~,,
Fig. 3. Construction of an areA-102 areA-300 double mutant. Exploiting the inability of strains carrying areA" mutations to utilise nitrogen sources other than ammonium and L-Gin, strains of relevant genotype areA-102 areA'-217 and areAr-150 areA-300 were crossed and progeny able to utilise S mM L-Art as nitrogen source were selected. One areA-102 areA-300 recombinant was obtained from 107 Arg-utilisingprogeny presumably as a result of pairing involving the 5' copy of the areA-300 duplication and a crossover (indicated by broken lines) in the 45-bp interval (Kudla etal., 1990) between the areA-102 and areAr-217 mutations. Additional crosses confirmed that the new phenotype (Table 1) segregates as an areA allele and its identity as an areA.102 areA-300 double mutant was confirmed by sequencing. The duplicated region comprising nt 1501 through 1917 is shown in white with first and last nt indicated.
the areA-300 single mutant indicating epistasy (e.g.,/~-Ala and chlorate sensitivities, respectively) or is intermediate indicating additivity (e.g., utiUsation of L-His and uric acid). The notable and totally unexpected exception is 2-pyrrolidone utilisation. Whereas areA-300 and, to a lesser extent, areA-102 both impair utilisation of 2-pyrrolidone, the double mutant utilises it nearly as well as wt. Phenotypic comparison with the other double mutant having the areA. 102 Val in the C.terminal 'finger' and the triple mutant having it in both 'fingers' would be fascinating. (e) Conclusion: are two 'fingers' better than one? The difficulty in selecting null (area r) mutations in the
DNA-binding region in areA-300 strains suggests one advantage to functional redundancy of this region. The mutually suppressive 2-pyrrol~done phenotype of the areA102 areA-300 double mutant hints at another possible advantage: two 'zinc fingers' can diverge and divergent 'fingers' might act synergistically. The selection of a didactylous mutant form ofareA is not without relevance to the evolution of the erythroid transcription factc?r GF-I/Eryfl/NF-EI which contains two (non-identicai) 'fingers' of sequence related to both that of area and each other (Tsai etal., 1989; Evans and Felsenfeld, 1989; Arst et ai., 1989; Trainor et ai., 1990; Zonet al., 1990; Kudla etal., 1990). As the two 'fingers' of areA-300 are separated by 114 an, it seems likely that they act independently and alternatively rather than jointly (for discussion of 'finger' spacing see Router et ai., 1990), a proposal supported by the additivity seen in certain aspects of the areA-102 areA-300 phenotype (Table I). Nevertheless, it is intriguing to consider whether this two-fingered form of area can recognise
We thank Linda Mitchell and Alan Sheerins for technical
assistance and Paul Hooley, Joan Tilburn, Wayne Davies, Brian T o m s e t t and Tim L a n g d o n for comments on the manuscript. This work was supported by the Science and Engineering Research Council (M.X.C. and H . N , A . ) a n d a E u r o p e a n Commission twinning grant (H.N.A.).
REFERENCES AI Taho, N.M., Sealy-Lewis, H.M. and Scazzocchio, C.: Suppressible alleles in a wide domain regulatory gone in AspergiUusaidulans. Curr. Genet. 8 (1984) 245-251. Arst Jr., H.N.: A near terminal pericentric inversion leads to nitrogen metabolite derepression in Aspergiilus nidulans. Mol. Gun. Goner. 188 (1982) 490-493. Arst Jr., H.N.: The areA gene mediating nitrogen metabolite repression in AspergUlus nidulans. In Nevalalnen, H. and Penttil~, M. (Eds.), Molecular Biologyof Filamentous Fungi. Foundation for Biotechnical and Industrial Fermentation Research, Helsinki, 1989, pp. 53-62. Arst Jr., H.N. and Cove, DJ.: Methylammonium resistance inAsperglUus nldulans. J. Bacteriol. 98 (1969) 1284-1293. Arst Jr., H.N. and Cove, D.J.: Nitrogen metnbolite repression in Asperglllus nldulans. Mol. Gun. Genet. 126 (1973) !11-141. Arst Jr., H.N. and Scazzoechio, C.: Formal genetics and molecular biology of the control of gone expression in Asperglllus nldulans. In Bennett, J.W. and Lasure, L.L. (Eds.), Gene Manipulations in Fungi. Academic Press, New York, 1985, pp. 309-343. Arst Jr., H.N., Tollervey, D.W. and Scaly.Lewis0 H.M.: A possible regulatory gone for the molybdenum-containingcofactor in Aspergillus nldulans. J. Gun. Microbiol. i28 (1982) 1083-1093. Arst Jr., H.N., Kudla, B., Martinez-Rossi, N., Caddick, M.X., Sibley, S, and Davies, R.W.: Asperglllus and mouse share a new class of'zinc finger' protein. Trends Goner. 5 (1989) 291. Caddick, M.X., Arst Jr., H.N., Taylor, L.H., Johnson, R.I. and Brownlee, A.G.: Cloning of the regulatory gene area mediating nitrogen metabolite repression in Asper~llus nidulans. EMBO J. 5 (1986) 1087-1090. Cove, DJ.: The induction and repression of nitrate reductase in the fungus Aspergillus nidulans. Binchim. Biophys. Acta I 13 (1966) 51-56. Das, G., Consaul, S. and Sherman, F.: A highly revertible c.rcl mutant of yeast contains a small tandem duplication. Genetics 120 (1988) 57-62. Evans, T. and Felsenfeld, G.: The erythroidospeeifictranscription factor Eryfl: a new finger protein. Cell 58 (1989) 877-885. Fu, Y.-H. and Marzinf, G.A.: nit-2, the major nitrogen regulatory gone of Neurospora crassa, encodes a protein with a putative zinc finger DNA-binding domain. Moi. Cell. Biol. 10 (1990) 1056-1065. Hynes, MJ.: Studies on the role of the area gene in the regulation of nitrogen catabolism in Asperglllus nidulans. AustoJ. Biol. Sci. 28 (1975) 301-313. Johnston, M. and Dover, J.: Mutational analysis of the GAL4-encoded transcriptional activator protein ofSaccharomyces cerevisiae.Genetics 120 (1988) 63-74. Katz, M.E. and Hynes, M.J.: Characterization of the amdR-controlled
127 lamA and lamB genes of Aspergillus m'dulans. Genetics 122 (1989) 331-339. Kinsey, J.A. and Fineham, J.R.S.: An unstable allele of the am locus of Neurospora crassa. Genetics 93 (1979) 5"/7-586. Kudla, B., Caddick, M.X., Langdon, T., Martinez-Russi, N.M., Bennett, C.F., Sibley, S., Davies, R.W. and Arst Jr., H.N.: The regulatory gene area mediating nitrogen metabolite repression in Aspergillus nidulans. Mutations affecting specificity ofgene activation alter a loop residue of a putative zinc finger. EMBO J. 9 (1990) 1355-1364. Rambusek, J.A. and Kinsey, J.A.: An unstable mutant gene of the am locus of Neurospora results from a small dupfication. Gene 2"/(1984) 101-10"/. Reuter, G., Giarre, M., Farah, L, Gausz, J., Spierer, A. and Spierer, E: Dependence of position-effect variegation in Drosophila on dose of a gene encoding an unusual zinc-finger protein. Nature 344 (1990) 219-223. Saiki, R., Sharf, S., Faloona, F., Mullis, K.B., Horn, G.T., Erlich, H.A. and Arnheim, N.: Enzymatic amplification of B-globin 8enomic sequences and restriction site analysis for diagnosis of sickle cell anemi& Science 230 (1985) 1350-1354. Shaffer, P.M. and Arst Jr., H.N.: Regulation of pyrimidine salvage in
Asper&iilus nidulans: a role for the major regulatory gene areA mediating nitrogen metabolite repression. Mol. Gen. Genet. 198 (1984) 139-145. Shaffer, P.M., Arst Jr.,H.N., Estherg, L, Femando, L, Ly, I".and Sitter, M,: An asparaginase of Asperg/bs n/du/ans is subject to oxygen repression in addition to nitrogen metabofite repression. Mol. Gen. Genet. 212 (1988) 337-341. Tralnor, C.D., Evans, T., Felsenfeld, (3. and Boguski, M.S.: Structure and evolution of a human erythroid transcription factor. Nature 343 (1990) 92-96. Tsai, S.-F., Martin, D.I.K., Zon, LL, D'Andrea, A.D., Wong, G.G. and Orkin, S.H.: Cloning of eDNA for the major DNA-binding protein of the erythrnid lineage through expression in mammalian cells. Nature 339 (1989) 446-451. Wiame, J.-M., Greuson, M. and Arst Jr., H.N.: Niuogen catabolite repression in yeasts and filamentous fungi. Adv. Microb. Physiol. 26 (1985) !-88, Zon, L.I., Tsai, S.-F., Burgess, S., Matsudaira, P., Brans, G.A.P. and Orkin, S.H.: The major human erythreid DNA-binding protein (GF-I): primary sequence and localization of the gene to the X chromosome. Proc. Natl. Acad. Sci. USA 87 (1990) 668-672.
NOTE ADDED IN PROOF
The major (possibly only) predicted class of null mutations occurring within the areA-300 duplicated region would be frameshift or chain-termination lesions located in the 5' copy of the region upstreamfrom or encoding the DNA-binding domain. The areAr-153 mutation has now been located as a -1 frameshift at nt 1653 of the 5' copy of the areA-300 duplication. As expected, the areAr-153 and areA-300 mutations co-revert at high frequency.
123
Elsevier GENE 03739
Nitrogen regulation in Aspergillus: are two fingers better than one? (Gene areA; Aspergiilus nidulans; didactylous mutation; duplication; erythroid transcription factor; nitrogen metabolite repression; recombinant DNA; zinc finger)
Mark X. Caddick" and Herbert N. Arst Jr. b ° Department of Genetics and Microbiology, University of Liverpool. Liverpool 1,69 3BX (U.K.) Tel. (+ 44-5 i) 7943632, and b Departmem of Bacteriology. Royal Postgraduate Medical School, London WI2 0NN (U.K.) Received by R.W. Davies: 14 June 1990 Revised: 30 June 1990 Accepted: ! July 1990
SUMMARY
The areA gene, mediating nitrogen metabolite repression in AspergiUus nidulans, encodes a positive-acting regulatory protein with a single putative DNA-binding 'zinc finger' which is remarkably similar to the two 'zinc fingers' of the major regulatory protein of vertebrate erythroid cells (GF-I/Eryfl/NF-EI). The areA-300 mutation alters the specificity of gene activation in that it elevates expression of certain structural genes whilst reducing expression of certain others. It is an 'in-frame' tandem duplication of 417 bp including the entire DNA-binding region. The consequences of areA didactyly are further explored by construction of a double mutant having an altered loop residue in the N-terminal 'finger'.
INTRODUCTION
The positive-acting regulatory gene areA mediates nitrogen metabolite repression in the metabolically versatile fungus, A. nidulans (Arst and Cove, 1973; Wiame et al., 1985; Caddick etal., 1986). Nitrogen metabolite (or 'ammonium') repression, probably acting at the level of transcription (Arst and Scazzocchio, 1985; Wiame et al., 1985; L. Gorfinkiel, G. Diallinas and C. Scazzocchio as cited in Kudla et al., 1990), enables preferential utilisation
Correspondence to: Dr. H.N. Arst, Department of Bacteriology, Royal Postgraduate Medical School, Ducane Road, London WI2 0NN (U.K.) Tel. ( + 44-81)7403243; Fax ( + 44.81)7403169. Abbreviations: aa, amino acid(s); areA, regulatory gene mediating nitrogen metaboUte repression; ~-Ala, ~.alanine; bp, base pair(s); GF-1/Eryfl/NF-EI, major vertebrate erythroid transcription factor; nt, nucleotide(s); oligo,oligodeoxyribonucleotide(s);PCR, polymerase chain reaction; r repressed (null) phenotype; wt, wild type. 0378-1i 19/90/$03.50 © 1990Elsevier SciencePublishers B.V.(BiomedicalDivision)
of the favoured nitrogen sources ammonium and L.glutamine. The derived aa sequence of areA contains a single putative DNA-binding 'zinc finger' with considerable sequence similarity to the two 'zinc fingers' of the major regulatory protein of vertebrate erythroid cells, GF-I/Eryfl/NF-EI (Arst et al., 1989; Trainor et al., 1990; Kudla et al., 1990). Mutant area phenotypes are extraordinarily diverse, including alleles which elevate expression of certain structural genes whilst having a null phenotype for certain others (Arst and Cove, 1973; Hynes, 1975; A! Taho et al., 1984; Shaffer and Arst, 1984; Arst and Scazzocchio, 1985). We have previously described (Kudla et al., 1990) a pair of such specificityalleles with approximatelyreciprocal or 'mirror image' phenotypes, each changing a conserved (Kudla et al., 1990; Fu and Marzluf, 1990)central residue of the 'finger' loop. Here we report that another specificity mutation is an 'in-frame' duplication of the entire DNAbinding region and examine the consequences of areA didactyly.
124 TABLE 1 Partial phenotype of the areA-102 areA-300 double mutant allele with the areA-102 mutation located in the 5' copy of the duplication Growth a on media b containing:
Relevant genotype
L-His ~
2-Pyrrolidone c
p-Ala ©
Uric acid c
Chlorate + ammonium c
areA-102 areA-300 areA-1O2 areA-300
+ + + +
+ +
+ + + -
+ + +
+ + + -
wt(areA +)
-
++
+ + +
++
++
a Growth tests were performed as described by Arst et aL (1982) and scored a~er two to three days' incubation at 37°C. Growth scores range from + + + (maximal growth) > + + > + > + > - (little or no growth) and are not necessarily equivalent on different media. At least 50 different strains of each genotype were tested. u Solid minimal medium (Cove. 1966) containing I ~ (w/v) D-glucose as carbon source was used. ° Nitrogen sources were present at 10 mM [2-pyrrolidone and ammonium (as the (+)-tartrate)], 5 mM (L-His and p-Ala) or 595 #M (uric acid). Chlorate (as the potassium salt) was present at 100 raM. A minus score for L-His. 2-pyrrolidone or uric acid utHisation indicates nitrogen-starved growth (Arst and Scazzocehi:J, 1985), whereas minus scores on ~-Ala and chlorate media reflect even less growth as a consequence of toxicity (Arst and Scazzocchio, 1985). Strains carrying areA-102 ( + areA-300) are inhibited by ~-Ala even when an alternative nitrogen source is present, probably due to excessive uptake (H.N.A., unpublished). Supplement levels ( > 33/zM) of meso-inositol reverse such/~-Ala toxicity, indicating that the target of toxicity is inositoi biosynthesis or a closely related function. Sensitivity to chlorate toxicity in the presence of ammoninm indicates nitrogen metabolite derepression of nitrate reductase synthesis (Arst and Cove, 1969; Arst, 1982; Kudla et ai., 1990).
in nitrogen-metabolite derepression of an asparaginase and nitrate and nitrite reductases but has a null phenotype for expression of activities necessary for utilisation of 2-pyrrolidone, L-methionine and L-leucine as nitrogen sources (ShatTer and Arst, 1984; Shaffer et al., 1988; HN.A. and P.M. Shaffer, unpublished). Additionally and unexpectedly, areA-300 seems to relieve repression of thymine-7hydroxylase and asparaginase syntheses by atmospheric oxygen (Shaffer and Arst, 1984; Shaffer et al., 1988).
RESULTS AND DISCUSSION
(a) The areA.300 mutation is a duplication of the entire putative DNA-bluding region The 4-nitroquinoline. 1-oxide.induced mutation areA-300 was selected for its ability to elevate pyrimidine-salvage pathway activity levels (Shaffer and Arst, 1984), In addition to enhancing uptake of pyrimidine derivatives and strongly derepressing synthesis of thymine-7-hydroxylase, it results
wt A+D
mmA
fx
~
B+E
, i
i
& A-300 ~J-
A+D} B+E } C+D
~
~J
D I
~ ....... i
D
, I
I
I
Fig, I. Amplification of regions within area using PCR. Oligo primers (17-20 nt long) A to E were used to amplify regions ofareA from wt and areA-300 genomic DNA by PCR as described by Kudla et aL (1990), except that annealing temperatures of 55-60 ° C were used. The relative position and orientation of the primers, the combinations used and their respective products are indicated. For areA-300, the primer combinations A + D and B + E produced two distinct products (indicated by continuous and dashed lines) visualised by gel electrophoresis and Southern hybridisation. The ratio ofprodocts varied with PCR conditions, shorter extension times generally favouring the shorter (wt) product. Primers C and D are divergently oriented, with a single nt overlap, in the wt and thus failed to give a PCR product with wt genomic DNA. Consistent with the presence of a duplication in areA-300, PCR with C + D amplified a region of418 bp which was sequenced directly (Kudla et al., 1990), enabling identification of the internal junction of the duplication (indicated by nt numbers, 1917 followed by 1501). Additional PCR products of areA-300 were also sequenced to confirm the fidelity of the entire duplicated DNA. The region duplicated is indicated by open bars; nonduplicated regions are shown as blackened bars.
125 Two striking genetic observations suggested that areA300 might be a duplication of the DNA-binding region. (i) When selection of loss-of-function (areA') mutations was attempted in areA-300 strains (Arst, 1989) by exploiting their derepression of nitrate reductase synthesis and consequent chlorate hypersensitivity (,~rst and Cove, 1969; Table I), areA-300 reverted to areA + at high frequency. Of 125 spontaneous chlorate resistance mutations associated with events involving areA, 102 were area + reversions as compared to 23 areA" forward mutations. High revertability is a feature of duplication mutations (Kinsey and Fincham, 1979; Rambosek and Kinsey, 1984; Das oral., 1988). (ii) Whereas 15 of 23 mapped areA" mutations selected in wt or other areA specificity mutant strains are located in the putative DNA-binding region, all 13 selected in areA-300 strains which have been mapped lie upstream from it. This suggests a predominance of frameshift and polypeptide chain-termination alleles. Localisation of a high proportion of loss-of-function mutations in the DNA-binding region is 49~ A
~1 A
P
S
R
P
A~S
P
G
G
T
K
N
G
E
Q
N
G
P
GCTGCTCCATCGCGGCCCGCC~GTCCAGGCGGGACCA~GAACGGAGAGCAGAATGGTCCC 1~,8o ,ILo^1 V I'o'.' 51 D,
~
~
•
T T (C) T N (C) F T 0 T T P L W R R N P E G A C C A u ( : T ' ~ C ~ T ~ C T TTACGC~GAC CACACCGCTGTGGCGGCGTAAC CCTGAAGGT
15~0 53~
~ ~ ' ) G L F L K ~T~CAACGCCTGCGGTTTGTTTTTGAAGTTG
L
H G V V R P .L_ CACGGCGTGGTGCGCCCACTG
1600 5S/~
~ ~ ~ ~ ,, v
z K K a s
R N s A N s '. A v
TCGCTCAAGACCGATGTGATCAAGAAGCGTAACCGCAACAGCGCCAATAGCCTTGCCGTT
characteristic of transcriptional activator genes (e.g., Johnston and Dover, 1988). Functional redundancy of the DNA-binding region due to a dupfication in areA-300 could account for the paucity of mutations in this region. Physical evidence for a dupfication in areA-300 strains was obtained using PCR (Saiki et aL, 1985). Pairs of primers giving a single PCR product with wt DNA gave two products, one abnormally large, using areA-300 DNA (Fig. 1). From the size difference between the larger and smaller PCR products obtained using the same pair of primers and from Southern-blot analysis (not shown) a duplication of 400-450 bp was estimated. Additionally, a pair of divergent primers, giving no product with wt template, gave a product with areA-300 template (Fig. I). PCR products were used directly to sequence both copies of the duplicated region plus flanking regions, indicating a 417-bp 'in-frame' tandem duplication (Fig. 2). The derived 139-aa sequence of the duplicated region encompasses the 52-aa 'finger'-containing sequence of similarity to the DNAbinding regions of the major erythroid transcription factor GF-1/Eryfl/NF-El (Arst etal., 1989; Kudla etal., 1990). The N-terminal copy of the duplication is effectively truncated (by the presence of the C-terminal copy) 80 aa from the C terminus. Truncations of 55-124 aa from the C terminus result in nitrogen metabolite derepression of activities under area control (Arst, 1982; Kudla et al., 1990). Therefore, separation of the N-terminal DNAbinding domain from sequences near the C terminus might be responsible for nitrogen metabolite derepression associated with areA-300.
1660 57~ G S 8 R V S ~ K S A it K N S V 0 O v T P GGCTCTTCACGGGTATCCAAAAAGTCAGCCCGCAAGAACTCGGTTCAGCAGGTAACTCCG 1720 59~
T A P T $ S R A 0 g H T T S r. S P P A M ACTGCGCCTACCTCGAGCCGGGCCCAAAGCAATACGACATCCGAGTCTCCCCCAGCAATG 1780 614 P G S S G R G S G Y V P I A A A P P K S CCGGGTAGCTCCGGCCGAGGCTCTGGAGTCGTTCCTATTGCGGCTGCCCCGCCTAAATCT
18a10 63al
$ S A A T T~S P G T N N G AGCTCAGCGGCGACTAC~CTCCGGG CACAAACAACGGT
1gOO
1917_T
Fi& 2. Nucleotidesequenceof the regionduplicatedin areA-300strains. The completesequenceof areA was reportedby Kudiaetal. (1990)and is in the EMBL,GenBankand DDBJ NucleotideSequenceDatabases under the accessionnumberX52491. Numberingofnt is fromthe start codon. The derivedaa sequenceis shownin the single-lettercode and numberedfromthe initiatorMet. Numberedarrowsindicatethe limits ofthe duplicatedregion.The 52-aaregionof sequencesimilaritywiththe erythroid transcription factor GF-I/Eryfl/NF-EI (Arst etal., 1989; Kudlaetal., 1990)is linedbetweennt and aa sequences,the Cysresidues of the 'finger'are circledand the areA-102sequencechangeis indicated at nt 1576and aa 526. Possiblyrelevantto the originof the duplication is that 6 of the 9 nt at its beginningare identicalto the correspondingnt immediatelyfollowingit.
(b) A double mutant in which the N . t e r m i n a l 'finger' o f ueA-300
contains a mutant residue
Beyond assessment of the phenotype of areA-300, one approach to determining the consequences of the areA-300 duplication is to examine double mutants in which one copy of the duplicate¢i sequence is mutated. The specificity mutation areA-102 (Arst and Cove, 1973; Hynes, 1975; Wiame et al., 1985; Arst and Scazzocchio, 1985; Katz and Hynes, 1989) converts the central leucine residue in the 'finger' loop to valine (Kudla etal., 1990). By using strains carrying suitably positioned area r mutations, a double mutant having the areA-102 mutation in the 5' copy of the duplicated sequence was obtained (Fig. 3), recognised by its unique phenotype (Table I) and confirmed by sequencing of PCR-amplified DNA (Fig. 2). The presence of the areA-102 sequence change in the 5' copy of the duplication was predicted since it could arise by a single crossover (Fig. 3), whereas a triple crossover would be required to construct the corresponding 3' mutated copy recombinant (not shown). The phenotype of the areA-102 areA-300 double mutant (Table I) generally resembles that of either the areA-102 or
126
and regulate expression of any genes not normally under area control.
ACKNOWLEDGEMENTS
t
,,
~w
~,~
J
*,~i,~, i i
,~,,
Fig. 3. Construction of an areA-102 areA-300 double mutant. Exploiting the inability of strains carrying areA" mutations to utilise nitrogen sources other than ammonium and L-Gin, strains of relevant genotype areA-102 areA'-217 and areAr-150 areA-300 were crossed and progeny able to utilise S mM L-Art as nitrogen source were selected. One areA-102 areA-300 recombinant was obtained from 107 Arg-utilisingprogeny presumably as a result of pairing involving the 5' copy of the areA-300 duplication and a crossover (indicated by broken lines) in the 45-bp interval (Kudla etal., 1990) between the areA-102 and areAr-217 mutations. Additional crosses confirmed that the new phenotype (Table 1) segregates as an areA allele and its identity as an areA.102 areA-300 double mutant was confirmed by sequencing. The duplicated region comprising nt 1501 through 1917 is shown in white with first and last nt indicated.
the areA-300 single mutant indicating epistasy (e.g.,/~-Ala and chlorate sensitivities, respectively) or is intermediate indicating additivity (e.g., utiUsation of L-His and uric acid). The notable and totally unexpected exception is 2-pyrrolidone utilisation. Whereas areA-300 and, to a lesser extent, areA-102 both impair utilisation of 2-pyrrolidone, the double mutant utilises it nearly as well as wt. Phenotypic comparison with the other double mutant having the areA. 102 Val in the C.terminal 'finger' and the triple mutant having it in both 'fingers' would be fascinating. (e) Conclusion: are two 'fingers' better than one? The difficulty in selecting null (area r) mutations in the
DNA-binding region in areA-300 strains suggests one advantage to functional redundancy of this region. The mutually suppressive 2-pyrrol~done phenotype of the areA102 areA-300 double mutant hints at another possible advantage: two 'zinc fingers' can diverge and divergent 'fingers' might act synergistically. The selection of a didactylous mutant form ofareA is not without relevance to the evolution of the erythroid transcription factc?r GF-I/Eryfl/NF-EI which contains two (non-identicai) 'fingers' of sequence related to both that of area and each other (Tsai etal., 1989; Evans and Felsenfeld, 1989; Arst et ai., 1989; Trainor et ai., 1990; Zonet al., 1990; Kudla etal., 1990). As the two 'fingers' of areA-300 are separated by 114 an, it seems likely that they act independently and alternatively rather than jointly (for discussion of 'finger' spacing see Router et ai., 1990), a proposal supported by the additivity seen in certain aspects of the areA-102 areA-300 phenotype (Table I). Nevertheless, it is intriguing to consider whether this two-fingered form of area can recognise
We thank Linda Mitchell and Alan Sheerins for technical
assistance and Paul Hooley, Joan Tilburn, Wayne Davies, Brian T o m s e t t and Tim L a n g d o n for comments on the manuscript. This work was supported by the Science and Engineering Research Council (M.X.C. and H . N , A . ) a n d a E u r o p e a n Commission twinning grant (H.N.A.).
REFERENCES AI Taho, N.M., Sealy-Lewis, H.M. and Scazzocchio, C.: Suppressible alleles in a wide domain regulatory gone in AspergiUusaidulans. Curr. Genet. 8 (1984) 245-251. Arst Jr., H.N.: A near terminal pericentric inversion leads to nitrogen metabolite derepression in Aspergiilus nidulans. Mol. Gun. Goner. 188 (1982) 490-493. Arst Jr., H.N.: The areA gene mediating nitrogen metabolite repression in AspergUlus nidulans. In Nevalalnen, H. and Penttil~, M. (Eds.), Molecular Biologyof Filamentous Fungi. Foundation for Biotechnical and Industrial Fermentation Research, Helsinki, 1989, pp. 53-62. Arst Jr., H.N. and Cove, DJ.: Methylammonium resistance inAsperglUus nldulans. J. Bacteriol. 98 (1969) 1284-1293. Arst Jr., H.N. and Cove, D.J.: Nitrogen metnbolite repression in Asperglllus nldulans. Mol. Gun. Genet. 126 (1973) !11-141. Arst Jr., H.N. and Scazzoechio, C.: Formal genetics and molecular biology of the control of gone expression in Asperglllus nldulans. In Bennett, J.W. and Lasure, L.L. (Eds.), Gene Manipulations in Fungi. Academic Press, New York, 1985, pp. 309-343. Arst Jr., H.N., Tollervey, D.W. and Scaly.Lewis0 H.M.: A possible regulatory gone for the molybdenum-containingcofactor in Aspergillus nldulans. J. Gun. Microbiol. i28 (1982) 1083-1093. Arst Jr., H.N., Kudla, B., Martinez-Rossi, N., Caddick, M.X., Sibley, S, and Davies, R.W.: Asperglllus and mouse share a new class of'zinc finger' protein. Trends Goner. 5 (1989) 291. Caddick, M.X., Arst Jr., H.N., Taylor, L.H., Johnson, R.I. and Brownlee, A.G.: Cloning of the regulatory gene area mediating nitrogen metabolite repression in Asper~llus nidulans. EMBO J. 5 (1986) 1087-1090. Cove, DJ.: The induction and repression of nitrate reductase in the fungus Aspergillus nidulans. Binchim. Biophys. Acta I 13 (1966) 51-56. Das, G., Consaul, S. and Sherman, F.: A highly revertible c.rcl mutant of yeast contains a small tandem duplication. Genetics 120 (1988) 57-62. Evans, T. and Felsenfeld, G.: The erythroidospeeifictranscription factor Eryfl: a new finger protein. Cell 58 (1989) 877-885. Fu, Y.-H. and Marzinf, G.A.: nit-2, the major nitrogen regulatory gone of Neurospora crassa, encodes a protein with a putative zinc finger DNA-binding domain. Moi. Cell. Biol. 10 (1990) 1056-1065. Hynes, MJ.: Studies on the role of the area gene in the regulation of nitrogen catabolism in Asperglllus nidulans. AustoJ. Biol. Sci. 28 (1975) 301-313. Johnston, M. and Dover, J.: Mutational analysis of the GAL4-encoded transcriptional activator protein ofSaccharomyces cerevisiae.Genetics 120 (1988) 63-74. Katz, M.E. and Hynes, M.J.: Characterization of the amdR-controlled
127 lamA and lamB genes of Aspergillus m'dulans. Genetics 122 (1989) 331-339. Kinsey, J.A. and Fineham, J.R.S.: An unstable allele of the am locus of Neurospora crassa. Genetics 93 (1979) 5"/7-586. Kudla, B., Caddick, M.X., Langdon, T., Martinez-Russi, N.M., Bennett, C.F., Sibley, S., Davies, R.W. and Arst Jr., H.N.: The regulatory gene area mediating nitrogen metabolite repression in Aspergillus nidulans. Mutations affecting specificity ofgene activation alter a loop residue of a putative zinc finger. EMBO J. 9 (1990) 1355-1364. Rambusek, J.A. and Kinsey, J.A.: An unstable mutant gene of the am locus of Neurospora results from a small dupfication. Gene 2"/(1984) 101-10"/. Reuter, G., Giarre, M., Farah, L, Gausz, J., Spierer, A. and Spierer, E: Dependence of position-effect variegation in Drosophila on dose of a gene encoding an unusual zinc-finger protein. Nature 344 (1990) 219-223. Saiki, R., Sharf, S., Faloona, F., Mullis, K.B., Horn, G.T., Erlich, H.A. and Arnheim, N.: Enzymatic amplification of B-globin 8enomic sequences and restriction site analysis for diagnosis of sickle cell anemi& Science 230 (1985) 1350-1354. Shaffer, P.M. and Arst Jr., H.N.: Regulation of pyrimidine salvage in
Asper&iilus nidulans: a role for the major regulatory gene areA mediating nitrogen metabolite repression. Mol. Gen. Genet. 198 (1984) 139-145. Shaffer, P.M., Arst Jr.,H.N., Estherg, L, Femando, L, Ly, I".and Sitter, M,: An asparaginase of Asperg/bs n/du/ans is subject to oxygen repression in addition to nitrogen metabofite repression. Mol. Gen. Genet. 212 (1988) 337-341. Tralnor, C.D., Evans, T., Felsenfeld, (3. and Boguski, M.S.: Structure and evolution of a human erythroid transcription factor. Nature 343 (1990) 92-96. Tsai, S.-F., Martin, D.I.K., Zon, LL, D'Andrea, A.D., Wong, G.G. and Orkin, S.H.: Cloning of eDNA for the major DNA-binding protein of the erythrnid lineage through expression in mammalian cells. Nature 339 (1989) 446-451. Wiame, J.-M., Greuson, M. and Arst Jr., H.N.: Niuogen catabolite repression in yeasts and filamentous fungi. Adv. Microb. Physiol. 26 (1985) !-88, Zon, L.I., Tsai, S.-F., Burgess, S., Matsudaira, P., Brans, G.A.P. and Orkin, S.H.: The major human erythreid DNA-binding protein (GF-I): primary sequence and localization of the gene to the X chromosome. Proc. Natl. Acad. Sci. USA 87 (1990) 668-672.
NOTE ADDED IN PROOF
The major (possibly only) predicted class of null mutations occurring within the areA-300 duplicated region would be frameshift or chain-termination lesions located in the 5' copy of the region upstreamfrom or encoding the DNA-binding domain. The areAr-153 mutation has now been located as a -1 frameshift at nt 1653 of the 5' copy of the areA-300 duplication. As expected, the areAr-153 and areA-300 mutations co-revert at high frequency.
