79
Gene, 42 ( 1986) 79-88 Elsevier GENE
1548
A cloned tryptophan-synthesis gene from the Ascomycete Escherichia coli, yeast and Aspergillus nidulans (Recombinant
DNA;
bacteriophage
plant
pathogen;
fungal
genetics;
C~e~Zi~~~las~eterostpuphus functions in
heterologous
complementation;
shuttle
vector;
IL)
B. Gillian Turgeon, W. Donald MacRae,
Robert C. Garber, G.R. Fink * and O.C. Yoder **
Department qf Plant Pathology, Cornell Univ., Ithaca, NY 14853, Tel. (607)255-3243: and * Whitehead Institute. 9 Cambridge Center, Cambridge, h&i 02142 [U.S.A.) Tel. j617)2.%-5215 (Received
October 23rd.
(Revision
received
1985)
and accepted
December
18th, 1985)
SUMMARY
A gene (TRPI ) in the tryptophan biosynthetic pathway of the fungal plant pathogen Cochliobolus heterostrophus was isolated by complementation of an Escherichia coli trpF mutant which lacked phosphoribosylanthranilate isomerase (PRAI) activity. The cloned gene also complemented an E. co& trpC mutant lacking indoleglycerolphosphate synthase (IGPS) activity, a yeast trpl mutant missing PRAI activity and an Aspergillus nidulans trpC mutant. It functioned in E. coli and A. niduluns without apparent rearrangement but in yeast only after the 5’ end of the gene was deleted. The gene was subcloned on a 4.65kb DNA fragment and the PRAI domain was localized to a 2.9-kb region. It showed homology to the A. nidulans trpC and Neurospora crassa #p-l genes. There was one predominant transcript of C. heterostrophus TRPI, the same size (2.6kb) as one of the two functional transcripts produced by A. niduIans trpC. The constitutive activity of the C. heterostrophus TRPl gene was high whereas that of the A. nidulans trpC gene was low.
iNTRODUCTiON
The tryptophan pathways in E. coli, yeast, and N. crussa are well understood (Schechtman and Yanofsky, 1983). There are five steps from choris.____ ** To
whom
correspondence
and reprint
requests
should
be
addressed. Abbreviations: serum
Ap, ampicillin;
albumin;
GAT.
illdoleglycerolphosphate
EDTA;
0378-I
isomerase;
sulfate;
0.015 M Na,
Sm,
BSA, bovine
amidotransferase;
synthetase;
phoribosylanthraniiat~ dodecyl
bp, base pair(s);
glutamine
IGPS,
kb, 1000 bp; PRAI, R. resistance;
streptomycin;
SSC,
SDS,
phossodium
0.15 M NaCi,
‘citrate, pH 7-8; TE, IO mM Tris (pH 7.5) 10 mM
[ 1,designates
I lY~86:$0?.50
0
plasmid-carrier
1986 Elsevier
state.
Science
Publishers
B.V.
mate to t~ptophan, catalyzed by seven enzymatic activities; the steps are the same in bacteria, yeast, and lilamentous fungi (Schechtman and Yanofsky, 1983). The seven enzymatic functions of E. cofi are encoded in a single operon and designated tlpA through trpG. In the filamentous fungi N. crassu and A. nidulu~s four unlinked genes encode four polypeptides, of which two are monofunctional, one is bifunctional, and one is trifunctional. The activities of the trifunctional gene (tip-I in N. crassa, trpC in A. nidulans) correspond to the trpG, trpC, and trpF functions of E. co& which encode CAT, IGPS, and PRAI, respectively (Schechtman and Yanofsky, 1983; Yelton et al., 1983). We isolated a gene encoding PRAI from the fila-
~Biomedi~~ Division)
mentous
maize
pathogen
gene is potentially construction
This
(b) Isolation and manipulation
of DNA
useful as a selectable marker in the
0ftr~sforInation
transformation study
C. heterostrophus.
technology
the molecular
vectors. We are using (Turgeon
were prepared
lysis (Birnboim
from E. cob’ by either the and
Daly,
boiling
(Holmes
C. heterostrophus. The strategy employed to clone the
nomic
DNAs
C. heterostrophus PRAI gene, complementation
A, niduluns and yeast were isolated
corresponding
has been
of pathogenicity
Plasmids alkaline
in
E. colt’ trpF mutant,
genetics
et al., 1985) to
used
gene from yeast (Struhl
N. crassa (Keesey and Yanofsky,
and DeMoss,
to clone
of an the
et al., 1979)
1982; Schechtman
1983) and A. nidufans (Yelton
et al.,
1983).
described
and Quigley, from
previously
1979) or the
1981) method.
C. heterostrophus, (Garber
Yelton et al., 1984; Sherman
Ge-
N. crassu,
and purified
and
Yoder,
as
1983;
et al., 1981). Restric-
tion enzyme digestions, hybridization analyses, and nick translations were performed according to standard
procedures
(Maniatis
et al.,
1982).
Filter
hybridizations were in 50 mM Tris buffer pH 7.5. containing 1 M NaCl. 1 mM EDTA, 10 x Denhardt’s and 10 iig salmon sperm DNA/ml at 65 -C; for lower stringency the temperature was SO’C. MATERIALS
AND METHODS
(c) Construction
of C. herewstroprkus genomic lib-
(a) Strains and media
raries
E. coli K-12 strain DHl [recA 1, e&A 1, gyrA96, thi-1, hsdR 17 (r, - , mk ), supE44] was used for
Genomic DNA was partially digested with Suu3A, size fractionated, and ligated into the BavrzHI site of either the yeast/E. cofi shuttle vector YEp24 (Botstein et al., 1979) or the i replacement vector EMBL4 (Frischauf et al., 1983). The plasmid library contained approx. 10J clones, 80”/, of which had inserts averaging 11 kb; the probability of finding any given sequence in the library was 0.98
library construction. Strain JA300 (thr, leuB6, thi, thyA, trpClI17, hsd~, hsdR) was used to isolate the C. heterostrophus TRPl gene and, atong with strain HBlOl [hsdS20(r,-,m,-), recA13,ara-14, pruA2, lacY1, galK2, rpsL20, Sm R, ~$5, mtl- 1, supE44] to propagate plasmids. Mutants of strain W3 1IO, each carrying a different mutation in the trp operon (Table I) were from Dr. C. Yanofsky. Since strain W3110 is hsdR +, all plasmids used in attempts to complement the various trp - mutations were first methylated by passage through strain DB6656 (pyrF: : Mu, rrp,,,,, IacZ, ,,,, hsdR _ hsdM * ). S. cere~~~iue strain F762 (trpl-A 1, ura3-52)
(Clarke and Carbon, 1976). The i, library consisted of 4 x 10’ clones with inserts of about 1%kb. (d) Transformation E. colt’ strains
and
A. nidulans strain UCDl (argB2, yA2, puhuA 1, metG, trpC801, biA1) were used to study the expression of the C. heterostrophus TRPl gene in heterologous fungal hosts. C. heterostrophus strain C3 (MA T-2. toxl ; ATCC 48330), described earlier (Leach et al., 1982) was the source of DNA and RNA used for all studies reported here. All microbial strains were stored at -70°C in glycerol (50’,!, for bacteria, 25 y0 for fungi) and recovered fresh for each experiment. Standard complete and minimal media for E. coli (Maniatis et al., 1982), yeast (Sherman et al., 1981), C. heterostrophus (Leach et al., 1982) and A. nidulans (Kafer, 1977; Barratt et al., 1965) were used where appropriate.
were transformed
by either
the
CaCl, (M~iatis et al., 1982) or the hexamine cobalt chloride (Hanahan, 1983) method, yeast by the lithium acetate protocol (Ito et al., 1983), and A. niduluns by the procedure of Yelton et al. (1984).
RESULTS
(a) Isolation of the C. heterostrophus from the plasmid library
TRPl
gene
Competent ceils of E. coii JA300 were transformed with 0.5 /lg C. heterostrophus plasmid library in YEp24 and plated on supplemented M9 medium
81
without
tryptophan.
After 2 days at 37 “C, 25 col-
onies were visible; 24 of these were Ap resistant. of the 24 grew vigorously lacking
tryptophan,
which transformed
Six
when returned
to medium
and five contained
a plasmid
strain
JA300
to Trp’
at high
frequency. One plasmid, designated pChTRP24, was labeled with [r-‘2P]dGTP and used to probe C. heterostrophus
genomic
DNA
and
ize
to
C. heterostrophus
lane a4) or to those fragments
sites in pChTRP24 rostrophus
of pChTRP24
gene
which (Fig. 1, enzyme
carried
by pChTRP24
is called
TRPl.
pChTRP24,
restriction
enzymes.
from the 3, library
hybridizing
genomic
at least
co-migrated
one
of the C. heterostrophus
TRPI
gene
with a
fragment from the cloned DNA (e.g. Fig. 1, lanes b3 and b4), identifying the isolated sequence as C. heterostrophus DNA. [ 32P]YEp24 did not hybrid-
To determine
if selection
for Trp’
in E. coli
caused rearrangement of the C. heterostrophus TRPl gene, approx. 7000 recombinant clones from a C. heterostrophus genomic library in the 1. vector EMBL4 were probed with pChTRP24. Seven plaques hybridized strongly and three of these were analyzed further. All three overlapped one end of the pChTRP24 insert and one overlapped and extended beyond the other end. Comparison of fragments from restriction digests of the EMBL4 clones with
2341234
1
(Fig. 1,
is shown in Fig. 2. The C. hete-
(b) Isolation
In every digest
DNA
contained only C. heterostrophus DNA lane a3 vs. lane b3). A map of restriction
both of which had been digested with each of several fragment
genomic
9.4.
6.6
0.6. b
a Fig. 1. Autoradiogram to fragments
presenting
of genomic
hybridization
C. heterosfrophus
sample was digested
with HindIlI,
in an 0.6”,, agarose
gel in Tris-acetate
tate,
0.002 M EDTA),
(Schleicher (a)YEp24
& Schuell, or
the fragments to
BA 85) and
probed
nitrocellulose
Lane 1, yeast
with
C. heterostrophus genomic
DNA
fragments (compare faint bands fragments. digestion.
of pChTRP24
DNA acepaper
“P-labeled
genomic
(2 icg): lane 2, YEp24 (50 ng); lane 3, pChTRP24 to C. hererostrophus genomic
Each
were separated
buffer (0.04 M Tris
transferred
(b)pChTRP24.
of pChTRP24
DNA.
DNA
(50 ng); lane 4,
( 10 pg). pChTRP24
hybridized
DNA (lane b4) and to restriction
which
did not hybridize
lanes a3, b3). The film was overexposed in lane b4, which represent
to YEp24 to show the
the vector-insert
border
Faint bands in other lanes are the result of incomplete Fragment
sizes (in kb) are specified on the left margin.
Fig. 2. Map ofpChTRP24. 8.2-kb
C. heferostrophus
portion tion;
Dashed insert.
of the yeast 2~ plasmid
containing
gene; 2~ is a
its origin of replica-
amp is the ApR gene from pBR322. TRPl lies between 3.5
and 8.2 kb. Clockwise
numbers
kb coordinates.
The shaded
cates a deletion
that occurred
yeast,
line is YEp24; solid line is the
URA3 is a yeast
within
of the transcript
was reverted
(Fig. 4) and a Trp’
the direction
of transcription;
have not been precisely
the
6.5 and 9.7 kb indi-
when pChTRP24
giving rise to pChTRP24d2
type. The arrow indicates
the circle represent
area between
mapped.
in
phenothe ends
82
AChTRPl-1
A A
I
A
.I.
AChTRPl-2
b
I
A
.I.
J
lI*
.
A
*I.
.
AChTRPl-3
Fig. 3. Comparison 1 library
I
A A
pChTRP24 insert
of the restriction
with [32P]pChTRP24. in the portions
map of the pChTRP24
extends
end. Vertical bars areEcoR1
of the 1 clones that overlapped
I
I
I
I
I
I
I
insert with those of three clones recovered
One of the clones (IChTRPl-1)
/z clones extend beyond the right-hand sites mapped
.
beyond
the left-hand
sites, dots are Sal1 sites, and triangles
pChTRP24
aligned
areHind
(c) Complementation
(d) Deletion mapping of pChTRP24
To determine how many E. coli trp functions could be complemented by the C. heterostrophus clone, trp - cells were transformed with pChTRP24 and selected for Ap resistance. Each plate oftransformed colonies was then printed to minimal medium to test
The recombinant
(Fig. 2) was
I of E. coli trp _ mutants
Complementation
Cells of each strain were transformed minimal
plasmid pChTRP24
digested with MndIII, EcoRI, BumHI, Sal1 or XhoI, and fragments from each digestion were self-ligated and used to transform E. coli strain JA300 to Ap resistance. The plasmids recovered were analyzed by restriction enzyme digestion and fractionation on agarose gels. Those with deletions in the
for tryptophan-independent growth. The C. heterostrophus sequence complemented E. coli trpF-
TABLE
all three
not trpD or trpE to test directly for because that enzyme E. coli (Schechtman
functional complementation of E. cofi had not undergone major rearrangement. These results are diagrammed in Fig. 3. of E. coli trp - mutants
insert;
sites. All restriction
with sites in pChTRP24.
(PRAI) and trpC (IGPS) but (Table I). It was not possible trpG + (CAT) activity in E. cofi is not essential for growth of and Yanofsky, 1983).
fragments of pChTRP24 digested with the same enzymes indicated that the sequence isolated by
a C. heterosrrophu.v
by probing
end of the pChTRP24
medium
with or without
Strain
by pChTRP24
with pChTRP24
and plated
on L agar containing
Ap. After one day, colonies
were printed
to
tryptophan. Relevant
Growth
on minimal
medium:
genotype With tryptophan
Without
JA300 rrpC1117
rrpF
+
+
W3110 drrpClO-16
trpC _, trpF _
+
+
W3 110 wpC782
trpC
+
W3 110 AtrpES
trpE
+
+ _
w3110 AtrpLD102
rrpE -. , trpD _ a
+
I’ Lack of trpD activity
was determined
indole but not on medium
supplemented
by showing
that W3 IlOAfrpLD 102[pChTRP24]
with anthranilate.
tryptophan
grew on minimal
medium
supplemented
with
83
t rpC-F+
t rpC+F-
t rpC-F‘
8 pChTRP24B
. . . . . . . . . . .‘B
pChTRP24H
.............
pChTRP24E
. . . . . . . . . . . . . . . . . E
pChTRP24X
B
x
m
.
H ............
.”
E . . . . . . . . . . . .
x . . . . . . . . . . . . . . -
B
s s . . . . . . . . . . .
B
pChTRP24S
B
B pChTRP24A2
’
. . . . . . . . . .
pChTRP24A2B.
Fig. 4. Deletion by digesting digestion represent
.’ -
.........
mapping
the plasmid
products, deletions.
tryptophan-independent
of the Trp functions with BarnHI
and recircularizing To assess
trpC _ F‘ , trpC_ Fm are mutants
C. heterostrophus DNA function in various tip
in pChTRP24.
Deletions
were introduced
either by reversion
(B), Hind111 (H), EcoRI (E), XhoI (X), or Sal1 (S), recovering it by ligation. Thick lines are C. heterosmphus sequences,
Trp function,
growth.
B
..........
+ indicates
E. coli cells were transformed normal growth;
- indicates
of E. coli (see Table I). kb markers
insert were tested for Trp mutants of E. coli.
The region of pChTRP24 necessary for trpC+ (IGPS) and trpF+ (PRAI) functions in E. cofi was localized between 3.6 and 8.3 kb on the map (Figs. 2 and 4). Both functions were retained when the remainder of the C. heterostrophus sequence, from 0 to 3.6 kb, was deleted (Fig. 4; pChTRP24B). If the C. heterostrophus sequence between 3.6 and 5.5 kb was deleted (Fig. 4, pChTRP24 H, E, and X), both IGPS and PRAI functions were lost. If all or part of the sequence between 5.3 and 8.3 kb was deleted (Fig. 4, pChTRP24 S, 42, and A2B), IGPS function was lost but PRAI function was retained. These observations determine the relative locations of the IGPS and PRAI domains on the clone and indicate that the function of the IGPS domain depends on an intact PRAI domain, but not vice versa.
with each
no growth;
are included
plasmid
of pChTRP24
the largest
fragment
in yeast or among
the
thin lines are YEp24, and dotted lines and transformants
42, plasmid isolated by reversion
were tested
for
in yeast. trpC + F-,
for pChTRP24.
(e) Homology to N. crassa and A. nidulans trp genes To test for homology
among
C. heterostrophus
TRPl, A. nidulans trpC and N. crassa trp-1, plasmids
containing the cloned genes from each organism were digested with enzymes which cut all or part of the insert out of the vector. The digestions were fractionated in agarose gels, Southern-blotted, and with 32P-labeled plasmids probed at 50°C pChTRP24,
pNC2 (N. crassa), or pHY201
(A. nidu-
lam).
Probing PvuII-digested pChTRP24 with pNC2, which carries N. crassa trp-I or with pHY201, which carries A. nidulans trpC, indicated that the three genes are homologous. For example, pNC2 and pHY201 hybridized to the 3.8-kb PvuII fragment of pChTRP24 (Fig. 5, arrowhead). This fragment contains only C. heterostrophus DNA and includes
84
123456
pNC2, and selected recovered
were
rate of one/l0
kig pChTRP24
6//lg of pNC2.
wild-
growth rates and morphologies
com-
plete from pChTRP24
genomes
which
had
received
and probed with
with [“P]pBR322 firmed
transformants ofA.
con-
that (not
their This homology that the
C. heterostrophus
TRPI
gene is functionally There was no homology
between DNA. (g) Expression
Fig. 5. Autoradiograms
representing
hybridization
of C. het~~
trophus TKPI DNA to A. niduluns rrpC and N. UXSSNrrp-I DNAs. details. Lane 1. i, DNA digested
See Fig. 1 for experimental
HindIII; digested digested and
lane 2, pNC2 with
EcoRI + XbaI;
with PvuII. Lanes
/1 DNA
[72P]pHY201; indicates
digested
digested
lane 6 probed
the 32%kb PvuII
C. heterosirophus
domains
and hybridizes
Psrl;
lane4,
with
pHY2OI
lanes 3, 5. and 6. pChTRP24
1, 2 and 3 probed
with
entirely
with
HirldIII;
with [‘*P]pNC?
lanes4,
5 probed
with [“P]pChTRP24. fragment
DNA,
of pChTRP24
spans
the IGPS
to probes containing
hith
Arrowhead which and
is
PRAI
N. UO.WI II~)-I and
A. niduluns trpC. Other bands in lanes 3 and 5 which hybridized to pChTRP24
contained
both vector
and insert sequences.
the IGPS and PRAI domains (Fig. 4). The heterologous hybridizations were less intense than the homologous hybridization, suggesting that there may be substantial sequence differences between C. heterostrophus TRPI and the other two genes. (f) Expression
of
C. heterostrophus
TRPI
A. nidulans Protoplasts of A. nidzduns transformed with
UCDl
in
TRPI in yeast
When of yeast F762 transformed with ‘, colonies arose at the of about 2OOO/LlgDNA. ’ colonies were Trp on medium lacking tryptophan. To obtain Trp’ colonies, the Ura+ Trp transformants were grown for 2 days on minimal agar medium plus tryptophan, then printed to the same medium without tryptophan. After 4 days about 50 colonies appeared, several of which were purified by streaking on minimal medium. Mitotic co-segregation of the yeast URA3 and C. heterostrophus TRPl genes was demonstrated by growing the cells to stationary phase in complete liquid medium, then plating on complete agar medium followed by printing to minimal medium supplemented with either tryptophan or uracil. About 10”; of the cells lost both URA3 and TRPI activities while the rest retained both genes, demonstrating that the TRPl activity was plasmid-encoded. Cells of E. coli JA300 (trpF ) were transformed with plasmid DNA isolated from twelve yeast revertants, selected for Ap resistance, and printed to supplemented M9 medium lacking tryptophan. All transformants grew vigorously, indicating that each plasmid retained PRAI activity. Restriction enzyme digests of plasmid DNA isolated from one E. co/i transformant (chosen at random) from each of the twelve E. coli transformations showed that two of the twelve clones tested had deletions of about 3 kb (Fig. 6). These two apparently identical clones, designated pChTRP24d 1 and pChTRP24d2, trans-
85
the construction
which had the single Hind111 on the
PvuII
fragment
nearest
tation
1) produced
the SP6 promoter
a transcript
(orien-
which hybridized
to
poly(A)+ RNA; the transcript from the opposite construction (orientation 2) did not hybridize. When C. heterostrophus poly(A)’ runoff transcript band
and
several
smaller
(Fig. 7, lanes 3,4).
of pChTRP24
that function
from twelve Trp + yeast colonies to transform
E.
coli
with EcoRI,
stained
with
details.
Lane 1, pChTRP24;
from
different
HirrdIII.
found to be Trp(pChTRP24dl about
fractionated bromide.
E. co/i clones;
All plasmids
pChTRP24
to Ap resistance.
digested
ethidium
carrying
Plasmid
were used were
gel, and
See Fig. 1 for experimental
lanes 2-13,
plasmids
lane 14. i. DNA except
DNAs
DNAs
on a 0.6’,,, agarose
were identical
in yeast,
in yeast.
recovered
digested
to pChTRP24 for those
and 42), each of which
in lanes 3 and IO
sustained
3 kb and was Trp + when transformed
with
and were
a deletion
of
back into yeast.
formed cells of yeast strain F762 to Trp’ at high frequency whereas the other ten (which were indistinguishable from pChTRP24) did not. Thus, although the native C. heterostrophus TRPl gene did not function in yeast, occasional mutations of the recombinant plasmid permitted expression in yeast. Failure to recover Trp + plasmids from some of the Trp + yeast colonies probably reflects differential segregation of pChTRP24 and its rearranged derivatives either in yeast or in E. coli cells. To determine if the deletion caused loss of trpC activity, cells of E. coli strains JA300 (trpF ), W3110trpC782 (trpC_), and W3110dtrpClO-16 (trpC trpF -) were transformed to Ap resistance with pChTRP24d2 and printed to supplemented medium without tryptophan. Transformed JA300 cells were Trp + , confirming trpF activity, whereas cells of trpC782 and AtrpClO-16 remained Trpeven though they were Ap resistant, indicating that pChTRP24d2 had lost trpC activity. (h) Transcriptional
analysis
The PvuII fragment between 3.5 and 7.75kb on the map of C. heterostrophus TRPl (Fig. 2) was inserted in both orientations into vector pSP64. Only
The
1 a single strong
weak bands strongest
slightly more intense
in RNA isolated medium
on
minimal indicating
substantial
hybridized
band
grown medium, Fig. 6. Revertants
RNA was probed with a
from orientation
than
was only
from fungus on
complete
constitutive
tran-
scription of TRPl. When A. nidulans RNA from cells grown on minimal medium was probed with a transcript derived from the A. nidulans trpC gene, a weak band was seen that corresponded in size to the C. heterostrophus main band, along with a stronger band about 200-bp shorter and several other small minor bands (Fig. 7, lane 2). No hybridizing bands were seen in RNA isolated from A. nidufans cells grown in complete medium, confirming that transcription of the trpC gene is very low when this fungus grows in a rich medium (Yelton et al., 1983). Sequence analysis of the A. nidulans trpC gene has shown that the two largest transcripts are authentic mRNAs which are translated; their sizes are 2.6 and 2.4-kb (Mullaney et al., 1985). The most abundant C. heterostrophus transcript must therefore be 2.6-kb since it comigrated with the largest A. nidulans transcript (Fig. 7). Markers included in the gel, however, sized the largest transcript at approx. 2.4-kb; this apparent discrepancy probably reflects the variability sometimes script lengths 1983). (i) Organization
encountered
(Mullaney
in determining
tran-
et al., 1985; Yelton
et al.,
of TRPZ
Because the runoff transcript from orientation 1 and not orientation 2 of the pSP64 constructions containing the C. heterostrophus TRPl gene hybridized to poly(A)+ RNA, we were able to determine the direction of transcription. The transcript which hybridized to C. heterostrophus poly(A) + RNA was by definition produced from the DNA coding strand. The transcript which failed to hybridize to C. heterostrophus poly(A) + RNA had the same sense as the native mRNA. We therefore concluded that the direction of transcription of the C. heterostrophus
86
buffer
pH 7.0. Electrophoresis
10 mM Na’ phosphate
was
buffer,
from the gels to nitrocellulose
paper
identified using RNA probes generated (Riboprobe,
Promega
C. heterostrophus TRPI,
219-
between
agarose
gels in
was transferred
by blotting
in 20 x SSC.
of C. heterostrophus TRPI and A. nidulans trpC were
Transcripts system
in 1.2%
pH 7.0. RNA
the
with an SP6 transcription
Biotec.
Madison,
PvuII fragment
WI).
For
of pChTRP24
3.5 and 7.75 kb on the map (Fig. 2), was electroeluted
from a gel and ligated
into the HincII site of the Riboprobe
plasmid
For A. nidulans trpC, a 3.3-kb BarnHI-
vector
SmaI fragment
pSP64.
of pKBY2 that spanned
from a gel and ligated pSP64 and pSP65. the protocol 1.25-2.5 Na
RNA probes
provided
RNA.
pH 6.5,50”,
Filters
0.05””
Ficoll,
25Opg
salmon
sperm
RNA/ml.
Filters
Na phosphate
were washed
sites of
according
to
Biotec. Specific activities were were probed
formamide,
TA, 0.05 “, BSA, denatured
restriction
were synthesized
by Promega
x 10s cpm/‘pg
phosphate
the trpC gene was eluted
into the appropriate
0.05”,
polyvinylpyrrolidone,
DNA/ml 3-5 times
pH 6.5, 50 mM NaCl,
in 50 mM
0.8 M NaCl, 1 mM EDand
5OOpg yeast
at 75°C
in 20 mM
1 mM EDTA
and 0.15,
SDS. of C. heterostrophus TRPI and A. nidulans
Fig. I. Transcripts trpC. Poly(A)‘RNA
from each fungus
glyoxal gel (McMaster cellulose
paper,
and Carmichael,
and probed
1977) blotted
with 32P-labeled
1 and 2, A. nidulans RNA
Lanes
was fractionated
on a to nitro-
(25 pg/lane)
probed
with
a
from A. nidulans trpC; lanes 3 and 4, C. heterostrophus
transcript
RNA (25 pg/lane)
probed
from C. heterostro-
with a transcript
phus TRPl. RNA in lanes 1 and 3 was from fungi grown in complete medium; indicate
lanes 2 and 4 from minimal
sizes (in kb) of cucumber
RNA was isolated
following
the method
(1969) with some modifications. was seeded
with conidia
with shaking was ground
(2
(250 rev./mm) in a mortar
One volume
of TE-saturated
for 48 h at 20°C. liquid
homogenized
in a Waring
The aqueous
then extracted
twice with chloroform
Nucleic acid was precipitated with 2.5 vol. of ethanol, RNA from DNA,
dissolved
14” triisopropyl
pellet,
with glass
alcohol
(24
of 0.15 M Na and
: 1). ace-
suspended
in
et al., 1982). To to a final concen-
held at 4’C contained
after
phenol - cresol,
(Maniatis
LiCl was added which
: 1) was added
blender
- isoamyl
centrifuged, H,O
of 2 M, then the solution The
50 mM EGTA,
with
in the presence
diethyl-pyrocarbonate-treated
centrifuged.
to
layer was removed
x g). reextracted
overnight
total
in 20 mM Tris buffer pH 7.5 containing
RNA,
and was
500 mM NaCl,
1mM EDTA and 0.1% SDS. Poly(A)‘RNA
was
from total RNA by passage
column (type 3,
Collaborative mendations. denatured presence
Research) For
through following
separation
for 1 h at 50°C
an oligo(dT)
with
of 50”,, dimethylsulfoxide
separated
the manufacturer’s
in gels,
RNA
I5?0 deionized
(j) Conclusions
grown
The mycelium
phenol - cresol (9
(17000
separate
medium
acid, at the rate of 1 g fresh mycelium/ml.
for 60 s at 4°C.
tration
Fungal
and Leaver
N, and transferred
acid, and
centrifugation
tate,
Markers
or complete
pH 8.5 containing
sulfonic
and the mixture
of Lovett
Minimal
6:~ p-aminosalycilic
naphthalene
beads
medium. virus RNAs.
104/ml) and the culture
x
under
cold 200 mM Tris buffer 250 mM NaCl,
mosaic
gene is the same as that of the pSP64 C. heterostroTRPl construction (orientation 2) which gave rise to a non-hybridizing transcript (see preceding section and Fig. 2). The order of functional domains on the C. heterostrophus sequence is thus 5’-IGPSPRAI-3 ’ and the direction of transcription (arrow in Fig. 2) is counterclockwise.
phus
runoff transcripts.
recoin
samples
were
glyoxal
in the
and 10 mM Na
phosphate
The C. heterostrophus TRPI gene, isolated by complementation of an E. coli trpF mutant, is functional in both prokaryotic and eukaryotic cells. This observation supports previous evidence that the activities encoded by the gene are highly conserved (Schechtman and Yanofsky, 1983; Yelton et al., 1983). The N. crassa trp-I gene is known to be trifunctional because it complements three different N. crassa trp mutants, each lacking IGPS, PRAI, or GAT activity (Schechtman and Yanofsky, 1983). A. mdulans trpC is also apparently trifunctional since its primary sequence has extensive homology with N. crassa trp-1 (Mullaney et al., 1985; Schechtman and Yanofsky, 1983). We have found that C. heterostrophus TRPl is similar to the corresponding N. crassa and A. nidulans genes. It clearly encodes IGPS and PRAI because it complements E. coli trpC and trpF mutants. It probably carries a GAT domain as well since the size of the C. heterostrophus
81
TRPl transcript
is the same as those of the N. crassa
trp-1 and A. nidulans trpC genes
1985; Schechtman
and Yanofsky,
There are also differences
(Mullaney
et al.,
1983).
among the three fungal
genes. First, the A. nidulans trpC gene is highly regulated whereas
C. heterostrophus TRPl is not. We ob-
Agriculture
Competitive
and Pioneer
Hi-Bred
was supported
Research
Grants
International.
Office,
W.D. MacRae
by a postdoctoral
fellowship
from
NSERC of Canada. The N. crassa plasmid pNC2 was from M. Schechtman, A. nidulans plasmids pHY201
served only a slight difference in the amount of TRPl
thank
transcript when C. heterostrophus was grown in minimal medium compared with the amount produced in
Barbara
and pKBY2 were from W. Timberlake. Peter
Mullin
Mosher
for technical
for preparation
We
assistance
and
of the manuscript.
complete medium, but no detectable trpC transcript when A. nidulans was grown in complete medium. Second, the fungal genes may differ in size. Sequencing of N. crassa trp-1 and A. nidulans trpC has revealed that all three functional domains are encoded by about 2300 bp and transcripts correspond to that length. Deletion subcloning indicates that up to 3 kb of the C. heterostrophus TRPl sequence may be required for both IGPS and PRAI activity in E. coli (Fig. 4) although the TRPl transcript is the same size as that of A. nidulans trpC. C. heterostrophus TRPl complements both trpC and trpF mutants of E. coli while cloned N. crassa trp-I complements only trpF, even though a trpC domain is present on the clone. The failure of N. crassa trp-1 to complement trpC in E. coli may be caused by lack of a binding site upstream of trpC that E. coli ribosomes can recognize (Schechtman and Yanofsky, 1983). Fusion of a bacterial ribosomeThird,
binding site to trp-1 enables it to complement both trpF and trpC mutations in E. cob (Schechtman and Yanofsky, 1983). PRAI activity of C. heterostrophus TRPl in E. coliis apparently initiated from within the coding region and not from the C. heterostrophus promoter, since the 5’ end of the gene can be eliminated and PRAI activity retained. It is likely that the C. heterostrophus TRPl promoter does not function in yeast, since unrearranged pChTRP24 did not confer a Trp’ phenotype on yeast cells that carried it, while an altered version of the plasmid gave rise to a Trp’ phenotype. The functional version appeared to be pChTRP24 with a 3-kb deletion at the 5’ end of the TRPl coding region.
REFERENCES Barratt,
R.W., Johnson,
mutant Birnboim,
53 (1965)
recombinant
extraction
plasmid
proce-
DNA. Nucl. Acids
Res. 7 (1979) 1513-1523. Botstein,
D., Falco,
S.C., Stewart,
S.E., Brennan,
M., Scherer,
D.T.. Struhl.
K. and Davis.
R.W.: Sterile
S., Stinchcomb.
host yeasts (SHY): a eukaryotic ment
for recombinant
system ofbiological
DNA
experiments.
contain-
Gene 8 (1979)
1l-24. Clarke,
L. and Carbon,
ColEl
hybrid
genome.
J.: A colony
plasmids
bank containing
representative
synthetic E. coli
of the entire
Cell 9 (1976) 91-99.
Frischauf,
A.M.,
Lambda
Lehrach,
replacement
H., Poustka, vectors
carrying
A. and
Murray,
polylinker
N.:
sequences.
J. Mol. Biol. 170 (1983) 827-842. Garber,
R.C.
and
Yoder,
O.C.:
filamentous
fungi and separation
ribosomal,
and plasmid
Isolation
of
into nuclear,
components.
DNA
from
mitochondrial,
Anal.
Biochem.
135
(1983) 416-422. Hanahan,
of Escherichia
D.: Studies on transformation
plasmids. Holmes,
coli with
J. Mol. Biol. 166 (1983) 557-580.
D.S. and Quigley,
preparation
M.: A rapid boiling method
of bacterial
plasmids.
Anal. Biochem.
for the
114 (1981)
193-197. Ito. H., Fukuda, of intact
Y.. Murata,
K. and Kimura,
yeast cells treated
A.: Transformation
with alkali cations.
J. Bacterial.
153 (1983) 163-168. Kafer,
map of Aspergillus nidulans, Adv.
E.: An 8.chromosome
Genet. 9: 105-145. Keesey Jr., J.K. and DeMoss,
J.A.: Cloning ofthe trp-I gene from
Neurospora
craSsa by complementation
Escherichia
coli. J. Bacterial.
of a frpC mutation
and in vitro culture Microbial.
Biochim. Maniatis,T., Spring
Biophys.
Harbor,
of
J.: High molecular
weight artefacts
in
from Blastocladiella at elevated temperatures.
Fritsch,
A Laboratory
for selection
of Cochliobolus heterostrophus.
128 (1982) 1719-1729.
J.S. and Leaver,
RNA extracted
in
152 (1982) 954-958.
Leach, J., Lang, B.R. and Yoder, O.C.: Methods J. Gen.
by grants from the the U.S. Dept. of
Genetics
H.C. and Daly, J.: A rapid alkaline
dure for screening
Lovett,
This work was supported National Science Foundation,
W.N.: Wild type and
nidulans.
233-246.
mutants
ACKNOWLEDGEMENTS
G.B. and Ogata,
of Aspergillus
stocks
Acta 195 (1969) 319-327. E.F. and Sambrook,
Manual.
J.: Molecular
Cold Spring Harbor
NY, 1982.
Cloning.
Laboratory,
Cold
88 McMastcr,
G.K. and Carmichael,
double-stranded
G.G.: Analysis
of single- and
nucleic acids on polyacrylamide
gels by using glyoxal and acridine
orange.
and agarose
Proc. Natl. Acad.
Sci. USA 74 (1977) 4835-483X. Mullaney,
E.J., Hamer,
Timberlake,
K.A., Yelton,
structure
deoxyribonucleic translation 237-25
M.M. and
of the rrpC gene from
Aspergillus nidzdans. Mol. Gen. Genet.
Rigby, P.W.J., Dieckmann,
199 (1985) 37-45.
M., Rhodes, C. and Berg, P.: Labeling
acid to high specific activity in vitro by nick
with DNA polymerase
1. 3. Mol. Biol. 113 (1977)
M.G.
functional expression
and
Yanofsky,
C.: Structure
of the tri-
trp-I gene from Neurosporu CY~SSCIand its aberrant in Escherichiu
coli. J. Mol. Appl. Genet. 2 (1983)
x3-99. Sherman. Harbor,
hybrid DNA molecules. Turgeon,
F., Fink, Cold
G.R.
and Hrcks,
Spring
Harbor
NY. 1981.
ofyeast:
S. and Davis, R.W.: Highautonomous
replication
of
Proc. Nat]. Acad. Sci. USA 76 (1979)
B.C., Garber,
J.B.: Methods Laboratory,
Cold
R.C. and Yoder, O.C.: Transformation
of the fungal maize pathogen the Aspergiilus nidulam
Cochlioholus heterosrrophus using
umdS
Yelton, M.M., Hamcr. Timberlake,
J.E., desouza,
W.E.: Developmental
201
rrpC
gene.
Proc.
E.R., Mullaney, regulation Natl.
Acad.
E.J. and
of the AsperSci. USA
80
(1983) 7576-7580. Yelton.
M.M.,
mation
Hamer,
of Aspergilus
J.E. and Timberlake,
Communicated
W.E.: Transfor-
nidukms by using a frpC plasmid.
Sci. USA 81 (1984) 1470-1474.
in Yeast Spring
gene. Mol. Gen. Genet.
(1985) 450-453.
Nat]. Acad.
Genetics.
D.T., Scherer,
transformation
gillus nidulans
I.
Schechtman.
frequency 1035-1039.
J.E., Roberti,
W.E.: Primary
Struhl, K., Stinchcomb,
by M.R. Culbertson.
Proc.