Cell, Vol . 35, 235 -242, November 1983, Copyright m 1983 by MIT
0092-8674/83/110235-08 $02 .0010
Isolation of the Transposable Maize
Controlling Elements Ac and Ds
N . Fedoroff, S . Wessler,* and M . Shure
Department of Embryology
Carnegie Institution of Washington
115 West University Parkway
Baltimore, Maryland 21210
Summary
Restriction endonuclease fragments containing part
of the Waxy (Wx) locus have been cloned from
strains with insertion mutations at the locus caused
by the controlling elements Activator (Ac) and Dissociation (Ds). Evidence is presented that the genetically defined Ac element corresponds to a 4 .3
kb insertion, while the two Ds elements correspond
to 4 .1 kb and 2 .0 kb insertions, all near the 3' end
of the Wx transcription unit. The 4.1 kb Ds is almost
completely homologous to the Ac element, differing
by a central deletion of less than 0 .2 kb. The 2.0 kb
Ds element is homologous to the ends of the Ac
element. Sequences homologous to the ends of the
Ac element are present in many copies in the genomes examined, while there are ten or fewer copies
of a sequence with homology to the center of the
cloned Ac element. The Ac element at the Wx locus
can be distinguished structurally from the other Aclike sequences in the genome .
Introduction
The controlling elements of maize were the first transposable elements discovered . The systematic genetic analysis
of mutations in maize caused by transposable elements
began with the work of Emerson (1914, 1917, 1929) and
Rhoades (1936, 1938, 1941, 1945) on unstable mutations
affecting pigment biosynthesis . But it was the elegant
studies of McClintock (1948, 1949) and Brink and Nilan
(1952) that provided the first evidence that genetic elements transpose and that unstable mutations could result
from the insertion of a transposable element . Although
transposable elements have since been identified in a wide
variety of organisms, the maize controlling elements remain
among the best-characterized from a genetic standpoint .
Several distinct families of controlling elements have been
identified . A given family of controlling elements contains
members that can and cannot transpose autonomously,
as well as members that differ in the frequency and
developmental timing of both transposition and the other
types of genetic rearrangements caused by the elements
(for reviews, see McClintock, 1951, 1956a, 1965 ; Fincham
and Sastry, 1974 ; Fedoroff, 1983) .
The Activator (Ac) and Dissociation (Ds) elements,
whose molecular isolation is described here, comprise a
family of controlling elements . The Ac element transposes
* Present address : Department of Botany, University of Georgia, Athens,
Georgia 30602 .
autonomously. Mutations caused by insertion of the element in or near a locus are unstable, reverting both
somatically and germinally at a high frequency . The Ds
element, which was initially identified as a site of chromosome dissociation or breakage (McClintock, 1945, 1946,
1947, 1948), cannot transpose autonomously and has
therefore been designated a nonautonomous element (Fedoroff, 1983) . It requires the presence of the autonomous
Ac element for both transposition and chromosome breakage (McClintock, 1951) . The observation that Ac insertion
alleles of a locus can give rise directly to Ds-like mutations
at the same locus suggests that Ds and Ac elements are
structurally related (McClintock, 1955, 1956b, 1962, 1963) .
There exist several unstable mutant alleles of the Waxy
(Wx) locus which are either Ac or Ds insertion mutations .
The Ac wx-m9, wx-m9, and wx-m6 alleles isolated by
McClintock (1952, 1963) were used for the present study
(see Experimental Procedures for a note on nomenclature) .
The phenotype of the Ac wx-m9 allelle is intermediate
between the dominant Wx and the recessive wx alleles .
The Ac wx-m9 allele reverts somatically both frequently
and relatively early in endosperm development . Germinally
stable Wx revertants of the Ac wx-m9 allele occur at a
high frequency, and a revertant strain was derived for use
in the present experiments . The wx-m9 allele is a derivative
of the Ac wx-m9 allele (McClintock, 1963) . It shows the
intermediate phenotype of the Ac wx-m9 allele, but is
unstable only in the presence of an Ac element elsewhere
in the genome . The wx-m9 allele therefore shows the
genetic behavior characteristic of a Ds insertion mutation
(McClintock, 1963) .
The wx-m6 allele arose by transposition to the Wx locus
of the Ds element first identified by McClintock (1947,
1948, 1952) . The Ds insertion site has been mapped
genetically between two mutations that give altered Wx
proteins (Nelson, 1968 ; Echt and Schwartz, 1981) and is
located near the 3' end of the Wx transcription unit (Shure
et al ., 1983) . The same Ds element is responsible for the
complex unstable mutations at the Shrunken locus that
have been studied in several laboratories (McClintock,
1952, 1953 ; Burr and Burr, 1980, 1981, 1982 ; Geiser et
al ., 1982 ; Fedoroff et al ., 1983 ; Courage-Tebbe et al .,
1983) .
In the present study, restriction endonuclease fragments
containing part of the Wx locus were cloned from strains
carrying the Ac wx-m9, wx-m9, and wx-m6 alleles, as well
as the newly derived revertant allele, designated Wx9-rl .
The fragments cloned from the strains with Ac and Ds
insertion mutations differed from the Wx progenitor alleles
and revertants by the presence of insertions varying in
length from 2 kb to 4 .3 kb . We present evidence that the
insertion sequences are all related and correspond to Ac
and Ds elements . We show that sequences homologous
to the ends of Ac are much more abundant in the genome
than sequences homologous to the middle of the element .
We also show that the genetically active Ac element at the
Wx locus in Ac wx-m9 allele can be distinguished from all
other Ac-like sequences in the same genome .
Cell
236
Figure 1. A Heteroduplex
between A EMBU
Phage DNAs Containing Wx Bgl II Fragments
Cloned from Maize DNA Carrying the AC wx-m9
and WxS-rl Alleles
Recombinant
A phage DNAs carrying Wx locus
Bgl II fragments cloned from the AC wx-m9 and
WxS-rl alleles in opposite orientation were denatured, partially renatured, and spread for electron
microscoprc
analysis. The double-stranded
regions correspond
to the maize DNA inserts and
the single-stranded
regions to the unpaired vector
arms.
d wx-m9
e wx-Ills
Figure 2. The Structure of the Eco RI and Bgl II Fragments of the Wx Locus
Cloned from Maize Strains Carrying Various Alleles of the Wx Locus
The relationship between the Eco RI Wx fragment described by Shure et
al. (1983) and the Bgl II fragments cloned from strains carrying the wx-m6,
WxS-rl, AC wx-m9, and wx-m9 alleles is indicated by the vertical alignment
of the restriction endonuclease cleavage site maps. As described in Shure
et al. (1983) and indicated by the arrow, the Eco RI fragment contains most
or all of the Wx transcriptron unit. The Bgl II fragments contain only the 3’
end of the transcription unit, but include the site of insertion of the AC and
Ds elements in the mutant alleles. The location and relative size of the AC
and Ds insertrons, whose structure is described in the text, are indicated
by the triangles below the fragments.
Results
The AC Element Is a 4.3 kb Insertion at the Wx
Locus in a Strain Homozygous for the AC wx-m9
Allele
To identify the AC element, we cloned genomic DNA
sequences homologous to a Wx cDNA plasmid (Shure et
al., 1983) from plants with and without AC insertions at the
locus (see Experimental Procedures for details). A 12.5 kb
Bgl II fragment with homology to the Wx cDNA was isolated
from DNA of plants carrying the WxS-rl revertant allele,
while the corresponding
Bgl II fragment isolated from DNA
of plants carrying the AC wx-m9 allele was approximately
17 kb in length. When phage DNAs with the two types of
Bgl II fragments were heteroduplexed,
the maize DNA
inserts were found to be homologous except for a region
of approximately 4 kb near one end (Figure 1). When we
examined heteroduplexes
between the AC wx-m9 Bgl II
fragment and a cloned Wx Eco RI fragment (Shure et al.,
1983) we found that the fragments also differed by a 4
kb sequence within the region of overlap between them
(Figure 2). Thus the AC wx-m9 allele contains a sequence
not present in either of two Wx alleles. In view of the fact
that one of the Wx alleles is a stable Wx revertant newly
derived from the AC wx-m9 allele, we conclude that the
insertion corresponds to the AC element causing the unstable mutation.
Detailed restriction endonuclease
mapping confirmed
the similarity between the AC wx-m9 and WxS-r? Bgl II
fragments and located the AC insertion site within a 0.23
kb Pst I fragment near the 3’ end of the mRNA coding
sequence (Figure 2). As illustrated in Figure 3, the cloned
AC wx-m9 Bgl II fragment has a 4.7 kb Pst I fragment with
an internal Eco RI site that is not present in the cloned
WxS-rl Bgl II fragment. Although the 0.23 kb Pst I fragment
that is replaced by the novel 4.7 kb Pst I fragment is not
apparent in Figure 3, its presence in the cloned WxS-rl
DNA and its absence in the cloned AC wx-m9 DNA were
established by analyzing Pst l-digested DNAs on polyacrylamide gels (data not shown). Detailed mapping of the AC
element showed it to be 4.3 kb in length and is described
below.
The Ds Element in the wxvn9 Allele Is Almost
Identical to the AC Element
We cloned a Bgl II fragment with homology to a Wx cDNA
plasmid from a strain homozygous for the wx-m9 allele.
The fragment is similar in structure to the Bgl II fragment
cloned from the AC wx-m9 allele (Figures 2c and 2d), but
is longer at the right end and contains two additional
internal Bgl II sites (see Experimental Procedures). Restriction endonuclease analyses of the fragment revealed the
presence of a 4.1 kb insertion within the same 0.23 kb Pst
I fragment within which the AC element is inserted in the
AC wx-m9 allele. This is shown diagrammatically in Figure
Id. In order to compare the insertions present in the AC
wx-m9 and wx-m9 alleles, we subcloned the Pst I fragments containing the insertions into plasmids (see Experimental Procedures), designating the subclones pAc9 and
pDs9, respectively. The maize Pst I fragments in the pAc9
and pDs9 plasmids gave well-matched heteroduplexes,
indicating a strong similarity between the sequences (Figure 4). However, when the plasmids were mapped with
restriction endonucleases, it became evident that the pAc9
lsolatron of Marze Elements AC and Ds
237
kb
3:z4.7=
2.62.1-
Figure 4. A Heteroduplex
between
Lrnearized
pAc9 and pDs9
Plasmrds carryrng the subcloned Wx Pst I fragments containing the Ac9
and Ds9 insertion sequences in opposrte orientations were linearized with
Cla I, denatured, renatured, and spread for electron macroscopic analysis.
0.8-
PstI
P8t1t
EcoRI
Figure 3. Blot Hybrtdrzatton Analysis of Bgl II Fragments
DNA Carrying the AC wx-m9 and WxS-rl Alleles
Cloned from Marze
Phage DNAs containing the AC wx-m9 and WxS-rl Bgl II fragments were
digested wrth Pst I alone or Pst I and Eco RI, fractionated
on an 0.7%
agarose gel, and transferred to nrtrocellulose. The blots were hybridized to
the Eco RI Wx genomic fragment (Rgure 2a). The WxS-rl Bgl II fragment
contains two additional Pst I fragments that are 0.23 and 0.17 kb in length
and are not detectable in the autoradrogram because of their size. The 0.23
kb fragment is missing In the AC wx-m9 Bgl II fragment, although the 0.17
kb fragment IS present (data not shown). The Pst I digest of the AC wx-m9
Bgl II fragment contarns a novel 4.7 kb fragment with limited homology the
probe that IS not present in the WxS-rl Bgl II fragment. The Pst I-Eco RI
digest of the AC wx-m9 DNA contains two shorter fragments of 2.6 and
2.1 kb that are not represented
rn the correspondrng
digest of WxS-rl
fragment.
and pDs9 plasmids were not identical (Figures 5a and 5b).
The cloned Pst I fragments differ from each other by the
absence from pDs9 of a seqeunce of less than 0.2 kb
present near the center of the AC element in pAc9. Thus
the AC wx-m9 and wx-m9 alleles have almost completely
homologous insertions at precisely the same site within
the transcription unit. Because the wx-m9 strain is a direct
derivative of the AC wx-m9 strain, but behaves genetically
like a Ds insertion mutation, we conclude that the AC-like
insertion at the locus is the Ds element. The elements have
been designated Ac9 and Ds9, respectively, to indicate
their isolation from the AC wx-m9 and wx-m9 alleles.
Frgure 5. Restriction Endonuclease Cleavage Sate Maps of Wx Pst I Fragments Containing the Ac9, Ds9. and Ds6 controllrng Element Insertions
The open boxes represent the Wx locus sequence, bounded by Pst I sites,
within whrch the insertion IS located. The solid central lines represent the
controlling element sequences. The orientation of the Wx locus fragment
into whrch the elements have inserted is indicated relative to the direction
of transcription (Figure 2a). The Ac9 and Ds9 elements are tndistingurshable
by restnction endonuclease mapping except for the absence of a sequence
of less than 0.2 kb, represented by the interruption in (b), from the Ds9
element that is present In the Ac9 element. The Ds6 element is Inserted rn
the opposrte orientatron to the Ac9 and Ds9 elements at a different site rn
the transcription unit.
The Ds Element in wx-m6 Is Also Related to the AC
Element
We cloned a Bgl II fragment containing part of the Wx
transcription unit from a strain homozygous for the wx-m6
allele. Its structure is shown in Figure 2e. The Wx allele
from which the Wx Eco RI fragment (Figure 2a) was cloned
is the progenitor of the wx-m6 allele. The structure of the
wx-m6 transcription unit differs from that of the progenitor
Wx allele by the presence of an insertion in a 1 kb Pst I
fragment near the 3’ end. The insertion, whose length was
estimated at 2.4 kb on the basis of blot hybridization data
Cell
230
(Shure et al., 1983) proved to be 2.0 kb in length. The
location and size of the insertion are indicated in Figure
2e. To compare the insertion detected in the wx-m6 allele
with the AC element, we first subcloned the Pst I fragment
containing the insertion, designating the plasmid pDs6
(see Experimental
Procedures).
To determine whether
there is homology between the maize sequences in p&6
and pAc9, we linearized and heteroduplexed
the plasmid
DNAs. Two representative heteroduplexes
are shown in
Figure 6. Since the insertion sites differ in the two alleles,
the flanking Wx sequences are not homologous and form
small loops at each end of the maize Pst I fragments in
the plasmids. The central 2 kb sequence of the maize Pst
I fragment in the p/Is6 plasmid is entirely homologous to
the ends of the AC element in pAc9. The central portion
of the AC element forms a large single-stranded loop. The
loop is approximately centered within the double-stranded
region, indicating that the 2 kb insertion in the wx-m6 allele,
designated Ds6, is homologous to a 1 kb sequence at
each end of the AC element. Further evidence that the Ds6
element resembles the termini of the Ac9 element is
provided by the similarities between the restriction endonuclease cleavage site maps shown in Figures 5a and 5c.
The AC and Ds Elements Are Not Internally
Repetitive and Are Inserted in Opposite
Orientations
The AC and Ds elements do not appear to have extensive
internal repetitions. The elements do not form fold-back
or stem-loop
structures after denaturation
(data not
shown). The restriction endonuclease
maps of the elements (Figure 5) similarly give no evidence of either long
internal or terminal redundancies.
We estimate that the
elements can have no more than 50 bp of terminally
redundant sequence.
The presence of unique restriction endonuclease cleavage sites near one end of the AC and Ds elements makes
it possible to determine the orientation of each of the three
elements within the Wx locus. Not unexpectedly, the Ac9
and Ds9 elements are both in the same orientation. The
Ds6 element is inserted in the opposite orientation to the
Ac9 and Ds9 elements.
Sequences Homologous to AC Termini Are More
Abundant in Genomic DNA than Are Sequences
Homologous to the Center of the Element
To assess the abundance of genomic sequences with
homology to the middle and the ends of the AC element,
we used a central Ava I-Eco RI fragment from Ac9 (see
Experimental Procedures) and the Ds6 element as probes.
There are sequences homologous to both the Ds6 element
and the central portion of the Ac9 element in all of the
maize genomic DNAs examined (Figures 7a and 7b).
Because recombinant phage DNAs were present in equivalent amounts and hybridized with the AC and Ds6 probes
to the same extent in the experiment whose results are
shown in Figure 7, the hybridization of the genomic DNAs
is directly comparable. There are many Bst Eli fragments
homologous to the Ds6 element in all of the maize DNAs
examined (Figure 7b). Moreover, the bands appear in a
continuous background
of hybridizing sequences, suggesting the presence of many more copies of sequences
with limited homology to the element. By contrast, the
maize genomic DNAs analyzed contain relatively few sequences with homology to the central portion of the Ac9
element (Figure 7a). The AC wx-m9 and wx-m9 DNAs
contain a fragment that comigrates with the cloned Bst Eli
fragment, as well as four to six additional fragments ranging in size from 3.6 kb to 25 kb (Figure 7a). The strain
from which the Wx Eco RI fragment was cloned and the
W23 x K5.5 hybrid strain contain three to four Bst Eli
fragments with homology to the AC probe. Thus sequences with homology to the central portion of the Ac9
element are much less prevalent in the maize genomic
DNAs analyzed than are sequences homologous to the
ends of the element.
The Structure of the AC-like Sequences in Maize
DNA
Figure 6. A Heteroduplex
between
pAc9 Element and pDs6
Piasmtds containing the subcloned Wx Pst I fragments with the Ac9 and
Ds6 insertions in the same orientation were ltnearized with Cla I, denatured,
renatured, and spread for electron microscopy.
The small loops near the
ends of the heteroduplex correspond to the unpaired Wx locus sequences
surrounding the controlling element insertion site and the large central loop
corresponds to the central portron of the AC element that is not homologous
to the Ds6 element.
To investigate the internal structure of the AC-like sequences in maize DNA, we cleaved genomic DNAs with
enzymes that cut within the element and probed with a
sequence from the center of the Ac9 element. When
genomic DNA from strains having and lacking a genetically
active AC element were digested with Hind Ill, which
cleaves a 1.6 kb fragment from the center of the cloned
Ac9 element, multiple copies of the fragment were detected (Figure 8a). We estimate that the number of copies
varies between four and eight to ten in the different DNAs
analyzed. Only about half of the AC-like sequences in each
lsolatron of Marze Elements AC and Ds
239
Genomic
Genomic DNA
DNA
kb
kb
3.6-
Q,
Cloned
DNA:
E
I
;
o, z
0,
E
#
;
Cloned
E(
DNA:
;
z
z
Cloned
DNA
Probe:
pAc9
-
:
Ix 5x
AC wx-m9
AvoI/EcoRI
fragment
Figure 7. Blot Hybridrzation Analysis
Restnction Endonuclease Bst Eli
lx 5x
AC wx-m9
pDs6
of Marze DNAs
AvoI
frogment
Digested
wrth the
Maize DNAs from plants havrng (AC wxm9) and lacking (Wx, W23 x K55,
wx-m9) an AC element were dtgested with Bst Eli, fractionated on an 0.7%
agarose gel, and transferred to nrtrocellulose. Mrxtures of X. laevrs DNA,
which shows no homology to the probes used, and an amount of hEMBL4
phage DNA containing the AC w-m9 Bgl II fragment corresponding
to one
or five genomic coptes were Included as Internal hybrrdizatron controls rn
both experrments. Falters were probed with 32P-labeled restriction endonuclease fragments. The AC probe used In (a) was the left Ava I-Eco RI
fragment from the center of the Ac9 element (Figure 5a). The probe used
In (b) was an Ava I fragment from the pDs6 plasmrd that contains most of
the Ds element (Figure 5~).
of the genomic DNAs yielded a 1.6 kb Hind III fragment.
Most of the additional sequences with homology to the
probe appeared on larger fragments, although all of the
strains also contained a Hind Ill fragment of about 1 .O kb
(Figure 8a). The 1.4 kb Hind Ill fragment derived from the
Ds9 element is detectable in wx-m9 genomic DNA, but is
less abundant than the 1.6 kb fragment. Results similar to
those obtained with Hind Ill were obtained when genomic
DNAs were doubly digested with Ava I and Eco RI (data
not shown). We conclude that there are several copies of
a sequence in all of the genomic DNAs examined whose
internal structure resembles that of the cloned Ac9 element.
In contrast, the results of analogous experiments carried
out with several other enzymes suggest that the genetically
active Ac9 element at the Wx locus and its Ds9 derivative
Acare structurally distinguishable from the other genomic
like sequences. Thus when the same genomic DNAs that
gave multiple Hind III fragments (Figure 8a) were digested
Frgure 8. Blot Hybrrdrzatron Analysis of Maize DNAs Digested
tton Endonucleases
Hind Ill (a) and Pvu II and Eco RI (b)
:
3
wrth Restnc-
The sources of the genomrc DNA are indicated above the autoradiogram.
The two nght-hand lanes tn each blot contarn a mtxture of X. laevrs genomic
DNA and an amount of the cloned AC wx-m9 and wx-m9 Egl II fragments
equrvafent to five genomrc copres of the element. Both blots were probed
wrth the labeled Ava I-Eco RI fragment from the center of the AC element
(see Experimental Procedures).
with Puv II and Eco RI, only the AC wx-m9 DNA yielded a
1.3 kb Pvu II-Eco RI fragment that comigrated with the
corresponding
fragment from the cloned AC element (Figure 8b). DNA from wx-m9 homozygotes lacked the 1.3 kb
fragment, but gave a 1 .I kb Pvu II-Eco RI fragment with
homology to the AC probe that comigrated with the corresponding fragment in the cloned Ds9 element (Figure 8b).
Thus the Ac9 and Ds9 elements at the WX locus are
distinguishable
from the other AC-like sequences, all of
which appear in larger fragments. Results similar to those
shown in Figure 8b were also obtained with Pvu II or Ava I
alone and with a mixture of both enzymes (data not
shown). The difference between the Ac9 and Ds9 elements at the Wx locus and the other AC-like sequences is
therefore not confined to a single restriction endonuclease
cleavage site nor to one end of the element. Thus despite
the similarity in internal structure of the several AC-like
sequences present in all of the genomes examined, the
single copy of the AC element at the Wx locus is distinguishable from the other AC-like sequences.
Discussion
AC and Ds Elements Are Structurally Related
We have presented evidence that the AC element in the
AC wx-m9 mutant allele (McClintock, 1963) is a 4.3 kb
Cell
240
sequence inserted near the 3' end of the transcription unit .
Wx locus fragments cloned from the two Ds alleles wx-m9
and wx-m6 (McClintock, 1952, 1963) also contain insertions within the transcription unit. The Ds9 insertion in the
wx-m9 allele is at the same site as the Ac element in the
Ac wx-m9 allele and almost identical to it . The Ds9 element
is missing a sequence of less than 0 .2 kb present in the
Ac element . Because the wx-m9 allele is a derivative of
the Ac wx-m9 allele (McClintock, 1963), it is likely that Ds9
arose directly from the Ac9 element by a deletion mutation .
The wx-m6 allele contains a 2 kb insertion at a site
approximately 0 .5 kb 5' to the Ac9 and Ds9 insertion site .
The 2 kb Ds6 element is homologous to the ends of the
Ac element . Moreover, its restriction endonuclease cleavage site map is indistinguishable from that of the Ds
element identified in association with a complex Ds-induced mutation at the Shrunken locus in maize (CourageTebbe et al ., 1983 ; H .-P . Doring, E, Tillman, and P. Starlinger, unpublished data) . Thus the Ac9, Ds9, and Ds6
elements are all structurally related to each other . Because
Ds elements are capable of transposing only in the presence of an Ac element located elsewhere in the genome,
they have structural information necessary for transposition, but are unable to produce a trans-acting substance
necessary for transposition . The small deletion that distinguishes the Ac9 and Ds9 elements may be within a
sequence encoding a protein required for transposition .
Sequences Homologous to the Termini of Ac Are
Abundant in Maize Genomic DNA
Because the Ds6 element is homologous to the ends of
the Ac element, we used it as a hybridization probe to
assess the genomic representation of Ac termini . The
several genomic DNAs analyzed contained many sequences homologous to the Ds6 element . Similar results
were reported by Geiser et al . (1982), who used the Ds
element associated with the Sh locus in the Ds-induced
sh-m5933 allele . This is not unexpected, since both Ds6
and the Ds element in the sh-m5933 allele originated from
the same genetically defined Ds element (McClintock,
1951, 1952, 1953) and the two elements appear to be
structurally similar . Thus the terminal sequences of the Ac
element are quite abundantly represented in the maize
DNAs that have been analyzed .
There Are Relatively Few Ac-like Sequences in
Maize Genomic DNA and They Are Distinguishable
In contrast to the results obtained with the Ds6 element, a
probe fragment from the center of the cloned Ac element
hybridized to a small number of maize DNA sequences
(four to ten) . About half of the Ac-like sequences in a given
strain resemble the cloned Ac element in structure, yielding
internal fragments that comigrate with those of the cloned
element. However, the Ac9 element and Ds9 elements at
the Wx locus can be distinguished from all of the other Aclike sequences in the Ac wx-m9 and wx-m9 genomic
DNAs . After digestion with certain restriction enzymes,
only the elements at the Wx locus yield fragments with
homology to Ac that comigrate with the fragments from
the cloned elements . The Ac and Ds elements at the Wx
locus are therefore unique . Moreover, the differences between the elements at the Wx locus and other Ac-like
sequences are neither confined to a single restriction
endonuclease cleavage site, nor to one end of the element .
The Relationship among the Ac-like Sequences in
Maize Genomic DNA
Although it is clear that sequences homologous to Ac
termini are substantially more abundant than sequences
homologous to the center of the element, we do not yet
understand the significance of this finding . It is undoubtedly
the case that genomes containing active Ac elements
accumulate sequences, such as Ds9, that represent deleted and otherwise mutated derivatives of the active
element . In this regard, there is a parallel between the P
elements of Drosophila, many of which appear to be
internally deleted, defective elements, and the Ds9 element
described here (Rubin et al ., 1982 ; Bingham et al ., 1982 ;
Spradling and Rubin, 1982 ; Rubin and Spradling, 1982) .
However, elements with small internal deletions resembling
Ds9 are not common in the maize genomes examined .
Most of the Ac-like sequences differ more radically from
the active element, suggesting either that they are derived
by larger deletions than the one that generated Ds9 or
have an altogether different structure . Moreover, Ac-like
sequences are present in the genome of strains that have
no active Ac elements . It is known that there are cryptic
controlling elements in maize DNA and that they can be
activated by agents that cause chromosome damage
(McClintock, 1946, 1950a and 1950b ; Neuffer, 1966 ; Bianchi et al ., 1969 ; Doerschug, 1973) . It is possible that some
of the Ac-like sequences detected in the present experiments correspond to such cryptic elements .
We also do not understand the basis of the difference
between the Ac9 and Ds9 elements and the several other
sequences that most closely resemble the cloned element .
It is possible that the elements at the Wx locus differ from
similar elements elsewhere simply because they are inserted in a transcriptionally active region and therefore
have different sequence modifications than they do at
other sites . Alternatively, it may be that Ac9 and Ds9
elements differ structurally from the other Ac-like sequences . The Ac9 element is the only genetically active
Ac element in the genome from which it was cloned and
the Ds9 element is a direct derivative of it . The difference
between these elements and other Ac-like genomic sequences may be the consequence of a rearrangement
that originally activated the Ac9 element from a cryptic
element . Should the underlying difference between the
Ac9 element and other Ac-like sequences prove structural,
its potential utility for cloning genes with Ac insertion
mutations is evident .
Experimental Procedures
Maize Strains
The W23 X K55 hybrid was obtained from E . H. Coe . Strains containing
the Ac wx-m9, wx-m9, and wx-m6 alleles were originally obtained from B .
Isolation of Maize Elements Ac and Ds
24 1
McClintock . The strains were either homozygous for the mutant allele of
the Wx locus or were made homozygous by outcrossing plants that were
heterozygous for a stable wx allele and the unstable allele of interest to
Wx/Wx or Wx/wx plants and recovering the unstable allele by self-pollination
of plants grown from Wx kernels . Appropriate genetic tests were carried
out to verify the genetic identity of the allele present in plants with Ds
mutations, but lacking an Ac element . Wx revertants were selected as
single kernels showing a Wx phenotype throughout the endosperm on ears
of plants homozygous for the Ac wx-m9 allele or heterozygous of the Ac
wx-m9 allele and a stable wx allele . Plants grown from Wx kernels were
self-pollinated, tested for the presence of Ac, and for Ac-activated breakage
of the short arm of chromosome 9, with concomitant loss of the Wx locus
during endosperm development . The Wx9-rl allele is a spontaneous revertant derived in this way from the Ac wx-m9 allele . It showed no evidence
in genetic tests of the presence of either an Ac or a Ds element at the Wx
locus.
It should be noted that the allele referred to as Ac wx-m9 in the present
communication was initially designated wx-m9 by McClintock (1963) . She
subsequently renamed it Ac wx-m9 in her records, distinguishing it from
the wx-m9 derivative that behaves genetically as a Ds insertion mutation .
The alleles were so designated upon receipt from B . McClintock and the
designations are retained in the present manuscript because they serve
the purpose of referring to both the common origin of these alleles and
their genetically distinct behavior,
Cloning of Genomic DNA Fragments
The cloning procedures and vectors were those described in Shure et al.
(1983), except that AEMBL4 DNA was digested with Bam HI and Sal I prior
to isolation of vector arms . Maize genomic DNA was isolated as described
in Shure et al. (1983) from leaf and immature tassel tissue of 6-week-old
plants. Total maize DNA was digested with Bgl II (BRL) using 5 U of
enzyme/µg of DNA under conditions specified by the enzyme supplier.
Although internal controls indicated that digestion of 32P-labeled phage k
DNA was complete under the conditions used, some of the cloned fragments obtained had internal Bgl II sites . This may be the result of simultaneous cloning of unrelated fragments or may be attributable to the fact that
certain Bgl II sites in maize DNA are only partially cleaved in some plants
(N . Fedoroff, unpublished data) . Total Bgl II-digested maize DNA was ligated
into the Barn HI cloning site of kEMBL4, the phage DNA was packaged in
vitro (Hohn, 1979) and the resulting recombinant phages were screened
by plaque hybridization (Benton and Davis, 1977) with the pWxO.4 cDNA
clone described by Shure et al . (1983) . Recombinant k phages containing
fragments with Ac and Ds insertions were found to grow poorly, invariably
producing very small plaques . To ensure that the cloned fragments were
representative of the original genomic sequences, multiple independent
clones were isolated and analyzed whenever possible . Moreover, restriction
endonuclease digests of cloned and genomic DNAs homologous to Wx
locus probes in the mutant strains were compared to verify the identity of
the cloned and genomic sequences .
After controlling element insertion sites had been identified by restriction
endonuclease mapping and blot hybridization analysis of recombinant
phage DNAs, the Pst I fragments of the Wx locus in which insertions had
been identified were subcloned into plasmid pKP32 (a kind of gift from
Keith Peden) . pKP32 is a derivative of pBR322 from which the Eco RI and
Ava I sites have been eliminated and which carries a deletion extending
from nucleotide 1641 to nucleotide 2490 of the pBR322 sequence . The
plasmids containing the insertion-bearing Pst I fragments subcloned from
the Ac wx-m9, wx-m9, and wx-m6 BgI II genomic fragments were designated pAc9, pDs9 and pDs6, respectively. The elements have been
designated Ac9 and Ds9 and Ds6 to indicate their genetic origins .
Electron Microscopy of Heteroduplexes
Phage DNAs or Cla I-linearized plasmid DNAs were denatured, annealed,
and spread for electron microscopic analysis as described by Davis et al .
(1971) .
Blot Hybridization Analysis
Hybridization analysis of restriction endonuclease digests of maize geomic
DNA (Southern, 1975) was done essentially as described by Fedoroff et al .
(1983) . The Ac probe used was the left Ava I-Eco RI fragment that includes
the sequence deleted in the Ds9 element . The fragment was purified from
an Ava I-Eco RI digest of pAc9 by preparative gel electrophoresis and
labeled with 32P by nick translation (Maniatis et al., 1975) to a specific
activity of 3 x 10 8-1 x 109 cpm/µg, using dATP and dCTP (Ame-sham,
specific activity >400 Ci/mmole) as labeled substrates . The Ds probe used
in the present experiments was the Ava I fragment from pDs6 that includes
almost all of the 2 kb Ds6 element, as well as a small amount of adjacent
Wx sequence .
Blot hybridization analysis of recombinant phage DNAs was done using
several subcloned DNA fragments, including the Eco RI Wx genomic
fragment subcloned in pBR322 (Figure 2a) and the 8 kb and 4 .6 kb Eco RI
fragments recovered from kEMBL4 containing a 12 .6 kb Bgl II fragment of
the Wx locus cloned from a strain containing the Wx9-rl allele (Figure 2b).
The two Eco RI fragments were obtained by virtue of the Eco RI sites
flanking the Bam HI cloning site in EMBL4 and correspond to Bgl II-Eco RI
fragments in genomic DNA (A-M . Frischauf, H . Lehrach, A-M . Poustka, and
N . Murray, unpublished data) .
Acknowledgments
We thank Dr. B. McClintock for genetic materials, Dr . M. Wu for assistance
with heteroduplex analysis, Scott Downing for technical assistance, and P.
Schmidt for preparation of the manuscript, This research was supported
by NSF grant PCM-8207708 and USDA grant 82-CRCR-1-1065 . M . S. was
a fellow of the Damon Runyan-Walter Winchell Cancer Fund (DRG-397-F[FT]) and S . W . was a fellow of the American Cancer Society (PF-1837) .
The costs of publication of this article were defrayed in part by the
payment of page charges . This article must therefore be hereby marked
"advertisement' in accordance with 18 U .S .C . Section 1734 solely to
indicate this fact .
Received July 8, 1983 ; revised August 19, 1983
References
Benton, W . D ., and Davis, R . W. (1977) . Screening kgt recombinant clones
by hybridization to single plaques in situ . Science 196, 180-182 .
Bianchi, A ., Salamini, F., and Restaino, F . (1969). Concomitant occurrence
of different controlling elements . Maize Genet . Coop . News Left . 43, 91 .
Bingham, P . M ., Kidwell, M . G ., and Rubin, G . M . (1982) . The molecular
basis of P-M hybrid dysgenesis : the role of the P element, a P-strainspecific transposon family . Cell 29, 995-1004 .
Brink, R . A ., and Nilan, R . A . (1951). The relation between light variegated
and medium variegated pericarp in maize . Genetics 37, 519-544.
Burr, B., and Burr, F . A . (1980). Detection of changes in maize DNA at the
Shrunken locus due to the intervention of Ds elements . Cold Spring Harbor
Symp . Quant. Biol . 45, 463-465.
Burr, B ., and Burr, F . A . (1981) . Controlling-element events at the Shrunken
locus in maize . Genetics 98, 143-156 .
Burr, B ., and Burr, F . A. (1982). Ds controlling elements of maize at the
Shrunken locus are large and dissimilar insertions . Cell 29, 977-986.
Courage-Tebbe, U ., Doring, H .-P ., Fedoroff, N ., and Starlinger, P . (1983).
The controlling element Ds at the Shrunken locus in Zea mays: structure
of the unstable sh-m5933 allele and several revertants, Cell 34, 383-393 .
Davis, R . W ., Simon, M ., and Davidson, N . (1971) . Electron microscope
heteroduplex methods for mapping regions of base sequence homology
in nucleic acids . In Methods in Enzymology, Volume 21, L . Grossman and
K . Moldave, eds . (New York : Academic Press), pp . 413-428.
Doerschug, E . B . (1973). Studies of Dotted, a regulatory element in maize .
Theor . Appl . Genet . 43, 183-189.
Echt, C . S ., and Schwartz, D. (1981) . Evidence for the inclusion of controlling
elements within the structural gene at the waxy locus in maize . Genetics
99,275-284 .
Emerson, R . A . (1914) . The inheritance of recurring somatic variation in
variegated ears of maize . Am . Nat . 48, 87-115 .
Emerson, R. A . (1917) . Genetical studies of variegated pericarp in maize .
Genetics 2, 1-35.
Emerson, R . A . (1929) . The frequency of somatic,rrigtation invariegated
pericarp of maize . Genetics 14, 488-511 .
Cell
242
Fedoroff, N . (1983) . Controlling elements in maize . In Mobile Genetic
Elements, J . Shapiro, ed . (New York: Academic Press), pp. 1-63 .
Fedoroff, N ., Mauvais, J ., and Chaleff, D . (1983) . Molecular studies on
mutations at the Shrunken locus in maize caused by the controlling element
Ds. J . Mot . Appl. Gen. 2, 11-29.
Fincham, J. R . S ., and Sastry, G . R . K . (1974) . Controlling elements in
maize . Ann . Rev. Genet . 8, 15-50 .
Geiser, M ., Weck, E ., Daring, H.-P ., Werr, W ., Courage-Tebbe, U ., Tillmann,
E ., and Starlinger, P. (1982) . Genomic clones of a wild-type allele and a
transposable element-induced mutant allele of the sucrose synthase gene
of Zea mays L . EMBO J . 1, 1455-1460 .
Hohn, B . (1979) In vitro packaging of k and cosmid DNA . Meth . Enzymol .
68,299-309 .
Maniatis, T ., Jeffrey, A., and Kleid, D . G . (1975) . Nucleotide sequence of
the rightward operator of phage a . Proc. Natl . Acad . Sci. USA 72, 11841188.
McClintock, B. (1945). Cytogenetic studies of maize and Neurospora .
Carnegie Inst. of Wash. Year Book 44, 108-112 .
McClintock, B. (1946) . Maize genetics . Carnegie Inst. of Wash . Year Book
45,176-186 .
McClintock, B . (1947). Cytogenetic studies of maize and Neurospora .
Carnegie Inst . of Wash . Year Book 46, 146-152 .
McClintock, B. (1948) . Mutable loci in maize . Carnegie Inst . of Wash . Year
Book 47, 155-169 .
McClintock, B . (1949) . Mutable loci in maize . Carnegie Inst, of Wash . Year
Book 48, 142-154 .
McClintock, B . (1 950a) . The origin and behavior of mutable loci in maize .
Proc . Nat. Acad . Sci . USA 36, 344-355 .
McClintock, B . (1950b) . Mutable loci in maize . Carnegie Inst. of Wash . Year
Book 49, 157-167 .
McClintock, B. (1951) . Chromosome organization and genic expression .
Cold Spring Habor Symp . Quant. Biol . 16, 13-47 .
McClintock, B. (1952) . Mutable loci in maize. Carnegie Inst . of Wash . Year
Book 51, 212-219 .
McClintock B . (1953). Mutation in maize . Carnegie Inst. of Wash . Year Book
52,227-237 .
McClintock, B . (1955). Controlled mutation in maize . Carnegie Inst. of Wash .
Year Book 54, 245-255 .
McClintock, B . (1956a) . Controlling elements and the gene . Cold Spring
Harbor Symp . Quant, Biol . 21, 197-216 .
McClintock, B . (1956b) . Mutation in maize . Carnegie Inst. of Wash . Year
Book 55, 323-332 .
McClintock, B . (1962). Topographical relations between elements of control
systems in maize. Carnegie Inst . of Wash . Year Book 61, 448-461 .
McClintock, B . (1963) . Further studies of gene-control systems in maize .
Carnegie Inst. of Wash . Year Book 62, 486-493 .
McClintock, B . (1965) . The control of gene action in maize . Brookhaven
Symp. Biol . 18, 162-184 .
Nelson, 0 . E . (1968) . The Waxy locus in maize . II . The location of the
controlling element alleles. Genetics 60, 507-524 .
Neuffer, M . G . (1966) . Stability of the suppressor element in two mutator
systems at the A 1 locus in maize . Genetics 53, 541-549 .
Rhoades, M. M. (1936) . The effect of varying gene dosage on aleurone
colour in maize . J . Genet . 33, 347-354.
Rhoades, M . M . (1938) . Effect of the Dt gene on the mutability of the a1
allele in maize . Genetics 23, 377-397 .
Rhoades, M . M (1941). The genetic control of mutability in maize . Cold
Spring Harbor Symp . Quant . Biol . 9, 138-144.
Rhoades, M . M . (1945) . On the genetic control of mutability in maize. Proc.
Nat . Acad . Sci . USA 31, 91-95 .
Rubin, G . M ., and Spradling, A . C . (1982) . Genetic transformation of
Drosophila with transposable element vectors . Science 218, 348-353 .
Rubin, G . M ., Kidwell, M . G ., and Bingham, P. M . (1982) . The molecular
basis of P-M hybrid dysgenesis : the nature of induced mutations . Cell 29,
987-994 .
Shure, M ., Wessler, S ., and Fedoroff, N . (1983) . Molecular identification and
isolation of the Waxy locus in maize, Cell 35, 225-233 .
Southern, E . M . (1975) . Detection of specific sequences among DNA
fragments separated by gel electrophoresis . J . Mot . Biol. 98, 503-517 .
Spradling . A . C., Rubin, G . M . (1982) . Transposition of cloned P elements
into Drosophila germline chromosomes . Science 218, 341-347 .