Mol Gen Genet (1992) 233:411-418 © Springer-Verlag 1992

A maize cryptic Ac-homologous sequence derived from an Activator transposable element does not transpose Jun-Yi Leu 1'2, Y. Henry Sun*, Yiu-Kay Lai 2, and Jychian Chen 1 1 Institute of Molecular Biology,Academia Sinica, Taipei, Taiwan z Institute of Life Science, Tsing Hua University, Hsinchu, Taiwan Received November 7, 1991

Summary. Sequences sharing homology to the transposable element Activator (Ac) are prevalent in the maize genome. A cryptic Ac-like DNA, cAc-ll, was isolated from the maize inbred line 4Co63 and sequenced. Cryptic Ac-11 has over 90% homology to known Ac sequences and contains an 11 bp inverted terminal repeat flanked by an 8 bp target site duplication, which are characteristics of Ac and Dissociation (Ds) transposable elements. Unlike the active Ac element, which encodes a transposase, the corresponding sequence in cAc-ll has no significant open reading frame. A 44 bp tandem repeat was found at one end of cAc- 11, which might be a result of aberrant transposition. The sequence data suggest that cAc-ll may represent a remnant of an Ac or a Ds element. Sequences homologous to cAc-11 can be detected in many maize inbred lines. In contrast to canonical Ac elements, cAc-ll DNA in the maize genome is hypermethylated and does not transpose even in the presence of an active Ac element.

Key words: Cryptic Ac-homologous sequence - Activator transposable element - Hypermethylation - Transposition - Maize

Introduction Transposable elements can cause a high rate of genetic instability, including spontaneous unstable mutations and chromosome rearrangements (for reviews, Berg and Howe 1989). The Ac/Ds family of transposable elements, first investigated by McClintock (1946), represents one of several two-unit interactive transposable element systems of maize. The autonomous Activator (Ac) element is capable of transposition, excision, and causing mutational changes, while the non-autonomous Dissociation (Ds) element causes genetic instability only in the presence of Ac activity. Some Ds elements can be derived from the Offprint requests to: J. Chen

autonomous Ac elements by mutational changes (Pohlman et al. 1984). All Ac elements that have been cloned and sequenced to date appear to be structurally indistinguishable (Pohlman et al. 1984; Miiller-Neumann et al. 1984). The Ac elements are 4.5 kb in size with 11 bp imperfect terminal inverted repeats. The element is flanked by an 8 bp direct repeat that is present only once at the target site prior to insertion of Ac. Ds elements are less conserved and can be broadly classified into two major categories: type I and type II. Ds2 elements are derived directly from Ac elements (D6ring et al. 1984); Ds 1 elements are members of the Ac/Ds family that have a similar 11 bp inverted terminal repeat but are structurally dissimilar to the Ac element (Sachs et al. 1983; Sutton et al. 1984). A large percentage of the genomes of plants comprise short, interspersed repetitive DNA sequences. Molecular evidence has shown that certain members of this middle repetitive fraction of the maize genome resemble transposable elements (Flavell 1982; Gupta et al. 1984; Chandler et al. 1986, 1988). Sequences homologous to most active transposable elements in maize are also found in all related stocks as dispersed repetitive sequences whether or not detectable active forms of these elements exist in the genome. These sequences may represent remnants of once active transposable elements, and they may potentiate the generation of new functional versions of transposable elements. It is possible to activate controlling elements in maize that previously had no detectable transposition activity, through mechanisms such as the breakage-fusion-bridge cycle (BFB) and agents that damage chromosomes (Peschke et al. 1987; McClintock 1978, 1984). This suggests that certain mechanisms, yet to be elucidated, may exist which change a cryptic sequence to an active element. Even in maize plants that have no detectable Ac activity, sequences homologous to all regions of active Ac elements exist (Dellaporta and Chomet 1985). Based on genomic blot analysis, at least the core region of the active Ac restriction map is essentially conserved among some cryptic copies (Dellaporta and Chomet 1985). It

412 has also been shown that active Ae elements are undermethylated in comparison with their inactive genomic counterparts (Dellaporta and Chomet 1985; Chen et al. 1987). Sequence data are not yet available to demonstrate the homology between the cryptic copies and the known active Ac elements. In this study we cloned and sequenced a cryptic Ae-like element. Sequence comparison of the cryptic Ac with active Ac elements indicates that the cryptic Ac is most likely to be a remnant of an Ae or a Ds element.

M a t e r i a l s and m e t h o d s

Maize lines and 9enome libraries. The 4Co63 lines homozygous for P ~ W W or P ~ V V were obtained from Dr. Irwin Greenblatt. A testcross of P ~ W W 4Co63 with a tester line that, carried a Ds insertion at the R locus (r-m3::Ds, kindly supplied by Dr. Jerry Kermicle) indicated that no active Ac element is present in the stock. Recombinant inbred lines were acquired from Dr. Benjamin Burr. Genomic DNA isolation and genomic DNA blot analysis were performed as previously described (Chen et al. 1987). A maize DNA library was constructed with the lambda vector DASH (Stratagene) and genomic DNA homozygous for P - V V in the 4Co63 background (Chen et al. 1987). Genomic DNA was digested with SstI and cloned into the lambda vector. This library was screened with the internal 1.6 kb HindIII fragment of Ac9 (Fedoroff et al. 1983). All probes were purified from agarose gels by electro-elution and were labeled with [a32p]dCTP (New England Nuclear) by random primer extension (Freinberg and Vogelstein 1984).

5 ktl 10xTaq Pol buffer [1 x T a q Pol buffer=67mM TRIS-HCI, pH 8.8, 3 mM MgC12, 16.6 mM (NH4)2SO4] , 2 gl of the four dNTPs at 10mM, 50 pmol of each primer, and 1 unit Taq polymerase. Forty-one cycles were performed (1 min at 94° C, 2 min at 55° C and 4 min at 72° C with the incubation time at 94° C extended to 2 rain in the first cycle and the time at 72° C extended to 9 min in the last cycle). Aliquots (15 gl) of each sample were analyzed by electrophoresis on 5 % polyacrylamide gels. For Southern blot analysis, 8 ktl of each sample was electrophoresed on 1.5 % agarose gels, transferred to Nytran papers and hybridized as described previously (Chen et al. 1987).

Results

Cloning and genomic blot analysis of the cryptic A c - l l

In order to investigate the relationship between cryptic Ac-like sequences and active Ac elements, we cloned a cryptic Ac sequence from maize inbred line 4Co63. The cDNA sequence data for the Ac element (Kunze et al. 1987) suggest that a transposase is encoded by Ac. The A

B Ac probe

D N A sequence analysis. Restriction fragments from lambda clone L-11 were subcloned into Stratagene Bluescript vectors according to standard procedures (Maniatis et al. 1982). Exo III nuclease (Erase a Base, Promega) was used to generate a sequential series of overlapping deletions. Double stranded plasmid DNA was sequenced by the dideoxynucleotide chain termination method (Sanger et al. 1977) using [35S]dATP (New England Nuclear) and Sequenase according to the manufacturer's instructions (UBS). Sequence analysis was performed using GCG programs. P C R amplification and hybridization analysis. Plasmid D11 comprises the 0.7 kb SstI-PstI fragment containing the 5" end flanking sequence of cAc-ll and the 3.1 kb BamH-I-PstI fragment containing the 3' end of cAc- 11 and flanking sequence in Bluescribe M13 plus vector (Stratagene). The following oligonucleotide primers were used for polymerase chain reaction (PCR) amplification: primer P 1, 5'-GTACTTGAACTGTAACGTCTGCTCT; primer P2,5'-GGATCGTATCGGTTTTCGATTACCG; primer P3, 5'-ACGGTTGGGAAAACAACTCTACC GT; primer P4,5'-AGACCTTAGACTATCTCCATCA GCT; primer P677,5'-CGACACATGGATGGCAAGA CAAAGT; and primer P678,5'-CATCAGCATCAGG CTTCCAGTCGAA. For PCR amplification (Oste 1988) 2 gg of genomic DNA was added in a 50 gl reaction with

F1 probe

-

~

d.

d.

....

-

kb

-23 -14

m

-9.4 -6.6

-4.5

-2.8

1

2

3 4 Sst I

5

6

7

8 9 Sst I

10

Fig. 1A and B. Southern blot analysisof genomicDNA of different maize inbred lines probed with an Ac or F1 probe (see Fig. 2A). Genomic DNA isolated from leaves of differentmaize inbred lines and cAc-11 lambda clone DNA were digestedwith SstI and probed with an Ac internal HindIII fragment or a cAc-11 flankingSmaIPstI fragmentprobe (F1). A Multiplefragmentshomologousto the Ac probe in maize inbred lines without Ac activity indicated the presence of cryptic Ac elements. B The 18 kb SstI fragment that could be identifiedby the F1 probe indicated that cAc-ll related sequences were present in the lines examined

413

a single-copy sequence in the maize genome (data not shown). With the flanking probe F1 and recombinant inbreds (Burr et al. 1988), we have mapped the cAc-ll to chromosome 4S-61 (linked to RFLP locus acol, data not shown).

internal 1.6 kb HindIII fragment of Ac represents the core coding region of the Ac element transposase. Eight to ten SstI fragments in the maize genome share homology with this Ac fragment (Fig. 1A, lanes 2, 3 and 4). We used this HindIII fragment of the Ac element as a probe to isolate Ac-like sequences from a maize genomic library constructed with SstI digested fragments of 4Co63 genomic D N A and the lambda vector DASH. One of the clones, L-11, which had an 18 kb SstI insert and also hybridized to fragments flanking the 1.6 kb HindIII fragment of the Ac element, was selected for further analysis. In order to confirm that the 18 kb SstI insert of clone L-11 was equivalent to one of the SstI fragments in the maize genome, we isolated a flanking probe from the L-11 clone. This L-11 flanking probe, designated F1 (0.5 kb SmaI-PstI fragment, see Fig. 2A), was shown to hybridize to the 18 kb SstI fragment of L-11 and an 18 kb genomic DNA fragment of three maize inbred lines (Fig. 1B). Southern blot analyses, using either the Ac or the F1 probe, showed that fragments of the same size could be detected in the L-11 clone and the maize genome (data not shown). These results indicated that L-11 represented a genuine genomic clone containing a cryptic Ac element, designated cAc-ll. From a copy number reconstitution experiment, this cryptic Ac-like sequence, as identified by the specific probe F1, was shown to be

A

DNA sequence of the cryptic A c - l l and sequence comparison of the cryptic A c - l l with the canonical Ac element

Subclones and serial Exo III nuclease-mung bean nuclease deletions of the L-11 clone were generated and sequenced (Genbank Database accession number X51640). The sequence data are summarized in Fig. 2. A sequence of 4069 bp was found to be homologous to the Ac element. Eight-basepair direct repeats were also found, flanking the imperfect inverted repeats that were very similar to the inverted repeats of the canonical Ac element and the known type I Ds elements (Fig. 2B). The 11 bp inverted repeats and 8 bp target site duplication are characteristics of the Ac transposable element family. The internal part of the cAc-ll sequence shared high homology with the canonical Ac sequence with exceptions in two regions (611-704 and 3401-3822 of cAc-11). Nucleotides 611-704 of the cAc-11 sequence (605-884 in

SstI to EcoRI fragment of clone L-11 i F1 probe SmaI PstI XbaI SalI EcoRI Hindllf SphI BamHI I

I

I

I

/

SalI

/

I

PstI EcoRI

I

! ii!iiii;

SstI

1

611 704

Ac element Hind]]]

P

'1

3401 3822 4069

BamHI

EcoRI Hindm

'

llll' 605 884

u

iii

~

,m,M . .

'

El]

non-homologous

4307 4565

II

III

II

III

I

Ii

I

I

III III III

II II II

III Ill

If II

III III

B

[] Transposase 3455

II II

m R

•

|

Cryptic

0 I

lkb I

2kb I

introns

mc-11 *

homology.>90 */o

1 I RNA /,230

I!1>1 J I

,

I<1

I l

ACCCTCAGCCT T G T A A G A C C G ~ T A G G G A T A A A A

ACGG J ]

, ACCCTCAGCCTT G T A A G A C C G ~ T A G G G A T A A A A

ACGG- J

*

I-TTTCATCTC T A ~ A A

I

Ac

INNNNNNNNITAGGGATGAAA- -q

rTT TCATCCCTG INN NNNNNN]

Ds

INNNNNNNNITAGGGATGAAA- -4

I-TT TCATCC CTAJNNNNNNNN]

I I

I>I

I

I


Fig. 2A and B. Summary of sequence data of cAc-11. A Part of the physical map of clone L-11 and of the canonical Ac element are presented to show the homologous region and fragments used as probes in Southern blot analysis. An R N A transcribed from the Ac element with four introns encodes a transposase. B Sequence of the border region of cAc-11, Ac and Ds elements showed the presence of similar 11 bp inverted repeats. The asterisk indicates the bases differing in the inverted repeats. The 8 bp direct repeats flanking the inverted repeats are boxed. A 44 bp duplication with a single base difference was found at one end of the cAc- 11 sequence

414

the Ac sequence) are in the fourth intron of the Ac transcript (Kunze et al. 1987) and 3401-3822 (3601 4307 in the Ac sequence) are near the transcription initiation site but precede the translation initiation site. Open reading frames observed in this cAc-ll sequence and the sequence assembled using the splicing sites of the Ac sequence did not encode a functional transposase. A 44 bp tandem duplication was found at one end of cAc-11 (Fig. 2B). This 44 bp tandem duplication included the sequences of the 11 bp inverted repeats, the 8 bp duplication flanking the inverted repeats, a 21 bp flanking sequence, and 4 bp inside the 11 bp inverted repeats. There is a single base difference in the 8 bp duplication. Hypermethylation of cryptic Ac-11 genomic DNA When maize genomic D N A was digested with methylation-sensitive restriction enzymes, cryptic Ac DNAs migrated in agarose gels as molecules with higher mole-

Sst I

Sal I

Pvu

+

+

Sst I

I - - I I

I

Hha

Sst I I

I

Pst I

+

I

+

Sst I

Sst I I

I

I

(,O 0

~ 0

1.O 0

r,D 0

0

0

0

0

I

I

cular weights than those of Ac DNAs. Therefore it has been suggested that the D N A of cryptic Ac-like sequences is hypermethylated in the maize genome (Chen et al. 1987). The methylation level of cAc-11 was examined by Southern blot analysis of maize genomic D N A digested with cytosine methylation sensitive restriction enzymes and probing with the F1 probe. The cloned cAc-11 D N A was not methylated and could be digested with methylation sensitive enzymes. If the restriction sites in the genomic D N A were methylated, the resulting fragments should be longer than those resulting from digestion of cloned DNA. From the sizes of the restriction fragments, it was observed that most of the methylatable restriction sites, including SalI, PvuI, ClaI, and SmaI, in cAc-ll were methylated in maize genomic D N A (Fig. 3). The presence of the 7.6 kb PstI fragment and the 16.7 kb PstI fragment (Fig. 3, lane 10) indicated that the PstI site 7.6 kb from the left SstI site (see Fig. 5) was partially methylated. The 18 kb SstI fragment could not be observed even with extra genomic D N A added to the reaction or upon extended exposure of the autoradiogram, indicating that the PstI site 16.7 kb from the left SstI site was not methylated. To detect the methylation of genomic D N A upstream of the SstI site, we examined a HindlII fragment with the

(.o 0

L)

F1 probe ,,

&

,,

&

,,

&

',

&

I m

n m

kb +

+

+

+

+

+

"0,

"0

"10

"0

"0

"0

Z

"!-

"1-

"!-

"1-

"r

3

4

5

6

23-'10

11.5--

t,-

t,"

t,'-

t,"

I~

r-

6.5--

kb

4.8--

/

2.9--

23 14

g

2.0

9.5

1 .1 --

0.8--

1

2

3

4

5

6

7

8

9

10

Fig. 3. Detection of genomic D N A methylation ofcAc-11 within the 18 kb SstI fragment. Genomic D N A of 4Co63 and D N A of clone L-11 were digested with assorted restriction enzymes as indicated and probed with the F1 probe. Fragments of larger size were found in lanes 4, 6 and 8 showing that the corresponding restriction enzyme sites were methylated in genomic cAc-11 DNA. The 7.6 and 16.7 kb PstI fragments in lane t0 indicated that the PstI site 7.6 kb from the SstI site is partially methylated and the PstI site 16.7 kb from the SstI site is not methylated

1

2

7

Fig. 4. Detection of genomic D N A methylation of cAc-11 within the upstream 20 kb HindlII fragment using Southern blot analysis. Genomic D N A of 4Co63 was digested with HindlII and the indicated enzymes and then probed with the F1 probe. The 14 kb fragment in lane 4 indicated that there is an HhaI site that is not methylated in this 20 kb region

415 .............

i.i!~ii!i

wrn

't II

iii•iiii!i•i!•ii!ii!ii•ii!i•i!iii•i!i!i!iiii!ii10.5 !i!iiiii•kbi!i• Ii~;~i~i~,i';~ii~,~;~'~i~

I'

!

CO 13.

'--

03

IIIIIIIIIIIllllll]lllliilllllllll

rr

~o

LU

0 0 ~0 LU'T'U.IXUJ

L

rc~rr

I=1 ~

.-o.oo

~ . o(.~o

m

XX

o°

LU

i

ill r~

°o

LU

11111 I LIF I

~_

e_

o

The flanking region of Sst I fragment

2 kb

cAcl 1 I

rr"

o

1~1

I Sst I fragment of L-11

I

I

,

= non-methylated site

*

= partial methylated site

I

Fig. 5. Summary of physical map and methylation state of 34.7 kb cAc-11 genomic DNA region. Within the 18 kb SstI fragment every methylatable site (listed under the line representing the DNA) we examined was methylated except for one partially methylated (**) PstI site and an unmethylated (*) PstI site. In the fragment up-

stream of the 18 kb Sst[ fragment, although no clone was available for physical mapping, only one unmethylated (*) HhaI site was found. Within this 16.7 kb region, methylatable restriction sites either did not exist or were methylated

F1 probe. This HindIII fragment overlaps with the left end of the SstI fragment and extends 16.7 kb left of the SstI site (see Fig. 5). Out of six methylation-sensitive restriction enzyme digestions, only HhaI gave a smaller fragment, indicating an unmethylated HhaI site in this HindII! fragment (Fig. 4). Since there is no corresponding clone for physical mapping, we cannot determine whether there are sites for these restriction enzymes in this 16.7 kb region or not. However, statistically it is unlikely that no other methylation-sensitive restriction enzyme sites exist within this 16.7 kb region. The failure to generate smaller fragments upon digestion suggested that these sites are hypermethylated, and therefore resistant to digestion. A summary of the physical map and methylation state of cAc-11 is shown in Fig. 5.

In m a i z e g e n o m e

Transposition o f the cryptic A c - l l

To examine whether cAc-11 is capable of transposition, we used the P C R to detect the possible excision products of transposition. F o u r oligonucleotide primers flanking the 5' and 3' border of cAc- 11 were synthesized and used in the P C R (see Fig. 6). If excision of cAc-11 occurred, a fragment o f 370 bp (280 + 90 bp) would be detected by primers P1 and P4. Otherwise, the P1 and P2 primers would detect a fragment o f 421 bp in size and the P3 and P4 primers would detect a 251 bp fragment. A deletion derivative of L-11, designated D11 was used as positive control (Fig. 6 and 7C) in which a 485 bp fragment should be synthesized in the PCR reaction with primers P1 and P4. The genomic D N A of the P - W W 4Co63 and P - V V 4Co63 maize lines was used as a PCR template. The P - V V 4 C o 6 3 genome contains an active Ac at the P locus but the P - W W 4Co63 genome contains no active Ac element. For both templates, the PCR product from

280 14.1 P1 ~ i ~- ~ ] [ - I

161 P3 1

IHH P2

p o s i t i v e c o n t r o l D1 1 485

EcoRI

Hi'nd~ B a r n H I

9O

III w

P4

PI:5'-GTACTTGAACTGTAACGTCTGCTCT-$' P2:5'-GGATCGTATCGGTTTTCGATTACCG-3' PS : 5'-ACGGTTGGGA AAACAACTCTACCGT-3'

P4

P4: 5'-AG ACCTT AG ACTATCTCC ATCAGCT-3'

Fig. 6. Sequences of primers used in the polymerase chain reaction (PCR) and physical map of DNA around the cAc-ll region. Primers (P1-P4) used in the PCR to detect the possible transposition events and the predicted fragment size determined from sequence data are shown. The positive control plasmid Dll is a deletion clone of L- 11 containing the 5' flanking and 3' end and flanking sequence of cAc-11 primers P1 and P2 was 421 bp in size (Fig. 7C, lanes 1 and 2) and that from primers P3 and P4 was 251 bp in size (Fig. 7C, lanes 4 and 5). Yet, when primers Pl and P4 were used for PCR, the expected 370 bp fragment was not detected even by Southern blot analysis (Fig. 7C, lanes 7 and 8). We could identify the expected 485 bp fragment that had been synthesized in the control reaction using primers P1 and P4 and with the D11 positive control D N A as template. Even at a low concentration of the D11 D N A (10 copies in a 50 I~1reaction), we could detect the 485 bp fragment by P C R and Southern blot analysis (Fig. 7C, lanes 11 and 12). Another control was included in which D N A isolated from a maize plant (JC191B) carrying two closely linked Ae elements at the P locus (J. Chen, unpublished data) was used as template and primers P677 and P678, which flanked one of the Ac

416 A

primers

B

1+2

I

primers

3+4

I (9

primers 1+2

primers 3+4

primers 1+4

>> .~.&

M ~.

& o~ ~:

bp

>

n'

T-

&

',

~

>

&

;

&

~

Z,

>

&

;

&

A

~

7

7

7

M

522 $27 104 309 242

1

2

3

4

5

6

7

8

9

10

11

12 13

1

2

3

4

5

6

7

8

9

10

11

1213

1

2

3

4

5

6

7

8

9

10 1 1 1 2

1314

Fig. 7A-C. PCR reaction to detect the presence of cAc-ll and possible transposition of cAc-ll in different maize inbred lines. Genomic DNA isolated from leaves of different maize inbred lines was used in the PCR with primers as indicated. The resulting PCR products were separated on a 5% polyacrylamide gel. The size marker is MspI digested pBR322 fragments. With primers P1 and P2 the expected PCR product is 421 bp (A and C, lanes l, 2, and 3). With primers P3 and P4 the expected PCR product is 251 bp (B and C, lanes 4, 5 and 6). With primers P1 and P4 the expected PCR product is 370 bp, if cAc-11 does transpose. C Dll is a positive control for PCR with primers P1 and P4. A fragment of 485 bp was

produced in the reaction with D11 as template. Lane 10, l0 s molecules of plasmid D 11 were used for PCR; lane 11, 103 molecules of plasmid D11 and 2 lag ofP-WW4Co63 DNA; lane 12, 10 molecules of plasmid Dll DNA and 2 lag of P-WW4Co63 DNA. The PCR product in lanes 11 and 12 could be detected by Southern blot analysis (data not shown). In lane 13, 2 lag of genomic DNA from a maize stock (JC 191B) carrying two closely linked Ac elements was used as template in the PCR. With primers P677 and P678, a 300 bp fragment was generated, indicating that the Ac element in this stock did transpose

elements, were used in the P C R (Fig. 7C, lane 13). A fragment o f about 300 bp should be detected if the Ac transposed. When multiple copies o f Ac are present in a maize genome, the apparent transposition frequency o f each Ae element is reduced (McClintock 1948); therefore less frequent transposition o f Ac elements is expected in the JC191B stock. We did detect the expected 300 bp fragment (Fig. 7C, lane 13) as the excision product of Ac transposition from the P locus in vivo. We also examined the transposition ofcAc-11 in other maize strains, including W22, W23, CM37, T232, CO159, Tx303, and B73. Transposition o f cAc-11 related sequences was not detected in the genomes of these maize strains (data not shown). These data indicated that cAc-11 did not transpose, either autonomously or non-autonomously.

same site, genomic DNAs of different maize strains were used as templates and PCR was carried out with the primers P1 and P2, and P3 and P4. Amplified fragments of similar size to that of cAc-11 were observed (Fig. 7A and 7B). In the P1, P2 products from Tx303 and CM37, although the bands o f expected size were very faint we could detect them unambigouosly by Southern blot analysis. The low amplification yield from Tx303 and CM37 D N A s with primers P1 and P2 might be due to sequence divergence from the primers used for PCR. There were some unexpected fragments generated in the P C R but they did not hybridize to the probe (data not shown). We sequenced part of the PCR product of 4Co63, W22, W23, T232, B73, and CO159. Identical sequences were found from all strains except a single base change in CO159 (a C to T transition at nucleotide 43 of the cAc-11 sequence; data not shown).

Presence o f sequences similar to c A c - l l in many maize inbred lines Discussion

Can we find c A c - l l at the same position in different maize strains? Southern blot analyses of genomic D N A s from several inbred maize lines digested with SstI and probed with the Ac probe (Fig. 1A) or the cAc-11 flanking sequence probe F 1 (Fig. 1B) showed that a similar 18 kb SstI fragment is present in all lines tested. To confirm that these cAc-11 related sequences were inserted at the

The c A c - l l is a remnant o f a transposed Ac or Ds element

Many Ac-homologous sequences have been observed in maize stocks regardless of the presence or absence of active Ae elements. These stocks contained sequences

417 homologous to both internal and terminal portions of Ac element. We isolated and sequenced a cryptic Ac, designated cAc-11, with more than 90% homology to canonical Ac elements. The 8 bp duplication flanking the cAc-11 and 11 bp terminal inverted repeats suggested that cAc- 11 was a remnant of a transposed Ac or Ds element. Chromosomal rearrangements associated with transposition have been reported in many different transposable element systems (Schwartz-Sommer et al. 1985; Klein et al. 1988; Robbins et al. 1989), sometimes creating deletions or duplications in the sequences flanking insertion sites. A 44 bp tandem duplication, including the 8 bp flanking duplication and 11 bp terminal inverted repeats, was found at one end of cAc-11. We suggest that this 44 bp duplication is a result of an aberrant transposition or of a recombination event. The cAc-11 could have been derived from either an active Ac element or a Ds element transposed into the present target site in the presence of an active Ac element. Mutations accumulated after the insertion led to the loss of transposition ability. Homology between cAc-ll and the Ac element is more than 90% except for two regions located in the non-coding regions of the Ac elements. We suggest that these two divergent regions in cAc-11 containing many repetitive sequences may be derived from recombinational events replacing the corresponding sequence in the Ac or Ds. Owing to their repetitive nature, we did not examine the distribution of these sequences in the maize genome. Southern blot and PCR analysis of many different maize inbred lines showed the presence of cAc-11 in all of them (Figs. 1 and 7). However, there is sequence variation among them as shown by the PCR product intensity (Fig. 7A and 7B) and the sequence of the PCR products. These data indicated that the transposition event that led the cAc- 11 to its present location occurred before the separation of these inbred lines. Cryptic A c - l l cannot transpose autonomously or non-autonomously

A testcross of P ~ W W 4 C o 6 3 with a tester line that carried a Ds insertion at the R locus indicated that no active Ac element is present in the stock. These data implied that cAc-ll could not trans-activate a Ds element, cAc-ll does not contain any significant protein-coding sequence, indicating that the autonomous transposition function encoded by the active Ac element is lost as a result of the accumulated mutations. Although the terminal inverted repeats and the overall sequence of cAc-11 are highly homologous to those of the canonical Ac element, we could not detect any cAc-11 excision events (Fig. 7C). Since Ac and Ds transposition are usually, if not always, associated with excision (Chen et al. 1987), these data suggested that cAc-ll either could not transpose or transposed at an extremely low frequency, perhaps due to the single base difference in each of the cAc- 11 terminal inverted repeats. However, the inverted repeats in Ac and Ds elements are not always identical (Pohlman et al. 1984; Sutton et al. 1984), suggesting that some base changes can be tolerated. The question whether the cAc-

11 terminal inverted repeats are capable of transposition can be tested in the heterologous system of a transgenic tobacco plant carrying Ac elements (Baker et al. 1986, 1987). It has been demonstrated that deletion of subterminal fragments at either end of Ac decreases or abolishes excision, indicating that sequences other than the terminal inverted repeats also play a role in transposition (Coupland et al. 1988; Kunze and Starlinger 1989). In cAc-11, some of the AAACGG motifs, proposed to be the binding sequence for transposase (Kunze and Starlinger 1989), had base substitutions or deletions. These alterations may decrease the ability to bind transposase and cause the inability to be transposed. This possibility can be tested by binding assays in vitro (Kunze and Starlinger 1989). Cryptic Ac~ll D N A in the maize genome is hypermethylated

The regulation of transposable activities by DNA modification has been observed in many transposable element systems such as Spin, M u and Ac elements (Schwartz and Dennis 1986; Schwartz 1989; Chandler and Walbot 1986; Chomet et al. 1987; Kunze et al. 1988), whose activities were correlated with hypomethylation. It has also been suggested that a hypermethylated Ac or Ds element may not be a substrate for transposase (Schwartz 1989). In the maize genome, 26 % of cytosine residues are methylated in most tissues, while in endosperm D N A 35% of cytosine residues are methylated (Bianchi and Viotti 1988). Within the 18 kb SstI fragment containing cAc-11, at least 11 out of 13 methylatable restriction sites are methylated. This methylation level (11/13 =84.6%), compared with the average 26%-35% of cytosine methylation, indicated that the genomic region surrounding cAc- 11 is hypermethylated. It is interesting that the inactivation caused by hypermethylation usually occurred within the Ac element sequence, while the flanking sequences of the inactive transposable elements were not modified significantly (Chomet et al. 1987; Banks et al. 1988). In the case of cAc-11 its flanking regions were mostly methylated. This implied that the modification is not transposable element specific. Genomic D N A blot analysis showed that the 3' end flanking region of cAc- 11 was constituted by repetitive sequences (data not shown). In the plant genome, many hypermethylated regions have been found to contain repetitive sequences (Gupta et al. 1984). It is likely that the ancestral cAc-ll might have been inserted into a hypermethylated region, causing its own hypermethylation and consequently the inability to transpose. If methylation is the sole reason for the inactivation ofcAc-11, then a prerequisite for the activation ofcAc-11 is the removal of D N A methylation. Transformed tobacco would be a suitable system for testing this hypothesis. The Ac and Ds elements have been shown to be functional in transgenic tobacco plants as they were in the maize genome (Baker et al. 1986, 1987; Hehl and Baker 1989; Jones et al. 1989). A hypomethylated cAc-ll can be

418 cloned and transformed into tobacco with or without active A c elements. T r a n s p o s i t i o n o f cAc-11 c a n t h e n be e x a m i n e d b y P C R as d e s c r i b e d p r e v i o u s l y . Acknowledgments. We are grateful to Huei-Fun Chang, Edward

Meehan, and Fei Chen for their technical assistance and critical reading of the manuscript. This work is supported by a grant (NSC 79-0203-B001-11) from the National Science Council Taiwan, Republic of China.

References

Baker B, Schell J, L6rz H, Fedoroff N (1986) Transposition of maize controlling element Activator in tobacco. Proc Natl Acad Sci USA 83 : 4844-4848 Baker B, Coupland G, Fedoroff N, Starlinger P, Schell J (1987) Phenotypic assay for excision of the maize controlling element Ac in tobacco. EMBO J 6:1547-1554 Banks JA, Masson P, Fedoroff N (1988) Molecular mechanisms in the developmental regulation of the maize Suppressor~mutator transposable element. Genes Dev 2: 1364-1380 Berg DE, Howe MM (1989) Mobile DNA. American Society for Microbiology, Washington, DC Bianchi MW, Viotti A (1988) DNA methylation and tissue-specific transcription of the storage protein genes of maize. Plant Mol Biol 11 : 203-214 Burr B, Burr FA, Thompson KH, Albertson MC, Stuber CW (1988) Gene mapping with recombinant inbreds in maize. Genetics 118 : 519-526 Chandler VL, Walbot V (1986) DNA modification of a maize transposable element correlates with loss of activity. Proc Natl Acad Sci USA 83:1767-1771 Chandler VL, Rivin C, Walbot V (1986) Stable non-mutator stocks of maize have sequences homologous to the M u l transposable element. Genetics 114: 1007-1021 Chandler VL, Talbert LE, Raymond F (1988) Sequence, genomic distribution and DNA modification of a M u l element from non-mutator maize stocks. Genetics 119:951-958 Chen J, Greenblatt IM, Dellaporta SL (1987) Transposition of Ac from the P locus of maize into unreplicated chromosomal sites. Genetics 117:109-116 Chomet PS, Wessler S, Dellaporta SL (1987) Inactivation of the maize transposable element Activator is associated with its DNA modification. EMBO J 6: 295-302 Coupland G, Baker B, Schell J, Starlinger P (1988) Characterization of the maize transposable element Ac by internal deletions. EMBO J 7:3653-3659 Dellaporta SL, Chomet PS (1985) The activation of maize controlling elements. In: Hohn B, Dennis ES (eds) Genetic flux in plants. Springer-Verlag, Wien, pp 169-216 D6ring HP, Tillmann E, Starlinger P (1984) DNA sequence of the maize transposable element Dissociation. Nature 307:127-130 Fedoroff N, Wessler S, Shure M (1983) Isolation of the transposable maize controlling elements Ac and Ds. Cell 35:235-242 Feinberg AP, Vogelstein B (1984) A technique for radiolabeling DNA restriction fragments to high specific activity. Anal Biochem 132:6-13 Flavell R (1982) In: Genome evolution. Dover GA, Flavell RB (eds) Academic Press, London, pp 301-323 Gupta M, Shepherd NS, Bertram I, Saedler H (1984) Repetitive sequences and their organization on genomic clones of Zea mays. EMBO J 3:133-139

Hehl R, Baker B (1989) Induced transposition of Ds by a stable Ac in crosses of transgenic tobacco plants. Mol Gen Genet 217:53-59 Jones JG, Carland FM, Maliga P, Dooner HK (1989) Visual detection of transposition of the maize element Activator in tobacco seedlings. Science 244: 204-207 Klein AS, Clancy M, Paje-Manalo L, Furtek DB, Hannah LC, Nelson OE (1988) The mutation bronze-mutable 4 Derivative 6856 in maize is caused by the insertion of a novel 6.7-kilobase pair transposon in the untranslated leader region of the bronze-l gene. Genetics 120:77%790 Kunze R, Starlinger P (1989) The putative transposase of transposable element Ac from Zea mays L. interacts with subterminal sequences of Ac. EMBO J 8:3177-3185 Kunze R, Stochaj U, Laufs J, Starlinger P (1987) Transcription of transposable element Activator of Zea mays L. EMBO J 6:1555-1563 Kunze R, Starlinger P, Schwartz D (1988) DNA methylation of the maize transposable element Ac interferes with its transcription. Mol Gen Genet 214:325-327 Maniatis T, Fritsch EF, Sambrook J (1982) Molecular cloning: A laboratory manual. Cold Spring Harbor Press, New York McClintock B (1946) Maize genetics. Carnegie Inst Washington Year Book 45:176-186 McClintock B (1948) Mutable loci in maize. Carnegie Inst Washington Year Book 47:155-169 McClintock B (1978) Mechanisms that rapidly reorganize the genome. Stadler Symp 10:25-47 McClintock B (1984) The significance of responses of the genome to challenge. Science 226: 792-801 Mfiller-Neumann M, Yoder JI, Starlinger P (1984) The DNA sequence of the transposable element Ac of Zea mays L. Mol Gen Genet 198:9-24 Oste C (1988) Polymerase chain reaction. Biotechniques 6:162-167 Peschke VM, Phillips RL, Gengenbach BG (1987) Discovery of transposable element activity among progeny of tissue culturederived maize plants. Science 238:804-807 Pohlman R, Fedoroff NV, Messing J (1984) The nucleotide sequence of the maize controlling element Activator. Cell 37: 635-644 Robbins TP, Carpenter R, Coen ES (1989) A chromosome rearrangement suggests that donor and recipient sites are associated during Tam3 transposition in Antirrhinum majus. EMBO J. 8:5-13 Sachs MM, Peacock WJ, Dennis ES, Gerlach WL (1983) Maize AciDs controlling element - a molecular viewpoint. Maydica 28 : 289-303 Sanger F, Nicklen S, Coulson AR (1977) DNA sequencing with chain terminating inhibitors. Proc Natl Acad Sci USA 74: 5463-5467 Schwartz D (1989) Gene-controlled cytosine demethylation in the promoter region of the Ac transposable element in maize. Proc Natl Acad Sci USA 86:2789-2793 Schwartz D, Dennis ES (1986) Transposase activity of the Ac controlling element in maize is regulated by its degree of methylation. Mol Gen Genet 205:476-482 Schwartz-Sommer Z, Gierl A, Cuypers H, Peterson PA, Saedler H (1985) Plant transposable elements generate the DNA sequence diversity needed in evolution. EMBO J 4:591-596 Sutton WD, Gerlach WL, Schwartz D, Peacock WJ (1984) Molecular analysis of Ds controlling element mutations at the adhl locus of maize. Science 223 : 1265-1268 C o m m u n i c a t e d b y H. S a e d l e r