Journal of General Microbiology (1989, 131, 951-958.
Printed in Great Britain
95 1
Three Restriction Endonucleases from Anabaena pos-a4uaae
B y P H I L I P R. W H I T E H E A D 7 A N D N I G E L L. B R O W N *
Department of Biochemistry, University of Bristol, University Walk, Bristol BS8 1 TD, UK
(Received 20 August 1984)
Three site-specific endonucleases, AflI, AflII and AflIII, have been partially purified from the
cyanobacterium Anabaenaflos-aquae CCAP 1403/13f. Their recognition and cleavage specificities have been determined to be:
5’-G&GAflII
AflIII
(“T) -C-C-3’
5‘-C-T-T-A-A-G-3‘
1
5’-A-C-Pu-Py-G-T-3’‘
1
AflII and AflIII are new specificities and may be useful in molecular cloning, as well as in the
analysis of DNA. The distribution of type I 1 restriction endonucleases in the cyanobacteria is
briefly discussed.
INTRODUCTION
Type I 1 restriction endonucleases are widely used in the analysis and restructuring of DNA
molecules, and provide good systems for the study of DNA-protein interactions. The current
compilation of restriction endonuclease activities (Roberts, 1984) contains 475 such activities,
representing a minimum of 103 different specificities from 356 bacterial strains. Enzymes with
new specificities are particularly useful in that they increase the range of DNA manipulations
that can be performed.
The cyanobacteria are a rich source of type I 1 restriction endonucleases. Both the filamentous
cyanobacteria (Anabaena, Nostoc) and the unicellular cyanobacteria (Aphanothece, Dactylococcopsis) contain these activities (Murray et al., 1976; Reaston et al., 1982; Whitehead & Brown,
1982; N. L. Brown, unpublished observations). In many cyanobacteria there are multiple
enzymes in a single strain, and one strain of Nostoc contains five such activities (Reaston et al.,
1982). Some of the enzymes are isoschizomers (i.e. enzymes with the same recognition
specificity occurring in different strains), and these often occur in a variety of combinations with
other endonucleases in different strains of cyanobacteria. For example, Nostoc sp. PCC 7524
(Reaston et al., 1982) contains enzymes with the recognition specificities of NspI, AvaI, A d ,
AsuII and SduI; Nostoc sp. PCC 7413 contains enzymes recognizing NspI and AvaII sites (M.
G . C. Duyvesteyn, J. Reaston & A. deWaard, unpublished observations reported in Roberts,
1984); Anabaena variabilis ATCC 27892 contains AvaI, AvaII and AvaIII (Murray et al., 1976;
Reaston et al., 1982); Anabaena subcylindrica contains AsuI, AsuII and an enzyme recognizing
the AcyI site (deWaard & Duyvesteyn, 1980); Anabaena cylindrica contains AcyI (deWaard et
al., 1978).
We have screened a number of bacterial strains as part of a search for novel restriction endonuclease activities. In this paper we describe the partial purification and characterization of
three type I 1 restriction endonucleases in the strain Anabaena floss-aquae CCAP 1403/13f.
t Present address: Biogen SA, Route de Troinex 3,
1227 Carouge, Geneva, Switzerland.
Abbreviations : Pu, purine nucleoside ; Py , pyrimidine nucleoside.
0001-2157 0 1985 SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 04:33:29
952
P. R. WHITEHEAD A N D N. L. BROWN
METHODS
DNA and enzymes. Bacteriophage lcI857Sum7 DNA, bacteriophage 4 x 1 74am3cs70 replicative form I (RFI)
DNA, plasmid pA03 DNA and pBR322 DNA were prepared as described previously (Brown et al., 1980).
Adenovirus-2 DNA was a gift of Dr Janet Arrand, St Mary's Hospital Medical School, London, UK, and
recombinant bacteriophage M13mp2 and M13mp7 DNAs containing fragments of Tn5Ol or pBR322 DNA were
gifts of Dr D. C. Fritzinger and R. D.Pridmore. Restriction endonucleases, DNA polymerase and other DNA
modifying enzymes were purchased from Boehringer, New England Biolabs or BRL.
Purification of restriction endonucleases from Anabaena flos-aquae. Packed cells of Anabaena flos-aquae
CCAP 1403/13f were provided by Professor A. E. Walsby, Department of Botany, Bristol University, UK, and
were stored at - 70 "C. These had been grown in the medium described by Walsby & Booker (1980) at 25 "C and
sparged with air under 500 lx illumination (rising to 1500lx later in growth). Frozen packed cells (1 1.5 g) were
sonicated in 20 ml extraction buffer (10 m~-2-mercaptoethanol,10 mM-Tris/HCl, pH 7.5) for 20 x 30 s at 60 W,
while maintaining the temperature below 4 "C.The cell suspension was centrifuged at l0000Og for 90 min. The
supernatant was made 1.0 M with respect to NaCl and applied to a Sepharose 4B column (50 x 2 cm in 1.0 M-NaCl,
10 m~-2-mercaptoethanol,10 mM-Tris/HCl, pH 7.5) in order to separate the endonucleases from the bacterial
DNA. Fractions containing endonuclease activity were pooled and dialysed against PC buffer (lo%, v/v, glycerol;
0.1 mM-EDTA; 10 mM-2-mercaptoethanol; 10 mwpotassium phosphate, pH 7-4) and applied to a phosphocellulose P-11 column (15 x 2cm) in PC buffer. The column was washed in PC buffer and developed with a
gradient (200 ml) of 0-1.0 M-KCl in PC buffer. Endonuclease-containing fractions were identified and pooled
according to their patterns of digestion on bacteriophage L DNA, and dialysed against PC buffer. Fractions
eluting at 0 - 1 4 4 ~ - N a C contained
l
the two enzymes AflI and AflII, and those eluting at 0.4-0.6~-NaCl
contained the two enzymes AflII and AflIII. These enzymes were separated on DEAE-cellulose columns
(15 x 1.5cm) developed with gradients (100ml) of 0-1.0M-NaCl in PC buffer. The separate enzymes were
dialysed against PC buffer and further purified by chromatography on heparin-Sepharose columns (5 x 1 cm)
developed with a gradient of 0-1.0 M-NaCl in PC buffer. The salt concentrations at which the enzymes elute from
the columns were measured using a conductivity meter, and are given in Table 1.
Endonuclease assays. Column fractions were assayed for site-specific endonuclease activity by incubating 1-5 pl
with 1 pg bacteriophage LcI857Sum7 DNA in 25 pl 10 m~-2-mercaptoethanol,10 m~-MgCl,, 10 mM-Tris/HCl,
pH 7.9, at 37 "C for between 0.5 and 16 h. Each reaction was stopped by adding EDTA to 10 mM, and analysed by
electrophoresis on 1 % agarose slab gels in 90 mM-Tris/borate, pH 8.3, 2.5 mM-EDTA, 0.5 pg ethidium bromide
ml-I . The ethidium bromide-stained DNA was visualized on a long-wave UV transilluminator (Ultra-violet
Products, San Gabriel, Calif., USA) and photographed on Polaroid type 667 film through a Schott-Jena KV470
filter.
The optimum conditions of pH, temperature and monovalent cation concentration for cleavage of DNA with
API, AflII and AflIII were determined by altering these parameters singly from the standard conditions (37 "C in
50 mM-NaC1, 10 mM-2-mercaptoethanol, 10 m~-MgCl,, 10 mM-Tris/HCl, pH 7.5). Assays were performed on
0.5 pg bacteriophage h1857Sam7 DNA for a time which did not quite give complete digestion of the product
under standard conditions.
A unit of endonuclease activity is defined as that required to digest to completion 1 pg of bacteriophage
LcI857Sam7 DNA in 1 h in a 50p1 reaction volume under optimal buffer and temperature conditions.
Determination of'the recognition and cleavage specificity of AflI, AflII and AflIII. The recognition and cleavage
specificities of the enzymes were determined by the method of Brown & Smith (1980). The numbers of cleavage
sites for each enzyme on a variety of fully sequenced DNA molecules were determined in both single digests and in
digests with another enzyme. These data were then compared with the computer-generated tables of Fuchs et al.
(1980) in order to indicate possible recognition sites. Single-stranded recombinant Ml3mp2 and M13mp7 DNA
derivatives (Gronenborn & Messing, 1978; Messing et al., 1981)containing a DNA fragment with a single site for
the restriction enzyme were obtained from a DNA-sequencing project in this laboratory (Brown et al., 1983; Diver
et al., 1983). The method for identifying the phosphodiester bonds cleaved by the restriction endonuclease is
described in detail elsewhere (Brown & Smith, 1980; Whitehead & Brown, 1982).
RESULTS
Partial purijcatwn of the site-spec@ endonuclease activities
Three different site-specific endonuclease activities were isolated from Anabaena Jlos-aquae
CCAP 1403/13f. The salt concentrations at which these activities eluted from the chromatographic media used are given in Table 1 . The protocol is designed to remove contaminating
nuclease activities, and not to maximize the specific activity of each endonuclease. The
approximate final yields of each enzyme per g wet weight of cells were: AJlI, 250 units; AJZII,
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 04:33:29
953
Restriction enzymes from Anabaena 30s-aquae
Table 1. Conditions for the elution of AJI restriction endonucleases from chromatographic media
Elution position*
Restriction
endonuclease
P hosphocellulose
P-11
AJI
AJII
AfrIII
0-1-0.4
0.1-0.6
DEAE-cellulose
Heparin-Sepharose
CL-6B
0.1-0-3
0.4-0.7
0-2-0.4
0.5-0-6
* Molarity of NaCl in PC buffer (lo%, v/v, glycerol; 0.1 mM-EDTA; 10 m~-2-mercaptoethanol; 10 mMpotassium phosphate, pH 7.4).
Table 2. Observed frequency of cleavage of sequenced DNA molecules by restriction
endonucleases from A. Jos-aquae
Minimum number of sites for
endonuclease*
DNA
lcI857Sam7
Adenovirus-2
sv40
pBR322
I$X 174am3cs70
M13
pA03
r
1
AfrI
AjIII
Aj?III
>25 (35)
>34 (73)
5 (6)
6 (8)
1 (1)
3 (3)
4 (4)
1 (1)
>17 (20)
>20 (25)
0 (0)
2 (2)
0 (0)
2 (2)
0 (0)
1 (1)
0 (0)
1 (1)
2 (2)
3 (3)
1 (1)
* These are the minimum numbers of sites, as two sites close together may not be distinguished from one
another. Numbers in parentheses give the number of sites predicted by computer search routines, and are equal to
or greater than the observed minimum number of sites, indicating that the enzymes are substantially free of
contaminating specific endonucleases.
300 units; Afr 111, 100 units. The absence of contaminating non-specific nucleases was
demonstrated by digestion of bacteriophage DNA with a 10-fold excess of enzyme for 1 h,
followed by gel electrophoresis of the products, which gave sharp bands on the gel; and also by
the DNA sequence analysis of the cleavage sites (see below). This latter method did show a very
small amount of contamination of one batch of AfrI 1 by a specific endonuclease (see Fig. 2 b) but
this was not seen in other preparations. Exonuclease contamination of the enzymes would give
rise to a series of bands in channels I and I 1 of the site location experiment due to the successive
removal of nucleotides from the fragment termini; and this was not seen.
The three endonucleases showed maximal activity at 37 "C in 50 mM-NaC1, 10 m ~ - 2 mercaptoethanol, 10 mM-MgCl,, 10 mM Tris/HCl, pH 7.5.
Determination of the recognition and cleavage specijicities
The observed minimum numbers of cleavage sites of the three endonucleases on a variety of
DNAs are given in Table 2. These data are compiled from both single-enzyme digests and
digests with a second enzyme. Fig. 1 shows the results of gel electrophoresis of the products of
digestion of bacteriophage L DNA or adenovirus-2 DNA with AJII, AfrII or AflIII.
The logic for determining the recognition and cleavage sequences is different for each of the
three enzymes, and will be considered separately.
The Afllsite. The sizes and number of fragments produced by AJII digestion of bacteriophage
L DNA were similar to those produced by AvaII. Digestion of pBR322 DNA with AflI or with
AvaII showed that the fragment patterns were identical, and a double digest of pBR322 with
both enzymes together produced no additional fragments (data not shown). The restriction
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 04:33:29
954
P . R . WHITEHEAD A N D N . L . BROWN
Fig. 1. Products of cleavage of bacteriophage lcI857Sarn7 DNA and adenovirus-2DNA with (a) ,431,
(b) ,4311 and (c) AJIII, separated by electrophoresis in 1 % agarose in 90m-Tris/borate, pH 8.3,
0-25mM-EDTA, 0.5 pg ethidium bromide ml-*.
fragment patterns produced by cleavage of fully sequenced DNAs with AfrI agree with the
patterns predicted for an enzyme recognizing the sequence
A recombinant DNA molecule containing a HgiAI fragment of pBR322 DNA cloned into
bacteriophage M13mp2 was provided by Dr D. C. Fritzinger. This contained the AvaII site at
position 1135 of pBR322 DNA (Sutcliffe, 1979). The single-stranded bacteriophage DNA was
used as a template to determine the site of AfrI cleavage in both DNA strands by the method of
Brown 8z Smith (1980). The autoradiographof the sequencing gel is shown in Fig. 2(u). The sites
of cleavage of AfrI within this sequence are:
1
5’-T-C-A-T-T-G-G-A-C-C-G-C-T-G-A-3’
3’-A-G-T-A-A-C-C-T-G-G-C-G-A-C-T-5’
t
demonstrating that AfrI cleaves the same phosphodiester linkages in this sequence as does
AvuII. Some radioactive fragment in channel I1 remains unprocessed by DNA polymerase; this
is probably due to AIpI nicking one strand of the DNA, and the nicked DNA not being a
substrate for the DNA polymerase reaction (Carter et ul., 1980).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 04:33:29
955
Restriction enzymes from Anabaena Jlos-aquae
I T C G A I I
(4
I
T C G A
I1
I T C G A
I1
(4
(b)
Fig. 2. Autoradiographsof sequencinggels locating the cleavage sites of (a)@I, (b)APII and (c) APIII
in the nucleotide sequencesof M13 recombinant DNAs (see text). Channels T, C, G and A are the basespecific tracks of the chain-terminationsequencingmethod (Sanger et d.,1977). Channel I locates the
phosphodiester bond cleaved in the sequencedstrand, and channel I1 locates that in the template strand
(see Brown & Smith, 1980).
The AJIII site. The two AJIII sites on bacteriophage 4X174am3cs70RF1 DNA (Table 1) were
mapped by standard double-digestionmethods to position 4400 ( & 20) and either position 2420
( 20) or 2920 ( k 20) on the 4x174 sequence (Sanger et al., 1978). Inspection of the sequence in
these regions, and reference to computer-generatedtables (Fuchs et al., 1980) revealed that the
sequence 5’-C-T-T-A-A-G-3’
is present at positions 4413 and 2914, and nowhere else in the
4x174 sequence. The observed AJIII cleavage patternsof other fully sequenced DNAs (Table 2)
were those predicted for an enzyme cleaving at this hexanucieotide sequence.
A recombinant bacteriophage M 13mp7 containing a TaqI fragment of transposon Tn501 was
identified which contained the sequence postulated to be the AflII recognition site. This is at
position 4764 of the complete Tn501 sequence (N. L. Brown, R. D. Pridmore & D. C. Fritzinger,
unpublished data; position 535 in that part of the Tn501 sequence published in Diver et al.,
1983). The autoradiograph from the cleavage site-location experiment (Brown & Smith, 1980)
using this recombinant DNA is shown in Fig. 2(b). The site of AJlII cleavage within this
sequence is shown by the solid arrows:
1
5’-G-G-C-G-G-C-T-T-A-A-G-T-G-C-Tk-3’
1
3’-C-C-G-C-C-G-A-A-T-T-CrA-C-G-A-G-5’
7 :
,4811 cleaves the hexanucleotide sequence symmetrically, to give fragments with 5’-terminal
extensions. On prolonged incubation faint secondary cleavage sites are observed (Fig. 2b) as
shown by the dotted arrows. These secondary cleavage sites are probably due to a minor
contaminating specific endonuclease activity. This contaminant was not reproducibly found,
and we have been unable to further characterize the activity. It has not been seen on normal
restriction digests, only in the very sensitive site-locationanalyses, and it does not interfere with
the normal use of AJIII.
The AJIIII site. The AJIIII cleavage sites on pBR322 DNA and on 4x174 RFI DNA were
mapped by standard double-digestionmethods to position 2480 ( f 20) on pBR322 and positions
220 (_+20)and 2159 (k20) on 4x174. The tables of Fuchs et al. (1980) suggested that the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 04:33:29
956
P. R. WHITEHEAD AND N . L. BROWN
sequence 5’-A-C-Pu-Py-G-T-3’
was the best candidate for the AflIII recognition sequence.
Members of this family of related sequencesare found in pBR322 DNA at position 248 1 (5’-AC-A-T-G-T-3’) and in 4x174 RFI DNA at positions 221 and 2146 (both 5’-A-C-G-C-G-T3’). Comparisonof the patterns of cleavage of fully-sequenced DNAs by AflIII with the patterns
predicted for an enzyme cleaving at this proposed site gave good correlation, indicating that this
is indeed the recognition site of AflIII.
A recombinant bacteriophage, M13mp2, containing the ThaI fragment from positions 21792520 of pBR322 DNA was used as template DNA for the site-location experiment. The
autoradiographshowing the results of this experiment is shown in Fig. 2(c). AflIII cleaves in the
sequence:
1
5’-T-G-C-T-C-A-C-A-T-G-T-T-C-T-T-T-3’
3‘A-C-G-A-G-T-G-T-A-C-
t
A- A-G- A- A- A-5‘
The cleavage sites are symmetrically placed within the hexanucleotide recognition site, and
AflIII cleaves to give 5’-tetranucleotide terminal extensions.
DISCUSSION
We have described the partial purification and characterization of three restriction endonucleases from Anabaenaflos-aquae CCAP 1403/13f. A fourth activity was detected but not
characterized. AflI recognizes and cleaves the sequence :
5’-GiG-
6)
-C-C-3‘
and has the same specificity as AvaII. The remaining activities, AflII and AflIII, are novel and
recognize the sequences:
S-C-T-T-A-A-G-3’
1
and
5’-A-C-Pu-Py-G-T-3’
1
respectively. Both of these enzymes cleave to give 5’-tetranucleotide terminal extensions, and
may be useful in molecular cloning, although the termini produced by AfIIII are degenerate and
cannot be annealed to all other AflIII fragments. Both AflII and AflIII are further additions to
the range of enzymes available for DNA manipulation.
A.flos-aquae CCAP 1403/13fcontains at least three (and possibly more than four) site-specific
endonuclease activities. One of these (AflI) has the same cleavage specificity as another
Anabaena endonuclease, AvaII (Hughes & Murray, 1980). Endonucleases with this specificity
are also found in the cyanobacteria Fremyella and Nostoc, as well as in Bacillus, Caryophanon,
Chloroflexus, Escherichia, Herpetosiphon and Salmonella (Roberts, 1984). There is no evidence to
suggest that these endonucleasesin the ‘AvaII family’ are proteins with a common evolutionary
origin, but it is tempting to speculate that these enzymes are very closely related and that they
and their corresponding modification methylases have been mobilized in some way. Such
mobilization could occur by the genes being plasmid-borne or bacteriophage-borne functions.
Other familiesof isoschizomericrestriction endonucleasesexist, such as the PstI family and the
EcoRII family (see Roberts, 1984), and these may also be the products of widely distributed
genes.
Another notable feature of the distribution of restriction endonucleases in the cyanobacteria
is that many strainscontain multiple restriction endonuclease activities, and that these activities
occur in different combinationsin different strains. This is likely to be one of the factors causing
difficultyin establishing a gene transfer system in the filamentous cyanobacteria (Wolk et al.,
1984). Duyvesteynet al. (1983) have shown that the variation in restriction endonucleasecontent
between strains may be important in the biological restriction of cyanophage DNA, and the
multiple endonuclease activities may be required for adequate protection against the large
number of cyanophages.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 04:33:29
Restriction enzymes from Anabaena flos-aquae
957
Lambert & Carr (1984) presented data which suggest that the filamentous and unicellular
cyanobacteria differ in the extent to which their DNA is modified. DNAs from the filamentous
genera (Anabaena, Nostoc, etc.) are extensively modified in a manner independent of the type I1
restriction/modification systems found in each strain, whereas DNAs from the unicellular
cyanobacteria (Anacystis, Aphanothece, etc.) show variation in the extent of modification. The
presence of other DNA-modifying enzymes in the cyanobacteria complicates the interpretation
of the functional role of multiple type I1 restriction/modification systems in the cyanobacteria.
It may be that the presence of multiple type I1 restriction/modificationsystems, in which
different specificities occur in different combinations in different strains, provides an 'index of
relatedness' between cyanobacteria and permits gene transfer in vivo between certain strains, but
not between others. Further characterization of the distribution of type I1 restrictionlmodification systems in the cyanobacteria is required to test this idea.
We are grateful to Annette Bees and Professor A. E. Walsby for providing cell paste from Anabaenajos-aquae,
and to the latter for discussion.We thank Dr D. C. Fritzinger, Dr Janet Arrand and R. D. Pridmore for providing
DNA. This work was supported by a Medical Research Council grant (no. G.9781047CB) to N.L.B., who is a
Royal Society EPA Cephalosporin Fund Senior Research Fellow.
REFERENCES
BROWN,
N. L. & SMITH,M. (1980). A general method
for defining restriction enzyme cleavage and recognition sites. Methods in Enzymology 65, 391404.
BROWN,
N. L., MCCLELLAND,
M. &WHITEHEAD,
P. R.
(1980). HgiAI : a restriction endonuclease from
Herpetosiphon giganteus HP1023. Gene 9, 49-68.
BROWN,N. L., FORD,S. J., PRIDMORE,
R. D. &
FRITZINGER,
D. C. (1983). Nucleotide sequence of a
gene from the Pseudomonas transposon TnSO1
encodingmercuric reductase. Biochemistry 22,40894095.
CARTER,J. A., CHATER,K. F., BRUTON,C. J. &
BROWN,N. L. (1980). A comparison of DNA
cleavage by restriction enzymes SalPI and Pst I.
Nucleic Acids Research 8, 4943-4954.
M. G. C. (1980). Are
DEWAARD,
A. & DUYVESTEYN,
sequence-specific deoxyribonucleases of value as
taxonomic markers of cyanobacterial species?
Archives of Microbiology 128, 242-247.
DEWAARD,
A., KORSUIZE,
J., VAN BEVEREN,
c. P. &
MAAT,J. (1978). A new sequence-specific endonuclease from Anabaena cylindrica. FEBS Letters 96,
106-1 10.
DIVER,W. P., GRINSTED,
J. G., FRITZINGER,
D. C.,
P. &
BROWN,
N. L., ALTENBUCHNER,
J., ROGOWSKY,
SCHMITT,
R. (1983). DNA sequencesof,and complementation by the tnpR genes of Tn21, Tn501 and
Tn1721. Molecular and General Genetics 191, 189193.
DUYVESTEYN,
M.G. C., KORSUIZE,
J., DEWAARD,
A.,
VONSHAK,
A. & WOLK,C. P. (1983). Sequencespecific endonucleases in strains of Anabaena and
Nostoc. Archives of Microbiology 134, 276-28 1.
FUCHS,C., ROSENVOLD,
E. C., HONIGMEN,
A. &
SZYBALSKI,
W. (1980). Identification of palindromic
sequences recognised by restriction endonucleases,
as based on the tabularized sequencing data for
seven viral and plasmid DNAs. Gene 10, 357370.
GRONENBORN,
B. & MESSING,
J. (1978). Methylation of
single-strandedDNA in vitro introduces new restric-
tion endonuclease cleavage sites. Nature, London
212, 375-377.
HUGHES,S. G. & MURRAY,
K.(1980). The nucleotide
sequences recognised by endonucleases AvaI and
AvaII from Anabaena variabilis. Biochemical Journal
185, 65-75.
LAMBERT,
G. R. & CARR,N. G. (1984). Resistance of
DNA from filamentous and unicellular cyanobacteria to restriction endonuclease cleavage. Bwchimica et biophysica acta 781, 45-55.
MESSING,J., CREA,R. & SEEBURG,
P. H. (1981). A
system for shotgun DNA sequencing. Nucleic Acids
Research 9, 309-32 1.
MURRAY,
K., HUGHES,S. G., BROWN,
J. S. &BRUCE,
S. A. (1976). Isolation and characterization of two
sequence-specific endonucleases from Anabaena
variabilis. Biochemical Journal 159, 317-322.
REASTON,
J., DUYVESTEYN,
M.G. C. & DEWAARD,
A.
(1982). Nostoc PCC 7524, a cyanobacterium which
contains five sequence-specific deoxyribonucleases.
Gene 20, 103-1 10.
ROBERTS,R. J. (1984). Restriction and modification
enzymes and their recognition sequences. Nucleic
Acids Research 12, t 167-r204.
S. & COULSON,
A. R. (1977).
SANGER,
F., NICKLEN,
DNA sequencingwith chain-terminating inhibitors.
Proceedingsof the Natwnal Academy of Sciences of the
United States of America 14, 54635467.
SANGER,
F., COULSON,
A. R., FRIEDMANN,
T., AIR,
G. M., BARRELL,
B. G., BROWN,N. L., FIDDES,
J. C., HUTCHISON,
C. A., 111, SLOCOMBE,
P. M. &
SMITH, M. (1978). The nucleotide sequence of
bacteriophage 4x174. Journal of Molecular Biology
125, 225-246.
SUTCLIFFE,
J. G.(1979). Complete nucleotide sequence
of the Escherichia coli plasmid pBR322. Cold Spring
Harbor Symposium in Quantitative Biology 43, 7790.
WALSBY,
A. E. &BOOKER,
M. J. (1980). Changes in the
buoyancy of a planktonic blue-green algae in
response to light intensity. British Phycological
Journal 15, 311-319.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 04:33:29
958
P . R . WHITEHEAD A N D N . L . BROWN
WHITEHEAD,
P. R. & BROWN,N. L. (1982). AhaIII: a
restriction endonuclease with a recognition site
containing only A :T basepairs. FEBS Letters 143,
296300.
WOLK,C. P., VONSHAK,A., KEHOE,P. & ELHAI,J.
(1984). Construction of shuttle vectors capable of
conjugative transfer from Escherichia coli to nitrogen-fixing filamentous cyanobacteria. Proceedings of
the National Academy of Sciences of the United States
ofAmerica 81, 1561-1565.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 19 Jun 2017 04:33:29