Small multicopy plasmids in three strains of halophilic archaea, SB3, GRB, and GN101, were found to be present in a cell as a population of related but not ...
Vol. 175, No. 4
JOURNAL OF BACTERIOLOGY, Feb. 1993, p. 1081-1086
0021-9193/93/041081-06$02.00/0 Copyright © 1993, American Society for Microbiology
Heterogeneity of Small Plasmids from Halophilic Archaea A. S. AKHMANOVA,t V. K. KAGRAMANOVA, AND A. S. MANKINt* of Molecular Biology and Bioorganic Chemistry, Moscow State University, Laboratory Belozersky 119899 Moscow, Russia Received 22 September 1992/Accepted 4 December 1992
Small multicopy plasmids in three strains of halophilic archaea, SB3, GRB, and GN101, were found to be present in a cell as a population of related but not identical sequences. Two types of heterogeneity were observed: macroheterogeneity, represented by two major plasmid sequence versions homologous to each other by 80%o, and microheterogeneity, in which individual plasmids differ by one or a few nucleotide substitutions.
Plasmids are self-replicating extrachromosomal genetic elements present in a cell in multiple copies. Usually, all copies of the same plasmid have the same nucleotide sequence. Similar but not identical plasmids generated by mutation are normally incompatible and rapidly segregate so that eventually only one plasmid variant is found in the cell culture (for reviews, see references 3 and 17). The presence of extrachromosomal circular DNA is characteristic of many species from the domain Archaea (23). Thus, several strains of halophilic archaea contain plasmids that range in size from 30 to 100 kbp and exist in a relatively low number of copies per cell (1, 4-6, 11, 18, 19). In addition, a small plasmid (about 1.7 kbp) with a high copy number has been found in three halobacterial isolates, SB3, GRB, and GN101 (5). The small plasmids from these three strains are closely related, as originally revealed by hybridization analysis (5) and later by sequencing (7-9, 14), but share no detectable homology either with the chromosome of the host strains or with DNAs of other halobacteria. The existence of a small amount of a single-stranded plasmid form in the cell may represent an intermediate generated by rolling-circle replication (22). While cloning and sequencing a small plasmid from the SB3 strain, we observed that some of the presumably identical recombinant plasmids isolated from the transformed Escherichia coli colonies in fact differed from each other by a few base changes (13). This unexpected phenomenon was examined in detail, and in this report we present evidence that the small plasmids in a halobacterial cell do not have a unique sequence but, instead, exist as a population of closely related sequences. Two main types of heterogeneity were detected: microheterogeneity, in which individual plasmids differ from each other by one or a few nucleotide substitutions, and macroheterogeneity, represented by two major plasmid versions exhibiting about 80% homology.
N. Hackett. Cells were grown in liquid complex media (2) at 370C with constant agitation or on 1.5% agar plates prepared with the same media diluted with one-fifth of a volume of water. E. coli plasmids were propagated in strains XL1 and DH5a. Cultures of E. coli cells were grown in 2 x YT medium in the presence of 50 pug of ampicillin per ml (20). All of the enzymes used in this work were from Biopreparat (Vilnius, Lithuania). Plasmid isolation and DNA analyses. Halobacterial colonies were isolated by a spreading technique (20), and cell cultures were grown from individual clones. Cells were harvested at the early stationary phase (A550, -1.2), and total plasmids were isolated from cultures by a conventional NaOH-sodium dodecyl sulfate procedure (20) with omission of lysozyme from the lysis buffer. RNA was removed by RNase treatment and polyethylene glycol precipitation (10), and the supercoiled form of a 1.75-kbp plasmid was further purified from a 0.8% agarose gel. Recovered plasmid DNA was subjected to restriction analyses, and the resulting DNA fragments were resolved in agarose or acrylamide gels. Gels were stained with ethidium bromide, and the relative intensity of DNA bands was quantified with an Ultrascan (LKB) scanner. Cloning and sequencing of halobacterial plasmids. Genetic engineering manipulations were performed as described elsewhere (20). A small plasmid from Halobacterium strain SB3 was linearized by SalI hydrolysis and ligated with the pUC19 vector, which was cut with the same enzyme. The ligation mixture was transformed into E. coli DH5aL cells. Recombinant plasmids were isolated from several individual colonies and analyzed by digestion with the Sau3A, TaqI, MspI, Eco47I, or BspRI restriction enzyme, which allowed discrimination of pHSB1-pUC19 from pHSB2-pUC19 (see Re-
sults). For sequencing, the insert of one of the pHSB2-pUC19 recombinant clones was excised and digested with MspI or TaqI and subfragments were cloned into the AccI site of the M13 mp9 vector. The nucleotide sequence of both strands of pHSB2 was determined from the resultant M13 recombinant bacteriophages or from the original pHSB2-pUC19 clone by the dideoxy-chain termination technique (21). Analysis of the sequencing results was done with the DNA Master program (16). Nucleotide sequence accession numbers. The GenBank accession numbers for the pHSB1 and pHSB2 sequences discussed here are X07128 and X66324, respectively.
MATERIALS AND METHODS
Strains, media, and enzymes. Halobacterium strains SB3, GRB, and GN101 (5) were kindly provided by F. Pfeifer and *
Corresponding author.
t Present address: Department of Molecular and Developmental
Genetics, Catholic University of Nijmegen, Toernooiveld, 6525 ED Nijmegen, The Netherlands. t Present address: Department of Medicinal Chemistry and Pharmacognosy, m/c 781, University of Illinois at Chicago, Chicago, IL 60680-6998. 1081
1082
AKHMANOVA ET AL.
1
2
J. BACTERIOL.
3 |
1845
1599
|-
1353
|*
1107
861
615
*
|
369
123
FIG. 1. Macroheterogeneity of the small plasmid from Halobacterium strain SB3. A 1.75-kbp plasmid isolated from a culture grown
from a single colony and further purified in agarose gel as described in Materials and Methods was hydrolyzed with restriction enzyme(s) Sail (lane 1), Sau3A (lane 2), or HindIII-Eco47I (lane 3) and analyzed in a 2% agarose gel. Mobility of marker DNA fragments and their molecular sizes (in base pairs) are indicated to the right of the gel.
RESULTS AND DISCUSSION
Macroheterogeneity of plasmid pHSB. The heterogeneity of small plasmid pHSB in halobacterial strain SB3 was first noticed during its sequencing (13). In those experiments, restriction fragments of pHSB were subcloned into vectors which were propagated in E. coli cells. In this shotgun sequencing approach, the same fragment is usually sequenced several times from individual isolates. Normally, this would yield identical sequences. However, this was not the case for pHSB, since we repeatedly noticed that DNA fragments which were presumably identical occasionally exhibited slightly different nucleotide sequences. We were able to show that at least one of the nucleotide variations was not a result of cloning in E. coli but was already present in the original pHSB preparation (13). To study this phenomenon in more detail, we analyzed the possible heterogeneity of the 1.75-kbp pHSB plasmid isolated from the individual colonies of Halobactenum strain SB3. Figure 1 represents restriction patterns generated by digestion of the plasmid with different restriction enzymes. While hydrolysis with SalI apparently linearizes pHSB into a single 1.75-kbp DNA fragment, digestion with Sau3A generates the main linear form and three additional bands (also comprising about 1.7 kbp). The same result was observed with a HindIII-Eco47I double digest in which the linear form is accompanied by four additional fragments which together roughly correspond to the full-length plasmid. One possible explanation for these results was that two plasmids, both with a size of ca. 1.75 kbp but differing in sequence, coexist in the cell. Specifically, one of the plasmids, designated pHSB1, would correspond to the linear plasmid form in lanes 2 and 3 in Fig. 1. The other, pHSB2, should have several Sau3A and Eco47I sites and would
account for shorter bands in these lanes. The approximate ratio of pHSB1 to pHSB2, as estimated from the band intensities on ethidium bromide-stained gels, was 2.5:1. This ratio did not exhibit noticeable variation between cultures grown from 10 individual colonies. To test the hypothesis of the coexistence of two plasmids in a cell, the 1.75-kbp Sall band (Fig. 1, lane 1) was cloned into the pUC19 vector. Sequence analysis of individual recombinant clones indeed revealed the presence of two plasmid forms, pHSB1 and pHSB2. Deduced sequences (Fig. 2) substantiated the above interpretation of restriction analysis. The 1,736-bp pHSB1 primary structure, corresponding to the pHSB sequence reported previously (7, 14), had unique SalI, HindIII, and Sau3A sites and no Eco47I site (and thus accounts for the linear plasmid bands in Fig. 1). In contrast, 1,781-bp plasmid pHSB2 had no HindIII, one Sall, three Sau3A, and five Eco47I sites, accounting for the shorter bands in the Sau3A and HindIII-Eco47I digests (only four pHSB2 bands are seen in the HindIII-Eco47I digest because the fifth fragment is only 11 bp long). Despite these differences, the overall degree of homology between plasmids pHSB1 and pHSB2 is very high (about 80%), so that their sequences can be readily aligned (Fig. 2). The most conserved part of the plasmids corresponds to a 316-codon open reading frame starting at the ATG at position 671 (pHSB1) or 716 (pHSB2). Interestingly, of 140 nucleotide substitutions within this open reading frame, most either do not change the encoded amino acid (66 of 110 mutated codons) or lead to conservative amino acid substitutions (another 16 codons), which suggests that the structure and function of the encoded protein are conserved. The putative protein, whose gene accounts for 55% of the plasmid length, is likely involved in plasmid maintenance, since sequences homologous to its gene are missing in the DNAs of strains which do not contain small multicopy plasmids (5) and characteristic sequence motifs of REP proteins mediating plasmid rolling-circle replication are present (12). Most of the differences between the pHSB1 and pHSB2 plasmids are clustered in the region between positions 300 and 550 (pHSB1 numbering), where a substantial number of nucleotide substitutions, insertions, and deletions are located. It is not known whether this region is of any functional significance or whether functional sites, including those involved in the control of plasmid replication, are confined to the more conserved plasmid sequences. The fact that pHSB1 and pHSB2 plasmid variants are found in the cultures grown from the individual halobacterial colonies strongly argues in favor of their coexistence in an individual cell. Difference between pHSB1 and pHSB2 sequences might be significant enough to explain their compatibility (especially if sequences responsible for replication control are located within the less conserved region). Alternatively, pHSB1 and pHSB2 might be generally incompatible but the process of their segregation is counterbalanced by some factors, such as plasmid exchange between cells within the population or selective pressure for the simultaneous presence of both plasmids. The latter, however, seems unlikely, since we were able to transform Halobacterium halobium R1, which lacks small plasmids, with an individual pHSB2 plasmid which was excised from the pHSB2-pUC19 clone and circularized, and we found that it could be maintained in halobacterial cells in the absence of pHSB1 (data not shown). In addition, a vector based on the pHSB2-equivalent plasmid from strain GRB1 (see below) propagated stably in H. halobium (15). Surprisingly, despite long homologous regions in pHSB1
VOL. 175, 1993
PLASMIDS IN HALOBACTERIA
I GATCAGTTACCATCGACACCG(;AACTTACGGG pHSBl pTCGACCTACCAGAT
1 1
GTCGACCTAC
pHSB2
ACAAATrCTGAUAGCTAAATG-CTAATCACCGACCAATAACT
pHSBl
101
TTAGTCA
101
TTMACGGTCAATT'CCGACTAATATTGATAAGCTATACCCTAATOGTAGCAGGTACTGATTTCTGACATACcGGACCAATAATGGACCAAT-ACT
pHSB2
200
GTCAACCACGGCAGCGACTAAGAACC
CCACGGCGAGAAGGCCGGCACAACCG
pHSB1
200
GGC-TCCACGC
-GTGCCCCGAGAAGGCCGGACAAAACCG
pHSB2
11 1111 11 1 1
1
11111111 I1
ACAGCT
CA
IIIIIII
AAC
G T ICIGCC-AC-ACCCGG-AGG A-C-T-TGACCCGGTTCCGTGCCTC-CCCCT
11 111 1
III
1111111
300
A
TGCGTCCGTCACAATCAGT-TrAACGCTCC
pHSB1
2
GAAGGGGCCGTACCCGCGGCCTGCAACCGGCAAACGAACCACCTTGCCTGGCTIGTTCACTCGGCCGTTCTTCGTCCGTCCGCTTAGGTCTTCCCGACG
pHSB2
393
G CTGTGACGGT--------
pHSBl
398
GCGAGACGGCrATA7GCTCCCCAGGTC
111111 III
I
TCCC-CT-------
ACG
C ACCGAGC
11'11 11
III
11 1 11111
I 1
GAAC -----AGG-
II
11 11
1 1111
CTCWTCAOGCCGCAACAGCCGGCAGGCGGGGGATCTCCCG'IUCTTCCAGCATCCCCGCAGTGAAGA
pHSB2
469
-TO= ---G---GGA---GTTCTCG--CAGCTGTTCTGGCC-GCCCCCCGGCGG ----
498
ATG0ICAAACCCGGAACGGCGCGCACOGCCCGCACCCCGAACACAGGCAACGCGTCCcCCCGGAACQGAGG:MC
552 597
AGACGCCGACAAAACIpHSB1CGACCGGCCT CGACGGCCT CTCGGCAGATAACCGfCCAGGCC pHSBl HIM 11 1111111111111 11 1111111111111 1111111111 11 11 1111 GCACACCTAATCAGCCpHGSGAACCGCCCGACCGGCCTGCCAACa pHSB2 CGGGCTrICAACGITAA CClCAGAACGAA
652
TAGCGC _C_
HIM111
697
I1
III
11
11
M A I R D O M _C _ _
PHSB1
111 l iii
11 II
L
B
R
D
L
B
CGCG
T
F
C¶I
D T
III
11
III CCGC-A
A V K A V T W
R
8
G
11 11
AAGGCTGI
C
CA
G
A
B
D
FQ
R
D Q
D
Y
W
8
D R
I F
T
F
P
Q
W Y
8
R
AC ACCGCTCACCC
897
ACGCAGG M R F T P
D D Q
B M L
R
Y
B
852
Q
T
a
I
B M
Y V
L
B
T
B
a
V
F
a
T
B
T
pHSB2
P
L
A
Q
R O T
I
V
V
B
N
L
B
Q S L I R a L R CCTCGAACGCGGCC
C
O
B
V
T
B
E R W _
A
C_
S8 T D B D
R
a
P
R
P
P
B H
L
L
D
B
55
L
P
LUH T 83 CGTCGAT
11 111111111
S8 L
W
R L
D M
F
a
XL _
a
1111111111111111111111111 1111111111111
CGGGAGCG Y R IB L Y A K A Q 8 L B R 0 L R B
A
pHSB1 11111 CGApHSB2
11111i111111111111111111111111111111111111 11111 11
A R
Y
11111111 1111111111111111
V
L
T O
H
W B
111
A V
pHSB1 pHSB2
X
R
R
CACCTCGAAGACCTGCTA=0TG AGGCC GpHAC1GCWnC CGCCCACGCCCACCG12 pHSBl 1 IIIIIIIIIIII 11 11111 1111111 1 11111111 111 11 11111 11 11 11 1 1111111 HIM 1111 _ C PHSB2 _CTCCGCIpGAACACTT GACCTG V T
L
A L
1052 1097
pHSB1
IS
1111 1111111111111111111 1111111 111111111 GAGGCC CGCTTCCAGAGACGACGAC H R
997
D
GAC
i111 111111111 1111111 11111111111 11111111111111 11111 11111 III 11111 CTCACACGTGC YMAIKR DOaMKL;RD BL T F D T 8 R AVUxA VI8 O
1 0 I
952
pHSB2
111111111111111111
752
797
111111
T
R
Y
A
8
T
8
L
V
D
D D R
a
B
B
R
a
W
L L
Y
8
P
R
A
P
MH
L
L I
I
P
F
D
B
B1
H
0
L
S8
L
Y V H
W
H
I
A V
B
R
R
a V F V
L
_CATCCTCGAACCGCACGAGAGCGGGTACGTACACATCCACClcGGClTCTGA 111111 1111111 11 11111111 1111 _TCGCCATCCATGAACCA
B a R B
A L A R V L 1 0
_A
P H B 80
B Y L A I HM
W
Y
H I H L a
V
V F V
pHSB1
pHSB2
1152
Q P V L D A H L D N C P T A a B D A H Q I L D I D GCG AGpH1GG CAGCC CCTCGACAACTGCCCGTACGAAG PHSBl
1197
GGGGCCGGIGTCGCGACAGTTCGAGCCGGGCTCACGCGCACCTGCGAAACJCCCGACGGCGGGCGAGGACGCCCATGACGaCGACGAAA E I R P
P V V A
o
1297
Q
B
111111111 11111 1 Y V
a
1252
D
B
A
F
IIIIIII IIIIIIIIIIIIII
1
Q
B
F
P
L
V
D A
RM
D A V R V R
PS
IIIIIIIIIII IIIIIIIIIIIIII 11
L
H
R
R
8
P
C
N
0
a
Y
T
A
N
B
8
D
a A
L
A
Y
I 1111 III
11
B
H
Y
F
D
A A Y M
L
B
Y
a
A
P
B
B
a
P
Y
A
H
V
A
R
Y
F
A
T
Y
W
A
GAGTACGGCGCTGAACCGOGCGAGATGCCOGCTCACGC 111111111 111111111111111 1111111111111111
8
a
~R Q
F
W
R
P
8
A
a
N
1452
A Q
S I B
B L M
BE M
P
Q
CACAGGAATCAGCCG
P
D
Q
B
L M Q
I V D A I
P
B
N V R A F
B
8 D D
11111111 1111111 11111 1497 CACAGGA AQCAAC A
P
B
B
B
D D R
R
A T M
W
W B M
8
O R a
A
Va I
A P
Q W
F R
D O
AAGAGACCcGAC
1111111111
A P O
8 V
B
Y
B L
a
Y
11ii111IIIIIIII 11111111111111I111111111 I M V O I A P I D 8 I EIB 8 R
8 L
R T
P
P
P
D L
8
M N
D
L O
pHSB1 pHSB2
a D V V
GGCC
11 11 111111111 11I1 O
pHSB1 pHSB2
a
_ 1111111111111111111 1397 GAGTQACG'CTCCGAOC CGITCGAGCAACGGCG I Y
pHSB2
N
COGGACGAGGACGCGGCGGGCGTCGGTCCACCCGTCGCGCAGCGGCGGCGTCGAGAACCTCGGTGGTACCTCGCOGCCTACA¶[GCGOGT 111111111111111111 1111 11111111111111111111111 I1111111111111111 11111111 11111111111111 11 _ O D D A V R VR RM P S R S a a v B N L a A Y L A A Y M A a B
1352
D I
pHSB1 pHSB2
I
P 0 0
TACACGGAACTCCAACGCCGCCACCGGGTICGTAAGACGCTCAGACGGCGGT-CCcr-CCCA
1552
OGACTAG TCAAAGCGCCGGOQCGGC
1597
CGAGGGACCCGICGGAGCCACGGAGCGACCcGTACACGAOGCGAACGCCGCCACCOIGTAAGACGCTCAGACGGCGQTCCCTICCCA V D P S I P R S 0 0
pHSB2
CGGAGAGGTCGAAGAAGTGGTCGGAGACCGAGGCGATGAGACTIYTGTCGGTGTGCCACAGCCTGTI'IAGATATAGATGAGT 1111 1111111 11111111111111I11111111 111111111 1I11I111111111I1111I1I1111
1736
pHSBl
1781
pHSB2
11111 11 B
1650 1697
1083
1 111
11
1 1111111
D P
Y
R R
11111111111
L R
T
P
P
pHSB1
1111111111111111
P
CCGGTI T=T
FIG. 2. Alignment of pHSB1 and pHSB2 sequences. Dashes indicate nucleotide deletions; vertical lines indicate nucleotides that are identical in both sequences. Nucleotide substitutions in the TaqI- and MspI- versions of pHSB1 are indicated above the pHSB1 sequence at positions 315 and 342. Amino acids are shown above (pHSB1) or below (pHSB2) corresponding codons in the conserved open reading frame.
1084
., _.kE AKHMANOVA ET AL.
TaqI-/MspI-
J. BA=rRIOL.
tion analyses with frequently cutting enzymes (MspI, TaqI, and BspRI) or by sequencing of a substantial part (more than G A T C G A T C one-third) of the plasmid sequence of TaqI- and MspIT T clones. This suggests that TaqI- and MspI- versions differ T from the main variant of pHSB1 by only one or a few base T G substitutions. G G Interestingly, all four possible combinations of Taq and Msp variations (TaqI IMspI+, TaqI-IMspI+, TaqI IMspl-, C and TaqI-IMspI-) were found in the collection of pHSB1A _ ._ A pUC19 clones analyzed. This indicates the coexistence of all four variants in halobacterial cells and might be explained by C C.W recombination between pHSB1 microversions. Note that "G .1 microheterogeneic sites are only 26 bp apart and thus the rate of recombination should be relatively high to account A ~ ~ ~ ~ ~ A for their shuffling. Other explanations, such as an unusually C VOW C high frequency of DNA replication errors at these sites, also C C cannot be excluded. G GG Since microvariations of the pHSB1 sequence at positions G _ X and 342 have been found because of their fortuitous 315 G G_ T_ .u T_ location within TaqI or MspI sites, it is possible that other T ~~~~~~~~~~~T sites of microheterogeneity exist in pHSB1 but escaped detection because they do not coincide with a restriction site. However, it must be noted that polymorphism at -*W-o T neither the other numerous TaqI and MspI sites nor any of the multiple BspR1 sites in pHSB1 was observed, suggesting that base substitutions are not randomly scattered along the length of the plasmid but occur at specific positions in the plasmid DNA. Also, we were not able to detect any restricFIG. 3. Nucleotide substitutions at TaqI and MspI sites of tion site polymorphism for plasmid pHSB2 by using enpHSB1 microheterogeneic versions. Portions of sequencing gels of zymes TaqI, MspI, and BspRl. TaqI'IMspI' and TaqlIWspl clones are shown. Variable posiHeterogeneity of small plasmids from other halobacterial tions are indicated by arrows and numbered. Dideoxy sequencing It is noteworthy that small-plasmid heterogeneity is strains. was performed on double-stranded DNAs of pHSB1-pUC19 recomnot confined to only one halobacterial strain. We also binant clones. observed both macro- and microheterogeneity for the plasmid from halobacterial strain GRB. While a reported nucleotide sequence of plasmid pGRB cloned in the E. coli vector (4) was almost identical to that of pHSB2, restriction analyand pHSB2, restriction analysis of the plasmid isolated from sis of the small plasmid in strain GRB additionally revealed SB3 cells or of more than 30 clones from the pHSB-pUC19 library revealed no recombinants between pHSB1 and the presence of a pHSB1-equivalent plasmid variant which pHSB2. This is remarkable, given the high rate of recombi- was indistinguishable from pHSB1 for several frequently nation characteristic of several halobacterial strains (4, 18), cutting restriction enzymes (data not shown). However, in and suggests that hybrid plasmids are not stable. contrast to the predominance of pHSB1 over pHSB2, the Microheterogeneity of plasmid pHSB1. The heterogeneity pHSB2 homolog in strain GRB was about five times more of pHSB plasmids is not limited to pHSB1 and pHSB2 abundant than the pHSB1 homolog (Fig. 4), as revealed by variants. Another type of sequence variability, microheterscanning of ethidium bromide-stained gels. As in strain SB3, ogeneity, was revealed by restriction fragment length poly- TaqI+/TaqI- and MspI+IMspI- microvariants of the pHSB1morphism when the collection of pHSB1-pUC19 recombi- equivalent plasmid coexist in GRB cells. nant clones was screened with restriction enzymes TaqI and The small plasmid from a third halobacterial strain, MspI, both of which have numerous cutting sites within the GN101, also exhibits sequence microheterogeneity: the 1.7pHSB1 sequence, 17 and 20, respectively. Surprisingly, kbp plasmids in this strain differ by the presence of either while some of the recombinant pHSBL-pUC19 plasmids one or two Sail sites (data not shown). could be cleaved at all of the TaqI sites predicted from the Stable maintenance of a heterogeneic plasmid population. pHSB1 and pUC19 sequences (TaqI' clones), other plas- Micro- and macroheterogeneity of the small plasmids in mids lacked the TaqI site at position 315 of pHSB1 (TaqI7 halobacteria described in this report is stably maintained. clones); similarly, only 8 of 10 pHSB1-pUC19 clones ana- This conclusion is supported by the fact that the same (or lyzed were cleaved at the predicted MspI site at position 342 very similar) types of heterogeneic plasmids were observed (MsplT clones). Notably, this restriction site polymorphism not only in individual colonies of the same strain but also in did not result from cloning in E. ccli, since we have strains which were isolated from geographically distant previously shown that Taq' and Taq- variants of pHSB1 locations (France for GRB and the United States for SB3 coexist in halobacterial cells (13). Sequencing of the corre[5]). Assuming that these strains have been evolving sepasponding regions of TaqIP and of MspI- pHSB1-pUC19 rately from each other for a considerable time (which is clonesT-TG revealed and C-)T substitutions at positions reflected in a difference in the spectra of the large plasmids 315 and 342, respectively, which map to the disrupted [5]), it is remarkable that during this period only the ratios restriction sites (Fig. 3). However, no other deviations from between major variants of the small plasmids have changed the pHSB1 primary structure were found either by restric(if it is not due just to difference in the genotypes of the host
Taql+/Mspl+
-
_
PLASMIDS IN HALOBACTERIA
VOL. 175, 1993
pHSB
pGRB
.41736
.4733 .4662
4386 FIG. 4. Difference in the ratios of the major variants of the 1.75-kbp plasmid between strains SB3 and GRB. Sau3A digests of the small plasmids from the two strains were analyzed in a nondenaturing 5% polyacrylamide gel. The 1,736-bp band represents a linear form of pHSB1, which contains one Sau3A site. Three shorter bands (733, 662, and 386 bp) correspond to pHSB2 (three Sau3A sites). Note the difference in relative intensity of the 1,736-bp band compared with the shorter bands between the two strains. The numbers on the right are molecular sizes in base pairs.
cells) while all of the macro- and microheterogeneic plasmid versions have been preserved. A question that remains unanswered is whether the same mechanism accounts for intracellular populational stability of macro- and microheterogeneity of the small halobacterial plasmids. On the one hand, the fact that both types of heterogeneity are found in the same cell is intriguing and can hardly be ascribed to pure coincidence. Thus, a common mechanism like, for example, efficient horizontal transfer of the plasmids within the cell population might explain the observed phenomena. On the other hand, it appears that while plasmid macrovariants do not recombine with each other, the microversions presumably do. This might imply different mechanisms of maintenance of macro- or microvariants, since efficient recombination can significantly slow down or even completely abolish plasmid segregation. Sequence heterogeneity of the small plasmids within the same cell, found for extremely halophilic archaea, has not been described for plasmids of other organisms. Our understanding of archaeal molecular biology is insufficient to permit us to distinguish whether our observations reflect a specific feature of halobacteria or represent a more general phenomenon. Recently developed vector transformation systems for halobacteria may help in unravelling some of the unexplained peculiarities of plasmid maintenance in archaea.
1085
ACKNOWLEDGMENTS Thanks are due to F. Pfeifer and N. Hackett for halobacterial strains and to D. W. Mount and B. Conrad for providing the DNA Master computer program. We are grateful to A. Kondrashev and P. Forterre for fruitful discussions; M. Ares, J. Hughes, P. Mitchel, and N. Grebenschikov for critical reading of and help in preparing the manuscript; and L. Baratova for support. REFERENCES 1. Cline, S. W., and W. F. Doolittle. 1992. Transformation of members of the genus Haloarcula with shuttle vectors based on Halobacterium halobium and Haloferax volcandi plasmid replicons. J. Bacteriol. 174:1076-1080. 2. Cline, S. W., W. L. Lam, R. L. Charlebois, L. C. Schalkwyk, and W. F. Doolittle. 1989. Transformation methods for halophilic archaebacteria. Can. J. Microbiol. 35:148-152. 3. Couturier, M., F. Bex, P. L. Bergquist, and W. K. Maas. 1988. Identification and classification of bacterial plasmids. Microbiol. Rev. 52:375-395. 4. DasSarma, S., T. Damerval, J. G. Jones, and N. Tandeau de Marsac. 1987. A plasmid-encoded gas vesicle protein gene in a halophilic archaebacterium. Mol. Microbiol. 1:365-370. 5. Ebert, K., W. Goebel, and F. Pfeifer. 1984. Homologies between heterogeneous extrachromosomal DNA populations of Halobacterium halobium and four new halobacterial isolates. Mol. Gen. Genet. 194:91-97. 6. Gutierrez, M. C., M. T. Garcia, A. Ventosa, J. J. Nieto, and F. Ruiz-Berraquero. 1986. Occurrence of megaplasmids in halobacteria. J. Appl. Bacteriol. 61:67-71. 7. Hackett, N. R., and S. DasSarma. 1989. Characterization of the small endogenous plasmid of Halobacterium strain SB3 and its use in transformation of H. halobium. Can. J. Microbiol. 35:86-91. 8. Hackett, N. R., M. P. Krebs, S. DasSarma, W. Goebel, U. L. Rajbhandary, and H. G. Khorana. 1990. Nucleotide sequence of a high copy number plasmid from Halobacterium strain GRB. Nucleic Acids Res. 18:3408. 9. Hall, M. J., and N. R. Hackett. 1989. DNA sequence of a small plasmid from Halobacterium strain GN101. Nucleic Acids Res. 17:10501. 10. Hattori, M., and Y. Sakaki. 1986. Dideoxy sequencing method using denatured plasmid templates. Anal. Biochem. 152:232238. 11. Holmes, M. L., and M. L. Dyall-Smith. 1990. A plasmid vector with a selectable marker for halophilic archaebacteria. J. Bacteriol. 172:756-761. 12. Ilyina, T. V., and E. V. Koonin. 1992. Conserved sequence motifs in the initiator proteins for rolling circle DNA replication encoded by diverse replicons from eubacteria, eucaryotes and archaebacteria. Nucleic Acids Res. 20:3279-3285. 13. Kagramanova, V. K., N. I. Derckacheva, and A. S. Mankin. 1989. Unusual nucleotide sequence heterogeneity of small multicopy pHSB plasmid from Halobacterium strain SB3, an archaebacterium. Can. J. Microbiol. 35:160-163. 14. Kagramanova, V. K., N. I. Derckacheva, and A. S. Mankin. 1989. The complete nucleotide sequence of the archaebacterial plasmid pHSB from Halobacterium strain SB3. Nucleic Acids Res. 16:4158. 15. Krebs, M. P., T. Hauss, P. M. Heyn, U. L. RajBhandary, and H. G. Khorana. 1991. Expression of the bacterioopsin gene in Halobacterium halobium using a multicopy plasmid. Proc. Natl. Acad. Sci. USA 88:859-863. 16. Mount, D. W., and B. Conrad. 1986. Improved programs for DNA and protein sequence analysis on the IBM personal computer and other standard computer systems. Nucleic Acids Res. 14:443-454. 17. Novick, R. P. 1987. Plasmid incompatibility. Microbiol. Rev. 51:381-395. 18. Pfeifer, F., U. Blaseio, and P. Ghahraman. 1988. Dynamic plasmid population in Halobacterium halobium. J. Bacteriol.
170:3718-3724. 19. Rosenshine, I., and M. Mevarech. 1989. Isolation and partial
1086
AKHMANOVA ET AL.
characterization of plasmids found in three Halobacterium volcandi isolates. Can. J. Microbiol. 35:92-95. 20. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 21. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463-5467.
J. BACTERIOL.
22. Sioud, M., G. Baldacci, P. Forterre, and A.-M. de Recondo. 1988. Novobiocin induces accumulation of a single strand of plasmid pGRB-1 in the archaebacterium Halobacterium GRB. Nucleic Acids Res. 16:7833-7842. 23. Woese, C. R., 0. Kandler, and M. L. Wheelis. 1990. Towards a natural system of organisms-proposal for the domains Archaea, Bacteria and Eucarya. Proc. Natl. Acad. Sci. USA 87:4576-4579.
