©
1995 Oxford University Press
4373-4378
Footnote
Genomic and cDNA sequence tags of the hyperthermophilic Archaeon
Pyrobaculum aerophilum
Genomic and cDNA sequence tags of the hyperthermophilic Archaeon Pyrobaculum aerophilum
Paul
Völkl
1,2
,
Peter
Markiewicz
1
,
Claudia
Baikalov
1
,
Sorel
Fitz-Gibbon
1
,
Karl O.
Stetter
2
and
Jeffrey H.
Miller
1,
*
1
Department of Microbiology and Molecular Genetics and the Molecular Biology
Institute, University of California, 405 Hilgard Avenue,
Los Angeles
, CA 90024,
USA
and
2
Archaeenzentrum, Universität Regensburg,
93053
Regensburg
,
Germany
Received October 7, 1996;
Accepted October 8, 1996
ABSTRACT
The hyperthermophilic archaeum,
Pyrobaculum aerophilum
, grows optimally at 100
o
C with a doubling time of 180 min. It is a member of the phylogenetically ancient
Thermoproteales
order, but differs significantly from all other members by its facultatively aerobic metabolism. Due to its
simple cultivation requirements and its nearly 100% plating efficiency, it was chosen as a model organism for studying the genome organization of hyperthermophilic ancient archaea. By a G+C content of the DNA of 52 mol%, sequence analysis was easily possible. At least some of
the mRNA of
P.aerophilum
carried poly-A tails facilitating the construction of a cDNA library. 245 sequence tags
of a poly-A primed cDNA library and 55 sequence tags from a 1-2 kb
Sau
3AI-fragment containing genomic library were analyzed and the corresponding amino acid sequences compared with protein sequences from databases. Fourteen percent of the
cDNA and >9% of genomic DNA sequence tags revealed significant similarities to
proteins in the databases. Matches were obtained to proteins from archaeal,
bacterial and eukaryal sources. Some sequences showed greatest similarity to eukaryal rather than to bacterial versions of proteins, other matches were found to proteins which had previously only been found in
eukaryotes.
INTRODUCTION
Hyperthermophilic archaea represent an interesting source for studying phylogeny and organization of ancestral life. Based on 16S rRNA analysis, the most ancient living organisms known are hyperthermophiles (
1
,
2
). Therefore, their study may provide some clues as to the genome organization
of the common ancestor of
Bacteria,
Archaea
and
Eukarya
as well as to the basic requirements for thermophily. However, due to the
difficult cultivation, especially of the deep branching and strictly anaerobic hyperthermophiles, these organisms are only poorly investigated. Most genetic studies with archaea were therefore done with the mesophilic
extreme Halophiles (
3
,
4
) and with aerobic representatives of
Sulfolobus
(
5
). Both represent the longest evolutionary lineages of the
Euryarchaeota
and of the exclusively hyperthermophilic organisms containing
Crenarchaeota
kingdom, respectively. But, for a better understanding of the basic characteristics of
phylogenetically ancient organisms it would be helpful to look at more slowly
evolving archaea too. Recently, within the so far strictly anaerobic
Thermoproteales
order, which is comprised of slowly evolving organisms within the
Crenarchaeota
, having short evolutionary lineages based on 16S rRNA phylogenetic analysis (
6
), a novel isolate was described which could serve as a model organism for the molecular investigation of hyperthermophilic archaea. This new isolate,
Pyrobaculum aerophilum
(type strain: IM2), grows optimally at 100oC and pH 7.0 and represents the only facultatively aerobic member of this
order (
7
). In contrast to other hyperthermophiles,
P.aerophilum
is therefore easy to grow either aerobically or anaerobically and can be plated
to form colonies within four days with up to 100% efficiency (
7
). With these features
P.aerophilum
is an ideal candidate for molecular investigation and genetic studies. As a
first step, sequence data from a large number of this organism's genes will provide invaluable comparative information and elucidate the early evolution of many biological
systems. In contrast to members of
Pyrodictium
and some other hyperthermophilic archaea (
1
) where due to their extremely high G+C content DNA sequencing is hampered,
P.aerophilum
has a G+C content of 52 mol% which is suitable for sequencing. The genome sizes of hyperthermophiles are universally small, ~2 Mb, which is less than half the size of an
E.coli
genome. Thus random sequencing rapidly identifies a large proportion of the
genes in the organism. The presence of poly-A mRNA in an organism makes it possible to reverse transcribe those genes
back into DNA, thus making them available for cloning and sequencing. Polyadenylation of mRNA was believed to be a eukaryotic feature until short
poly-A tails were also found at the 3'-terminal ends of bacterial mRNAs (
8
). In archaea, polyadenylation of mRNA has so far only been described in
Methanococcus vannielii
, a mesophilic archaeum within the
Euryarchaeota
kingdom (
9
). Based on this background, we tried to isolate poly-A mRNA from the crenarchaeal member
P.aerophilum
and constructed genomic and cDNA libraries. Out of these libraries we generated
sequence tags from randomly chosen clones and categorized them as potential
homologs to known protein sequences. This sequence tag method has been
successfully applied for studying human brain specific transcripts (
10
) and for the investigation of the genome of the eubacterial extreme thermophile
Thermotoga maritima
(
11
). The sequence tags may further be used for creating a physical map of the genome, phylogenetic analysis, and `reverse genetics' (
10
). In the following study, DNA sequence tags were translated into protein
sequences and compared with those available in the GenBank, EMBL and PIR
databases. With this method we were able to identify numerous genes, and at the
same time we generated sequence data usable for further studies such as full-length gene sequencing. Additionally, the results of this study allow us
to draw conclusions about the relationship of single proteins of
P.aerophilum
with those of phylogenetically diverse organisms.
MATERIALS AND METHODS
Strains and culture conditions
P.aerophilum
was grown aerobically in BS medium at 97oC as described previously (
7
). Cells were harvested in the late exponential growth phase by centrifugation and the cell masses were stored at -80oC until use.
Escherichia coli
strains used were XL1-Blue (
recA1
,
endA1
,
gyrA96
,
thi-1
,
hsdR17
,
supE44
,
relA1
,
lac
, [F'
proAB
lacIq
Z[Delta]
M15
, Tn
10(tetr)
]) and SURE (
mcrA
, [Delta](
mcrBC
-
hsdRMS-mrr)171
,
supE44
,
thi-1
, l-,
gyrA96
,
relA
, lac,
recB
,
recJ,
sbcC
,
umuC
::Tn
5
(
kanr
),
uvrC
, [F'
proAB
lacIqZDM15
, Tn
10(tetr)
]) (Stratagene, La Jolla, CA).
Escherichia coli
strains were grown on Luria Bertani medium (LB) prepared as described by Miller
(
12
), LB was supplemented with 0.2% maltose and 10 mM magnesium sulfate when
strains were grown before and during Lambda phage infection. NCY broth
contained 5 g/l NaCl, 2 g/l MgSO
4
.7H
2
O, 5 g/l yeast extract (Difco) and 10 g/l NZ amine (casein hydrolysate; Sigma).
The pH was adjusted to 7.5 with NaOH. Agar plates contained 1.5% agar (Difco).
Top agarose contained 0.7% agarose (BioRad) instead of agar. SM buffer
contained 5.8 g NaCl, 2.0 g MgSO
4
.7H
2
O, 50 ml 1 M Tris-HCl, pH 7.5, and 5 ml 2% gelatin per liter. Ampicillin was added to 100 [mu]g/ml and tetracycline to 20 [mu]g/ml medium as described by Miller (
12
). Tetracycline was not added to media supplemented with maltose and magnesium
sulfate.
DNA isolation
DNA from
P.aerophilum
was purified by a method described by Sharp and Williams (
13
) and modified by Kim
et al
. (
11
). Approximately 0.7 g of frozen cell paste was suspended in 10 ml dH
2
O. Protease K was added to a final concentration of 1.5 mg/ml and incubated at
37oC. After 1 h, 1.2 ml of a 4% SDS solution was added and the sample was incubated at 55oC for 1 h. The lysate was chilled on ice, and extracted with 12 ml phenol-chloroform-isoamyl alcohol (24:24:1 by volume). The phases were separated
by spinning at 17 000
g
for 10 min at room temperature. The upper aqueous phase was extracted three
additional times then the DNA was precipitated by adding 0.1 vol 3 M sodium
acetate and 2.5 vol absolute ethanol. After incubation for 10 min at room temperature, the precipitated DNA was spooled out with a closed end Pasteur pipette, rinsed
in 70% ethanol, transferred into an Eppendorf tube, and dried under vacuum for
5 min. The dry pellet was resuspended in 5 ml TE buffer (pH 8.0). RNA was
digested with RNaseA at a final concentration of 50 [mu]g/ml at 37oC. After 1 h, 1.5 mg/ml Protease K was added. After incubation at 37oC for 1 h, 0.1 vol 2% deoxycholate was added and incubation was continued for another hour. The solution was extracted four times
with 1 vol phenol-chloroform-isoamyl alcohol, ethanol precipitated, dried and resuspended in 1 ml TE buffer pH 8.0. The final yield was 0.67 mg DNA, as calculated by OD
260
measurement.
Plasmid-DNA isolation
Plasmid DNA was isolated by the alkaline lysis procedure as described by Kraft
et al
. (
14
). Prior to sequencing the plasmid DNA was analyzed by restriction mapping and agarose gel analysis as described by Sambrook
et al
. (
15
).
RNA isolation
RNA was isolated from a frozen
P.aerophilum
cell paste from a culture grown aerobically in a 300 l fermentor. The Stratagene (La Jolla, CA) single step RNA isolation Kit using a guanidinium thiocyanate protocol (
16
) was used. Crude RNA was further purified by CsCl gradient ultra
centrifugation. The RNA, in a volume of 3.7 ml denaturation solution D (4 M
guanidinium thiocyanate, 25 mM sodium citrate, pH 7.0, 0.5% sarcosyl, 0.1 M 2-mercaptoethanol) was carefully layered over 8.3 ml 5.7 M CsCl, 10 mM EDTA
in a 15 mm * 102 mm polyallomer ultracentrifuge tube and pelleted by centrifugation
in a SW28 rotor at 25 000 r.p.m. for 24 h at 20oC. The supernatant was removed and the RNA pellet washed with 1 ml 75%
ethanol, air dried and resuspended in 1 ml 10 mM Tris-HCl, pH 7.5, 0.5 M KCl (oligo-dT binding buffer). The recovery was ~25% of the crude RNA.
Purification of poly-A RNA
Polyadenylated RNA was separated from CsCl-purified RNA by binding on oligo-dT cellulose (New England Biolabs). Oligo-dT cellulose (100 mg) was washed in an Eppendorf tube with 1
ml 0.1 N NaOH, quickly spun down, and the supernatant removed. The oligo-dT cellulose was neutralized by washing in 1 ml elution buffer (10 mM Tris-HCl, pH 7.5) until the pH was 7.5 (5 times). Finally the cellulose was equilibrated with 1 ml binding buffer (10 mM Tris-HCl, pH 7.5, 0.5 M KCl) for 1 h at 4oC. The CsCl purified RNA in 1 ml binding buffer was
mixed with the oligo-dT cellulose and slowly swirled at 4oC for 2 h. After brief centrifugation at 4oC, the supernatant was removed and the oligo-dT cellulose washed 5 times with 1 ml binding buffer. Poly-A RNA was eluted by adding 1 ml elution buffer and incubating at 45oC for 15 min. The elution was repeated 2 times.
The concentration of RNA in all wash and elution steps was followed by ethidium bromide spot tests. The three elution samples were combined and ethanol precipitated by adding 0.1 vol 3 M sodium acetate and 2.8 vol absolute ethanol. The samples were incubated at -70oC for 12 h and centrifuged at 25 000 r.p.m. in a SW28 rotor at 4oC for 2 h. The poly-A RNA pellet was washed with 75% ethanol, dried and
dissolved in 50 [mu]l dH
2
O.
Genomic library construction
Partial digestions of
P.aerophilum
DNA were made with different dilutions of
Sau
3AI endonuclease incubated exactly 1 h at 37oC (
15
). The restriction fragments were analyzed on a 1% TAE agarose gel. A digestion of 10 [mu]g DNA with 0.2 U
Sau
3AI for 1 h at 37oC yielded DNA fragments ranging in length from 500-2500 base pairs. A digestion with 0.04-0.08 U
Sau
3AI for 1 h at 37oC gave fragments 3-15 kb in size. From these two sets of restriction digests, fragments
of 1-2 kb and 3-10 kb in size were excised from a preparative 1% TAE agarose gel,
and the DNA recovered using the GeneClean DNA Isolation kit (BIO101, La Jolla, CA). The fragments were ligated into pBluescript II (SK-) vector (Stratagene) digested with
Bam
HI and dephosphorylated with calf intestine alkaline phosphatase (0.25 U/10 [mu]g vector DNA, 30 min, 37oC). Ligation products were introduced into
E.coli
strain `SURE` by electroporation using a BioRad electroporator (BioRad,
Richmond, CA). Transformants were selected by plating on LB agar plates supplemented with ampicillin (100 [mu]g/ml), X-gal (40 [mu]g/ml) and IPTG (0.1 mM). Single colonies were grown in 3 ml liquid
LB/amp medium, and the cell masses used for plasmid isolation.
cDNA library construction
cDNA libraries were constructed using the Uni-ZAP XR cDNA Cloning kit (Stratagene, La Jolla, CA). The protocol was
followed as described by the manufacturer. The construction of a poly-T primed library was performed with 5 [mu]g of purified poly-A RNA using a 50 base oligonucleotide containing an 18 base
oligo-dT 3'-end, a `GAGA' 5'-end and an internal
Xho
I site as primer for reverse transcription. For creating a randomly primed library, 20 [mu]g of crude RNA was used with a 46 base random primer containing an internal
Xho
I site (Stratagene).
Sequencing and data analysis
Clones from the oligo-dT primed library and genomic libraries were sequenced by the Sanger chain termination method (
17
) using a Sequenase Version 2.0 kit (US Biochemical) and [[alpha]-
32
P]dATP (NEN). Sequencing products were separated on 6% polyacrylamide-urea gels at two intervals to obtain overlapping sequencing runs. Genomic
clones were sequenced from both directions with the SK and KS primers
(Stratagene). For sequencing oligo-dT primed clones SK and M13 -20 primers (Stratagene) were used.
Sequences were analyzed for similarity to known proteins with the NCBI BLAST
program (
18
), using the NCBI non-redundant database containing GenBank, PIR, SwissProt and EMBL. Database sequences were retrieved using BLAST retrieve or Internet Gopher (
19
).
RESULTS
cDNA libraries
To obtain sequence information of expressed genes in
P.aerophilum
two cDNA libraries were constructed. For the first library, polyadenylated RNA
was isolated from crude RNA by binding to oligo-dT cellulose. About 1% of the crude RNA had bound and could be recovered
from the oligo-dT cellulose indicating the presence of abundant poly-A RNA in the cell. For the second library, crude RNA was used as template for randomly primed reverse
transcription. Both cDNA libraries were cloned unidirectionally into Lambda ZAP
II (Stratagene). The number of clones obtained from the poly-T primed library was 3.2 * 10
6
, while the randomly primed library yielded only 4.5 * 10
4
clones. The analysis of the first 18 poly-A cDNA clones by
Not
I/
Apa
I restriction enzyme digestion revealed an average insert size of 1 kb, ranging
from 0.6-2.0 kb. In contrast, the randomly primed cDNA clones had very short or no
inserts and were therefore not further used for random sequencing analysis.
On average, from a single clone a sequence tag of ~310 nucleotides in length could be determined by one sequence reaction. Preliminary analysis of 245 oligo dT-primed cDNA sequence tags (abbreviated as ESPAT:
e
xpressed
s
equence of
P
.
a
erophilum from oligo-d
T
primed library), by comparison of the translated DNA sequence with known
protein sequences from PIR, SwissProt and GenBank databases resulted in strong
matches to 34 different proteins from other organisms, which is a match rate of
~14%. Matches were found to proteins of various functions including several
intermediary metabolism proteins, DNA/RNA related proteins, and surface/transport proteins (Table
1
). Other non-metabolic matches included the eukaryotic elongation factor ef-2, human DNA polymerase replication factor C and one clone (ESPAT-71) that weakly matched a wide variety of DNA-binding proteins. In this case the closest matches were
with eukaryotic proteins (not shown). Sequence tag ESPAT-136 had strong similarity to archaeal and eukaryotic translation
elongation factor ef-2 showing the histidine which, if post-translationally modified, would be the target of diphtheria toxin.
The remaining clones had open reading frames whose amino acid sequence did not correspond to any known protein sequence in the databases. However, these sequences were
not ribosomal DNA, and frequently showed short matches to protein `motifs',
such as nucleotide binding folds or cysteine patterns similar to those found in
iron-sulfur cluster containing proteins.
Genomic libraries
In order to study DNA sequences from transcribed and non- transcribed regions of the
P.aerophilum
genome a 1-2 kb
Sau
3AI genomic library was constructed. The average insert size of the genomic
clones was 1.5 kb. With an estimated
P.aerophilum
genome size of 2 * 10
6
base pairs, a given gene would be represented in a non-biased library of this size (>4000 clones) with >95% probability. From a
total of 30 different clones 55 sequence tags, with an average length of 320
nucleotides, were analyzed by comparing translated sequences with amino acid
sequences in databases using the program BLAST (
18
). Five sequence tags exhibited notable similarity to known protein sequences in
the databases (Table
1
). This is a match rate of ~9%, in contrast to the 14% match rate of the cDNA sequence tags. As in the
oligo-dT primed library, some clones exhibited greater similarity to eukaryotic
rather than to bacterial proteins.
In order to verify some of the putative matches, sequences up- and downstream of the matching region of the nitrate reductase and the ribokinase were generated and analyzed. For all sequences the
similarities continued in these elongated sequence stretches (not shown).
Similarly, starting with the similarity found by sequence tag GSPA-35, the entire gene encoding a subtilisin like serine protease was cloned
and the DNA sequence determined together with flanking regions (
20
). The genomic DNA sequences were not interrupted by introns as shown by
comparison with the corresponding cDNA sequences.
Table 1
Summary of cDNA and genomic sequence tags of
P.aerophilum
exhibiting similarities to known protein sequences
|
Clone
a
|
Sequence
b
|
Matching protein (organism)
c
|
Identity
d
|
Accession
e
|
|
1. Homologies related to DNA/RNA-metabolism
|
|
ESPAT-99 SK
|
410
|
*DNA pol. replication factor C (Human)
|
37/59 (48%)
|
SP:P35249
|
|
ESPAT-136 SK
|
345
|
(*)Elongation factor ef-2 (
Sulfolobus
)
|
69/117 (59%)
|
SP:P23112
|
|
ESPAT-145 SK
|
650
|
Rec
F protein (
S.thyphimurium
)
|
25/40 (62%)
|
SP:P24900
|
|
ESPAT-217 SK
|
400
|
DNA-ligase (
Desulfurolobus
)
|
21/53 (40%)
|
SP:Q02093
|
|
ESPAT-217 M13
|
400
|
Ribokinase (
E.coli
)
|
38/92 (41%)
|
SP:P05054
|
|
ESPAT-258 SK
|
337
|
50S ribosomal protein (
Mc. vannielii
)
|
16/30 (53%)
|
SP:P15824
|
|
2. Homologies to proteins with metabolic function
|
|
ESPAT-33 SK
|
226
|
Aconitate hydratase (
Bacillus
)
|
19/33 (57%)
|
SP:P09339
|
|
ESPAT-80 SK
|
349
|
*Carbamoyl phosphatase (Yeast)
|
13/18 (47%)
|
SP:P07259
|
|
ESPAT-91 SK
|
320
|
Salty Thiosulfate Reductase [
Salmonella
]
|
14/27 (25%)
|
SP:P37600
|
|
ESPAT-224 M13
|
400
|
Polysulfide reductase B(
Wollinella
)
|
31/50 (49%)
|
SP:P31076
|
|
ESPAT-207 SK
|
324
|
Molybdopterin biosynth. (
Moe
A) (
E.coli
)
|
15/25 (60%)
|
SP:P12281
|
|
ESPAT-215 SK
|
290
|
Formate dehydrogenase (
E.coli
)
|
21/60 (35%)
|
SP:P24183
|
|
ESPAT-224 SK
|
322
|
Dimethy sulfoxide reductase (
E.coli
)
|
38/75 (50%)
|
SP:P18776
|
|
ESPAT-238 SK
|
246
|
Disulphide oxidoreductase [
Entamoeba
]
|
12/22 (43%)
|
gi[brvbar]895831
|
|
ESPAT-249 SK
|
273
|
Dihydroxyacid dehydratase [
Clostridium
]
|
16/28 (43%)
|
SP:P31959
|
|
ESPAT-257 SK
|
302
|
Resp. nitrate reductase A (
E.coli
)
|
47/72 (65%)
|
SP:P09152
|
|
ESPAT-310 SK
|
318
|
*NADH plastochin. oxidored. (
Marchantia
)
|
17/51 (33%)
|
SP:P12131
|
|
ESPAT-317 M13
|
306
|
Glutamate dehydrogenase (
Clostridium
)
|
42/82 (51%)
|
SP:P27346
|
|
ESPAT-330 SK
|
379
|
Glucose-1 dehydrogenase A (
Bacillus
)
|
25/70 (35%)
|
SP:P10528
|
|
|
|
7 alpha-hydroxysteroid hydrogenase (Eubac.)
|
24/54 (44%)
|
PIR:A42468
|
|
GSPA-35 SK
|
463
|
Subtilisin (
Bacillus
)
|
33/66 (50%)
|
SP:P00780
|
|
GSPA-70 SK
|
320
|
*Homoaconitase (
Yeast
)
|
32/72 (37%)
|
SP:P49367
|
|
GSPA-81 KS
|
301
|
GMP synthase (
Bacillus subtilis
)
|
11/28 (39%)
|
GP:S88687_1
|
|
3. Homologies to surface/transport proteins
|
|
ESPAT-53 SK
|
298
|
*Erythrocyte membrane prot. 7 (Human)
|
45/99 (46%)
|
SP:P27105
|
|
ESPAT-111 SK
|
357
|
drr
A: Daunorubicin res. prot. (
S.peucetius
)
|
45/94 (40%)
|
SP:P32010
|
|
ESPAT-260 M13
|
300
|
ABC Transporter (
Haemophilus
)
|
16/33 (22%)
|
SP:P44531
|
|
ESPAT-299 M13
|
291
|
*Cu 2+-transporting ATPase, P-type (Human)
|
30/66 (15%)
|
PIR:JC2465
|
|
ESPAT-314 SK
|
292
|
Daunorubicin res. ATP bind. prot. (
Strep
)
|
16/32 (50%)
|
SP:P32010
|
|
ESPAT-322 M13
|
300
|
Inner membran protein lack (
Agrobacterium
)
|
29/49 (59%)
|
SP:Q01937
|
|
GSPA-15 KS
|
293
|
ATP-dependent transporter [
Cyanophora
]
|
25/41 (23%)
|
SP:P48255
|
|
4. Homologies to proteins of miscellaneous function
|
|
ESPAT-235 M13
|
274
|
Cation-transporting ATPase [
Synechococcus
]
|
17/28 (54%)
|
SP:P37279
|
|
ESPAT-246 SK
|
264
|
*Adenosylhomocysteinase (
C.elegans
)
|
32/61 (52%)
|
SP:P27604
|
|
ESPAT-309 M13
|
263
|
HisF protein (
Azospirillum
)
|
25/41 (60%)
|
SP:P26721
|
|
ESPAT-326 M13
|
308
|
*Activator 1 37 KD subunit (Human)
|
22/41 (53%)
|
PIR:A45253
|
|
GSPA-15 MD1
|
387
|
Ribonuclease E (
E.coli
)
|
14/35 (40%)
|
PIR:JG0009
|
a
ESPAT,
e
xpressed
s
equence from
P
yrobaculum
a
erophilum oligo-d
T
primed cDNA library followed by number of clone and primer used for sequencing.
GSPA,
g
enomic
s
equence of
P
.
a
erophilum
.
b
Number of bases sequenced.
c
Most homologous protein. Eukaryotic matches are indicated by an asterisk (*).
d
Numbers are: identical amino acids/amino acids in the most similar area. In
parenthesis: percentages of identical amino acids within the total sequence
tag.
e
Accession ID of the matching protein sequence: SP, swissprot; GP, genpept; PIR,
PIR databases.
Sequence similarities of
Thermotoga maritima
In a similar sequencing project using the hyperthermophilic bacterium
Thermotoga maritima
, 32% of 244 cDNA and 14% of 21 genomic DNA sequence tags revealed similarities
to known protein sequences (
11
). With two exceptions (AVP-3 vacuolar H
+
-phosphatase from
Arabidopsis
and aspartate aminoacyl tRNA synthase from
Saccharomyces
), strong matches were exclusively to eubacterial proteins. The basic results of
the two sequence tag approaches from
T.maritima
and
P.aerophilum
are summarized in Table
2
.
Table 2
Comparison of the results of the random sequence tag approaches from the
bacterium
T.maritima
and the archaeum
P.aerophilum
|
Subject
|
Thermotoga
|
Pyrobaculum
|
|
|
maritima
|
aerophilum
|
|
Total of sequence tags analyzed
|
265
|
300
|
|
expressed sequences
|
244
|
245
|
|
genomic sequences
|
21
|
55
|
|
Match rate of expressed sequence tags
|
32%
|
14%
|
|
Match rate of genomic sequence tags
|
14%
|
9%
|
|
Matches specific to eukaryotic proteins
|
2/52
|
9/34
|
DISCUSSION
A critical requirement for genetic analysis is that single cells of an organism
form visible colonies on defined media with high efficiency and in a reasonable
time. Unfortunately, the deep branching and therefore phylogenetically most interesting archaea have extremely low plating efficiencies. Additionally, they are strictly
anaerobic and due to their high growth temperatures and other features
difficult to grow and to handle. With the new isolate
P.aerophilum
a suitable candidate for genetic studies of hyperthermophilic primitive archaea
is available for the first time. By constructing an oligo-dT primed cDNA library
P.aerophilum
has been shown to possess polyadenylated mRNA. Each of the 245 clones sequenced
to date was unique and also did not contain ribosomal DNA, indicating that
polyadenylation is widespread among different mRNA species of
P.aerophilum
and is therefore useful for the construction of a highly representative cDNA
library. With the sequence tag strategy a large amount of sequence information
was accumulated, creating a broad basis for further studies. Using cloned
sequences, it may be possible to introduce defined mutations into the genome of
this organism, and thereby analyze the function of the gene by `reverse genetics.' Furthermore, the identified genes could be used for creating a map of the
P.aerophilum
genome.
In the poly-A cDNA library, the match rate to known proteins was 14%, while the rate
in the genomic libraries was lower (9%), which is expected due to the
additional regulatory and non-coding DNA. Despite their lower match rate, the genomic clones are useful
when analyzing non-coding regions of the genome. Introns in archaeal protein genes are
unknown, although they have been shown to occur within some ribosomal genes
like the 16S rRNA gene of
P.aerophilum
(
21
). If they occur at all they can't be very frequent because of the small genome
sizes. The genomic sequence encoding a serine type protease in
P.aerophilum
did not contain an intervening sequence when compared with the corresponding
cDNA sequence (
20
). Furthermore, the genome seems to be organized in polycistronic transcription
units as in the intron-scarce eubacteria, as was previously seen for Archaeal Methanogen genes (
22
).
The cDNA match rate for
P.aerophilum
was lower than for the hyperthermophilic eubacterium
Thermotoga maritima
, which also employed a sequence tag approach (
11
). In this study, >30% of cDNA sequence tags could be assigned as likely
homologs of genes known from other organisms. Similar projects, such as
sequencing human brain cDNAs (
10
) or cDNAs of the nematode
Caenorhabditis elegans
(
23
), yielded around 20% and up to 40% known sequences, respectively. It may be
that
P.aerophilum
contains many genes which are unique to its lineage, which do not exist in
eukaryotes or eubacteria. But, like in the other cases the real number of
homologous sequences is probably higher because with a 100 amino acid long
sequence tag an unconserved region of an otherwise homologous protein will not
be detected. The clearest difference to the results from
T.maritima
was that several matches from
P.aerophilum
were to eukaryotic rather than to prokaryotic proteins, where with few
exceptions, the sequence tags from
T.maritima
matched exclusively eubacterial proteins. The elongation factor ef-2 from
P.aerophilum
was clearly of the eukaryotic type. It exhibited highest similarity to
Sulfolobus
but, it still had 40% identity to the human elongation factor ef-2. This finding is absolutely in line with the phylogenetic positions of
P.aerophilum
and supports the three domain concept of life (
6
,
24
). In other cases amino acid sequence similarities were highest to eukaryotic proteins for which no prokaryotic equivalent is known, such as the erythrocyte membrane protein. The finding of both
similarities to typically eukaryotic and similarities to bacterial proteins
reflect the phylogenetic position of archaea sharing both eubacterial and
eukaryotic features. The study of archaea will in many cases provide the missing link in our understanding of the relationship between homologous systems in diverse organisms (
25
).
ACKNOWLEDGEMENTS
We would like to thank Jean J. Lee for excellent technical assistance. This work
was supported in part by grants of the Office of Naval Research (ONR) to J.H.M.
(N00014-95-1-0938) and the Deutsche Forschungsgemeinschaft to K.O.S.
REFERENCES
1 Stetter,K.O. (1992) In Tran Than Van,J.K., Mounolou,J.C., Schneider,J. and McKay,C. (eds), Colloque Interdisciplinaire du Comite National de la Recherche Scientifique, Frontiers of Life, pp. 195-219.
2 Woese,C.R. (1987) Microbiol. Rev., 51, 221-271. MEDLINE Abstract
3 Pfeifer,F., Offner,S., Kruger,K., Ghahraman,P. et. al. (1994) Systematic Appl. Microbiol., 16, 569-577.
4 Tchelet,R. and Mevarech,M. (1994) Systematic Appl. Microbiol., 16, 578-581.
5 Schleper,C., Roder,R., Singer,T. and Zillig,W. (1994) Mol. Gen. Genet., 243, 91-96. MEDLINE Abstract
6 Woese,C.R., Kandler,O. and Wheelis,M.L. (1990) Proc. Natl Acad. Sci. USA, 87, 4576-4579. MEDLINE Abstract
7 Völkl,P., Huber,R., Drobner,E., Rachel,R., Burggraf,S., Trincone,A. and Stetter,K.O. (1993) Appl. Environ. Microbiol., 59, 2918-2926.
8 Gopalakrishna,Y., Langley,D. and Sarkar,N. (1981) Nucleic Acids Res., 9, 3545-3554. MEDLINE Abstract
9 Brown,J.W. and Reeve,J.N. (1985) J. Bacteriol., 162, 909-917. MEDLINE Abstract
10 Adams,M.D., Kelley,J.M., Gocayne,J.D., Dubnick,M., Polymeropoulos,M.H., Xiao,H., Merril,C.R., Wu,A., Olde,B., Moreno,R.F. et al. (1991) Science, 252, 1651-1656. MEDLINE Abstract
11 Kim,C.W., Markiewicz,P., Lee,J.J., Schierle,C.F. and Miller,J.H. (1993) J. Mol. Biol., 231, 960-981.
12 Miller,J.H. (1972) Experiments in Molecular Genetics, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
13 Sharp,R.J. and Williams,R.A.D. (1988) Appl. Environ. Microbiol., 54, 2049-2053.
14 Kraft,R., Tardiff,J., Krauter,K.S. and Leinwand,L.A. (1988) Biotechniques, 6, 544-546. MEDLINE Abstract
15 Sambrook,J., Fritsch,E.F. and Maniatis,T. (1989) Molecular Cloning: A Laboratory Manual, 2nd Edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
16 Chomczynski,P. and Sacchi,N. (1987) Anal. Biochem., 162, 156-159. MEDLINE Abstract
17 Sanger,F., Nicklen,S. and Coulson,A.R. (1992) Biotechnology, 24, 104-108.
18 Altschul,S.F., Gish,W., Miller,W., Myers,E.W. and Lipman,D.J. (1990) J. Mol. Biol., 215, 403-410.
19 Krol,E. (1992) In Krol,E. (ed.), The Whole Internet User's Guide and Catalog. O'Reilly & Associates, Sebastopol, CA, pp. 189-210.
20 Völkl,P., Markiewicz,P., Stetter,K.O. and Miller,J.H. (1994) Protein Sci., 3, 1329-1340.
21 Burggraf,S., Larsen,N., Woese,C.R. and Stetter,K.O. (1993) Proc. Natl Acad. Sci. USA, 90, 2547-2550. MEDLINE Abstract
22 Reeve,J.N., Hamilton,P.T., Beckler,G.S., Morris,C. J. and Clarke,C.H. (1986) System. Appl. Microbiol., 7, 5-12.
23 Bargmann,C.I. (1992) Nature Genet., 1, 79-80. MEDLINE Abstract
24 Cammarano,P., Palm,P., Creti,R., Ceccarelli,E., Sanangelantoni,A.M. and Tiboni,O. (1992) J. Mol. Evol., 34, 396-405. MEDLINE Abstract
25 Keeling,P.J. and Doolittle,W.F. (1995) Proc. Natl Acad. Sci. USA, 92, 5761-5764. MEDLINE Abstract
Return
*To whom correspondence should be addressed. Tel: +1 310 825 8460; Fax: +1 310
206 3088; Email: jhmiller@ewald.mbi.ucla.edu
