RESEARCH ARTICLE
2747
Characterisation of a cluster of genes encoding
Theileria annulata AT hook DNA-binding proteins and
evidence for localisation to the host cell nucleus
David G. Swan1,*, Rowena Stern1, Sue McKellar1, Kirsten Phillips2, Chris A. L. Oura1, Tülin Ilhan Karagenc3,
Laura Stadler1 and Brian R. Shiels1
1Department of Veterinary Parasitology, University of Glasgow, Bearsden Road, Glasgow, G61 1QH, UK
2Department of Molecular Recognition, The Hannah Institute, Mauchline Road, Ayr KA6, Scotland, UK
3Adnan Menderes University, Faculty of Veterinary Medicine, Department of Parasitology, Isikli, Aydin, Turkey
*Author for correspondence (e-mail:[email protected])
Accepted 27 April 2001
Journal of Cell Science 114, 2747-2754 (2001) © The Company of Biologists Ltd
SUMMARY
Infection of bovine leukocytes by the apicomplexan
parasite Theileria annulata results in alteration of host cell
gene expression and stimulation of host cell proliferation.
At present, the parasite-derived factors involved in these
processes are unknown. Recently, we described the
characterisation of a parasite gene (TashAT2), whose
polypeptide product bears AT hook DNA-binding motifs
and may be transported from the parasite to the host
nucleus. We now describe the isolation of a further two
genes (TashAT1 and TashAT3) that are very closely related
to TashAT2. All three TashAT genes are located together in
a tight cluster, interspersed by two further small open
reading frames, all facing head to tail. TashAT2 was shown
to be expressed in all T. annulata cell lines examined,
whereas TashAT1 and TashAT3 were expressed in the
sporozoite stage of the parasite, and also in infected cell
lines, where their expression was found to vary between
different cell lines. Evidence for transport was provided by
antisera raised against TashAT1 and TashAT3 that reacted
with the host nucleus of T. annulata-infected cells.
Reactivity was particularly strong against the host nuclei
of the T. annulata-infected cloned cell line D7B12, which is
attenuated for differentiation. A polypeptide in the size
range predicted for TashAT3 was preferentially detected in
host enriched D7B12 nuclear extracts. DNA-binding
analysis demonstrated that fusion proteins containing the
AT hook region of either TashAT1 or TashAT2 bound
preferentially to AT rich DNA.
INTRODUCTION
of genes encoding transcription factors that are implicated in
the control of cell division or apoptosis (Dobbelaere et al.,
1988; Baylis et al., 1995; Ole-MoiYoi et al., 1993; Botteron
and Dobbelaere, 1998; Heussler et al., 1999). However,
although induction of host cell division is known to be
Theileria dependent (McHardy et al., 1985), little is known
about how the parasite directly modulates leukocyte gene
expression or stimulates the host cell to divide.
It has been suggested that as the macroschizont
differentiates into the extracellular merozoite, parasite factors
involved in host cell division may be downregulated, resulting
in the cessation of cell division (Carrington et al., 1995). Thus,
the association between parasite and host cellular division
would be uncoupled, owing to the removal of the signal that
initiates proliferation of the infected lymphocyte. We have
previously identified a small gene family whose expression is
downregulated during differentiation to the merozoite in T.
annulata (Swan et al., 1999). One member of this family,
TashAT2 encodes a gene product that bears a predicted AT
hook motif DNA-binding domain. Furthermore, experimental
data suggest that TashAT2 is transported from the parasite to
the host nucleus, implying a role in the modification of host
Theileria annulata and Theileria parva are closely related
parasites that belong to the phylum apicomplexa. Both
parasite species are transmitted from a feeding tick to their
bovine host, and invade white blood cells (Dschunkowsky
and Luhs, 1904). The parasite differentiates into the
multinucleate macroschizont and stimulates the host cell to
undergo unlimited division (Hulliger, 1965). In vivo, infected
cells proliferate within the lymph node that drains the site
of tick attachment and then spread throughout the bovine
host, infiltrating other major organ systems of the body, such
as the gastrointestinal tract, kidney and lungs (Irvin and
Morrison, 1987). The presence of Theileria-infected cells in
these organs is often accompanied by the presence of
petecheal haemorrhages that contain infected cells. These
resemble multicentric lymphosarcomas, and injection of
Theileria-infected cells into SCID (Fell et al., 1990) or
athymic mice (Irvin et al., 1975) results in the formation of
tumours.
Parasite infection of the host cell is associated with
modulation of leukocyte gene expression, including a number
Key words: Apicomplexan parasite, Host/parasite interaction, AT
hook DNA-binding motif, Theileria annulata
2748
JOURNAL OF CELL SCIENCE 114 (15)
cell gene expression. We present further characterisation of
TashAT2 and two other members of the gene family, TashAT1
and TashAT3. All three genes form part of a cluster, encode
AT hook DNA-binding motifs (Johnson et al., 1988) and are
very closely related in sequence. We present evidence to
suggest that TashAT3 and, possibly, TashAT1 could be
transported to the host cell nucleus and discuss the possible
implications.
MATERIALS AND METHODS
Cell culture
The T. annulata (Hidirseyh), T. annulata (Diyarbakir) (Pipano and
Shkap, 1979), T. annulata (Ankara) (TaA2) macroschizont-infected
cell lines (Shiels et al., 1992) and the cloned cell lines derived from
TaA2 (D7, E3, C9 and D7B12) were maintained in vitro at 37°C or
induced to differentiate as previously described (Shiels et al., 1992).
BL20, an uninfected bovine lymphosarcoma cell line (Morzaria et al.,
1982), and TBL20, a T. annulata-infected cell line derived from BL20
(Shiels et al., 1986), were cultured as for D7, except that myoclone
super plus foetal bovine serum (Gibco BRL) was substituted for heatinactivated foetal bovine serum (Sigma).
Cloning and northern blot analysis
A fragment of TashAT1 had previously been isolated from a λgt11
library of genomic DNA derived from merozoites of the D7 infected
cloned cell line (Swan et al., 1999). The TashAT1 fragment was used
to screen a λDASHII library of D7 genomic DNA using standard
protocols. Two λDASHII overlapping clones, based on their
restriction maps and hybridisation profiles, were isolated and a 13.4
kbp region sequenced on both strands. DNA sequencing was
performed on a Licor 4000 automated DNA sequencer according to
the manufacturer’s protocol. DNA and protein analyses were
performed using the GCG sequence analysis package (Devereux et
al., 1984).
RNA from in vitro cultured cells and from purified T. annulata
(Ankara) piroplasms was isolated by the Triazol reagent, according to
the manufacturer’s instructions (Sigma). Sporozoite RNA was isolated
from T. annulata (Ankara)-infected Hyaloma anatolicum ticks as
described by Williamson et al. (Williamson et al., 1989). RNA was
size fractionated by electrophoresis through a formaldehyde-agarose
gel and analysed by Northern blotting as previously described (Shiels
et al., 1994). Hybridisation was carried out overnight at 65°C
according to the method of Church and Gilbert (Church and Gilbert,
1984).
Generation of fusion protein and immunoblotting
A 362 bp fragment of TashAT1 starting 63 bp from the first putative
translational start site was PCR amplified using ampliTaq polymerase
and the primers; 5′-tttaggatccgtaaaatttgcttcttcc-3′ and 5′gaaggaattctggtggaattttaataaa-3′. The PCR product was then subcloned
into the vector pGEX-2TK (Pharmacia), expressed in Escherichia coli
strain JM109 as a glutathione-S-transferase (GST) fusion protein and
purified on a glutathione-sepharose column using the Pharmacia
protocol. Antisera (anti-TashAT1/3) to the fusion protein were raised
in New Zealand White rabbits and immunoblotting was carried out as
described by Swan et al. (Swan et al., 1999). Signal was detected by
ECL using the method provided by the suppliers (Pierce and
Warriner).
DNA-binding analysis of TashAT1
A λgt11 library of D7 genomic DNA or purified λgt11 clones were
induced to express fusion protein under standard conditions.
Nitrocellulose filters were laid on top of the developing plaques,
which were then incubated at 37°C for 4 hours. The filters were
removed from the plates, washed briefly in TNE 50 (10 mM TrisCl
pH 7.5, 50 mM NaCl, 1 mM EDTA, 0.1% NP40), then incubated
overnight in blocking buffer (2.5% dried milk, 25 mM Hepes (pH 8.0),
1 mM DTT, 10% glycerol, 50 mM NaCl, 1 mM EDTA, 0.1% NP40)
at 4°C. The filters were then rinsed at 4°C in 1× binding buffer (1×
BB; 25 mM Hepes pH 7.9, 3 mM MgCl2, 4 mM KCl) and the
expressed fusion protein denatured by incubation in 1×BB containing
6 M guanidinium HCl, 1 mM DTT twice for 5 minutes. Renaturation
was then carried out by immersing the filters for 5 minutes in
denaturation buffer, which was sequentially diluted 1 in 2 with 1×BB,
1 mM DTT, four times. The renatured filters were washed in 1×BB,
1 mM DTT, incubated in 1×BB, 5% dried milk for 30 minutes, rinsed
in 1×BB, 0.25% dried milk and washed briefly with HNE (10 mM
Hepes pH 7.9, 50 mM NaCl, 1 mM EDTA, 1 mM DTT).
Hybridisation was then carried out in HNE buffer containing 500 µg
ml−1 poly dIdC filtered through a 0.22 µm filter. Double stranded
concatenated probes were generated by ligating oligonucleotides
radiolabelled at the 5′ end with 32P using T4 polynucleotide kinase.
The oligonucleotides were: CAT1, 5′atgcGCACACAATTTGTAGGGCGAC3′; CAT1m2, 5′atgcGCACACAATTACGAGGGCGAC3′;
CAT2, 5′ atgcAGATAAACATGCACACAATTTGTA3′; or CAT3,
5′atgcAGGGCGAC3′ previously annealed to their respective
complementary oligonucleotides. Each oligonucleotide had a four
base overhang at the 5′ end, either atgc (or gcat in the reverse
complement) shown in lower case, for concatenation. Probes were
added to the filters and incubated at 4°C for 60 minutes, after which
the filters were given three 5 minute washes in HNE buffer, followed
by exposure to X-ray film.
DNA-binding analysis of TashAT2
The procedure for determining TashAT2 DNA-binding specificity
was based on the method used by Pollock and Treisman (Pollock
and Treisman, 1990), but used glutathione-sepharose instead of
immunoprecipitation to bind the DNA-protein complex. In brief, a
double-stranded 76 mer oligonucleotide with a 26 base pair random
core flanked by two specific 25 base sequences (primer F and primer
R) was PCR amplified and labelled using 32P dCTP. A part of
TashAT2 containing the AT hook domain (amino-acid residues 296541) linked to GST as a fusion protein (GST-TashAT2) (Swan et al.,
1999), was purified and approximately 3 µg of GST-TashAT2 or 3
µg of purified GST alone used per binding cycle. 100 µl of a 50%
solution of glutathione-sepharose was washed twice in phosphate
buffered saline (PBS) before adding 100 µl of PBS containing the
fusion protein. The mixture was then rotated at room temperature
for 15 minutes and the beads washed three times in binding buffer
(10 mM Hepes pH 7.9, 25 mM KCl, 1 mM EDTA, 50 mM NaCl,
0.1% NP40 plus protease inhibitors) to remove excess protein. 0.4
ng of radiolabelled probe was added to the beads in 25 µl of binding
buffer and rotated for 30 minutes at 4°C. The beads were washed
in binding buffer three times, followed by phenol/chloroform
extraction. The DNA was then ethanol precipitated from the
aqueous phase, washed in 70% ethanol and resuspended in 20 µl
of TE pH 8.0. 10 µl were taken for PCR amplification using
primer F and primer R, and the cycle of protein to DNA binding
repeated four times. To test the ability of the amplified sequences to
bind GST-TashAT, an electrophoretic mobility shift reaction
(EMSA) was set up containing 10 µl of the radiolabelled DNA
that was purified during each cycle, 0.5 µg of protein, 1×BB, 5%
Ficoll, 1 µg of poly dGdC:dGdC in a 40 µl reaction. The reaction
mix was incubated for 30 minutes at 4°C. Electrophoretic separation
was then carried out on a 4% polyacrylamide gel cast in 0.5×TBE
and run in 0.5×TBE buffer. A mobility shifted band from cycle 4
was excised, eluted into dH2O, PCR amplified and the EMSA
repeated. The resulting mobility shifted band was excised, PCR
amplified and subcloned into the vector pGEMT-easy for
sequencing.
T. annulata DNA-binding proteins
BglII
HindIII
EcoRI
EcoRI
HindIII
BglII
EcoRV
SpeI
EcoRI
EcoRI
HindIII
EcoRV
SpeI
EcoRI
EcoRI
SpeI
KpnI
EcoRI
HindIII
EcoRI
RESULTS
2749
Cloning and sequence characterisation of TashAT1
and TashAT3
A fragment of TashAT1 was originally isolated from a λgt11
library representing T. annulata genomic DNA isolated from
1 Kb
TashAT2
TashAT3
TashAT1
the cloned infected cell line D7 (Swan et al., 1999). This
fragment was used to screen a λDashII library of D7 parasite
Fig. 1. Restriction map and schematic of the TashAT cluster.
genomic DNA and several hybridising clones were isolated.
Two overlapping clones were selected, restriction mapped and
Although the AT hook motif containing domains of TashAT1/3
13.4 kbp of DNA sequence obtained. The restriction pattern of
and TashAT2 are very similar, TashAT1/3 contains a 77 amino
the 13.4 kbp contig coincided with the profile obtained from
acid residue insertion between AT hooks 2 and 4 (Fig. 2C).
D7 genomic DNA restriction digested and probed with internal
Within this insertion, in addition to AT hook 3, TashAT1/3 also
fragments from the TashAT cluster (data not shown). Five
contains two peptides of sequence RPRK that could contribute
ORFs were detected (Fig. 1), three of which shared extensive
to DNA binding. Immediately upstream of the AT hook domain
sequence homology. ORF1 encodes TashAT2 (Swan et al.,
1999); ORFs 2 and 4 will be described
elsewhere (L.S., D.G.S. and B.R.S.,
A
unpublished) and ORFs 3 and 5 are
MMVVLKLSHI IFTLFLYRVK FASSEILYLD NLDNPNFYTI KIVEDRLTKI 50
designated
TashAT1
and
TashAT3,
MILSTPEDKI TEIRSKRKLI WGSDRGEYVK CFTRFSFESS DKTLITIEIG 100
NAVDEAMKFI YVSGNFYKYI NKSEFEDYYK SFCSVFIKIP PGKLPIPRLK 150
respectively. TashAT1 and TashAT3 encode
KNVKTEKVDK RKLKRDRQRK DKPQSEQHDK NVDIVSQSLA EEGIDLEKKI 200
ORFs of 1401 bp and 2985 bp respectively.
VGREEPTQQT EKQQEPTELE PETIPVELES DDEEIDESNV SKPKESDGIL 250
TQNRYTQTDI QEIEDIGIQT EIHELENIVT QTDIQTKESS IQTDIQEVED 300
Comparison of the DNA sequence of all
IDTQTDIQEL ENIGIQTIGN FSDITEVTKK HEQPEVPKRR PGRPRKQKPE 350
three ORFs revealed a remarkable degree of
PEQPKRKRGR PRKQKYETKK TWLLRPRNMK TETKKTWLLR PRKQKPEPEQ 400
conservation to each other. TashAT1 is
PKRKRGRPRK QKYETKKTWL LRPRNMKTET KKTWLLRPRK HKPEPEQPKR 450
S T*466
98.9% identical to the first 1401 bp of
KRGRPRKQKP EPESDHSEES TQPHPQEQET EDSIKALGPS PEKRPFSFDI 500
TashAT3. Overall TashAT2 is 83.6%
YCEDRDAEDE LRRRAKRFRS EPLESHEQED TTDAGVSSGA GAPPPPGDGS 550
EPSDGPGDCP PPEQDQDDTV LVQLNKERIL YLEDPGSNKG VNYEHEINDG 600
identical to TashAT3 and 67.5% identical to
IPTLIIRAKP HKTITHIFEN GLIICEAEKG SKLLSLSAFS YYNEFILVEI 650
TashAT1. However, TashAT3, from position
IFKTPMASYN RYFRKHGGDW KEVNIKDFEV YYQDLRGNVI MIEEMMVDIS 700
1320-2715 bp of the ORF is 99.9% identical
LPIDPSKIFS KTSEKNGIIT TILMPLPGFF IKQVTNGANI VWSSNFRRCL 750
AITMTSRNGD MPRLLMLEIS GPNDGVTEEL HFHRIGDKFS LISSKLYDNV 800
to base pairs 1386-2781 of TashAT2. A
VYEESPDYDF ESTTKSPQLI ETDHSSIFIS SKDSISSEST NEQTPIEKFT 850
comparison of the translation product of
LQPETIHLEI SSDEEEPIDL SIKHKSKAPE SVAEPTEPET ITLDLLSSHD 900
EEHEDIDLSD IELQISSDDE SHEDHTHQDL IILGSEPSEE SHTEDNNSST 950
TashAT1 with that of TashAT3, and of
IVRTESTKPQ PDISTEDTHT QTDTDIDKKT RSSLPLKKRP YKQD* 994
TashAT2 with TashAT3, is shown in Fig.
2A,B, respectively. An analysis of the
B
translation products of both TashAT1 and
TashAT3 155 TEKVDKRKLKRDRQRKDKPQ
TashAT2 258 TEKVDKRKLKGDRQRKDKQE
TashAT3 revealed the presence of a potential
signal peptide at the 5′ end of TashAT1/3
TashAT3 SEQHDKNVDIVSQSLAEEGIDLEKKIVGRE-EPTQQTEKQQEPTELEPETIPVELESDDE
and detected possible AT hook DNATashAT2 SEQHDKNVDIVAQALAEEGIDLEKEIVGREVDKIIEKYKITKETQTDIPTGSIETQTDIQ
binding motifs (Fig. 2A,C). AT hooks are
TashAT3 EIDESNVSKPKESDGILTQNRYTQTDIQEIEDIGIQTEIHELENIVTQTDIQTKESSIQT
small (8-10 amino acid residues), semiTashAT2 QLE--NIDT--QTDIQEVEDIETQTDLPT-GSIEIQTDIQEVENIDTQTDIPTGSIETQT
conserved basic motifs rich in K, R and P,
TashAT3 DIQEVEDIDTQTDIQELENIGIQTIGNFSDITEVTKKHEQPEVPKRRPGRPRKQKPEPEQ
with a G present in a highly conserved
TashAT2 DIQEVEDIDIQTDIQEVEDIGIQTIGNFSDITEVTKKHEKPEVPKRRPGRPRKHKPEPEQ
region of the motif (Johnson et al., 1988);
TashAT3 PKRKRGRPRKQKYETKKTWLLRPRNMKTETKKTWLLRPRKQKPEPEQPKRKRGRPRKQKY
they bind preferentially to the minor groove
TashAT2 PKRKRGRPRK-------------------------------------------------of AT-rich DNA (Reeves and Nissen, 1990)
TashAT3 ETKKTWLLRPRNMKTETKKTWLLRPRKHKPEPEQPKRKRGRPRKQKPEPESDHSEESTQP 473-905
and have been found in a variety of
TashAT2 ---------------------------HKPEPEQPKRKRGRPRKQKPEPESDHSEESTQP 495-927
regulatory factors with diverse functions
(Aravind and Landsman, 1998). TashAT1/3
Fig. 2. Sequence analysis of TashAT1 and TashAT3 and their
C
have four AT hook motifs (Fig. 2C)
comparison with TashAT2. (A) Sequence of the predicted translation
1. KRRPGRPR
products of TashAT1 and TashAT3. The potential signal sequence is in
contained in a 120 amino acid residue
2. KRKRGRPR
3. KRKRGRPR
italics. The AT hook DNA-binding domain is in bold and the putative
domain, motifs 2, 3 and 4 are identical to
4. KRKRGRPR
transcriptional transactivation domain is underlined. Where the
each other, and to motifs 2 and 3 of
5. KRGRGRPR
sequence of TashAT1 differs from that of TashAT3 is shown above the
6.
KRPRGRPK
TashAT2, which was previously determined
relevant sequence of TashAT3. The asterisks mark the end of each
7. RKPRGRPK
to contain three AT hook motifs (Swan et al.,
sequence. (B) A comparison of TashAT2 with TashAT3. Identical
1999). All four AT hook motifs of
residues are shown in bold. The numbers at the end of the comparison denote the region
TashAT1/3 reflect the semi-conserved basic
which is 100% identical between the two sequences. A gap generated by the analysis in
nature of a typical AT hook (Fig. 2C). Motifs
the sequence of TashAT2 is shown by a series of dashes. (C) A comparison of the AT
2, 3 and 4 contain the core sequence RGRP
hook motifs of TashAT1 and TashAT3 with those of the HMGI(Y) protein. Rows 1-4 are
which is present in almost all AT hook
the AT hook motifs of TashAT1/3. Rows 5-7 are those of HMGI(Y). Accession Numbers
for the TashAT cluster: TashAT1, AJ291829; TashAT2, AJ132045; TashAT3, AJ291830.
motifs (Aravind and Landsman, 1998).
2750
JOURNAL OF CELL SCIENCE 114 (15)
Fig. 3. Indirect Immunofluorescence assays using antisera to TashAT1/3. (A-E) Analysed using anti-TashAT1/3 antiserum; (F) Analysed using
pre-immune serum. The cells used were BL20 (A), TBL20 (B), D7B12 (C); D7 at 37°C (D), D7 at 41°C for 7 days (E), D7B12 (F). Scale bar:
10 µm.
of TashAT1/3 there is a region composed of small imperfect
repeats, rich in glutamic acid, glutamine, aspartate and
threonine residues (Fig. 2A), that is similar in composition to
that described for transcriptional transactivation domains
(Triezenberg, 1995).
Immunofluorescence indicates the host nucleus as
a possible location for TashAT1/3
Antisera were initially raised against the N terminus of
TashAT1; however, it was subsequently determined that this
sequence is also present in TashAT3. Therefore, all
immunofluorescence studies carried out with this antisera
cannot distinguish between TashAT1 and TashAT3.
Immunofluorescence analysis of COS7 cells transfected with
TashAT2 showed no reactivity using anti-TashAT1/3 indicating
specific detection of TashAT1/3 (data not shown).
When anti-TashAT1/3 was used in IFAT analysis of the
T. annulata-infected cells, D7B12 and D7 (Fig. 3C,D),
immunoreactivity was observed against the macroschizont,
displaying a punctate pattern of dots that may represent
reactivity with or the region surrounding parasite nuclei, and
against the host cell nucleus (Fig. 3C,D). In
D7B12 cells, host nuclear reactivity was very
bright in some cells, although a large variation
in immunoreactivity was observed within each
cell population. There was no reactivity with
the uninfected control, BL20 (Fig. 3E),
Fig. 4. (A) Northern analysis of T. annulatainfected and uninfected cell lines probed with the
entire TashAT3 gene. (B) The same northern blots
probed with 2P3 (large subunit rRNA of T.
annulata). DIY, Diyarbakir; Hid, Hiderseyh; P,
passage number; Pi, piroplasms; Sp, sporozoite.
whereas with T. annulata-infected BL20 cells (TBL20) (Fig.
3B), faint reactivity with the host nucleus was obtained. During
a differentiation time course of D7 cells, anti-TashAT1/3
reactivity against the host nuclei diminished overall, although
some cells showed clear fluorescence (Fig. 3E). These cells
probably represented undifferentiated macroschizont-infected
cells as the differentiation process is known to be stochastic
(Shiels et al., 1994).
Differential expression of TashAT mRNA
In order to examine the relevance of the TashAT cluster to T.
annulata-infected cells in general, we determined the RNA
expression profile of all three TashAT genes in a range of
Theileria-infected cell lines, and in the sporoblast/sporozoite
and piroplasm stages of the parasite life cycle. The in vitro
infected cell lines analysed were the T. annulata (Ankara)infected cell line TaA2 and cloned cell lines derived from TaA2
(D7, D7B12, C9, and E3); low and high passage cell lines
infected with T. annulata (Hidirseyh) and T. annulata
(Diyarbakir); and the bovine lymphosarcoma cell line, BL20
and its T. annulata (Ankara)-infected counterpart, TBL20.
T. annulata DNA-binding proteins
2751
There was no signal detected in piroplasm RNA by any of the
TashAT probes. Hybridisation of the same blots with the gene
encoding the parasite large subunit rRNA (Fig. 4B) confirmed
that the mRNA levels of all three TashAT genes can vary
depending on the cell line and passage number.
Fig. 5. Western analysis of cell and nuclear extracts using antiTashAT1/3 antiserum. (1) D7B12 total cell extract. (2) D7B12 host
nuclear fraction. (3) D7B12 parasite nuclear fraction. (4) BL20
nuclear fraction.
mRNA species at 2.1 kb, 3.6 kb and 4 kb have been deduced
to correspond to TashAT1, 2 and 3 respectively, using probes
derived from each of the individual TashAT genes, which
gave a more specific signal (Swan et al., 1999; D.G.S.,
unpublished).
Northern blots were probed with the entire radiolabelled
TashAT3 gene, which, out of the three TashATs, has the best
overall homology with the other two TashATs. TaA2 and the
cloned cell lines derived from it expressed the three mRNA
species detected at 4 kb, 3.6 kb and 2.1 kb, corresponding to
TashAT2, TashAT3 and TashAT1, respectively (Fig. 4A). Each
of the cloned cell lines had essentially the same expression
profile as the parent cell line, TaA2. TashAT1 and TashAT2
mRNA levels were more abundant than those of the TashAT3
RNA species; except in the case of D7B12 where TashAT3
mRNA levels were higher, and TashAT2 mRNA levels
decreased slightly (Fig. 4A). TashAT1 expression did not alter
significantly and was the most highly expressed of the TashATs
in TaA2 and the cloned cell lines derived from TaA2. In
marked contrast, cell lines T. annulata (Hidirseyh), and T.
annulata (Diyarbakir) expressed TashAT1 at low to barely
detectable mRNA levels (Fig. 4A). TashAT2 and
TashAT3 mRNA species were present in these cell
lines, although the levels of TashAT3 mRNA was
fainter. TashAT2, however, was the only TashAT
message detected in TBL20s. The TashAT probe was
also used in hybridisation analysis of T. annulata
(Ankara) sporoblast/sporozoite RNA and piroplasm
RNA (Fig. 4A). The results indicated that TashAT3 and
TashAT1 were expressed by sporoblast/sporozoites,
and at this stage of the parasite life cycle, the TashAT3
message was detected at the highest level (Fig. 4A).
Fig. 6. DNA-binding analysis of TashAT2.
(A) Electrophoretic mobility shift assays carried out with
the radiolabelled PCR product obtained after each round of
binding to the GST-TashAT fusion protein. Lanes 1-4
represent cycles 1-4. Lane 5 represents cycle 4 and lane 6
represents the major band from cycle 4 excised and PCR
amplified. (B) DNA sequence obtained after excising and
PCR amplifying the band from lane 6 and sub cloning into
pGEM7ZF. The TAAAT/ATTTA motif is shown in bold.
Immunoblot analysis using anti-TashAT1/3
To determine whether anti-TashAT1/3 reactivity against the
host nucleus in D7B12 cells was due to recognition of
TashAT1, TashAT3, or both, immunoblot analysis was carried
out on extracts of whole cells and host or parasite enriched
nuclear fractions (Fig. 5). A faint band at 180 kDa and a
stronger band at 66 kDa were specifically detected in whole
cell extracts, relative to pre-immune serum, by anti-TashAT1/3.
Partitioning of host and parasite nuclear fractions, showed a
clear enrichment of the 180 kDa band in the host nuclear
fraction, while the 66 kDa band was detected at increased
levels in the parasite-enriched nuclear fraction. Assessment of
the origin of these polypeptides, host or parasite, was
performed by analysis of an uninfected BL20 extract. Faint
recognition of a band at 66 kDa, along with a band of 125 kDa,
indicated that the anti-TashAT1/3 antisera crossreacted with
bovine-derived polypeptides. It was concluded that the most
likely candidate for a parasite-derived molecule specifically
detected by the antisera was the 180 kDa polypeptide. From
the predicted size of the ORFs, this polypeptide is unlikely to
be encoded by TashAT1 but could be derived from TashAT3.
TashAT polypeptides bind specifically to AT rich
DNA
To test TashAT2 for DNA binding, a recombinant protein
representing the AT hook region and an upstream basic region
of TashAT2 fused to the GST gene (GST-TashAT2) was used
in DNA-binding assays with a radiolabelled randomised
double-stranded oligonucleotide. Cycles of binding followed
by PCR amplification were carried out in an attempt to enrich
for DNA with affinity for GST-TashAT2. Protein-DNA
complexes obtained with the enriched DNA sequences were
visualised by mobility shift gel electrophoresis and after four
binding cycles, an increase in the DNA-protein complex was
clearly visible (Fig. 6A). The complex from cycle 4 was eluted
2752
JOURNAL OF CELL SCIENCE 114 (15)
from the mobility shift gel, PCR amplified and a further
mobility shift (Fig. 6A) selected for a more specific DNAprotein complex. DNA eluted from this complex was PCR
amplified and subcloned. The sequences obtained from the
inserts of 12 subclones are shown in Fig. 6B. Although the
sequences are not the same, they are all AT rich and show
general similarities. Thus, there are three AT-rich regions
discernible in each sequence separated by one to four G or C
nucleotides, and a high proportion of the AT-rich regions
contain the sequence ATTTA or TAAAT.
As part of a separate investigation into stage differentiation
of T. annulata, a λgt11 library of D7 genomic DNA had been
screened for proteins that bind a DNA motif located upstream
of the Tams1 gene, which encodes the major T. annulata
merozoite surface protein (Shiels et al., 2000). One of the
concatenated oligonucleotide probes, CAT1 (Fig. 7A), bound to
a λgt11 clone that expressed a fragment of TashAT1 containing
the AT hook DNA-binding domain. To test whether this binding
was sequence specific and whether it related to the specific
mobility shifts observed for the Tams1 promoter (Shiels et al.,
2000), and to determine which regions of the probe were
required for binding to the TashAT1 fusion protein, the purified
lambda clone was probed with three more concatenated
oligonucleotides (Fig. 7A). CAT1M2 is the same as CAT1 but
with a three base pair change, CAT3 is composed of the GC 3′
region of CAT1 and CAT2 overlaps with CAT1 with 10 extra
bases 5′ and missing the GC-rich region of CAT3. This
demonstrated that concatenated CAT3 bound weakly to the
TashAT1 λgt11 clone in comparison with the CAT1, CAT1M2
and CAT2 probes. Binding reactions using the CAT1M2 (Fig.
7B, panel A) and CAT3 (Fig. 7B, panel B) probes are shown.
Neither probe displayed affinity for a control λgt11clone
purified from the same library (Fig. 7B, panels C,D) that was
shown to express recombinant polypeptide by reactivity with
bovine antiserum raised against the parasite (data not shown).
DISCUSSION
T. annulata infection of the bovine host cell alters bovine gene
expression and induces host cell proliferation until the parasite
progresses towards differentiation to the merozoite. This
results in a marked reduction, and eventual cessation, of host
cell division, followed by the destruction of the host cell as
merozoites are released. Previous work and this study have
characterised a family of DNA-binding proteins encoded by
the parasite that display characteristics suggestive of an
involvement in the regulation of host cell gene expression and,
perhaps, leukocyte proliferation. All three members of the
TashAT family possess putative signal sequences as well as
nuclear localisation signals, demonstrating that they contain
the structural information that would allow them to be secreted
from the macroschizont and be transported into the host cell
nucleus. TashAT gene expression is downregulated during
differentiation at a time point coincident with the initial
reduction in host cell division, and immunofluorescence
studies suggest that the location of at least two of the TashATs,
TashAT2 (Swan et al., 1999) and TashAT3, is likely to be the
host nucleus. As well as bearing AT hook motifs, TashAT2 has
been shown in this study to bind preferentially to AT-rich DNA.
This was demonstrated by the binding of the AT-rich double
Fig. 7. DNA-binding analysis of TashAT1. (A) Double stranded
oligonucleotides used in the binding analysis. (B) Panels A and B
show binding of CAT1M2 and CAT3 probes respectively to a λgt11
clone expressing a fragment of TashAT1 bearing the AT hook motif.
Panels C and D show binding of CAT1M2 and CAT3, respectively, to
a λgt11 clone expressing an unrelated T. annulata gene.
stranded oligonucleotides by the AT hook domain of TashAT2.
Moreover, variations of the sequence TAAAT flanked by GCrich sequences were often present and repeated in the binding
site, perhaps reflecting the presence of three AT hook motifs
in TashAT2, which would increase the binding affinity.
Interestingly, the binding sites determined for the TashAT2
fusion protein are very similar to the DNA-binding sites
determined for the AT hook DNA-binding HMGI(Y) protein
in the enhancer/promoter region of the human β-interferon
gene (Du et al., 1993). However, the DNA sequence bound in
vivo by the complete TashAT2 polypeptide could be more
specific and could depend on other host or parasite co-factors
binding in a complex. TashAT1 can also bind preferentially to
AT-rich DNA and, although the CAT1 and CAT1M2 probes
contain a GC-rich region, a concatenated probe representing
this region on its own (CAT3) did not bind to the TashAT1
λgt11 expression clone. By contrast, an upstream probe, CAT2,
which represented the AT-rich region of CAT1 showed binding
activity. These results suggest that the AT-rich region of the
CAT1 probe is required for binding. CAT1 forms part of the
putative promoter region of Tams1, but there is no further
evidence to date, that TashAT proteins bind specifically to the
CAT1 region and regulate expression of Tams1. In fact,
changing three bases of CAT1 did not alter the binding
characteristics of TashAT1 significantly, whereas this alteration
did abolish specific binding by polypeptides in T. annulata
parasite enriched nuclear extracts (Shiels et al., 2000). One
possibility for the detection of TashAT1 by the CAT1 motif is
that concatenated DNA-binding probes can isolate proteins that
T. annulata DNA-binding proteins
bind to related DNA motifs, as shown by the isolation of
HMGI(Y) by a screen with a concatenated octamer motif
(Eckner and Birnstiel, 1989).
The AT hook domains of TashAT2 and TashAT1/3 are very
similar, but there are some notable differences. TashAT2 has
three AT hook motifs, whereas TashAT1/3 has four. Also
present in TashAT1/3 are two small basic repeats, similar in
sequence to regions found in the HMG1/2 DNA-binding
domains (Landsman and Bustin, 1993). It could be proposed,
therefore, that TashAT1/3 has a stronger affinity for DNA than
TashAT2 and that their sequence specificities will be different.
The individual members of the TashAT family show
remarkable sequence conservation. In particular, TashAT1 is
virtually identical to the 5′ part of TashAT3, and TashAT2 and
TashAT3 are very similar over a region beginning at the AT
hook domain. This suggests a very recent duplication event has
occurred, possibly owing to adaptation to in vitro cell culture.
Southern blotting of DNA derived from a range of in vitro cell
lines and in vivo derived piroplasm DNA, however, indicates
the existence of all three gene copies (data not shown).
Therefore, it would appear that gene duplication has occurred
in vivo and is not unique to a few in vitro infected cell lines.
Identical copies of genes have been found in other
apicomplexan parasites; there are two identical copies of the
elongation factor α in Plasmodium knowlesi and Plasmodium
berghei (Vinkenoog et al., 1998), of the rhoptry associated
protein from Babesia bovis (Suarez et al., 1998) and of the
Rop2 gene from Toxoplasma gondii (Beckers et al., 1997).
Northern blot analysis of various cell lines indicated that
in general, TashAT2 was found to be the most consistently
and highly expressed of the genes. This may be related to
the previous detection of anti-TashAT2 reactivity in
macroschizont-infected in vivo derived cells (D. G. S. and B.
R. S., unpublished). Analysis of RNA from tick derived
sporozoites showed differences in the TashAT expression
profile. TashAT3 is the major TashAT RNA to be expressed in
sporoblasts/sporozoites, while the RNA species representing
TashAT1 was also detected. The relationship between these
expression patterns and the functional role of TashAT1/3 is not
clear but their primary role might be during sporoblastogenesis
in the tick salivary gland. Alternatively, they could function to
allow the establishment of the parasite following sporozoite
invasion of the bovine leukocyte. Either of these roles could
involve transport to the host cell nucleus and modulation of tick
or bovine gene expression.
The levels of TashAT1 and TashAT3 mRNA were found to
be plastic across different cell lines. Thus, TashAT3 was less
abundant than TashAT2 in the majority of cell lines, but more
abundant in the D7B12 cell line. A higher level of TashAT3
expression in D7B12 cells was supported by immunoblot and
immunofluorescence data. In a similar fashion, TashAT1
mRNA levels were extremely low in the non-cloned cell lines.
The reasons for this plasticity in expression are unclear but may
be related to the derivation of individual cell lines or clones.
Furthermore, it is possible that the host cell background can
influence the expression of the TashAT gene cluster, as
expression levels were significantly lower in the TBL20 line
derived from sporozoite infection of previously immortalised
BL20 cells. One possible consequence of the plasticity of
TashAT expression is that it may relate to the observed
differences in bovine gene expression displayed by individual
2753
cell lines and different passages of the same cell line or clone
(Adamson et al., 2000; Sutherland et al., 1996).
The presence of three TashAT genes in a small cluster,
coupled to the possible presence of TashAT2 and TashAT1/3 in
a host nucleus, provokes the idea that at least two (and possibly
more) parasite genes are involved in modulation of the host
environment by altering the control of leukocyte or tick cell
gene expression. Whether these proteins target different subsets
of genes is unknown, but modulation of host cell environment
could include induction of host cell division, as rearrangements
of the AT hook that contains HRX gene fused to a variety of
partners are common in several types of leukaemia (Waring and
Cleary, 1997). Furthermore, the archetypal AT hook-containing
protein, HMGI(Y) and the closely related HMGIC are
upregulated in proliferating, non-differentiated cells, and
chromosomal translocations involving HMGI genes are found
in many types of neoplasia (Hess, 1998). Thus, AT hook DNAbinding proteins play a very important role in the control of
eukaryotic gene regulation and proliferation. It is therefore not
unreasonable to speculate that the TashAT cluster performs
similar tasks in Theileria-infected cells.
We are grateful to Professor Duncan Brown and the C.T.V.M,
University of Edinburgh, for providing the T. annulata-infected cell
lines; Diyarbakir and Hidirseyh. D. G. S., K. P. and C. O. were
supported by grants from the Wellcome Trust, R. S. by a BBSRC
studentship, and L. S. by a MRC studentship.
REFERENCES
Adamson, R., Logan, M., Kinnaird, J., Langsley, G. and Hall, R. (2000).
Loss of matrizx metalloproteinase 9 activity in Theielria annulataattenuated cells is at the transcriptional level and is associated with
differentially expressed AP-1 species. Mol. Biochem. Parasitol. 106, 51-61.
Aravind, L. and Landsman, D. (1998). AT-hook motifs identified in a wide
variety of DNA-binding proteins. Nucleic Acids Res. 26, 4413-4421.
Baylis, H., Megson, A. and Hall, R. (1995). Infection with Theileria annulata
induces expression of matrix metalloproteinase 9 and transcription factor
AP-1 in bovine leukocytes. Mol. Biochem. Parasitol. 69, 211-222.
Beckers, C. J. M., Wakefield, T. and Joiner, K. A. (1997). The expression
of Toxoplasma proteins in Neospora caninum and the identification of a gene
encoding a novel rhoptry protein. Mol. Biochem. Parasitol. 89, 209-223.
Botteron, C. and Dobbelaere, D. (1998). AP-1 and ATF-2 are constitutively
activated via the JNK pathway in Theileria parva-transformed cells.
Biochem. Biophys. Res. Commun. 246, 418-421.
Carrington, M., Allsop, B., Baylis, H., Malu, N. M., Shochat, Y. and Sohal,
S. (1995). Lymphoproliferation caused by Theileria parva and Theileria
annulata. In Molecular Approaches to Parasitology, pp. 43-56. New York:
Wiley Liss.
Church, G. M. and Gilbert, W. (1984). Genomic sequencing. Proc. Natl.
Acad. Sci. USA 81, 1991-1995.
Devereux, J., Haeberli, P. and Smithies, O. (1984). A comprehensive set of
sequence analysis programs for the VAX. Nucleic Acids Res. 12, 387-395.
Dobbelaere, D., Coquerelle, T. M., Roditi. I., Eichhorn. M. and Williams,
R. O. (1988). Theileria parva infection induces autocrine growth of bovine
lymphocytes. Proc. Natl. Acad. Sci. USA 85, 4730-4734.
Dschunkowsky, E. and Luhs, J. (1904). Die Piroplasmosen der Rinder.
Centralblatt fur Bakteriologie, Parasitenkunde. Infektion. Hyg. 35, 486-493.
Du, W., Thanos, D. and Maniatis, T. (1993). Mechanism of transcriptional
synergism between distinct virus-inducible enhancer elements. Cell 74, 887898.
Eckner, R. and Birnstiel, M. L. (1989). Cloning of cDNAs coding for human
HMG I and HMG Y proteins: both are capable of binding to the octamer
motif. Nucleic Acids Res. 17, 5947-5959.
Fell, A. H., Preston, P. M. and Ansell, J. D. (1990). Establishment of
Theileria-infected bovine cell-lines in scid mice. Parasite Immunol. 12, 335339.
2754
JOURNAL OF CELL SCIENCE 114 (15)
Hess, J. L. (1998). Chromosomal translocations in benign tumors: the HMGI
proteins. J. Clin. Pathol. 109, 251-261.
Heussler, V. T., Machado, J., Jr, Fernandez, P.C., Botteron, C., Chen, CG., Pearse, M. J. and Dobbelaere D. A. E. (1999). The intracellular
parasite Theileria parva protects T-cells from apoptosis. Proc. Natl. Acad.
Sci. USA 96, 7312-7317.
Hulliger, L. (1965). Cultivation of three species of Theileria in lymphoid cells
in vitro. J. Protozool. 12, 649-655.
Irvin, A. D. and Morrison, W. I. (1987). Immunopathology, Immunology
and Immunoprophylaxis of Theileria infection. In Immune Responses
in
Parasitic
Infections:
Immunopathology,
Immunology
and
Immunoprophylaxis (ed. E. J. L. Soulsby), pp. 223-274. Boca Raton FL:
CRC Press.
Irvin, A. D., Brown, C. G. D. and Kanhai, G. K. (1975). Comparative growth
of Bovine lymphosarcoma cells and lymphoid cells infected with Theileria
parva in athymic (nude) mice. Nature 255, 713-714.
Johnson, K. R., Lehn, D. A., Elton, T. S., Barr, P. J. and Reeves, R. (1988).
Complete murine cDNA sequence, genomic structure and tissue expression
of the high mobility group protein HMG-I(Y). J. Biol. Chem. 34, 1833818342.
Landsman, D. and Bustin, M. (1993). A signature for the HMG-1 box DNAbinding proteins. BioEssays 15, 539-546.
McHardy, N., Wekesa, L. S., Hudson, A. T. and Randall A. W. (1985). Antitheilerial activity of BW720C (buparvaquone): a comparison with
parvaquone. Res. Vet. Sci. 39, 29-33.
Morzaria, S. P., Roeder, P. L., Roberts, D. H., Chasey, D. and Drew, T. W.
(1982). Characteristics of a continuous suspension cell line derived from a
calf with sporadic bovine leukosis. In Fifth International Symposium on
Bovine Leukosis (ed. O. C. Straub), pp. 519-528. Luxembourg: Commission
of the European Communities.
Ole-MoiYoi, O. K., Brown, W. C., Lams, K. P., Nayer, A., Tsukamoto, T.
and Macklin M. D. (1993). Evidence for the induction of casein kinase II
in bovine lymphocytes by the intracellular protozoan parasite Theileria
parva. EMBO J. 12, 1621-1631.
Pipano, E. and Shkap, V. (1979). Attenuation of two Turkish strains of
Theileria annulata. J. Protozool. 26, 80.
Pollock, R. and Treisman, R. (1990). A sensitive method for the
determination of protein-DNA binding specificities. Nucleic Acid Res. 18,
6197-6204.
Reeves, R. and Nissen, M. S. (1990). The AT-DNA-binding domain of
mammalian high mobility group I chromosomal proteins. J. Biol. Chem.
265, 8573-8582.
Shiels, B. R., McDougal, C., Tait, A. and Brown, C. G. D. (1986).
Identification of infection-associated antigens in Theileria annulata
transformed cells. Parasite Immunol. 8, 69-77.
Shiels, B. R., Kinnaird, J. H., McKellar, S., Dickson, J., Ben Miled, L.,
Melrose, R., Brown, D. and Tait, A. (1992). Disruption of synchrony
between parasite growth and host cell division is a determinant of
differentiation to the merozoite in Theileria annulata. J. Cell Sci.107, 99107.
Shiels, B. R., Smyth, A., Dickson, J., McKellar, S., Tetley, L., Fujisaki, K.,
Hutchison, B. and Kinnaird, J. H. (1994). A stoichiometric model of stage
differentiation in the protozoan parasite Theileria annulata. Mol. Cell. Diff.
2, 101-125.
Shiels, B. R., Fox, M., Mckellar, S., Kinnaird, J. and Swan, D. G. (2000).
An upstream element of the Tams1 gene is a site of DNA-protein interactions
during differentiation to the merozoite in Theileria annulata. J. Cell Sci.
113, 2243-2252.
Suarez, C. E., Palmer, G. H., Hötzel, I., Hines, S. A. and McElwain, T. F.
(1998). Sequence and functional analysis of the intergenic regions
separating Babesial rhoptry-associated protein-1 (rap-1) genes. Exp.
Parasitol. 90, 189-194.
Sutherland, I. A., Shiels, B. R., Jackson, L., Brown, D. J., Brown C. G. D.
and Preston, P. M. (1996). Theileria annulata: altered gene expression and
clonal selection during continuous in vitro culture. Exp. Parasitol. 83, 125133.
Swan, D. G., Phillips, K., Tait, A. and Shiels, B. R. (1999). Evidence for
localisation of a Theileria parasite AT hook DNA-binding protein to the
nucleus of immortalised bovine host cells. Mol. Biochem. Parasitol. 101,
117-129.
Triezenberg, S. J. (1995). Structure and function of transcriptional activation
domains. Curr. Opin. Gen. Dev. 5, 1990-1996.
Vinkenoog, R., Aparecida-Sperança, M., van Breemen, O., Ramesar, J.,
Williamson, D. H., Ross-MacDonald, P. B., Thomas, A. W., Janse, C. J.,
del Portillo, H. A. and Waters, A. P. (1998). Malaria parasites contain two
identical copies of an elongation factor 1 alpha gene. Mol. Biochem.
Parasitol. 94, 1-12.
Waring, P. M. and Cleary, M. L. (1997). Disruption of a homolog of trithorax
by 11q23 translocations: leukemogenic and transcriptional implications.
Curr. Top. Microbiol. Immunol. 220, 1-23.
Williamson, S., Tait, A., Brown, D., Walker, A., Beck, P., Shiels, B.,
Fletcher, J. and Hall, R. (1989). Theileria annulata sporozoite surface
antigen expressed in Escherichia coli elicits neutralising antibody. Proc.
Natl. Acad. Sci. USA 86, 4639-4643.