Journal of General Virology (1993), 74, 2437-2443. Printed in Great Britain
2437
Nucleotide sequence evidence for the occurrence of three distinct
whitefly-transmitted geminiviruses in cassava
Y. G. H o n g , 1 D. J. Robinson I and B. D. Harrison 2.
1 Scottish Crop Research Institute, Invergowrie, Dundee DD2 5DA and 2Department of Biological Sciences,
University of Dundee, Dundee DD1 4HN, U.K.
The complete nucleotide sequence of the D N A of Indian
cassava mosaic virus (ICMV) and a key part of that of a
group B isolate of African cassava mosaic virus from
Malawi (ACMV-M) were determined and compared at
the nucleotide and encoded amino acid levels with the
published sequences of an ACMV group A isolate
(ACMV-K) and other whitefly-transmitted geminiviruses (WTGs). The D N A of ICMV consists of two
circular single-stranded molecules, DNA-A [2815 nucleotides (nt)] and DNA-B (2645 nt), which differ substantially in sequence from the genome components of
ACMV-K (DNA-A 70 %, DNA-B 47 % sequence identity) and other WTGs. ICMV DNA-A contains eight
open reading frames (ORFs) encoding proteins of > 100
amino acid residues, of which four ORFs (one genome
sense, three complementary sense) are comparable to
those of other WTGs. DNA-B contains one O R F in
each sense, as in other WTGs. None of the putative viral
proteins are more similar in amino acid sequence to the
proteins of A C M V - K than to those of another WTG.
The coat protein of ACMV-M is more like that of
tomato yellow leaf curl virus from Sardinia (86%
sequence identity) than those of ICMV or ACMV-K.
The intergenic regions of ACMV-K, ACMV-M and
ICMV DNAs differ in size, and largely in sequence,
except for two 30 to 40 nt sequences which are also
conserved in other W T G s and can form stem-loop
structures. The intergenic region of ICMV D N A
contains three copies of a 41 nt sequence, and that of
ACMV-M D N A contains an imperfect repeat of a 34 nt
sequence which resembles the repeated sequence in
ICMV DNA. The differences between ACMV-K,
ACMV-M and ICMV are considered great enough to
justify their separation as isolates of three distinct
W T G s : African cassava mosaic virus, East African
cassava mosaic virus and Indian cassava mosaic virus.
Introduction
isolates could be assigned to three discrete clusters on
the basis of their patterns of reaction (Harrison &
Robinson, 1988). Evidence for these groupings has been
strengthened by serological tests on virus isolates from
cassava in 11 additional countries in tropical Africa or
the Indian subcontinent (Harrison et al., 1991b; M. M.
Swanson & B. D. Harrison, unpublished). Group A
isolates react with at least 15 of the MAbs and occur in
West Africa, Burundi, Chad and Uganda, and in the
western parts of Kenya and Tanzania. Group B isolates
react with five to nine of the MAbs and are found in
Malawi, Madagascar, Zimbabwe and the eastern parts
of Kenya and Tanzania. Group C isolates occur in India
and Sri Lanka. They react with only two or three of
the MAbs and are considered to represent a separate
geminivirus, which was named Indian cassava mosaic
virus (ICMV; Harrison et al., 1991b).
These groupings are associated with biological differences. For example, whereas group A isolates are
readily transmitted to, and maintained in, Nicotiana
benthamiana, in which they accumulate best at approx.
Cassava mosaic, the most economically important virus
disease of a root crop in Africa, is caused by infection
with a whitefly-transmitted geminivirus (WTG), African
cassava mosaic virus (ACMV; Bock & Woods, 1983). A
similar disease occurring in southern India is likewise
associated with a WTG, particles o f which are serologically related to those of ACMV (Bock & Harrison,
1985). However, serological relationships among different W T G s are common, as shown by tests with
polyclonal antisera (Roberts et al., 1984) or monoclonal
antibodies (MAbs; Harrison et al., 1991a; Swanson et
al., 1992a, b). Tests on 87 virus isolates from mosaicaffected cassava in 10 countries with 17 MAbs to the
West Kenyan type strain of ACMV showed that the
The ICMV nucleotide sequences presented in this paper have been
submitted to the EMBL database and assigned the accessionnumbers
Z24758 (DNA-A) and Z24759 (DNA-B).
0001-1764 © 1993 SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 16:53:13
2438
Y. G. H o n g , D. J. R o b i n s o n a n d B. D. H a r r i s o n
23 °C, g r o u p B isolates (such as A C M V - C ) are transmitted to N . b e n t h a m i a n a with difficulty a n d only rarely
at t e m p e r a t u r e s below 30 °C, a n d they are h a r d or
impossible to m a i n t a i n i n this species. Also, they a t t a i n
only low c o n c e n t r a t i o n s in N . benthamiana, even at
30 °C. G r o u p C isolates can be m a i n t a i n e d in N.
b e n t h a m i a n a b u t m u l t i p l y best at 30 °C ( R o b i n s o n et al.,
1984; H a r r i s o n et al., 1987). The groupings are also
s u p p o r t e d by the results of nucleic acid h y b r i d i z a t i o n
tests with probes for the two circular s s D N A molecules
which m a k e u p the viral genome. T h u s whereas a D N A
p r o b e for D N A - A ( D N A - 1 ) o f the g r o u p A type isolate
of A C M V (from western K e n y a , d e n o t e d A C M V - K )
detected all isolates tested, a full-length p r o b e for D N A B ( D N A - 2 ) reacted only weakly, if at all, with the D N A
of isolates in groups B or C ( R o b i n s o n et al., 1984;
H a r r i s o n et al., 1987). The D N A - A p r o b e also detected
other W T G s whereas the D N A - B p r o b e did n o t (Roberts
et al., 1984). W e have n o w explored the relationships
a m o n g the virus isolates f r o m cassava by nucleotide
sequence c o m p a r i s o n s . T h e sequences of the two D N A
species of A C M V - K (group A) were d e t e r m i n e d b y
Stanley & G a y (1983). W e have d e t e r m i n e d the complete
sequences of the two D N A species o f a n isolate of
I C M V , together with key regions of the g e n o m e of a
g r o u p B isolate o f A C M V from M a l a w i ( A C M V - M ) .
T h e results provide further evidence o f the distinctness of
the three isolates, a n d show that the differences a m o n g
t h e m are c o m p a r a b l e in degree to those a m o n g different
geminiviruses. They also show that the nucleotide
sequences o f I C M V a n d A C M V - M c o n t a i n previously
u n d e s c r i b e d repeats in the n o n - c o d i n g region which
occurs in b o t h D N A - A a n d D N A - B .
Methods
Virus sources. The stock culture of ICMV was obtained from a
vegetativelypropagated cassava plant derived from a stake collected in
1986 at Trivandrum, Kerala, India. The virus was transmitted by
inoculation of sap to N. benthamiana and maintained in this host by
mechanical inoculation. ACMV isolate M (ACMV-M) was from a
cassava plant collected at Jalawe, Rumphi, Malawi, and propagated
vegetatively in the glasshouse at Dundee. The viruses were kept under
licence from the Scottish Office Agriculture and Fisheries Department.
Extraction o f I C M V DNA. ICMV ssDNA was extracted as described
by Robinson et al. (1984) from purified virus particles (Sequeira &
Harrison, 1982). ICMV dsDNA was extracted from infected N.
benthamiana leaves and further purified by CsCI gradient centrifugation
(Sambrook et al., 1989).
Preparation of total DNA from cassava infected with ACMV-M.
About 50 mg cassava leaf tissue was ground with a pestle and mortar,
and 250 gl of ice-coldextraction buffer (50 mM-Tris-HC1pH 8.0, 5 mMEDTA, 0"1% BSA, 1"25M-NaC1, 0.1% 2-mercaptoethanol) and 25 gl
10 % SDS were added. After incubation at 65 °C for 5 min, the mixture
was extracted once each with phenol :chloroform (4:1, v/v) and with
chloroform. Sodium acetate (3 M, pH 5"2, 0'1 vol.) and two volumes of
ethanol were added to the aqueous phase, which was kept at - 7 0 °C
for 20 rain. DNA was pelleted by centrifugation at 12000g for 10 rain.
The DNA was washed with 70% ethanol, dried in a vacuum and
redissolved in 200 gl of TE (10 mN-Tris HCI, 1 mM-EDTA, pH 8.0).
Cloning o f l C M V DNA. DNA extracted from ICMV particles was
made double-stranded with the Klenow fragment of Escherichia coli
DNA polymerase I, using as primer an oligonucleotide [5' d(GTAATATTA) 3'] which is complementary to a sequence that is conserved
within the intergenic region of all geminivirusessequenced so far. The
products were cut with EcoRI and cloned in pUC19. A part of ICMV
DNA-B not represented among these clones was synthesized by PCR,
using primers deduced from the adjacent sequences, and was cloned in
pUC19 after digestion with restriction enzymes. Full-length DNA
clones of both DNA-A and DNA-B were also constructed by inserting
dsDNA, purified from infected plants, in the BglII and BamHI sites,
respectively, of pUC19.
PCR and cloning of the coat protein gene and common region of
ACMV-M. A virus-specific DNA fragment was amplified from total
DNA from infected plant tissue by PCR using two degenerate primers
whose sequences were derived from conserved regions in published
sequences of WTG genomes. The 20-mer, P1 (5' d[GG(C/T)CACGA(T/C/A)TTAAT(T/G)AG(G/A)GAI3'), corresponded to residues
397 to 416, and the 24-mer,P2 (5' d[GTTGAT(C/T)ATGTATTGT(A/
T)(T/C)(G/C/A)ATGTA] 3"), to residues 1276 to 1299 in ACMV-K
DNA-A (Stanley & Gay, 1983).
Reaction mixtures (100 gl) contained 1 to 2 pg total DNA, 75 pmol
each of P1 and P2, 100 gM of each dNTP, 2.5 mM-MgCI~,50 mM-KC1,
0.1 mg/ml gelatin, 10 mM-Tris HC1 pH 8.0 and 2-5 units of Taq DNA
polymerase (Cambio). The reaction mixture was overlaid with 50 gl of
light oil, and PCR was initiated with 1 cycle of 94 °C for 2 rain, 54 °C
for 1.5 rain, 72 °C for 2 min, followed by 30 cycles of 94 °C for 30 s,
54 °C for 1 min and 72 °C for 1.5 min, and finally by 1 cycle of 72 °C
for 5 min, in a Cambio Intelligent Heating Block. After electrophoresis
in a 1.0% LMT agarose (Gibco BRL) gel, the PCR product was
recovered by extraction with phenol--chloroform, then treated successively with proteinase K and T4 DNA polymerase (Sambrook et al.,
1989), and cloned into the HincII site of pUC19.
DNA sequencing. Nucleotide sequences were determined by the
dideoxynucleotide chain termination method (Sanger et al., 1977)
which used Klenow fragment polymerase (Pharmacia) or Sequenase
(US Biochemicals). Single-stranded DNA, obtained from subclones in
M 13rap18 or M 13rap19, and double-stranded plasmid DNA were used
as templates. Compressions in some sequences were resolved by
replacing dGTP with 7-deaza-dGTP in the sequencing reaction.
Data analysis. Sequences were aligned and analysed using the
UWGCG computer programs (Devereux et al., 1984). Sequencesof the
followinggeminiviruseswere used in the comparisons: abutilon mosaic
(AbMV; Frischmuth et al., 1990), ACMV-K (Stanley & Gay, 1983),
ACMV Nigerian isolate (ACMV-N; Morris et al., 1990), bean golden
mosaic (BGMV; Howarth et al., 1985), potato yellowmosaic (PYMV;
Coutts et al., 1991), squash leaf curl (SqLCV; Lazarowitz & Lazdins,
1991), tomato golden mosaic (TGMV; Hamilton et al., 1984), tomato
leaf curl-Australia (TLCV-A; Dry et al., 1993), tomato yellowleaf curlIsrael (TYLCV-I; Navot et al., 1991) and TYLCV-Sardinia (TYLCVS; Kheyr-Pour et al., 1992) viruses.
Results
Organization of the
ICMV
genorne
I C M V was f o u n d to have two species of circular s s D N A ,
D N A - A a n d D N A - B . Each species was completely
sequenced in b o t h directions. D N A - A was sequenced
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 16:53:13
Three geminiviruses from cassava
(a)
,-
*
,kTA
~
340
~
~
.
r
c
N
A
=
~
X
It
P
a
J~ D
=
n
Z
Z
e
=
T
P
c
a
c
B
K
V
~
]~
X
R
,
L
N
~
Y
D
J
P
Y
=
400
J
It
400
x x v e T v n v ~ x Q s w ~ e e e z s x x e e w Y e x Y n e e o v p s o c s o
p
C
K
VQ
v
K
g
Z
D
F
G
B V V N N F D W ~
y
A
J
I~
N
A
CA
~
T
y
K
N
J
p
Z
J
J
H
Z
L
g
K
Z
WNp
Q
&
I
N
&
L
e
V
lq
V
Y
D V V H
R
A
T
SN
I
G
X
VlIC
~
Z
X
~
S
~'~
NX
P
~ T A T V K I ~ N E R
le
$e
~q V
~
X
Z
N
R
I~
y
V
~
V
Y
y
~
fir
~
D
(~
V
V
T
levx
J
R
Q
X
r
l,
~v
D
R
Y
Q
V
O
K
Jt
G
X
*
~
N
' ~ ' A O A A ~ A ' e ~ A
z
Q KN
N
Z
z
fl
C
Z
v
Ni I
•
fl
P
R
V
X
Y
~
Y
l
P
l~
g
lq
C
G
A
Jt
--
:---
L
p
C oN
Q
T
H
~
P
o
t
~
P V D
Q
•
R
Y
C
x
•
O
K W N A ~ V ~ Q
Y
B
~g I~
T
I[
N
A
OQ
L
~
L
Y
Y
~
N
L
1~20
X
I
L
W
S
O
•
v
:
~
v
'r
z
~
D
0
N
T
Z
.T, 1~ o
Y
R
J
D
~
Y
y
~
L
x
J V I " / ~ A q ' N ~ ~ k ' I ~ C ' G
•
CA
Q
X
X
L
V
P
D
N
v
V
B
1(10
Q
Z
R
X
K
F
R
W
X
0
Z
O
R
L
H
K
B
M
O
,~
R
~
A
P
V
A
A
AE ~
R
D
0 p VQ ~0 I N Pi, ~'t 0B ~
~'
fJ
B
L
P
~'
p
N
E
y
(b)
5
~
W
y
L
X
T
Lo R
Z
I
"
~'t'Xq'A,k
p
A
G
l
d
It
i
~
R
dR y
R
dt
K
D
Q
F
G
p
D
p
T
•
H
N ~ N
L
VN
Z
J
le
T
O
T V ~
•
K
Y
B A T
K
PL
P
JY P O L F O
I~
A
o
0
D
~
Z
~
VII
~
J8
k
g
3F p
~
Q
P
L
~
:
¥
•
~
Vtl'
~
J
v
N
Z
I~
~
T
JC 1~
:
~
O
-
~
~
T
"4
K
HA
n
K
,
II
V
O
X
~
G
,
O
•
~
w
II
~
p
V
R]P
y
K
dr
J
°
N
m
•
•
C
~
-
~
L
~
I(
,
~
e
V
B
~
.
y
T
K
~
Y
N
x
x
x
a
A
y
~r
p
I~
•
T
r
r
pit
n
It
X
L
A
R
It
N
.~
F
G
O
D
~
A
R
~.
~'
5"
Y
J~
T
y
~
•
X
PV
X
T]e
L
dl
O
~
~
V
G
D
Y
•
D
•
S
P
if
L
O
L
L
F
T
C
V
LV
Jt
D
K
Z
N A B
E
DR
P
TV
G
P
Z
0D
L~Y
R
R
F
O
,
~
O
R
Y
P
Z
W
T
g
F
K
D
D
G
O
~,
IB D
V
J
•
0
L
|
]i(
Y
C
K
Y
V
T
t
Y
r
C
Z
A
X
~
c
~
v
T
~
S
Z
Z
D
,'L~'.A,r
.
,z.~',Je,,~~
N
~
D
,
~
__
Y A D L
'~
~
CAq,CC"Z,'pCAJ k L ~
&
HO
M
'
V
L
120
K
it
'
'/
~
~
z
~
'
A
0
D
~
k
't'~A'rk
OOGAT
1080
1200
JI
lOOO
~
2400
2520
2645
Fig. 1. Nucleotide and encoded amino acid sequences of I C M V D N A - A (a) and ICMV DNA-B (b).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 16:53:13
2439
2440
Y. G. Hong, D. J. Robinson and B. D. Harrison
2815/1
(((~
A
A+ ~ A R 0
ALlt(I
2000"~~/AR1
A-)1
2000"v~
~_~/,./~
lOOO
£
AL5
AL4
-1ooo
AL3
2645/1
2645/1
2000
shown in black (Fig. 2). The amino acid sequences of the
putative proteins encoded by these ORFs are shown in
Fig. 1.
2815/1
R
IBR 1
1000
2000
~ ~ 0 0 0
Fig. 2. Arrangement of ORFs in ICMV DNA-A (2815 nt) and DNAB (2645nt) shown in the genome (+) and complementary( - ) senses.
ORFs in black (ALl, AL2, AL3, AR1, BL1 and BR1) are conservedin
most, and those in grey (AL0, AL4, AL5 and AR0) are conservedin a
few other WTGs. Potential promoter sequences (V), polyadenylation
signals (V) and the common region (CR) are also indicated.
from three contiguous cDNA clones bounded by EcoRI
sites. Sequences at the junctions were confirmed by
sequencing appropriate fragments of a full-length clone
of DNA-A. More than half of DNA-B was sequenced
from two cDNA clones which abutted at an EcoRI site.
The sequences at the junction, and of the gap remaining,
were determined from clones of the products obtained by
PCR using primers based on the known sequences.
Subsequently the part of a full-length-clone that overlapped both ends of the larger cDNA clone was
sequenced. The sequences from the full-length clone and
of the PCR-derived product differed in only four out of
1630 nucleotides, and only three of these differences
resulted in an amino acid change. Where such nucleotide
differences occurred, the version occurring in the fulllength clone is given precedence.
The final sequences [DNA-A, 2815 nucleotides (nt);
DNA-B, 2645 nt] are shown in Fig. 1. As in other WTGs,
the sequences of DNA-A and DNA-B are different
except for the common (intergenic) region (labelled CR
in Fig. 2), which is shared. The first nucleotide in this
common region is defined as number 1 in each sequence.
In total, the two DNA species contain 10 open reading
frames (ORFs), encoding polypeptides of more than 100
amino acid residues, in either the genome ( + ) sense or
complementary ( - ) sense strand (Fig. 2). The ORFs
shown in black in Fig. 2 are found in several other
WTGs, and those in dark grey are shared with a few
other WTGs (such as ACMV-K). Potential promoter
sequences (TATA or CAAT boxes) and polyadenylation
signals (A/GATAA) are available for all the ORFs
Comparison of DNA sequences of I C M V and other
geminiviruses
When the nucleotide sequences are compared with
those of other geminiviruses, such as ICMV, ACMV-K,
BGMV from Puerto Rico and TYLCV-S (which lacks a
DNA-B), the sequence of ICMV DNA-A is equally
similar (70 %) to the comparable genome components of
ACMV-K and TYLCV-S, but is less similar to BGMV
DNA-A (62%), especially at nucleotides 1 to 500 but
also at other points in the sequence. As with other
WTGs, much less similarity is found among the DNA-B
sequences: ICMV and ACMV-K, 47 %, and ICMV and
BGMV, 46%. Moreover, no part of DNA-B is well
conserved although the region at about nucleotides 1800
to 2000 is somewhat more conserved than the rest of the
sequence.
Similarities of I C M V gene products with those of other
whitefly-transmitted geminiviruses
The putative products of the six ORFs (black in Fig. 2)
found in ICMV and 10 other WTGs were compared.
Table 1 shows the percentage amino acid sequence
identities obtained in these comparisons. Among the
proteins encoded by ICMV DNA, those derived from
genes AR1 and ALl are the most strongly conserved
when compared with the proteins of other WTGs. The
proteins encoded by the ICMV genes are most similar to
those of TYLCV-I (AR1), TLCV-A (ALl, AL2, AL3),
SqLCV (BR1) and TGMV (BL1). Thus none of the
encoded proteins is more similar to those of ACMV-K
than to the proteins of another WTG. As expected from
comparisons of the nucleotide sequences, the products of
the ORFs in DNA-B are much less conserved between
ICMV and other WTGs than those of the DNA-A ORFs
(Table 1). Also, the amino acid sequence identities
obtained by comparing ICMV with other WTGs were all
markedly less than those obtained by comparing ACMVK and ACMV-N (93 to 97% for different ORFs). In
addition, it was found that although TLCV-A, TYLCVI and TYLCV-S have only one genome segment (Navot
et al., 1991; Kheyr-Pour et al., 1992; Dry et al., 1993),
the proteins encoded are as similar to those of ICMV and
other WTGs with bipartite genomes as are the proteins
of many of the bipartite genome viruses to one another.
In general, the data summarized in Table 1 indicate
that ICMV is substantially different from other WTGs
and support the separation of ICMV and ACMVK/ACMV-N as two distinct viruses.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 16:53:13
Three geminiviruses from cassava
Table 1. Percentage amino acid sequence identity of
putative products of ORFs of some whitefly-transmitted
geminiviruses compared with I C M V
Viruses compared
ORF
ICMV/other
ICMV/ACMV-K ACMV-K/ACMV-N WTGs*(range)
AR1
ALl
AL2
AL3
BR1
BL1
78
73
59
63
29
42
96
97
94
93
93
97
74-79
52-81
4%70
49-68
2%33
3646
* AbMV, BGMV, PYMV, SqLCV, TGMV, TLCV-A, TYLCV-I
and TYLCV-S.
1
ACMV
-M
.
.
5~ K'RPGD I I I STPVS
.
.
KVRRRLNFD~
50
p~I~IRVVAPTVRVT*
RS KLWTNRPM
AC~/V-K
.............
G .............
R..AT
.... H. ~N.KRA.V
ACMV-N
.............
G .............
R..AT
.... H..N.KRA.M
ICMV
..... A .......
G .............
SS.AAV
51
ACMV-M
.
.
YRKPKMYRMYRS
AC~4V- K
ACMV-N
.
.... T .........
I.R ..........
.... M .........
I.R ..........
N...RW
A(3WIV - M
GI TRRVGKRWCVKS
I01
.
.
.
F ........
L. ICKVI..,T..P
F ........
L. ICKVI..
150
KKQNRTNRVMFFLVRDRRPYGPS
I ..........
ACMV-N
.L .........
I ......................
IC~V
.L ..........................
151
.
.
,T° .I
.
.L .........
ACMV-
,T..P
F.S.H..V.I.K.M.I..
I YI LGKIWMDENT
L..T
........
N.I.Y.L
.......
NA
N...Y.L
......
IT'/"
TK .... S ...........
.
I
i00
A(~MV - K
ACMV-M
.....
YEQRDDVKHTGMVRCVSDVARGS
.....................
VDK*
.
200
PQDFGQVFNMFDNEPSTATVENDLRDRy~/LRKFYATVVGGPSGMKEQAL
K
......
ACMV-N
......
I(I~V
.....
I ............
I .......
F ......
I ............
IE ......
F ......
E ................
201
Af3vIV - M
.
.
MH .........
.
H ......
H...I
LYC
......
...........
WH...T..QYAS
.
K
.... Y.L.H..T
ACMV-
N
.... Y.L.H..T
IC~V
.R .... V..Y
ACMV-M
YFYDAVTN
ACMV-K
.... SIG.
AC~V-N
.... SIG.
ICMV
.... S.S.
.....
250
VKRFFRI~QEQAKI'ENHTENALLLY~%CTHAS~TLK
ACMV-
IR I
..... AG ...............................
..... AG ...............................
.... Q..AG
..........
(82 %) and ICMV (80 %). Of the viruses listed in Table
l, PYMV has the coat protein with the least similar
sequence (71% identity) to that of ACMV-M. Comparison of the sequences of the four W T G isolates from
cassava shows that amino acid changes among them
occur at many points in the coat protein, except that
there are very few changes between the sequences of
A C M V - K and A C M V - N (Fig. 3). The sequence data
therefore support the serological data (Harrison et al.,
1991 b) in distinguishing the ACMV group B isolate from
group A and ICMV isolates. All four coat protein ORFs,
like those of several other WTGs, end with two stop
codons, TAATAA.
Nucleotide sequences of I C M V and A C M V - M common
regions
....
....
K.*QS
.
PDVPKGCEGPCKVQS
ICf4V
......
2441
M ....................
258
Fig. 3. Differencesamong the amino acid sequencesof the coat proteins
of ACMV-M, ACMV-K, ACMV-N and ICMV. The asterisks mark
gaps inserted to maintain alignment.
Coat protein sequence of A C M V - M
A cloned 1600 nt fragment of ACMV-M DNA, which
contained both the coat protein gene (ORF AR1) and
the intergenic region, was sequenced in both directions
and each nucleotide was determined at least four times.
The deduced amino acid sequence of ACMV-M coat
protein (Fig. 3) is similar in size (257 residues) to the
sequences of the coat proteins of ICMV (256 residues),
and A C M V - K and ACMV-N (both 258 residues).
However, among WTGs, the ACMV-M sequence is
most similar to the comparable sequences of TYLCV-S
(86 % identity) and TYLCV-I (85 %), and somewhat less
similar to the sequences of ACMV-N (83 %), A C M V - K
Another indication of the extent of relationship between
two WTGs can be obtained by comparing the nucleotide
sequences of the common regions in their DNA. In
different geminiviruses these sequences are substantially
different, apart from a sequence of about 30 nt that is
capable of forming the stem-loop structure which is
conserved in all geminiviruses.
Among the cassava geminiviruses, the common region
differs considerably in length. In A C M V - K it is 196 nt
whereas DNA-A and DNA-B of ICMV share a stretch
of 404 nt (Fig. 4) with only four differences. In ACMVM, however, the common region (defined as the sequence
between O R F A L l and AR0; see Fig. 2) is about 225 nt.
Although some regions of similarity occur outside the
stem-loop sequence, the sequences of the common
regions of the three isolates are substantially different
(Fig. 4). Nevertheless the sequence similarities of the
common regions of the cassava geminiviruses (ACMVK / I C M V 64%, A C M V - K / A C M V - M 63%, ACMVM / I C M V 59 %) are mostly somewhat greater than those
between any o f the cassava viruses on the one hand and
other WTGs, such as BGMV, T G M V and TYLCV-S, on
the other (44 to 59 %).
The differences in size of the common regions of
the cassava geminivirus DNAs are associated with the
occurrence of repeated sequences. In ICMV, both DNAA and DNA-B have a 41 nt sequence which occurs three
times with only two nucleotide changes (Fig. 4).
Similarly, in ACMV-M, a 34 nt sequence is repeated with
only five changes (Fig. 4).
The common regions of all previously sequenced
geminivirus DNAs contain a sequence predicted to form
a stable stem-loop structure. The equivalent structure,
including the sequence T A A T A T T A C in its loop, is also
found in ICMV and ACMV-M ( - 8 4 - 8 k J/tool; Fig. 4).
In addition, formation of another such structure in the
D N A of both viruses and of A C M V - K can be predicted
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 16:53:13
2442
Y. G. Hong, D. J. Robinson and B. D. Harrison
Af~IV - K (A)
1
.
.
.
.
...... CTC.AACTAGAGACACTCTTGAGCATCTCCTCCTATTAATTC-GAG
II I ITITIII
I
11 11111
I IIIIIill
A C I M V - M (A)
ATG CGAGCGAATTC-GAGACACCTCGGTTGATGTCCTCTACTGAATTq~AG
IC~MV (A)
TTCTCTCTC
A(3MV- K (A)
51
ACATTATATAGGTGTCT
A C I M V - M (A)
ACAATATATAGTTGTCT
IC~V(A)
T CAATATATACATGAGTAC
CAAA~TAGATGTAAATAATGGAA
i01
.
.
...... TAA~TTAAAAGG
.
A , C ~ V - K (A)
ACMV-M(A)
TTTTCAAAATTTGAACCAAAA
i
I 11
II[I
li
CTCAATCGGTACTCAGATAGTTrAGCC
[
II
C C CATATTAGGTAC
i00
.
.C T A A A T G G C A T T C T T G T A A T A A G ~ C T r
Ill IIIIII1 IIIII I fill11111
IIIIIIIII
50
i 11111 I I I 1
•C C A A A T G G C A T A A T G G T A A T T A T G T A G A T C T A
II I ~IIlllI[lIIl
IIII II
I
....
.
CTCAAAA~CAGAACACCCAAGGGG
150
I[llllIl fill II
t lIIIllll
I(3MV (A)
... A T A T A A ~ T r C A A A A G C
A C M V - K (A)
151
CCAACCGTATAATATTAC
.G C
II
. . . . . . . . . . . . . . . . . . . . . . . .
GG
Itl 11
I1
........................
CGGTTC-GCC CCGCCCCC
GG
200
...............
llI lllllllliIIIIIIII llIII llIII
. ~ C M V - M (A)
C CATC CGTATAATATTAC
IC~/V (A)
CCATCCGTATAATATTACCGGA~CCCCCC'~CTTTATGGTGGTC
C GGR TGGCt"GCGCCCGRAARA
G R R . G~'~gGA .
A C M V - K (A)
201
.............................
ACMV-M
. .CCC~CTAGGRTGOCCO~GRARTGTGGTCCt"61~CC_~C
[llIIIlllllllI[llHllllilIlllilll
I III[l
.
TTTAATGTGGTC
.
250
CCCGCGCAC
llllllllIlllll]lll
(A)
IIIIII [[[IIIIIIIIIII
I [IIIIIIII I
ICSIV (A)
CCCCCCAOgTGGA~gGcogCGCCCCCCTt~'TTATGGTGGTC~CCCCC~Og
ACI~/V- K (A)
251
TACT~
300
I If
A C I M V - M (A)
%~fGTT
I C ~ V (A)
TGGA TGGCCGCGCCCCCCG
301
.
.
.
.
C CCCCACTCAGAACGCTCGCTCAAAGCTTGTATATGTGTGGTCCCCCTTT
350
I C M V (A)
351
400
I C M V (A)
I%AGTACTTC-GGCAACAAGTCGR~C-CCTGTACAATGTGGGACC~-Fx'~'ACTA
IC~4V (A)
401
AACGA~Ff
ill
~'~-~u ~ ~
CG2~-aAGATGT
435
CCCTGAGTCT~TCCACGGTTTCCGGTG
Fig. 4• Comparisonof the nucleotidesequencesof the common regions
of ACMV-K DNA-A [ACMV-K(A)], ACMV-M DNA-A [ACMVM(A)] and ICMV DNA-A [ICMV(A)]. The sequences are aligned to
show identities. The conserved sequences that can form the two
stem-loop structures are underlined. The sequences that occur three
times in ICMV DNA, and twicein ACMV-M DNA, are shown in bold
italics•
(Fig• 4). This contains the sequence T A T A T A in the loop
and is somewhat less stable (--42.4 to - 5 4 . 6 kJ/mol).
Discussion
The genome organization of ICMV is typical for WTGs
with bipartite genomes, and contains six ORFs that are
equivalent to those of other viruses of this kind. In
addition it contains two ORFs (AR0 and AL0) that are
found both in A C M V - K and in the monopartite genome
WTG, TYLCV-S, but do not occur in all WTGs. Despite
these general similarities, the nucleotide sequences o f
ICMV D N A - A and DNA-B, respectively, are only 62 to
70 % and < 50 % similar to the sequences of A C M V - K
and other WTGs. Also, the putative gene products of
ICMV are substantially different (29 to 81% amino acid
identity) from those of other WTGs. These differences,
together with the large and consistent differences between
the epitope profiles of their particles (Harrison &
Robinson, 1988), and the biological differences already
mentioned (Harrison et al., 1987), support the view that
ICMV and A C M V - K are distinct WTGs. Their distinctness is further emphasized by the inability of ICMV
and A C M V - K to form viable pseudo-recombinants by
reassortment of their genome segments (J. Stanley,
personal communication), and by the remarkably small
amount of variation in epitope profile found e i t h e r
among ACMV isolates from West Africa or among
ICMV isolates from India (Harrison & Robinson, 1988)•
All these comparisons indicate that ICMV and ACMVK are consistently different, and justify the status of two
distinct viruses.
Among WTGs in general, it is becoming increasingly
difficult to decide whether two virus isolates are best
considered strains of the same virus or distinct viruses.
The problem is exacerbated by the difficulty in establishing the host ranges o f the many non-sap-transmissible
WTGs. These host ranges must perforce be determined
by inoculation with virus-carrying vector whiteflies and
can be artificially restricted by the feeding preferences of
the vector, with the result that apparently non-overlapping host ranges can in some instances be shown, by
more exhaustive work (such as tests by agroinoculation),
to overlap or even to coincide• Also, comparison of the
amino acid sequences of W T G proteins (Table 1) shows
that great similarities can exist between viruses that are
biologically very different. One pointer to the status of
the relationship between two W T G isolates is whether or
not they can form pseudo-recombinants. Another seems
to be provided by comparison of the nucleotide sequences
of their common (intergenic) regions. Although this
sequence typically includes two regions which can form
stem-loop structures, the loops of which are strongly
conserved, other regions differ greatly in different WTGs.
When judged on the size or sequence of their common
regions, both ICMV and ACMV-M are substantially
more different from each other and from A C M V - K than
would be expected for strains of the same virus,
reinforcing the conclusions drawn from comparisons of
the putative viral polypeptides. G o o d taxonomic grounds
therefore exist for giving them different names, and we
suggest that isolates closely similar to ACMV-M should
be called East African cassava mosaic virus (EACMV).
However, although it is important for virologists and
diagnosticians to distinguish between ACMV, EACMV
and ICMV, the value o f this distinction to those
concerned with cassava as a crop is uncertain• This will
depend on whether the three viruses differ in other
properties, such as relative transmissibility by different
whitefly biotypes or species, interaction with virus
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 16:53:13
Three geminiviruses from cassava
resistance genes in cassava or occurrence in plant species
other than cassava.
Accepting that the function o f some viral gene
products, and possibly of some viral non-coding nucleotide sequences, necessitates interactions with host components, the occurrence of related viruses in the same
plant species provides an opportunity to assess the
evidence of adaptation of virus to host species. Among
the putative viral polypeptides of A C M V - K and ICMV,
none are more similar to one another than to those of
WTGs from hosts other than cassava. However, the
common regions of ACMV, EACMV and ICMV share,
as well as the stem-loop regions (nucleotides 147 to 179
and 52 to 62 in Fig. 4) found in other WTGs, some
additional motifs which are not found in other WTGs.
These are A A A T G G C A T (residues 71 to 79), AATT T G A A (residues 108 to 115) and G T G G T C C C C
(residues 235 to 243). Whether these sequences interact
with cassava-specific materials remains to be discovered.
Perhaps the most unexpected features of the nucleotide
sequences of ICMV and EACMV are the three repeats of
41 nucleotides in the common region of ICMV and the
twice (imperfectly) repeated similar sequence in
EACMV. The function of these reiterations is unknown.
We thank P. F. McGrath for providing the source of)~CMV-M and
J. Stanley for allowing us to quote his unpublished results. The work
was funded in part by the EC [Contract TS2A-0137-C(CD)].
References
BOCK, K. R. & HARRISON, B. D. (1985). African cassava mosaic virus.
AAB Descriptions of Plant Viruses, no. 297.
BOCK, K.R. & WOODS, R.D. (1983). Etiology of African cassava
mosaic disease. Plant Disease 67, 994-995.
COUTTS, R. H. A., COFFIN, R.S., ROBERTS, E . J . F . & HAMILTON,
W. D. O. (1991). The nucleotide sequence of the infectious cloned
DNA components of potato yellow mosaic virus. Journal of General
Virology 72, 1515 1520.
DEVEREUX,J., HAEBERLI,P. & SMITHI]?S,O. (1984). A comprehensive set
of sequence analysis programs for the VAX. Nucleic Acids Research
12, 387 395.
DRY, I.B., RIGDEN, J.E., KRAKE, L.R., MULLINEAUX, P.M. &
REZAIAN, M.A. (1993). Nucleotide sequence and genome organization of tomato leaf curl geminivirus. Journal of General Virology
74, 147 151.
FRISCHMUTH, T., ZIMMAT, G. & JESKE, H. (1990). The nucleotide
sequence of abutilon mosaic virus reveals prokaryotic as well as
eukaryotic features. Virology 178, 461-467.
HAMILTON,W. D. O., STEIN,V. E., COUTTS,R. H. A. & BUCK, K. W.
(1984). Complete nucleotide sequence of the infectious cloned DNA
components of tomato golden mosaic virus : potential coding regions
and regulatory sequences. EMBO Journal 3, 2197-2205.
2443
HARRISON, B.D. & ROBINSON, D.J. (1988). Molecular variation
in vector-borne plant viruses: epidemiological significance. Philosophical Transactions of the Royal Society of London B 321,
447-462.
HARRISON, B. D., LENNON, A. M., MASSALSKI,P. R., ROBINSON, D. J.
& THOMAS, J.E. (1987). Geographical variation in geminivirus
isolates associated with cassava mosaic disease. Report of the Scottish
Crop Research Institute for 1986, 179-180.
HARRISON, B. D., MUNIYAPPA,V., SWANSON,M. M., ROBERTS,I. M. &
ROBINSON, D. J. (1991 a). Recognition and differentiation of seven
whitefly-transmitted geminiviruses from India and their relationships
to African cassava mosaic and Thailand mung bean yellow mosaic
viruses. Annals of Applied Biology 118, 299-308.
HARRISON, B. D., SWANSON,M. M., MCGRATH, P. F. & FARGETTE, D.
(1991b). Patterns of antigenic variation in whitefly-transmitted
geminiviruses. Report of the Scottish Crop Research Institute for
1990, 88 90.
HOWARTH, A. J., CATON, J., BOSSERT, M. & GOODMAN, R. M. (1985).
Nucleotide sequence of bean golden mosaic virus and a model for
gene regulation in geminiviruses. Proceedings of the National
Academy of Sciences, U.S.A. 82, 3572-3576.
KHEYR-PouR, A., BENDAHMANE, M., MATZEIT, V., ACCOTTO, J.P.,
CRESPI, S. & GRONENBORN, B. (1992). Tomato yellow leaf curl virus
from Sardinia is a whitefly-transmitted monopartite geminivirus.
Nucleic Acids Research 19, 6763-6769.
LAZAROWITZ, S. G. &; LAZDINS, I. B. (1991). Infectivity and complete
nucleotide sequence of the cloned genomic components of a bipartite
squash leaf curl geminivirus with a broad host range phenotype.
Virology 180, 58-69.
MORRIS, B., COATES, L., LOWE, S., RICHARDSON,K. & EDDY, P. (1990).
Nucleotide sequence of the infectious cloned DNA components of
African cassava mosaic virus (Nigerian strain). Nucleic Acids
Research 18, 19~198.
NAVOT, N., PICHERSKY, E., ZEIDAN, M., ZAMIR, D. & CZOSNEK, H.
(1991). Tomato yellow leaf curl virus: a whitefly-transmitted
geminivirus with a single genomic molecule. Virology 185, 151 161.
ROBERTS, I. M., ROBINSON,D. J. & HAgglSON, B. D. (1984). Serological
relationships and genome homologies among geminiviruses. Journal
of General Virology 65, 1723 1730.
ROBINSON, O. J., HARRISON, B. D., SEQUEIRA,J. C. • DUNCAN, G. H.
(1984). Detection of strains of African cassava mosaic virus by
nucleic acid hybridisation and some effects of temperature on their
multiplication. Annals" of Applied Biology 105, 483-493.
SAMBROOK, J., FRITSCH, E.F. & MANIATIS, T. (1989). Molecular
Cloning: A Laboratory Manual, 2nd edn. New York: Cold Spring
Harbor Laboratory.
SANGER, F., NICKLEN, S. & COULSON, A. R. (1977). DNA sequencing
with chain-terminating inhibitors. Proceedings of the National
Academy of Sciences, U.S.A. 74, 5463-5467.
SEQLrEIRA, J.C. & HARRISON, B.D. (1982). Serological studies on
cassava latent virus. Annals of Applied Biology 101, 33~'2.
STANLEY, J. & GAY, M. R. (1983). Nucleotide sequence of cassava
latent virus DNA. Nature, London 301,260-262.
SWANSON, M.M., VARMA, A., MUNIYAPPA, V. ~¢ HARRISON, B.O.
(1992a). Comparative epitope profiles of the particle proteins of
whitefly-transmitted geminiviruses from nine crop legumes in India.
Annals of Applied Biology 120, 425-433.
SWANSON,M.M., BROWN, J. K., POULOS, B.T. & HARRISON,B. D.
(1992b). Genome affinities and epitope profiles of whitefly-transmitted geminiviruses from the Americas. Annals' of Applied Biology
121,285-296.
(Received 15 April 1993; Accepted 12 July 1993)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 16:53:13