volume 13 Number 8 1985
Nucleic Acids Research
Eleven new sequence variants of citrus exocortis viroid and the correlation of sequence with
pathogenicity
Jane E.Visvader and Robert H.Symons
Adelaide University Centre for Gene Technology, Department of Biochemistry, University of
Adelaide, Adelaide, South Australia 5000, Australia
Received 4 March 1985; Accepted 22 March 1985
ABSTRACT
Full-length double-stranded cDNA was prepared from p u r i f i e d c i r c u l a r RNA
of two new Australian f i e l d isolates of c i t r u s exocortis v i r o i d (CEV) using
two synthetic oligodeoxynucleotide primers. The cDNA was then cloned into the
phage vector M13mp9 for sequence analysis. Sequencing of nine cDNA clones of
isolate CEV-0E30 and eleven cDNA clones of isolate CEV-J indicated that both
isolates consisted of a mixture of v i r o i d species and led to the discovery of
eleven new sequence variants of CEV. These new variants, together with the
six reported previously, form two classes of sequence which d i f f e r by a
minimum of 26 nucleotides in a t o t a l of 370 to 375 residues. These two
classes correlate with two b i o l o g i c a l l y d i s t i n c t groups when propagated on
tomato plants where one produces severe symptoms and the other gives rise to
mild symptoms. Two regions of the native structure of CEV, comprising 18% of
the t o t a l residues, d i f f e r between the sequence variants of mild and severe
i s o l a t e s . Whether or not both of these regions are essential for the
variation i n pathogenicity has yet to be determined.
INTRODUCTION:
Viroids are an unusual group of autonomously replicating plant pathogens
and consist of a low molecular weight, c i r c u l a r , single-stranded RNA molecule
which exists as a highly base-paired, rod-like structure (1,2). No viroid
appears to encode protein products, either in vivo (3) or in vitro (4,5).
These infectious agents must therefore interact d i r e c t l y with host factors,
via sequence and structural signals, in order to replicate and exert their
pathogenic effects. Sequence comparisons of naturally occurring variants of
the same viroid are of importance in defining the conserved and variable
features of the viroid molecule and may reflect regions that have a role in
viroid replication and symptom expression in the host plant.
Citrus exocortis viroid (CEV) is the causative agent of the exocortis
disease of citrus (6). In addition to citrus species (family Rutaceae), CEV
also infects members of the Solanaceae, Compositae, Cucurbitaceae, Leguminosae
and Umbelliferae families (7). An isolate of CEV is defined as the viroid RNA
obtained from a single citrus tree in the f i e l d and may consist of one or more
© IRL Press Limited, Oxford, England.
2907
Nucleic Acids Research
sequence v a r i a n t s .
CEV-A, our reference Australian i s o l a t e (8) and a
Californian i s o l a t e , CEV-C ( 9 ) , were both found to be mixtures of at least two
RNA species.
In a d d i t i o n , we have sequenced the major, or only variant
two other Australian i s o l a t e s , CEV-DE25 and CEV-OE26 (10).
in
A l l of the s i x
sequence variants of CEV reported so far contain 371 residues.
In t h i s paper
we report the sequences of CEV variants found in two new Australian i s o l a t e s ,
CEV-J and CEV-0E30.
Sequence analysis revealed that both isolates contained a
complex mixture of variants and has defined 11 new sequence variants which
vary i n length from 370 t o 375 residues.
Thus, 17 n a t u r a l l y
occurring
sequence variants of CEV have now been determined.
D i f f e r e n t f i e l d isolates of potato spindle tuber v i r o i d (PSTV) e x i s t
which produce symptoms varying from mild t o lethal on tomato plants
(1).
These changes in pathogenicity have been correlated with nucleotide
differences occurring i n one region of the PSTV molecule ( 1 ) .
causes d i f f e r e n t pathological effects on certain hosts ( 7 ) .
Like PSTV, CEV
Although CEV i s
usually associated with stunting of growth and scaling of bark of c i t r u s
species propagated on Poncirus t r i f o l i a t a rootstock, dwarfing can sometimes
occur without scaling (11,12).
The symptoms of exocortis can take months to
appear on sensitive indicator plants such as Etrog c i t r o n (Citrus medica) or
several years for orange trees (C. sinensis) grafted on P.
rootstock ( 6 ) .
trifoliata
A more convenient host t o study s t r u c t u r e - f u n c t i o n
relationships of CEV i s tomato (Lycopersicon escuientum) on which symptoms
appear two to three weeks a f t e r i n o c u l a t i o n .
We have propagated f i v e
d i f f e r e n t isolates of CEV on tomato and have i d e n t i f i e d two d i s t i n c t
b i o l o g i c a l groups, severe isolates and mild i s o l a t e s .
A comparison of the
sequence variants of mild and severe isolates has allowed the determination of
regions of the CEV molecule which are associated with pathogenicity.
MATERIALS AND METHODS
Materials
Avian myeloblastosis virus reverse transcriptase was obtained from
Molecular Genetic Resources ( F l o r i d a ) and bacteriophage T4 DNA ligase and
polynucleotide kinase from Boehringer.
The Klenow fragment of E. c o l i DNA
poiymerase I and a-32P-dATP and a-32P-dCTP (both 2,400 Ci/mmol) were obtained
from BRESA Pty. Limited (Adelaide).
CEV-specific oligodeoxynucleotides were
kindly synthesized by Stephen Rogers and Derek Skingle.
The phage M13-
s p e c i f i c sequencing oligodeoxynucleotide primer was from New England Biolabs.
2908
Nucleic Acids Research
Propagation of CEV Isolates and Purification of CEV
Isolate CEV-DE30 was kindly provided by Dr. M. Schwingharaer as a leaf
sample of a Bellamy navel orange on P. t r i f o l i a t a rootstock (Yanco, New South
Wales; budline 033). Isolate CEV-J was a leaf sample of a Leng navel orange
tree on P. t r i f o l i a t a rootstock and was kindly provided by R.H. Lloyd,
Paringi, New South Wales. Both trees were dwarfed and showed marked scaling
of the bark of the rootstock. Isolates CEV-A, CEV-DE25 and CEV-0E26 have been
described previously (8,10). Extracts (13) of infected leaves were inoculated
onto one week old tomato seedlings (L. esculentum cv. Grosse Lisse) and grown
under a r t i f i c i a l l i g h t at 28°C for 14 h and in the dark about 22°C for 10 h.
Partially purified nucleic acid extracts of tomato leaves harvested three to
four weeks after inoculation were prepared (13) and used either for further
propagation or for purification of CEV (13).
The biological properties of isolates CEV-A, CEV-DE25, CEV-DE26, CEV-DE30
and CEV-J were examined on tomato. Plants were harvested four weeks after
inoculation and nucleic acids extracted (14) from 15 g of infected leaves.
The levels of CEV in these extracts were compared by Northern blot analysis
(14).
Synthesis and Cloning of Full-Length Double-Stranded cDNA of CEV
The procedure used to construct full-length cDNA clones of CEV (Fig. 1)
was similar to that of Fields and Winter (15) for the preparation of f u l l length cDNA clones of influenza virus RNAs. First strand cDNA was synthesized
from 1 pg of circular CEV and 2.2 ug of DNA primer d(CGAAAGGAAGGAGACGAGCTCCTG)
for 1.5 h at 44°C with 27 units of reverse transcriptase in a volume of 20 y l ;
the concentration of KC1 was reduced from 70 mM to 35 mM. The products were
denatured at 100°C for 2 minutes, incubated with 0.1 ug/ml RNase A at 37°C for
1 h and then fractionated by electrophoresis on a 3% polyacrylamide gel in 90
mM Tris base, 90 mM boric acid, 2 mM EDTA, 8 M urea, pH 8.3. Full-length cDNA
was eluted, 5'-phosphorylated with ATP and polynucleotide kinase, hybridized
with 0.27 ug of 5'-phosphorylated primer, d(CTGCTGGCTCCACAUCCGA), and
incubated with the Klenow fragment of DNA polymerase I for 1 h at 25°C. The
reaction was terminated by extraction with water-saturated phenol:chloroform
(1:1), followed by ether washing, ethanol precipitation and, f i n a l l y , precipitation with 5 mM spermine (16) to remove unhybridized DNA primer and residual
unincorporated deoxynucleotides. After blunt-end ligation of the doublestranded cDNA into the Smal site of the replicative form of the phage vector
M13mp9 (17) using T4 DNA ligase, E. coli JM101 was transformed and singlestranded phage M13 DNA prepared from recombinant plaques. Phage DNA was
2909
Nucleic Acids Research
sequenced by the dideoxy nucleotide chain termination technique (18) using an
M13 specific DNA primer (17-mer). Gel compressions were resolved by the use
of dITP (19) in the sequencing reactions or by electrophoresis of reaction
mixtures in 6% polyaery 1 amide gels containing 40% formamide and 8 M urea.
RESULTS
Biological Properties of Five Isolates of CEV on Tomato
Propagation of isolates CEV-A (8), CEV-OE25 (10), CEV-0E26 (10), CEV-DE30
and CEV-J on tomato has identified two groups of isolates according to t h e i r
biological properties: (1) severe isolates which produce severe stunting and
leaf epinasty, and (2) mild isolates which are essentially symptomless. The
results are summarized in Table 1 . When equivalent amounts of extract from
plants infected with each isolate of CEV were examined by Northern blot
analysis after denaturation of the nucleic acids by glyoxalation (20), no
significant differences were found in the concentration of CEV (data not
shown). This finding indicated that variation in symptom expression from mild
to severe was independent of the level of CEV in the host plant. Differences
in pathogenicity could therefore be d i r e c t l y correlated with changes in
nucleotide sequence of the variants.
Construction of Full-Length cDNA Clones
Preliminary sequencing of the viroid p u r i f i e d from tomatoes infected with
the CEV-0E30 and CEV-J isolates using partial enzymatic cleavage methods (21)
revealed that the viroids were mixtures of RNA species and thus could not be
sequenced by direct methods. In order to determine the sequences of the
variants in each i s o l a t e , i t was necessary to construct f u l l - l e n g t h cONA
clones ( F i g . 1 ) . F i r s t strand cDNA synthesis was i n i t i a t e d on circular CEV
with a synthetic oligonucleotide (24-mer) which hybridized to residues 175-198
(8) at the right-hand end of the proposed native structure of CEV. For second
strand synthesis, a synthetic oligonucleotide (19-mer) corresponding to
residues 199-217 was used as primer. The double-stranded cDNA was then bluntend ligated into the Smal site of the phage vector M13mp9 to produce f u l l length clones. The length of the oligonucleotide and/or the site of priming
were found to be important c r i t e r i a for e f f i c i e n t synthesis of f i r s t strand
cDNA. I n i t i a l l y , a short DNA primer (15-mer), complementary to part of the
highly conserved central region of viroids (22), was used but i t was found to
prime poorly. In contrast, the 24-mer DNA primed very e f f i c i e n t l y at the
right-hand end of CEV where there i s less secondary structure and few G-C base
2910
Nucleic Acids Research
TABLE 1 :
Biological
P r o p e r t i e s on Tomato S e e d l i n g s o f F i v e I s o l a t e s o f CEV
REFERENCE
ISOLATE
PATHOGENICITY ON TOMATO
( c v . Gross
pairs.
Lisse)
CEV-A
8
Severe
CEV-DE25
10
Severe
CEV-DE26
10
Mild
CEV-DE30
T h i s work
CEV-J
T h i s work
Mild
Severe
The complete sequences of 20 cDNA clones were determined by the
dideoxy chain termination procedure (18).
Sequences of Variants in Isolates CEV-DE30 and CEV-J
Four variants were found in the nine cDNA clones of CEV-DE30 that were
sequenced, while nine variants were found in 11 cDNA clones of CEV-J
sequenced. These variants f a l l into two classes on the basis of their
sequence: Class A sequences are very similar to that of CEV-A (8) while
Class B sequences can be closely aligned with that of CEV-DE26 (10) which
differs from CEV-A by 27 nucleotide changes (Fig. 2 ) . The nucleotide changes
occurring in the sequence variants of the CEV-DE30 and CEV-J isolates relative
to either of the CEV-A or CEV-0E26 isolates are shown in Fig. 3. Sequence
changes are indicated with respect to CEV-A i f the variant belongs to the
so 0 /r
Cool to RT
\
s—x
(
\
dNTPs
Reverse Transcriptase
B
A
24 ntier primer
electrophoresis
3
T4 ONA l l g u e
Sma I
M13 mp»3
'
5
'
1. 19mer primer c
2. dNTPs * Klenow
3'
RF
Figure 1: Outline of the procedure used in the construction of full-length
cONA clones of CEV. See Materials and Methods for d e t a i l s . The site of
priming of the 24-mer DNA primer at the right-hand end of the predicted native
structure of CEV is indicated by an arrow.
2911
Nucleic Acids Research
2%
is
B :a
«
1:8
:s
Us
W
III
si!
3:3
Sil
S.'S
:::=
jii
Is :s
H—8
iii
S|
S.'S
3:s
-O
in -a
o t-
sit
o o
1- 01
a.—
o
3
c c
•a
0.0
a> —.
W
(0
<
o
2912
SIS,
CO
lis
CO
CO
<
s d
3:5
I- O
•
Z3
US
CT T3 O J
•r
CU1
Nucleic Acids Research
Class A of sequence or relative to CEV-OE26 if the variant belongs to Class B.
The nucleotide differences occurring in each sequence variant are summarized
in Table 2 where variants (V) of CEV-DE30 are designated V-a to V-d, and those
of CEV-J are designated V-a to V-h.
The four sequence variants of isolate CEV-0E30 belong to Class B and
differ in only one to three residues. Isolate CEV-J exhibited greater
complexity and consisted of seven sequence variants of Class A (CEV-J, V-a to
V-g) and two of Class B (CEV-J, V-h and V-i). Only one variant was identical
between the CEV-0E30 and CEV-J isolates. The variant RNAs of CEV-0E30 and
CEV-J range in size from 370 to 375 residues (Table 2 ) . In contrast to the
five sequence variants of PSTV which are all 359 residues long (1), it would
appear that strict conservation of the size of the CEV molecule is not an
essential feature for replication.
Only one of the sequence variants of isolates CEV-DE30 and CEV-J which
have been determined so far has been found previously. Thus, these two
isolates have defined 11 new sequence variants of CEV. The complexity of CEVDE30 and CEV-J suggests that the sequencing of further full-length cDNA clones
will define even more new variants.
Apart from the sequence heterogeneity occurring at six residues (residues
250, 251, 260, 263, 264 and 278) in the central region of the native structures of CEV-0E30 and CEV-J (Fig. 3 ) , the nucleotide differences are restricted
to two regions on either side of the centre of the molecule. The majority of
the nucleotide changes occurring in the variants of each sequence class alter
the predicted secondary structure relative to that of the reference sequence
of each class (Fig. 2 ) . Although the nucleotide changes at residues 260, 263,
264 and 278 are present in the centre of the native structure, they do not
occur among the residues that constitute the highly conserved central region
found in CEV, PSTV, chrysanthemum stunt viroid (CSV), coconut cadang cadang
viroid (CCCV), tomato apical stunt viroid (TASV) and tomato planto macho
viroid'(TPMV) (22,23).
Seventeen Sequence Variants of CEV and their Correlation with Pathogenicity
Sequence analysis of six different isolates of CEV has defined 17
variants which appear to form two classes of sequence, Class A and Class B
(Figs. 2, 3 ) . The two classes correlate with differences in pathogenicity on
tomato plants: Class A sequences are found in severe isolates of CEV (CEV-A,
CEV-DE25 and CEV-J) while Class B sequences are found in mild isolates (CEVDE26 and CEV-0E30) as well as in CEV-J. CEV-J, a mixture of both Class A and
B sequences, is associated with severe symptoms since the Class A sequence is
2913
UCUU *G6U0CC U %eu aC Ue»CCI) 8
g
f
if.
I
' '
SO
371
39
50
/ii
^jt-300
.y. 300
A
Au ^ C C
Jf ^300
CEV-DE3O SEQUENCE VARIANTS
.L.-.-W
1
2.
c*iuiKtf
*ASAGSC«S CS 66S8**MAS t '
i r /i
eeu8C*6«C*CS***»M*
CEV-J SEQUENCE VARIANTS
C
KCIT *Mtracei^»Miew«cii"ecg«cj.M a t * i»«
\cc»u^uUt9tccM»t
ccctu*iatsah e*McctfCv^EBC| e B
/
\
1
1.
CLASS B SEQUENCES
I
U
JSO
80*i)CU
\
t
37i
c
CEV-J SEQUENCE VARIANTS
CLASS A SEQUENCES
6
c,
*
8
C(KG
C
* « £ CC6«86
WO
COGS««« *««ir e
^B'GUCUCUCUUCB,
if.
*
-k
fc
e
^250
S MSU C k6BnCGBB6SS * S
^.250
6c
\
if. if. if.
1. .
"' A
C
if.jf.if.wa
WO
*
150
C
U66 C
U
I
200
C
C
C
200
aC6UC(Clltlac
200
CC*GC68*6*»»
iC C 8 C60AH C tWBBC
150
150
tC *«6°CSSfiMSfi* e « C a«C«Kfi6l) e M C 6 eGMtl C
"ii\ I
* « e CCBBB 6
jtActgUctc,
K
s s
*0CC CCGS6fi
100
«SBAB C U
U
IG6ABCU
D «ccuc g a
Nucleic Acids Research
present. Although the biological properties of the Californian isolate of CEV
(CEV-C) sequenced by Gross et al. (9) are unknown, we predict that it would
cause severe symptoms on tomato, on the basis of its Class A sequence.
Monomeric full-length cDNA clones of the sequence variants CEV-A and CEVAM (10) in the BamHI site of the vector pSP6-4 (24) are infectious on tomato
where they produce severe symptoms (unpublished results). Both these variants
are Class A sequences and thus confirm the correlation between Class A
sequence and the severe phenotype. Unfortunately, the monomeric full-length
cDNA clones of isolates CEV-J and CEV-DE30 described in this paper are not
infectious on tomato. This is presumably a consequence of the site in the CEV
molecule (right-hand end) used in the construction of these clones. This lack
of infectivity precluded the definitive confirmation of the correlation
between Class B sequences and the mild phenotype.
Sequence heterogeneity within the 17 sequence variants of CEV occurs in
the central region and in two regions on either side of this. The nucleotide
changes located in the central area occur in both severe and mild isolates
(Fig. 3) and are thus independent of changes in pathogenicity. Severe and
mild isolates differ in two regions, designated P L and PR (P = pathogenicity),
of the native structure (Figs. 3,4). It is concluded that either one or both
of these regions are responsible for the variation in pathogenicity.
Potential Translation Products of CEV Variants
Potential polypeptide products of the plus and minus strands of each
sequence variant were analysed since any conserved polypeptide sequence may
indicate a possible functional role in vivo. We have previously reported the
putative translation products of CEV-A (8) where both AUG and GUG were used as
possible initiation codons. Four out of the 17 variants considered here
contain one AUG in the plus strand while all 17 variants contain one AUG in
the minus strand. There are no fully conserved or even partially conserved
AUG-initiated polypeptides in the minus strand. Only one potential
polypeptide product of 15 amino acids is conserved between the CEV variants;
this initiates with a GUG at residue 339 and terminates at residue 13 (UGA) of
the plus strand. Since a 5'-proximal AUG initiation codon appears essential
for the initiation of translation of eukaryotic RNA (25,26), it seems highly
Figure 3: Classification of the sequence variants of isolates CEV-DE30 and
CEV-J into Class A and Class B sequences. The sites of the nucleotide differences occurring in the sequence variants of each isolate, relative to either
CEV-A (Class A) or CEV-DE26 (Class B) are indicated by a star. Isolate CEV-J
contains both Class A and Class B sequences whereas isolate CEV-DE3O contains
Class B sequences only.
2915
Nucleic Acids Research
TABLE 2: SEQUENCE HETEROGENEITY OF VAR IANTS IN ISOLATES CEV-DE30 AND CEV-J
CEV
ISOLATE
SEQUENCE VARIANTS
(V) DETERMINED
FOR EACH ISOLATE
SIZE
(RESIDUES)
CEV-0E30
(Class B)
V-a
V-b
V-c
V-d
371
370
371
372
CEV-J
(Class A)
V-a
371
V-b
371
V-c
371
V-d
372
V-e
372
V-f
374
V-g
375
V-h
V-i
371
371
CEV-J
(Class B)
NUCLEOTIDE CHANGES RELATIVE
TO EITHER CEV-A (CLASS A) OR
CEV-DE26 (CLASS B ) *
301 U + A
250 -A, 301 U + A
263 A -• C
46 G * A, 70-71 + G, 263 A * C,
301 U + A
234 U + A, 264 G + U, 278 A + U,
301 A * G
50 G + A, 234 U + A, 264 G + U,
278 A + U, 301 A + G
70 -G, 129-130 +U, 234 U + A,
263-264 AG • CU, 278 A • U,
313 G + A
129-130 +U, 234 U • A,
263-264 AG •>• CU, 278 A + U
129-130 +U, 234 U + A,
251 G + C, 263-264 AG -> CU,
278 A + U, 313 G • A, 321 A + U,
129-130 +U, 132-133 +C,
134-135 +G, 234 U * A, 260 U + C,
263-264 AG + CU, 278 A • U,
313 G + A
129-130 +U, 132-133 +C,
134-135 +G, 231 G + C, 234 U + A,
235-236 +C, 263-264 AG + CU,
278 A • U, 313 G + A
301 U + A
251 G + C, 301 U • A
* Residue numbers at the sites of change in either CEV-A (Class A) or CEV-0E26
(Class B). Nucleotide changes in CEV-J (Class A) are indicated relative to
CEV-A while nucleotide changes in CEV-J (Class 8) and CEV-DE30 (Class B) are
indicated relative to CEV-DE26. Nucleotide exchanges are represented by X •»• Y
where X occurs in CEV-A or CEV-DE26 and Y occurs in variants of isolates CEV-J
or CEV-DE30. An insertion i n a sequence variant is represented by + Y and a
deletion by - X.
2916
Nucleic Acids Research
72
c
/A
324
287
P,
128
139
n
o
238
227
PD
Figure 4: Diagram showing the two regions, P L and P R (boxed), of the native
structure of CEV that differ between severe and mild isolates (Table 1 ) .
Severe symptoms on tomato correlate with Class A sequences and mild symptoms
with Class B sequences. The residue numbers on the outside of the boundaries
of the two regions are indicated and are the same for both Class A and Class B
sequences (Fig. 2 ) .
unlikely that CEV encodes any functional polypeptide products. These findings
are consistent with the lack of conservation between possible translation
products of both the plus and minus strands of the closely related viroids,
CEV, PSTV and CSV (8,21). These data, therefore, provide further evidence
that viroids rely entirely on host-encoded components for their replication.
DISCUSSION
Sequence analysis of full-length cDNA clones of the CEV-A, CEV-DE30 and
CEV-J isolates has indicated that each isolate is a mixture of RNA species.
It is feasible that all naturally-occurring viroid infections contain a
mixture of two or more sequence variants. If this is a general feature of
viroid infection then i t has important implications in the correlation of
viroid sequence with a particular biological property. Fortunately, the
ability to prepare infectious full-length cDNA clones of viroids (27-30,
unpublished results) allows experiments requiring infection by a single
sequence variant to be carried out.
The large number of sequence variants occurring in some isolates of
CEV, in particular CEV-J, could be accounted for by either, (1) a high copy
error rate (31) by an RNA polymerase replicating a single RNA species, or (2)
infection of one plant by several sequence variants during propagation of
citrus varieties by grafting or during regular practices such as pruning
(6). The long potential l i f e of citrus trees in the field (over 60 years)
would allow the accumulation of sequence variants in each tree by either
route. It is not known i f each of the sequence variants of CEV replicate with
a similar efficiency or i f a mixture is required for complementation of
function to enable infection. The second possibility appears unlikely since
2917
Nucleic Acids Research
certain full-length cDNA clones (i.e. unique species) of PSTV (27,30), hop
stunt viroid (28,29) and CEV (unpublished results) are infectious.
Two regions of the native structure of CEV, P L and P R (Fig. 4 ) , were
found to differ in sequence between the two biological groups defined by Class
A and Class B sequences (Fig. 2 ) . These two regions fall within the P (pathogenicity) and V (variable) domains of viroids described by Keese and Symons
(32). The P|_ region is characterised by an oligo(A) sequence conserved in all
viroids and has been found to correlate with variation in pathogenicity of
PSTV on tomato (1). This suggests that at least the P L region of CEV is
associated with symptom expression. The PR region of CEV is the most variable
domain between all viroids (32). Although variation occurs at two bases in
the Ppj region of the five PSTV variants sequenced so far (1), there is no
correlation with pathogenicity. However, it remains to be determined whether
or not the PR region is associated with pathogenicity in the host plant. One
approach to resolving this problem would be to construct full-length chimeric
cDNA clones of CEV with the two possible arrangements of the P L and P R regions
derived from mild and severe sequence variants.
Sanger (1) has recently proposed a mechanism for the involvement of the
left-hand variable region of PSTV in pathogenicity, whereby this region
becomes increasingly unstable, and hence is more accessible to interaction
with host factors, as pathogenicity increases. Since CEV and PSTV share
approximately 60% sequence homology (8,9) and possess similar host ranges and
biological properties (6,7), the free energies (AG) of the predicted secondary
structures of the corresponding Pj_ regions of sequence variants within severe
and mild isolates of CEV were calculated (33; D. Riesner, personal communication)
In direct contrast to PSTV, it was found that the stability of the P L region
(residues 48-72, 297-324) of severe variants (AG -51 to -69 KJ/mol) was
considerably greater than that of the mild variants (AG +7 to -5 KJ/mol).
Furthermore, there does not appear to be any consensus sequence in the Pj_
regions of the mild and severe forms of PSTV and CEV. It is therefore
possible that pathogenicity is influenced by a number of different sites in
the P L region and/or the PR of CEV rather than a single site. In vitro
mutagenesis of the P L and PR regions of full-length cDNA clones of CEV, in
conjunction with infectivity studies, should define the site(s) that play a
role in determining viroid pathogenicity.
The correlation between two classes of CEV sequence variants and
differences in pathogenicity on tomato, may not apply to citrus. Isolates
CEV-0E30 and CEV-0E26 both produce mild symptoms on tomato but differ on
2918
Nucleic Acids Research
citrus where CEV-0E30 is associated with scaling of the bark of the rootstock
and dwarfing of the orange tree and CEV-DE26 is associated with dwarfing only
(M. Schwinghamer, personal communication).
Furthermore, CEV-OE25, which
d i f f e r s in both sequence and symptom expression on tomato relative to CEV-DE3O
and CEV-DE26, is associated with identical symptoms to CEV-DE26 on orange
trees on P. t r i f o i i a t a rootstock (M. Schwinghamer, personal communication).
CEV presumably interacts with homologous cellular components in c i t r u s and
tomato, but the pathological effects d i f f e r according to the host.
ACKNOWLEDGEMENTS
We thank D r . M. Schwinghamer and R.H. Lloyd f o r s u p p l y i n g CEV-infected
c i t r u s samples, D r . D. Riesner f o r unpublished thermodynamc d a t a , Stephen
Rogers and D r . Derek S k i n g l e f o r s y n t h e t i c o l i g o n u c l e o t i d e s , Paul Keese f o r
d i s c u s s i o n s , J e n n i f e r Cassady and Sharon Freund f o r a s s i s t a n c e and O r . R . I . B .
Francki f o r glasshouse f a c i l i t i e s .
This work was supported by t h e A u s t r a l i a n
Research Grants Scheme and by a Commonwealth Government Grant t o the Adelaide
U n i v e r s i t y Centre f o r Gene Technology.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
Sanger, H.L. (1984) i n The Microbe: Part I , V i r u s e s , Mahy, B.W.J. and
P a t t i s o n , J . R . , E d s . , p p . 281-334. Society f o r General Microbiology
Symposium 36, Cambridge U n i v e r s i t y Press, Cambridge.
Diener, T.O. (1983). Adv. Virus Res. 28_, 241-283.
Conejero, V. and Semancik, J . S . (1977). Virology ]]_, 221-232.
Davies, J . W . , Kaesberg, P. and Diener, T.O. (1974). V i r o l o g y 6 1 ,
281-286.
Semancik, J . S . , Conejero, V. and Gerhart, J . (1977). V i r o l o g y 8 0 ,
218-221.
Diener, T.O. (1979). V i r o i d s and V i r o i d Diseases, W i l e y , New York.
Runia, W.Th. and P e t e r s , D. (1980). Neth. J . P I . Path. J56_, 135-146.
Visvader, J . E . , Gould, A . R . , Bruening, G.E. and Symons R.H. (1982).
FEBS L e t t 137, 288-292.
Gross, H . J . , Krupp, G . , Domdey, H . , Raba, M., Jank, P . , Lossow, C ,
A l b e r t y , H . , Ramm, K. and Sanger, H.L. (1982). Eur. J . Biochem. 1 2 1 ,
249-257.
Visvader, J . E . and Symons, R.H. (1983). Virology 130, 232-237.
Fraser, L.R. and Broadbent, P. (1979). Virus and Related Diseases of
C i t r u s i n New South Wales, Department of A g r i c u l t u r e , New South Wales.
Bowman, F.T. (1956). Citrus-Growing i n A u s t r a l i a , Angus and Robertson,
Sydney.
P a l u k a i t i s , P. and Symons, R.H. (1980). J . Gen. V i r o l . 4 6 , 477-489.
Hutchins, C , Keese, P . , Visvader, J . E . , Rathjen, P . , McTrfnes, J . and
Symons, R.H. (1985). Plant M o l . B i o l . , i n press.
F i e l d s , S. and W i n t e r , G. (1982). Cell 28, 303-313.
Hoopes, B.C. and McClure, W.R. (1981). Nucleic Acids Res. 2.. 5493-5504.
Messing, J . and V i e i r a , J . (1982). Gene 19, 269-276.
2919
Nucleic Acids Research
18.
19.
20.
21.
22.
23.
24.
25.
Sanger, F . , Coulson, A.R., B a r r e l l , B . G . , S m i t h , A . J . H . and Roe, B.A.
(1980) J . M o l . B i o l . 143, 161-178.
Symons, R.H. (1979). Nucleic Acids Res. 7 , 825-838.
McMaster, G.K. and Carmichael, G.G. (19777. Proc. N a t l . Acad. S c i . USA
]±, 4835-4838.
H a s e l o f f , J . and Symons, R.H. (1981). Nucleic Acids Res. 9_, 2741-2752.
H a s e l o f f , J . , Mohamed, N.A. and Symons, R.H. (1982). Nature 299, 316321.
K i e f e r , M.C., Owens, R.A. and Diener, T.O. (1983). Proc. N a t l . Acad.
S c i . USA 8 0 , 6234-6238.
M e l t o n , D7A~., K r i e g , P.A., R e b a g i i a t i , M.R., M a n i a t i s , T . , Z i n n , K. and
Green, M.R. ( 1 9 8 4 ) . Nucleic Acids Res. _12_, 7035-7056.
B a r a l l e , F.E. and Brownlee, G.G. (1978). Nature 274, 84-87.
26.
Kozak, M. (1978).
27.
Cress, D . E . , K i e f e r , M.C. and Owens, R.A. (1983). Nucleic Acids Res.
Uj 6821-6835.
Ohno, T . , Ishikawa, M., Takamatsu, N . , Meshi, T . , Okada, Y . , Sano, T.
and S h i k a t a , E. (1983). Proc. Japan Academy J59_, Ser. B 251-254.
Meshi, T . , Ishikawa, M., Ohno, T . , Okada, Y . , Sano, T . , Ueda, T . and
S h i k a t a , E. (1984). J . Biochem. 9 5 , 1521-1524.
T a b l e r , M. and Sanger, H.L. (1984T7 EMBO J . 2, 3055-3062.
H o l l a n d , J . , S p i n d l e r , K., Horodyski, F . , Grabau, E . , N i c h o l , S. and
Van de P o l , S. (1982). Science 215, 1577-1585.
Keese, P. and Symons, R.H. (1985T7" P r o f . N a t l . Acad. S c i . USA,
in press.
H a s e l o f f , J . and Symons, R.H. (1982). Nucleic Acids Res. 10, 3681-3691.
28.
29.
30.
31.
32.
33.
2920
Cell _15_, 1109-1123.