Diagnosis of Flame Chlorosis
`by Reverse Transcription-Polymerase Chain Reaction (RT-PCR)
`
`S. Haber, R. T. Rymerson, and J. D. Procunier, Agriculture and Agri-Food Canada, Research Centre, Winnipeg,
`MB R3T 2M9, G. Murray, Department of Microbiology, University of Manitoba, Winnipeg MB R3T 2N2, and
`S. E. Cvitkovitch, Agriculture and Agri-Food Canada, Research Centre, Winnipeg, MB R3T 2M9
`
`ABSTRACT
`Haber, S., Rymerson, R. T., Procunier, J. D., Murray, G., and Cvitkovitch, S. E. 1995. Diagno(cid:173)
`sis of flame chlorosis by reverse transcription-polymerase chain reaction (RT-PCR). Plant Dis.
`79:626-630.
`
`Flame chlorosis (FC), a viruslike disease of cereals, is associated specifically with double(cid:173)
`stranded RNAs (FCdsRNAs). The FCdsRNAs can be detected by dot blot hybridization assay
`that is adequate for detecting FC-RNA in symptomatic areas of infected leaf tissue, but has
`proved insufficiently sensitive to detect FC-RNA in small (:::;10 mg) quantities of suspect root
`tissue or mycelium of candidate fungal vectors. A reverse transcription-polymerase chain reac(cid:173)
`tion assay to detect FC-RNA (FC-RT-PCR) was developed to improve sensitivity. Total RNA
`was extracted from milligram quantities of test tissue, reverse transcribed, and amplified by the
`PCR. The primer pairs #86 [F:5'-CTATTCGCTTGGCTCAGATCG-3' and R:5'-CCAGAGTA(cid:173)
`GTGACTAGAACAGC-3'] or #307 [F:5'-GTGAAAGTCTTGAGGATGC-3' and R:5'-TTCA(cid:173)
`TCTCTATTGGCACCACG-3'] were used. These primer pairs had been determined from a
`consensus 821-bp FC sequence covering an open reading frame (GenBank No. X59248), and
`were predicted to yield 358- and 347-bp DNA fragments, respectively. Sensitivity of specific
`FC-RNA detection was further enhanced by hybridization of the RT-PCR product to digoxi(cid:173)
`genin-labeled riboprobe (digFC-RNA) used in the earlier FC dot blot hybridization assay; the
`digoxigenin label was subsequently reported with enzyme-linked antidigoxigenin antibody and
`chromogenic substrate.
`
`Flame chlorosis (FC) is a soilborne, vi(cid:173)
`ruslike disease of cereals (5) that is asso(cid:173)
`ciated with specific double-stranded RNAs
`(FCdsRNAs) (6,7). We previously dem(cid:173)
`onstrated a diagnostic assay that exploited
`stringent hybridization of labeled probes
`to disease-specific RNA (FC-RNA) (7).
`Diagnosis based on serology was not fea(cid:173)
`sible because we had not been able to find
`virus particles in diseased tissues (1). The
`dot blot hybridization assay (FC-DBA)
`showed clearly that FC-RNA (predomi(cid:173)
`nantly in the form of FCdsRNA) was pres(cid:173)
`ent at much higher levels in the areas of
`yellow variegation on symptomatic leaves
`than in adjacent green areas (7). Indeed,
`FCdsRNA made up as much as 1 % of the
`dry cell mass in small, localized areas of
`FC leaves that were intensely chlorotic,
`and was readily detected in crude extracts
`made from 1.5-mm disks (10 to 15 µg dry
`tissue mass) (7). By contrast, similar
`
`Description and use of proprietary commercial
`products and processes does not indicate endorse(cid:173)
`ment by Agriculture and Agri-Food Canada, its
`staff or agencies.
`
`Publication No. 1616 Agriculture and Agri-Food
`Canada Winnipeg Research Centre 195 Dafoe Rd.,
`Winnipeg R3T 2M9, Canada.
`
`Accepted for publication 14 February 1995.
`
`© 1995 Department of Agriculture and Agri-Food,
`Government of Canada
`
`626 Plant Disease/ Vol. 79 No. 6
`
`quantities of root tissue from FC plants did
`not usually give positive results in FC(cid:173)
`DBA even though electron microscopy
`showed that some root cells had FC(cid:173)
`specific cytological alterations as extensive
`as
`those found
`in mesophyll cells of
`symptomatic leaf tissue ( 1 ). In root, unlike
`leaf, tissue there are no visible symptoms
`to guide the selection of samples.
`In addition to addressing the need for
`more sensitive and definitive diagnosis, we
`used the approach of detecting FC-RNA to
`search for a fungal vector of the soilbome
`FC agent (4). Based on the observation
`that FC symptoms and FC-RNA appeared
`as early as the one-leaf stage, we hypothe(cid:173)
`sized that a candidate FC vector should
`readily infect leaf and root initials in the
`earliest stages of seed germination (4).
`Pythium spp. meet this requirement in that
`they infect cereal embryos within 24 to 48
`h after planting (3), and FC-DBA showed
`that certain Pythium isolates possessed
`FCdsRNA (4; S. Haber, unpublished). The
`amount of FC-RNA in the Pythium iso(cid:173)
`lates, however, was much lower than in
`chlorotic areas of FC leaf tissue, so it was
`necessary to culture fungal mycelium in
`liters of medium in order to obtain suffi(cid:173)
`cient nucleic acid extract for analysis. To
`analyze diverse fungal isolates from FC
`and non-FC soils more rapidly, it was
`clearly desirable to be able to detect FC(cid:173)
`RNA in extracts prepared at the scale of
`single 1.5-ml microvials from small quan(cid:173)
`tities of mycelium.
`
`When part or all of the sequence of the
`diagnostic RNA is known, it is straightfor(cid:173)
`ward to exploit the coupled reactions of
`reverse transcription-polymerase chain re(cid:173)
`action (RT-PCR) for amplification (15).
`Making a reverse transcript from dsRNA,
`however, may be problematic given its high
`melting temperature (14). This difficulty can
`if disease-specific single(cid:173)
`be overcome
`stranded RNA (ssRNA) also is present, even
`if it constitutes only a small proportion of
`the total ssRNA pool. We demonstrate here
`that tissues with FCdsRNA (6,7,12) also
`contain PC-specific ssRNAs (FCssRNAs),
`which hybridize to riboprobes made from
`cDNA clones of FCdsRNAs.
`The cDNA clones of FCdsRNA that
`were developed for the dot blot hybridiza(cid:173)
`tion assay, FCcDNA-21 and -23 (7), es(cid:173)
`tablish
`together with other FCcDNA
`clones a consensus sequence of 821 bp
`(GenBank No. X59248) that contains an
`open reading frame. This sequence has no
`significant homology with any sequence
`deposited to date in the GenBank and
`EMBL nucleic acid sequence databases.
`Therefore, strong evidence for the pres(cid:173)
`ence of FCssRNA in the original RNA
`extract is shown when specific primers
`from X59248 amplify reverse transcripts
`under stringent annealing conditions and
`yield fragments of predicted sizes. Subse(cid:173)
`quent stringent hybridization of the DNA
`product with PC-specific digoxigenin(cid:173)
`labeled riboprobe (digFC-RNA) confirms
`the presence of FCssRNA in the original
`RNA extract and further enhances the
`sensitivity of detection.
`
`MATERIALS AND METHODS
`Preparation of FC-specific nucleic
`acids. For developing the RT-PCR proto(cid:173)
`cols, FC-RNA, quantified by spectropho(cid:173)
`tometry, was prepared by run-off tran(cid:173)
`scription of cDNA clones using T7 DNA(cid:173)
`directed RNA polymerase and the proto(cid:173)
`cols supplied with the Boehringer Mann(cid:173)
`heim DIG-labeling system (Boehringer
`Mannheim Canada, Laval QC). FCssRNA
`was also prepared from infected leaf tissue
`and from cultured Pythium mycelium us(cid:173)
`ing a method adapted from a protocol for
`the purification of wheat RNA (10). Tissue
`was lyophilized, then ground to a fine
`powder with dry ice in a small mortar.
`About IO mg was transferred to a 1.5-rnl
`microvial containing 400 µl of a freshly
`mixed emulsion of 200 µl each of Solution
`A (1,000 parts phenol, 140 parts n-cresol,
`
`Illumina Ex. 1118
`IPR Petition - USP 10,435,742
`
`

`

`digFC#21 probe
`
`.
`-
`
`. . . . . . . . . . :>
`
`.
`- 5•
`
`S
`~ 0~
`~••••••••••••••••••••••••••••••••••••••••••=•• FCssRNA
`-
`-
`-
`-
`-
`........ ••••••••••••• ............ ~ -
`5' ,__ _______________________________ ..-______________ 3•
`+-~A
`#307R
`s
`=~
`#86F
`
`~
`
`--
`
`#307F
`
`5' ••
`
`3• ~
`
`-
`-
`-
`-
`dlgFC#23 probe
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`-
`
`0
`
`100
`
`200
`
`300
`
`500
`400
`Nucleotides
`Fig. 1. Schematic diagram showing relationship of flame chlorosis-specific primer pairs and digoxigenin-labeled RNA probes to the consensus 821-nu(cid:173)
`cleotide flame chlorosis sequence (GenBank No. XS9248).
`
`600
`
`700
`
`800 821
`
`1 part 8-hydroxyquinoline (wt/vol/wt),
`saturated with Solution B (IO mM Tris(cid:173)
`HCI pH 7.4, 50 mM NaCl, 0.5% [wt/vol]
`napthalene-1,5-disulphonic acid sodium
`salt). This emulsion/tissue powder mixture
`was then continuously inverted at 4°C for
`IO min before separating the phases by
`microcentrifugation (12,000 rpm) for 5
`min. The upper aqueous phase was re(cid:173)
`moved, the volume estimated, and 4-
`aminosalicylate sodium salt (Cat. No.
`A3505, Sigma Chemical Co., St. Louis,
`Mo.) added to a final concentration of 6%
`(wt/vol). After two more extractions with
`Solution A, the RNA in the aqueous phase
`was precipitated by adding 2.5 volumes of
`ice-cold ethanol, incubating at -20°C for
`JO min, and centrifuging at 12,000 rpm for
`20 min at 4°C. The precipitate was resus(cid:173)
`pended in 100 µI of sterile, RNAse-free
`water, extracted with 20 µI of water(cid:173)
`saturated phenol, and the aqueous phase
`collected for subsequent RT-PCR (9).
`RT-PCR. Conditions for RT-PCR were
`adapted from the protocol of Zimmeonann
`(15), which describes the use of a single
`heat-stable enzyme from Thennus thenno(cid:173)
`philus (Tth) for both RT and PCR. The
`following components were added in se(cid:173)
`quence to reaction vials in a PCR thermo(cid:173)
`cycler (Model TC!, Perkin-Elmer Cetus,
`Norwalk, Conn.) at 70°C: 1) 11.4 µl RNA/
`water mixture from the above RNA micro(cid:173)
`extraction; 2) 2.0 µI reverse transcription
`buffer (100 mM Tris-HCI, pH 8.3, 900 mM
`KCI); 3) 1.0 µI (adjusted to 100 ng/µ1) each
`of the forward and reverse primer of pair
`#86 [F:5'-CTAITCGCITGGCTCAGATC(cid:173)
`G-3' and R:5'-CCAGAGTAGTGACTAGA(cid:173)
`ACAGC-3'] or pair #307 [F:5'-GTGAAAG(cid:173)
`TCITGAGGATGC-3' and R:5'-ITCATCT(cid:173)
`CTAITGGCACCACG-3'] (cf. Fig. I); 4)
`2.0 µI of 9 mM MnCl2; 5) 1.6 µI of a mix(cid:173)
`ture of 10 mM dGTP, 10 mM dATP, 10 mM
`dTTP, and 10 mM dCTP. After overlaying
`this RNA/primer mix with 50 µI of mineral
`oil, 4 units (= I µ!) of Tth enzyme
`(Amersham "Tet-z," Boehringer Mannheim
`''Tth" or Perkin-Elmer "ffth" may be used)
`were added, and the combined 20-µl volume
`briefly spun down before incubating at
`
`70°C for JO min. To change the enzyme's
`activity from reverse transcriptase to DNA
`polymerase, 80 µI of DNA polymerase
`buffer (10 mM Tris-HCI, pH 8.3, 100 mM
`KC!, 1.88 mM MgC12, 0.75 mM EGTA and
`5% glycerol) was added to each 20-µl RT
`reaction,
`the combined 100-µl volume
`briefly centrifuged to ensure mixing, and an
`additional 50 µI of mineral oil added. This
`mixture was then subjected to the following
`regime in the PCR thennocycler: 35 cycles
`of I min at 94°C, I min at 55°C, I min at
`75°C (extended I s for each cycle). After
`the final cycle, there was an additional 5
`min at 75°C, before cooling to 4 °C.
`Agarose gel electrophoresis of PCR
`products. Products of FC-RT-PCR reac(cid:173)
`tions were electrophoresed in I% agarose
`gels (Mini Sub-Cell, or Wide Mini Sub(cid:173)
`Cell, Bio-Rad Laboratories, Richmond,
`Calif.) in Tris-acetate EDTA (TAE; 13)
`buffer at 40 V for 2 h, and stained with
`ethidium bromide (13). Sizes of FC-RT(cid:173)
`PCR products were estimated by compari(cid:173)
`son with the migration of the 511-, 396-,
`344-, 298-, and 220-bp bands of a I-kb
`DNA ladder (Life Technologies, Gaithers(cid:173)
`burg Md.).
`Combined FC-RT-PCR and dot blot
`assay (FC-RT-PCR-DBA). Products of
`FC-RT-PCR reactions were denatured in
`0.03 N NaOH, neutralized with an equal
`volume of I M Tris-HCI, pH 7.5, and dot
`blotted to nylon membranes using a mani(cid:173)
`fold (Pierce Easy-Titer, Model 77000,
`Pierce, Rockford, Ill.). The membrane was
`probed with digFC-RNA, and developed
`with anti-digoxigenin alkaline phosphatase
`(antiDIG-AP) and chromogenic substrate
`(7).
`Southern blotting of FC-RT-PCR
`DNA and detection with DIG-labeled
`FC-riboprobe (digFC-RNA). After aga(cid:173)
`rose gel electrophoresis., FC-RT-PCR DNA
`fragments were blotted to nylon membrane
`(Nytran, Schleicher & Schuell, Keene,
`N.H.) with a vacuum-blotting apparatus
`(Model 785, Bio-Rad) following
`the
`manufacturer's instructions. The blotted
`nucleic acids were fixed to the membrane
`by irradiation for 3 min in a microwave
`
`abc de fgh
`
`Fig. 2. Presence of flame chlorosis- specific
`double-stranded and single-stranded RNA (FC(cid:173)
`dsRNA, FCssRNA) in diseased leaf tissue.
`Topologies of RNA had been confirmed as
`described (6,12). a) ssRNA stained with ethid(cid:173)
`ium bromide; b) Northern blot of (a) probed
`with anti-message (cf. Fig. I) digoxigenin(cid:173)
`labeled
`flame chlorosis-specific riboprobe,
`digFC#21-RNA; c) dsRNA stained with ethid(cid:173)
`ium bromide; d) Northern blot of (c) probed
`with digFC#21-RNA; e) ssRNA stained with
`ethidium bromide; f) Northern blot of (e)
`probed with message-sense digoxigenin-labeled
`flame chlorosis-specific riboprobe, digFC#23-
`RNA; g) dsRNA stained with ethidium bro(cid:173)
`mide; h) Northern blot of (h) probed with
`digFC#23-RNA. Bars on right indicate posi(cid:173)
`tions of marker bands from I-kb ladder: 3,054,
`2,036, 1,636, 1,0 I 8 and SI 7 /S06 bp (top to
`bottom).
`
`oven (Litton-Moffat 450 W). The blot was
`then hybridized with digFC-RNA ribo(cid:173)
`probe (cf. Fig. I), and developed with
`antiDIG-AP and chromogenic substrate
`(7).
`
`RESULTS AND DISCUSSION
`FCssRNA as effective template in RT(cid:173)
`PCR. The two primer sets, #86 and #307
`(cf. Fig. I), were effective in the RT-PCR
`amplification of FC-RNA that had been
`made by T7 run-off transcription from
`
`Plant Disease/ June 1995 627
`
`

`

`cDNA clones FC#23 and #21 (7), respec(cid:173)
`tively, (data not shown). In symptomatic
`plant
`tissue, FCssRNA as well
`as
`FCdsRNA is present, as demonstrated by a
`b C
`
`A a
`
`B
`
`comparison of Northern blots of ds- and
`ssRNA extracted from FC leaves (Fig. 2).
`The message-sense digFC-RNA riboprobe
`made from clone FC#23 (cf. Fig. I ) hy(cid:173)
`bridizes only with FCdsRNA, while the
`anti-message sense riboprobe made from
`clone FC#21 hybridizes with both FCds(cid:173)
`RNA and FCssRNA (Fig. 2).
`The FCssRNA present constitutes only a
`small proportion of the ssRNA pool in
`diseased leaf tissue (Fig. 2A,B), whereas
`FCdsRNA accounts for all dsRNA de(cid:173)
`tected in FC leaf tissue (6). Moreover,
`FCdsRNA bands may be discerned di(cid:173)
`rectly in gel electrophoresis of total nu(cid:173)
`cleic acid extracted from FC leaf tissue
`(6). Only FCssRNA present in the pool of
`total extracted RNA can be amplified by
`FC-RT-PCR, as FCdsRNA cannot be
`melted and reverse-transcribed with our
`protocol (data not shown). When RNA
`extracted from FC leaves was used as sub(cid:173)
`strate in RT-PCR with primer pairs #86
`and #307, unique fragments were pro(cid:173)
`duced of the predicted sizes, 358 and 347
`bp, respectively (Fig. 3A). The specificity
`of the amplification was further confirmed
`by Southern blot hybridization at high
`stringency (7) with FC#21 (digFC-RNA)
`riboprobe (Fig. 38 ).
`Sensitivity or detection, and potential
`for diagnosis from tis.ffle containing low
`amounts or FCRNA. In young, heallby
`cereal plant leaves, total RNA is (to the
`limit of detection) al.I in ssRNA form, and
`constitutes about 3 x 1 o-4 of the dry tissue
`mass (J. D. Procunier, unpublished). A.I(cid:173)
`t.hough FCdsRNA may constitute as much
`as 1 % of the dry cell mass in small local(cid:173)
`ized areas of infected leaves (7), it does
`not greatly increase the total amount of
`
`nucleic acid that can be isolated from bulk
`leaf tissue (cf. Fig. 4A in reference 6).
`When primer pair #307 was used, FC-RT(cid:173)
`PCR followed by dot or Southern blotting
`and hybridization to FC#21 riboprobe de(cid:173)
`tected :FCssRNA in as little as I 00 fg RNA
`purified from FC leaf tissue (Fig. 4A,B).
`FCssRNA could therefore be expected to
`be detected from as little as I mg of test
`tissue if FCssRNA made up as much as
`lQ-6 of total RNA. Indeed, FC-RT-PCR
`followed by hybridization with FC#21
`riboprobe detected FC-RNA in 10 µl of a
`100-µl extract made from 10 mg of ly(cid:173)
`ophilized tissue of Pythium arrhenomanes
`Drechs. (Fig. 5; cf. reference 4).
`Primer pair #86 was less effective than
`#307 (Figs. 3A,B). Computer analysis of
`X59248, the sequence on which the selec(cid:173)
`tion for FC-RT-PCR primers is based, in(cid:173)
`dicates that reverse transcription must pro(cid:173)
`ceed through more RNA stem structures
`(not shown), and that the stem structures
`predicted to exist in the sequence covered
`by the. #86 primers are more stable (11 )
`than is the case with primer pair #307. To
`the extent that reverse transcription is im(cid:173)
`peded in the first step of RT-PCR, the dif(cid:173)
`ferences in RNA secondary structure may
`contribute to the clear differences seen in
`overall efficiency of amplification. Even if
`the initial differences in the relative ef(cid:173)
`ficiencies of the RT step are small, the
`exponential effect of subsequent PCR
`amplification can result in substantial dif(cid:173)
`ferences in yield.
`An a.ltemative explanation may be sim(cid:173)
`ply that the primers themselves account for
`the differences in sensitivity. This has been
`demonstrated recently in auempts to opti(cid:173)
`mize amplification of gene elements of
`
`a
`
`C d e
`
`f g h
`
`j
`
`k
`
`A
`B
`
`(cid:141)
`
`Fig. 3. Flame chlorosis-specific primer pairs
`#86 and #307 effective in amplifying flame
`chlorosis single-stranded RNA (FCssRNA) by
`reverse transcription-polymerase chain reaction
`(RT-PCR). (A) DNA product of RT-PCR ampl.i(cid:173)
`ficat.ion of FCssRNA extracted from diseased
`leaf tissue: n) RT-PCR with #86 primer; b) I ·kb
`DNA ladder; c) RT-PCR with #307 primer. (B)
`Southern blot of gel shown in (A) probed with
`flame chlorosis- specific
`digoxigenin-Jo.beled
`riboprobe, dmgFC#2 1-RNA. The 1,636-bp and
`396-bp bands of the I-kb DNA ladder hybridize
`to digFC#21-RNA and are indicated by square
`dots.
`
`628 Plant Disease / Vol. 79 No. 6
`
`Fig. 4. Titration of name cblorosis- specific reverse transcription-polymerase chain reaction (FC-RT(cid:173)
`PCR). Aliquots (10 µI) of PCR reactions employing primer pair #307 (cf. Methods) containing the
`equivalent of a) 10 ng RNA; b) I ng; c) 100 pg; d) 10 pg; e) I pg; f) 100 fg; g) 10 fg; h) I fg; i) 100
`ng; j) 10 ag; k) no RNA were (A) spotted onto a nylon membrane, or (B) separated on an agarose gel
`and blotted to a nylon membrane. Both filters wen: probed with digoxigenin-labeled name chlorosis(cid:173)
`specific riboprobe, digFC#21-RNA.
`
`

`

`mycobacteria (8). In amplifying the same
`sequence region, new combinations of
`primers derived from slightly different po(cid:173)
`sitions of the original target sequence
`achieved 100- to 1,000-fold higher sensi(cid:173)
`tivity (8).
`Variants of FC-RT-PCR for different
`diagnostic formats. One of the advan(cid:173)
`tages of the approach outlined here is that
`it combines sensitivity of detection with
`specificity of diagnosis, and allows choice
`among several variants offering different
`combinations of speed, convenience, and
`sensitivity. The simplest, fastest but also
`least sensitive variant is the direct visuali(cid:173)
`zation by gel electrophoresis and ethidium
`bromide staining of a DNA band of the
`predicted size. In the course of developing
`
`a
`
`b
`
`C
`
`d
`
`e
`
`f
`
`Fig. 5. Detection of single-stranded flame
`chlorosis RNA in Pythium mycelium. RNA
`extracted from 10 mg of lyophilized mycelium
`powder of Pythium arrhenorru.mes Drechs. (a
`through d) (cf. reference 4), or negative control
`Pyrenophora teres Drechs. (0 was subjected to
`flame chlorosis-specific reverse transcription(cid:173)
`polymerase chain reaction (FC-RT-PCR) as
`described in Methods. From the 100-µI RT-PCR
`reaction volume: a) 50 µI; b) 5 µI; c) I liter;
`d) 0.1 µI and O 50 µJ were assayed by dot blot
`hybridization (6); e) I µI plasmid DNA contain(cid:173)
`ing 0.3 µg flame chlorosis- specific DNA insert
`(positive control).
`
`the RT-PCR protocols we established that
`this variant of FC-RT-PCR could detect 10
`pg and I pg of run-off transcript FC-RNA
`using primer pairs #86 and #307, respec(cid:173)
`tively (data not shown). When these FC(cid:173)
`RT-PCR products are dot blotted and
`probed with digFC-RNA riboprobe (7) it
`is the stringent hybridization to a known
`probe, rather than the predicted size of an
`amplified product, that ensures that spe(cid:173)
`cific FC-RNA is being detected (Fig. 4A).
`Southern blotting of FC-RT-PCR products
`followed by hybridization with digFC(cid:173)
`RNA riboprobe combines the sensitivity of
`dot blotting and two independent criteria,
`predicted size and specific hybridization,
`in ensuring specificity (Fig. 4B). The ad(cid:173)
`ditional sensitivity afforded by dot blotting
`and Southern blotting of FC-RT-PCR
`products in practice allows FC-RNA to be
`diagnosed from 1 mg of lyophilized my(cid:173)
`celium, compared with the 200 mg re(cid:173)
`quired for the conventional hybridization
`assay (4,7).
`In some applications, one may be will(cid:173)
`ing to sacrifice some sensitivity to obtain
`the information of Southern analysis more
`quickly and easily. Instead of identifying
`the PC-specific DNA product by blotting
`and hybridizing to DIG-labeled probe, the
`DNA product can be labeled in the PCR
`reaction itself using DIG-labeled deoxyri(cid:173)
`(Boehringer Mannheim).
`bonucleotides
`After agarose gel electrophoresis, DIG(cid:173)
`labeled PC-specific DNA fragments can be
`blotted to nylon membrane and, after
`blocking, reported directly with enzyme(cid:173)
`labeled antiDIG antibody. Alternatively,
`DIG-labeled FC-specific DNA fragments
`can be detected in situ, eliminating both
`the blotting and hybri.dization steps (S.
`Haber, unpublished). In preliminary trials
`we have found these variants of FC-RT(cid:173)
`PCR to be intermediate in sensitivity be(cid:173)
`tween simple ethidium bromide staining
`and dot or Southern blot hybridization to
`DIG-labeled riboprobe (S. Haber, unpub(cid:173)
`lished). The results of the in situ variant
`are available within 4 h of completing
`agarose gel electrophoresis, and the purple
`bands against a clear background are eas(cid:173)
`ily quantified with
`inexpensive gel(cid:173)
`scanning apparatus (e.g., ISCO Model
`1520, ISCO Instruments, Lincoln, Nebr.).
`If, on the other hand, much greater de(cid:173)
`tection sensitivity is desired, the FC-RT(cid:173)
`PCR product could be further amplified in
`a second round of PCR using nested prim(cid:173)
`ers and Taq polymerase (15), and the
`products of the nested PCR hybridized to
`labeled probe in dot or Southern blots.
`Use of RT-PCR to provide evidence
`that dsRNA is a product of viral infec•
`tion. The use of RT-PCR extends the prin(cid:173)
`ciple of diagnosing disease by detecting
`specific RNA. Detection and characteriza(cid:173)
`tion of dsRNA has been advanced as a
`general approach for diagnosing infection
`with ssRNA viruses (12), and it has proved
`essential in cases of infections with RNA
`
`viruses or viruslike agents not associated
`with the production of protein-coated virus
`particles
`(6,7). If the disease-specific
`dsRNA is a replicative intermediate of an
`infectious RNA rather than, for example,
`the product of simultaneous transcription
`off both strands of circular dsDNA (14),
`an ssRNA counterpart must also occur.
`Thus, when the nucleotide sequence of
`cDNA of the disease-specific dsRNA is at
`least partially known, RT-PCR can be used
`to amplify portions of the ssRNA counter(cid:173)
`part to the disease-specific dsRNA. Detec(cid:173)
`tion of the amplified dsDNA product, on
`the basis of its predicted size and homol(cid:173)
`ogy to specific probes, confirms the pres(cid:173)
`ence of disease-specific ssRNA when it
`might not have been detectable otherwise.
`An example of a specific question that this
`approach might shed light on is whether
`the dsRNA composing the putative avo(cid:173)
`cado virus I is linked to viral infection, or
`perhaps the product of unusual transcrip(cid:173)
`tion from the avocado genome (2).
`We are now exploiting FC-RT-PCR to
`screen root tissue, and isolates of Pythium
`spp. and other fungi for RNA sequences
`specific to the FC viruslike agent.
`ACKNOWLEDGMENTS
`We thank B. Gillis, E. Mueller. and M. Wolf for
`excellent technical assistance, Reginald Sims for
`preparing figures and plates, and G. Casey for
`synthesis of oligonucleotide primers. D. J. S. Barr,
`J. Gilben, and A. Tekauz provided fungal cultures.
`We are also indebted to N. K. Howes, R. P. Bod(cid:173)
`naryk, and J. A. Dodds for helpful, critical discus(cid:173)
`sions.
`
`LITERATURE CITED
`I. Chong, J., and Haber, S. I 992. Cytological
`alterations associated with flame chlorosis, a
`novel viruslike disease of barley, wheat, and
`oat. Phytopathology 82:815-821.
`2. Cook, G., Kotz~. J. M., and Ne!, L.H. 1994.
`Double-stranded RNA comprising the puta(cid:173)
`tive avocado virus I has a high degree of se(cid:173)
`quence homology to the avocado genome.
`Plant Pathol. 43:913-916.
`3. Cook. R. J., and Veseth, R. J. 1991. The actual
`yield of wheat: limiting effects of pests and
`diseases. Pages 41-59 in: Wheat Health Man(cid:173)
`agement. American Pbytopathological Soci(cid:173)
`ety, St. Paul, Mn.
`4. Haber, S., Barr. D. J. S., Chong, J., Tekauz,
`A., and Kim, W. 1993. Flame cblorosis is
`linked to Pythium spp. Pages 115-1 18 in:
`Proc. Symp. Inter. Work. Group Plant Vir.
`Fung. Vee., 2nd.
`5. Haber, S., Barr, D. J. S .. and Platford, R. G.
`1991. Observations on the distribution of
`flame chlorosis in Manitoba and its associa(cid:173)
`tion with cenain zoosporic fungi and the in(cid:173)
`tense cultivation of cereals. Can. J. Plant Pa(cid:173)
`thol. 13:241-246.
`6. Haber, S., Kim, W., Gillespie, R., and Tekauz,
`A. 1990. Flame chlorosis: A new. virus-like,
`soil-transmitted disease of barley in Mani(cid:173)
`toba. J. Phytopathol. 129:245-256.
`7. Haber, S., Wakarcbuk, D.A., Cvitkovitch, S.
`E., and Murray, G. 1992. Diagnosis of flame
`chlorosis, a viruslike disease of cereals ,by
`detection of disease-specific RNA with di(cid:173)
`goxigenin-labeled RNA probes. Plant Dis.
`76:590-594.
`8. He, Q., Marjamaki, M ., Soini. H., Mensola,
`J ., and Viljanen, M. K. 1994. Primers are de(cid:173)
`cisive for sensitivity of PCR. BioTechniques
`17:82-87.
`
`Plant Disease/ June 1995 629
`
`

`

`9. Katcher, H. L., and Schwartz, I. 1994. A
`distinctive propeny of Tth DNA polymerase:
`Enzymatic amplification in the presence of
`phenol. BioTechniques 16:84-92.
`10. Koller, B., Delius, H., and Oyer, T. A. 1982.
`The organization of chloroplast DNA in wheat
`and maize in the region containing the LS
`gene. Eur. J. Biochem. 122: 17-23.
`11. Kulinski, T., Bratek-Wiewiorowska, M. D.,
`Zielenlciewicz, P., Zolkiewski, M., and Zielen(cid:173)
`kiewicz, W. 1991. Comparative calorimetric
`
`studies on the dynamic conformation of plant
`5S rRNA: II. Structural interpretation of the
`thermal unfolding patterns for lupin seeds and
`wheat germ. Nucleic Acids Res. 19:2449-
`2445.
`12. Morris, T. J., and Dodds, J. A. 1979. Isolation
`and analysis of double-stranded RNA from vi(cid:173)
`rus-infected plant and fungal tissue. Phytopa(cid:173)
`thology 69:854-858.
`13. Sambrook, J., Fritsch, E. F., and Maniatis, T.
`1989. Gel electrophoresis of DNA. Pages 6.3-
`
`6.20 in: Molecular Cloning: A Laboratory
`Manual. 2nd ed. Cold Spring Harbor Labora(cid:173)
`tory, Cold Spring Harbor, New York.
`14. Wakarchuk, D. A., and Hamilton. R. I. 1985.
`Cellular double-stranded RNA in Phaseolus
`vulgaris. Plant Mol. Biol, 5:S5-72.
`15. Zimmermann, K., Pischinger, K. and Mann(cid:173)
`halter, J. W. 1992. Rapid nonradioactive de(cid:173)
`tection of HIV-I RNA from a single-cell
`equivalent by reverse transcription PCR with
`nested primers. BioTechniques 15:806-808.
`
`630 Plant Disease/ Vol. 79 No. 6
`
`