i65

J. gen. Virol. (I976), 33, I65-I67 Printed in Great Britain

The Detection of Viruses by Enzyme-Linked Immunosorbent Assay (ELISA) (Accepted 28 Jzote I976) SUMMARY

The use of enzyme-linked antibodies for the detection of two morphologically different plant viruses is described. The technique is extremely sensitive, enabling assay of the viruses at concentrations as low as IO to xoo ng/ml both in purified preparations and in crude plant extracts. Enzyme-labelled antibodies have been used for the quantitative determination of specific antibodies in various human and animal diseases, using methods based on principles outlined by Engvall & Perlmann (I97I, I972) and recently reviewed by Voller, Bidwell & Bartlett (I976). However, the potential of such methods for the direct detection and assay of infectious agents such as viruses has been overlooked. We describe here the first application of enzyme-linked immunosorbent assay (ELISA) for the detection and assay of two plant viruses. The assays were made using the microplate method described by Voller et al. 0974). The viruses studied were arabis mosaic virus (AMV) which has isometric particles of diam. approx. 3o nm, and plum pox virus (PPV) which has filamentous particles approx. 75o × zo nm. An antiserum to each virus was prepared in rabbits using density gradient purified virus emulsified with Freund's incomplete adjuvant for AMV, and with both incomplete and complete adjuvant in successive injections for PPV. The antisera gave titres in ring interface precipitin tests of I :I 6 ooo against AMV and of 1:4ooo against PPV. Each antiserum reacted only slightly (~ I : 4) in tube precipitin tests against healthy plant extracts. The globulin fraction of each antiserum was precipitated by adding an equal vol. of 36 % Na2SO4. The precipitate was washed in I8 % N%SO4 and stored at - 7 o °C. Part of the globulin fraction of each serum was conjugated with alkaline phosphatase, E.C. 3- I. 3- 1 (Sigma type VII), using glutaraldehyde as the coupling agent (Avrameas, I969). For the test the unlabelled globulin fractions of the specific antisera were diluted to Io/zg protein/ml in o'o5 M-carbonate buffer, pH 9.6; 2o0 #1 quantities were added to each well in a polystyrene microhaemagglutinationplate (Cooke M29AR, Dynatech Laboratories) which was then placed at 37 °C to allow adsorption of the y-globulin to the well surface. After 5 h the wells were emptied and washed by flooding for 3 min with o'I5 M-phosphate buffered saline, pH 7"1, containing o'o5 % Tween zo (PBS-Tween). The operation was repeated 3 times for each washing step after which the plate was shaken dry. The test samples, 2oo/zl quantities of purified virus or of leaf extracts in PBS-Tween, were then added to duplicate wells and the plates placed at 4 °C. After 18 h the plates were washed as before and 2o0 l~l of a predetermined dilution of the enzyme-labelled antibody conjugate was added to each well. The plates were incubated for 4 h at 37 °C. After a further washing step 3oo #1 of the enzyme substrate (p-nitrophenyl phosphate, I mg/ml in Io % diethanolamine buffer, pH 9"8) was added at room temperature. After 3o min the reaction was stopped by

166

Short communications I

I

I

I

I

I

1"4

3.5

1"2

3.0

1"0

g

2.0

0"6

1.5

0"4

1.0

0 16 Reciprocal of dilution ( × 50) 2

4

Fig. I

I

0.5

. . . . . .

1

I

2.5

0"8

"2

I

8

I,

10-2

I

I

1 0 - 3 10-4 Dilution

I

10-s

Fig. 2

Fig. t. The detection of AMV by ELISA in plant extracts. Leaves were extracted with 2 vol. (w/v) of phosphate buffer and subsequent dilutions were made in PBS-Tween: • 0 , infected cucumber extract; • A, infected N. clevelandii extract; • II, infected C. quinoa extract; © - - - © , extracts of corresponding healthy plants and of SLRV-infected cucumber and N. ctevelandii.

Fig. 2. The detection of PPV by ELISA in purified preparations and in plant extracts. Leaves were extracted with 4 vol. (w/v) of phosphate buffer and subsequent dilutions were made in PBS-Tween: • .', purified preparation, initial concentration was 60 #g/ml; • • , infected iV. clevelandii extract; O - - - C), extracts of healthy and of hop mosaic virus infected N. clevelandik the addition o f 50/zl o f 3 M-NaOH to each well and the extinction at 400 n m of the contents of each well was recorded. The sensitivity and specificity o f E L I S A for detecting A M V was investigated using dilutions of purified preparations both of this virus and o f strawberry latent ringspot virus (SLRV). Both viruses are members o f the nepovirus group and are transmitted by the same vector, but they are serologically unrelated. A M V was readily detected at concentrations as low as 80 ng/ml which was the lowest concentration tested (E4oo = 0-58). There was no significant reaction with any o f the S L R V samples (IOO to I-O #g/ml, E400 < 0"07). In another experiment tests were made using sap from plants o f three different herbaceous species. Leaves were g r o u n d with twice their weight o f neutral phosphate buffer and the extracts were centrifuged at low speed to remove chloroplasts and other large cell debris. The partially clarified extracts were diluted in PBS-Tween for the tests. A M V was detected in extracts from infected tissues at very lligh dilutions, whereas similarly treated extracts f r o m uninfected plants and f r o m plants infected with S L R V gave no detectable reaction (Fig. I). F o r the tests with PPV, dilutions of a purified virus preparation and o f extracts o f leaves f r o m infected Nicotiana clevelandii plants were assayed. As controls, tests were made with leaf extracts f r o m healthy plants and also f r o m N. clevelandii plants infected with h o p mosaic virus. This was included as an example of another aphid-borne virus with filamentous particles, but one not related serologically to PPV. PPV was detected in the purified virus preparation at a concentration less than Io ng]ml and in leaf extracts diluted to lO -4 (Fig. 2). There were no significant reactions with the extracts f r o m either the uninfected or the h o p mosaic virus infected plants.

Short communications

I67

These results show that two quite different viruses can be detected readily and specifically by enzyme immunoassay. The versatility and relative simplicity of the assay, its suitability for use with crude plant extracts and its extreme sensitivity in comparison with more conventional plant virus serological techniques suggest numerous applications in the detection and assay of plant viruses. In a simplificd version, with visual rather than spectrophotometric assessment of results, ELISA could be used for inexpensive and rapid field tests which might be of great benefit for epidemiological studies, particularly for filamentous viruses or for other viruses that occur in low concentrations or that seldom cause obvious symptoms.

Nuffield Institute of Comparative Medicine, Zoological Society of London, Regent's Park London, N. W. I East Mailing Research Station, East Mailing, Maidstone, Kent U.K.

A. VOLLER ANN BARTLETT D. E. BIDWELL M . F . CLARK A . N . ADAMS

REFERENCES AVRAMEAS,S. (1969). Coupling of enzymes to proteins with gluteraldehyde. Immunochemistry 6, 43-52. ENGVALL, E. & PERLMANN, P. (I97t). Enzyme-linked immunosorbent assay (ELISA). Quantitative assay of immunoglobulin G. Immunochemistry 8, 871-874. ENGVALL, E. & PERLMANN, P. (t972). Enzyme-linked immunosorbent assay, ELISA. III. Quantitation of specific antibodies by enzyme-labelled anti-immunoglobulin in antigen-coated tubes. Journal of Immunology zo9, I29-I35. rOLLER, A., BIDWELL, D., HULDT, G. & ENGVALL,E. (I974)- A microplate method of enzyme-linked immunosorbent assay and its application to malaria. Bulletin of the Worm Health Organization 5x, 2o9-2I I. rOLLER, A., BIDWELL,O. E. & aARTLETT, A. (1976). Enzyme immunoassays in diagnostic medicine: theory and practice. Bulletin of the World Health Organization (in the press).

(Received 19 May I976)