415
J. gen. Virol. (I98O), 5I, 415-419
Printed in Great Britain
Further Studies on the U s e of Protein A in Immune Electron
Microscopy for Detecting Virus Particles
(Accepted 30 June I98O)
SUMMARY
The immune electron microscopic technique which involves pre-coating electron
microscope grids with protein A before coating them with the specific antiserum,
has been found suitable for detecting isometric insect and plant viruses. With the
three virus-antibody combinations tested, the optimum antiserum dilution for protein A plus antiserum treatment was found to be I : IOO or less, whereas in the case
of grids treated with antiserum alone it ranged from i : iooo to i : 2000 although
the titre of the antisera ranged from 1:5~2 to 1:4o96. The increase in the number
of particles on grids treated with protein A plus antiserum over those treated with
antiserum alone, at each optimal antiserum dilution, was 25.7-fold (sugarcane
mosaic virus), 2-I-fold (tobacco mosaic virus) and 1.6- and 2"4-fold (Erysimum
latent virus - two different sap dilutions). Protein A plus antiserum-coated grids
can be stored for up to 6 months at 4 °C, but not at room temperature or in a
desiccator, and still retain about 25 ~ of their activity, sufficient to detect any virus
using the electron microscope. Antisera preserved in glycerol can be used successfully for detecting viruses by immune electron microscopy.
In a previous paper (Shukla & Gough, i979) we reported a modification of the current
immune electron microscopic (IEM) techniques (Derrick, 1973; M ilne & Luisoni, 1977) for
detecting two elongated plant viruses, sugarcane mosaic virus (SCMV) and tobacco mosaic
virus (TMV), which involved pre-coating electron microscope specimen grids with protein A
(a cell wall protein of Staphylococcus aureus) before coating them with the specific antiserum.
Although the conditions were not optimal for strict comparison, the method trapped more
SCMV and TMV than grids treated with antiserum alone. It appeared particularly suitable
for virus particles in low concentration in plant extracts.
In this article, we show that our techrdque is also suitable for detecting isometric viruses
belonging to different taxonomic groups. The possibility of storing protein A plus antiserumtreated grids for later use, as well as the optimum antiserum dilution for trapping maximum
number of particles on a grid, as claimed recently by other workers (Derrick & Brlansky,
I976; Paliwal, 1977; Milne & Lesemann, 1978; Brlansky & Derrick, I979), have also been
investigated.
The four plant viruses used were broad bean wilt virus (BBWV) type strain (Taylor &
Stubbs, 1972), Erysimum latent virus (ELV; Shukla et al. I98O), a common strain of TMV
and the Johnson grass strain of SCMV. They were propagated either in an environmentcontrolled growth cabinet or in a glasshouse in Blackeye cowpea, Chinese Cabbage, Turkish
tobacco and sweetcorn cv. Iochief respectively. Virus extracts were prepared from systemically infected leaves from plants infected for about 4 weeks (cowpea, Chinese cabbage and
sweetcorn) or about 6 months (tobacco) by grinding the leaf tissue using a pestle and mortar
in IO ml/g (BBWV), IOO, 200 or 1ooo ml/g (ELV), IOOOml/g (TMV) or 5 ml/g (SCMV) of
o.i M-phosphate buffer pH 7 (PB). The SCMV and TMV antisera were the same as described previously (Shukla & Gough, 1979). The BBWV antiserum was obtained from the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 31 Jul 2017 22:32:08
OO22-I317/8o/00oo-4079 $02.oO~198o SGM
416
Short communications
Plant Research Institute, Victoria and had a titre of I : IO24 in agar diffusion tests (Taylor
et al. I968). The ELV antiserum was prepared in a rabbit by a series of four intravenous
injections (4 mg virus~m1 H20/injection ) over a period of 6 weeks and the blood was
collected from the marginal ear vein I week after the last injection. This antiserum had a
titre of 1:4o96 in agar diffusion tests.
Sackbrood virus (SBV) was extracted by grinding the diseased bee larvae (collected from
several ~hives in Victoria) in 0.05 M-tris buffer pH 7"6 plus o.I % 2-mercaptoethanol in the
ratio of 1:3 (w/v). The extract was emulsified with o'5 vol. diethyl ether plus o'25 vol. CC14
and subjected to one cycle of differential centrifugation (low speed, 12ooo g for Io min;
high speed, 16oooo g for 6o min). The pellets were resuspended in o.o2 M-tris buffer p H 7"6.
The SBV antiserum was prepared by Dr L. Bailey of the Rothamsted Experimental Station,
Harpenden, U.K. and had a titre of 1:256 in agar diffusion tests.
Urdess otherwise stated, the methods for preparing and processing the grids were essentially the same as described in an earlier paper (Shukla & Gough, I979). The antiserum dilutions used for decorating the virus particles were: ELV, i :200; TMV, BBWV, I : ioo; SBV,
i:32 and SCMV, 1:5. Dilutions of the antisera for coating grids and decorating virus
particles were prepared in PB. The grids were processed in the laboratory at about 2o °C.
Due to the time-con,suming process of counting the virus particles, the mean number of
particles was calculated from photographs of one randomly selected square on each of the
two grids per treatment. The magnifications used for elongated and isometric viruses were
× 4ooo and × 23000 respectively.
When a low dilution of antiserum (SBV, I : 32; ELV, I : 4o; BBWV, I : IOO; the dilutions
were chosen randomly) was used, the use of protein A in conjunction with the specific
antiserum significantly increased the number of particles of the three isometric viruses compared to the grids coated with antiserum alone. The mean particle counts on grids coated
with protein A plus antiserum/antiserum alone were 7o64/756 (ELV, sap dilution I:IOO),
373/33 (BBWV) and 3523/2r8 (SBV), giving an increase of 9"3-fold, r 1.3-fold and I6.2-fold
respectively. It should be pointed out that the above experiment was performed before
determining the optimum antiserum dilutions required for the two techniques. This is
evident from the results of our next experiment (Table I). Therefore, the values obtained
are only indicative and may be considered as rough estimates.
In order to determine the optimum antiserum dilution for the two techniques, protein A
plus antiserum and antiserum alone, we tested three viruses representing different morphological groups at six different antiserum concentrations ranging from undiluted to I : 5ooo.
It can be seen (Table I) that grids coated with protein A plus antiserum trapped more
particles for all the three viruses at each antiserum concentration, except ELV at r :5ooo,
compared to the grids treated with antiserum alone. The optimum antiserum dilution for
protein A plus antiserum treatment was found to be I : Ioo or less, whereas in the case of
grids coated with antiserum alone, it ranged from I:IOOO to 1:2ooo. The increase in the
number of particles on grids treated with protein A plus antiserum over those treated with
antiserum alone, at each optimum antiserum dilution, was 25-7-fold (SCMV), :~.i-fold
(TMV), r.6-fold (ELV at I : 2oo sap dilution) and 2"4-fold (ELV at I : Iooo sap dilution).
Except for SCMV antiserum, which trapped a maximum number of particles when undiluted, the antisera to the other two viruses always trapped more particles at a dilution
o f I : IO0.
To determine whether pretreated grids could be stored for later use, freshly prepared
grids were coated with protein A and undiluted SCMV antiserum. Undiluted antiserum was
used due to the fact that it trapped a maximum number of particles with protein A plus
antiserum treatment (Table I). The grids were washed with 2o drops each of PB and water,
drained and air-dried. They were then placed in grid boxes and stored at room temperature,
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 31 Jul 2017 22:32:08
Short communications
417
Table I. Effect of antiserum dilution on the number of S C M V, T M V and EL V particles on
grids coated with antiserum alone and protein ,4 + antiserum
Virus
Sap
dilution
SCMV I" 5
TMV
ELV
ELV
Antiserum
titre
Mean number of particles at various antiserum dilutions
,
Treatment
Undiluted
I : IOO
x :5oo
I : tooo
1:2ooo
1:5ooo
1:512
Antiserum
I39+34
t82+Io
112+ 7
I94+I3
I 7 I + I 6 I46_+I
Protein A +
antiserum 4992+I8O 36o9+243 2784+492 386+I74
493+223 263+4o
I:IOOO t:2o48 Antiserum
82+24
4oi+56
424+55
4o8+1IO
5o6+52 355+-t3o
Protein A +
antiserum
8954-27 lO54-+35 Io55+_1o6 767+72
582+32 393+_85
1:2oo 1:4o96 Antiserum 9774-625 23314-32I 31834-9o9 32344-924 243o_+96 2424+78
Protein A +
antiserum 252o+138 5292+384 4872+_666 3322+1o95 2881+359 502+-229
I:iooo I:4O96 Antiserum 32o-+I8
4144-19
477+-1o6 6554-II
623+I3 554+_1o3
Protein A+
antiserum
490+_8o 1551+127 II57+283 II984-117
896+I12
218+_7o
in a desiccator placed in the laboratory and in a refrigerator at 4 °C. The virus-trapping
ability of the stored grids was tested at different intervals up to 6 months. To activate the
stored grids, 5 #1 PB was added, the grids incubated for 5 min, then drained. The addition of
the SCMV suspension to the grids and the subsequent steps were the same as described by
Shukla & Gough (1979). Under all the three conditions there was a progressive decrease in
the number of particles as the storage time was lengthened (Table 2). However, grids stored
at 4 °C trapped significantly more particles at all of the four times tested than those stored
in a desiccator and at room temperature; the desiccator-stored grids proved slightly better
than those stored at room temperature. Paliwal 0977) reported that grids treated with
barley yellow dwarf virus antiserum could be stored at room temperature in a desiccator up
to 6 weeks without losing their capacity to trap the virus particles efficiently, but after 8
weeks he observed a sharp decline in the number of particles trapped. Our results (Table 2)
suggest that protein A plus antiserum-treated grids can be stored for up to 6 months at
4 °C, but not at room temperature or in a desiccator, while still retaining 25 ~ of their
activity. We observed an average of 518 SCMV particles per micrographic area after 6
months storage. This number is sufficient to detect any virus using the electron microscope.
The results presented in this paper and those of our previous paper (Shukla & Gough,
1979) have shown that our IEM technique is suitable for detecting elongated (SCMV, TMV)
as well as isometric (BBWV, ELV, SBV) insect and plant viruses. Although the increase in
the number of particles obtained with our technique compared to the Derrick technique
differed from one virus to another, we always obtained a much greater increase with SCMV
than with ELV or TMV but the reason(s) for this are not known (Table i ; Shukla & Gough,
1979). Derrick (I973) obtained a lower increase (2o-fold) with potato virus Y (PVY) than
with TMV (4o-fold) using antiserum-coated and uncoated grids. TMV and ELV, although
different taxonomically, can be considered similar as far as their stability and high concentration in sap are concerned. In contrast, PVY and SCMV (both potyviruses) are less
stable and occur in low concentrations in plants. This indicates that our technique may be
superior to the Derrick technique and its various modifications (Derrick, 1973; Milne &
Lesemann, 1978; Brlansky & Derrick, I979), especially for detecting viruses present at a
low concentration in crude extracts. Although protein A plus antiserum treatment trapped
more particles of SCMV, TMV and ELV at each antiserum concentration, except ELV at
1 : 5ooo, compared to the grids treated with antiserum alone (Table I), it was observed that
the magnitude of the increase was far less than that obtained with these viruses using antiDownloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 31 Jul 2017 22:32:08
Short communications
418
Table 2. Number of S C M V particles on protein .4 + undiluted antiserum-coated grids under
different conditions/br different durations
Duration
(months)
,
Fresh
Mean number of particles
,-~
Room
Desiccator
,
4 °C
2407--+I49
2313---+81
916_+193
124-+54
456--+35
207-+42
2013-+ I53
I074+_186
3
2758-+ I5z
63-+43
zio-+62
I833-+87
6
2oto-+ 138
22-+ io
68-+ I0
I
2
518-+ 8
serum dilution of I : IOO or less in our earlier experiment (see also Shukla & Gough, I979).
This may have been due to the non-optimal antiserum dilution used for the Derrick technique. It was further observed that the optimum antiserum dilution for protein A plus
antiserum-treated grids was much lower than that for grids coated with antiserum alone
(Table I).
It is well established that in an IEM technique it is the virus-specific IgG molecules which
are responsible for trapping virus particles on grids, although a very low level of particlebinding is generally obtained due to non-specific factors, such as serum proteins, hydrophilic films on grids etc. It is also well recognized that protein A binds to the Fc portion of
IgG molecules; two IgG molecules bind to one molecule of protein A (Forsgren & Sj6quist,
I966; Sj6quist et al. I972). Therefore, grids treated with protein A plus antiserum would
be expected to trap many more particles at all antiserum dilutions, compared to those
treated with antiserum alone. However, our present results have shown that this is not so.
The trend in particle counts obtained at various antiserum dilutions withtheproteinAsystem
(Table r) seems logical in two ways: first, fewer particles were always obtained when the
antiserum was undiluted, With the exception of SCMV antiserum, compared to the next dilution of I : Ioo; this may be due to the high concentration of serum proteins in the undiluted
antiserum competing for the binding sites on the grids (Milne & Lesemann, I978); second,
with the increase in antiserum dilution the number of IgG molecules would be expected
to fall gradually, which in turn should bind fewer and fewer particles; this trend was clearly
observed in the protein A system with the three viruses tested. What is not clearly understood
is the fact that with the Derrick technique a peak of particle binding at antiserum dilutions
between I : I ooo to r :5000 (with different virus-antibody combinations) is obtained irrespective of the antiserum titre (Table I ; Paliwat, 1977; Milne & Lesemann, 1978). On the other
hand, Derrick & Brlansky 0976) obtained a similar number of PVY particles (8I 9 to r I87)
on grids treated with various antiserum dilutions ranging from ~ : too to 1:32000. The titre
of the antiserum was I:z56. Derrick & Brlansky 0976) suggested that the number of
attachment sites on a grid would be constant for a given antiserum without regard to the
dilution of antiserum used to prepare the grid, as long as it is sufficient to coat the grids
completely with a film of serum proteins, within the adsorption time being used. Another
explanation for this fact can be drawn from the results of Milne & Lesemann (I978) who
found that antiserum dilutions between I : 8oo and I : 3zoo were most effective in trapping
oat sterile dwarf virus particles. At concentrations above I : 8o0, inhibition of trapping became progressively stronger due to serum proteins competing for sites on the grid, and
beyond I:3zoo there were apparently insufficient antibody molecules to trap all the virus
available. Our results obtained with the Derrick method (Table I) are very similar to those
of Milne & Lesemann 0978) and perhaps the explanation given by these authors may
also apply in our case.
It has been observed during the course of our experiments that the grids from the same
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 31 Jul 2017 22:32:08
Short communications
419
batches receiving identical t r e a t m e n t s behave very differently as far as the n u m b e r o f
particles t r a p p e d is c o n c e r n e d (for instance, Tables I a n d 2). A s suggested b y M i l n e &
L e s e m a n n 0 9 7 8 ) , this m a y be due to v a r i o u s u n k n o w n factors n o t t a k e n into a c c o u n t when
a p p l y i n g the m e t h o d .
T h e B B W V a n t i s e r u m used in the present w o r k was diluted w i t h an e q u a l v o l u m e o f
glycerol. G r i d s c o a t e d with this a n t i s e r u m b e h a v e d n o r m a l l y except t h a t there was slightly
m o r e c o n t a m i n a t i o n . This did n o t affect detection o f the virus particles, so t h a t antisera
preserved in glycerol can be used successfully to detect viruses b y I E M .
I t was f o u n d t h a t p r o t e i n A s o l u t i o n can be stored frozen for long p e r i o d s w i t h o u t losing
its activity. A n average o f 745 E L V particles (antiserum a n d sap dilutions - I : Iooo) were
o b t a i n e d p e r m i c r o g r a p h i c a r e a using an 18 m o n t h - o l d p r o t e i n A solution, which h a d been
r e p e a t e d l y frozen a n d thawed, c o m p a r e d to 840 particles t r a p p e d using freshly p r e p a r e d
p r o t e i n A solution.
W e are m o s t grateful to D r R. G. G a r r e t t for p r o v i d i n g B B W V a n d its a n t i s e r u m a n d to
M r C. R e i n g a n u m for the SBV extract a n d its antiserum.
KEITH H. GOUGH
DHARMA D. SHUKLA
Division o f Protein Chemistry
CSIRO, Parkville (Melbourne)
Victoria 3052, Australia
REFERENCES
BRLANSKY,R. H. & DERRICK, K. S. (1979). Detection of seedborne plant viruses using serologically specific
electron microscopy. Phytopathology 69, 96-1oo.
DERRleK, r. s. (1973). Quantitative assay for plant viruses using serologically specific electron microscopy.
Virology 56, 652-653 .
DERRICK,K. S. & BRLANSKY,R. H. (1976). Assay of viruses and mycoplasmas using serologically specific electron microscopy. Phytopathology 66, 815-820.
FORSOREH,A. & SJ6QUlST,J. (1966). 'Protein A' from Staphylococcus aureus. I. Pseudo-immune reaction with
human y-globulin. Journal of Immunology 97, 822-827.
MILNE,R. G. & LESEMANN,D. E. (1978). An immunoelectron microscopic investigation of oat sterile dwarf and
related viruses. Virology 9° , 299-3o4.
MILNE, R. G. & LUISONI,E. (I977). Rapid immune electron microscopy of virus preparations. Methods in
Virology 6, 265-28 I.
PALIWAL,Y. C. (1977). Rapid diagnosis of barley yellow dwarf virus in plants using serologically specific
electron microscopy. Phytopathologische Zeitschrift 89, 25-36.
SHUKLA,D. D. & COUGH,K. H. (1979). The use of protein A, from Staphylococcus aureus, in immune electron
microscopy for detecting plant virus particles. Journal of General Virology 45, 533-536.
SHUKLA, D. D., KOENIG, R., GOUGH, K. H., HUTH, W. & LESEMANN, D.-E. (1980). Erysimum l a t e n t v i r u s - f u r t h e r
characterization as a tymovirus. Phytopathology 7o, 382-384.
SJ~QUIST, J., MELOUN, B. & HJELM, H. ( I 9 7 2 ) . Protein A isolated from Staphylococcus aureus after digestion
with lysozyme. European Journal of Biochemistry 29, 572-578.
TAYLOR,R. H. & STUBBS,L. L. (1972). Broad bean wilt virus. CMI/AAB Descriptions of P/ant Viruses, No. 81,
4 PPTAYLOR,R. n., SMITH,P. R., REIHGANUM,C. &GIBBS,g. J. 0968). Purification and properties of broad bean
wilt virus. Australian Journal of Biological Sciences 2x, 929-935.
(Received 2I December I979)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 31 Jul 2017 22:32:08