I97

J. gen. Virol. (1977), 35, ~97-I99 Printed in Great Britain

Role of Sialic Acid in Infection with Vesicular Stomatitis Virus (Accepted 20 December I976) SUMMARY

The role of sialic acid in the infection of tissue culture ceils and mice with vesicular stomatitis virus has been studied. No loss of infectivity of the Indiana serotype of the virus was detected by incubating with neuraminidase although the virus particles had lost sialic acid as judged by their ability to inhibit the agglutination of red blood cells by influenza virus. The results did not depend on the type of cell used for growth and assay of the virus since essentially similar findings were made in BHK cells, L cells or mice. Similar results were obtained with Brazil virus, a subtype of the Indiana serotype and with the New Jersey serotype. We consider that the sialic acid of the virus which is removed by neuraminidase does not play a major role in the infectivity of the virus. In the last few years the biological role of sialic acid in two alphaviruses, Semliki Forest virus (Kennedy, 1974) and Sindbis virus (Stollar et al. I976) and a rhabdovirus, vesicular stomatitis virus (VSV; Schloemer & Wagner, I974, I975 a, b) has been investigated. Kennedy 0974) and Stollar et al. (I976) could find no role for the sialic acid in the two closely related alphaviruses. These results were in agreement with some preliminary experiments we had done with the Indiana serotype of VSV grown in BHK cells and assayed either in BHK cells or in mice. However, Schloemer & Wagner 0974, I975a) showed that the sialic acid on the surface glycoprotein of VSV was directly involved in the attachment of the virus to L cells and its agglutination of red blood cells. In their experiments removal of the sialic acid with neuraminidase resulted in a Ioo-fold drop in infectivity and a loss of the ability of the particles to agglutinate red cells. These biological activities were restored to their original level by resialylation in vitro of the neuraminidase treated particles. The apparent discrepancy between our results and those described by Schloemer & Wagner prompted us to re-investigate the problem and in particular to determine whether (1) the individual virus isolates and (2) the cells in which they were grown and assayed were crucial factors. We have used three different VSV strains: (a) the Indiana serotype, (b) Brazil virus, a sub-type of the Indiana serotype (Federer, Burrows & Brooksby, I967) and (c) the New Jersey serotype. These were grown in monolayers of BHK 2I cells, clone I3 in Eagle's medium at 37 °C. Virus yields of about Io o.5 1Ds0/ml (measured by plaque assay in BHK monolayers) or Io 1° LDs0/ml (measured by i.e. inoculation of 7-day-old mice) were regularly obtained. Each virus was treated with 50 to I5o units/ml of neuraminidase (Behringwerke) at 37 °C in the presence of o.oi M-CaC12 and o.1 M-acetate, pH 6.2, and samples titrated at intervals in BHK cells or in mice. Little or no loss of infectivity was observed even after incubation for I4o min. Similar results were also obtained when virus purified by sucrose gradient centrifugation was used (Table I). As a control of the activity of the enzyme used, the virus preparations were tested for their activity in inhibiting the haemagglutination of red blood cells by influenza virus (Compans, 1974). Using equine influenza virus (A1, Prague) the purified VSV preparations had titres of about 640 to 128o and neuraminidase treatment under the conditions specified

Short communications

I98

Table i. Effect of neuraminidase on the infectivity of three serotypes of VSV in mice and BHK 2I cells and on their inhibition of influenza virus haemagglutinin

Serotype Indiana Control Neuraminidase 30 min I4o min New Jersey Control Neuraminidase 3o rain Brazil Control Neuraminidase 30 min I40 min

Reciprocal titre of haemagglutinin inhibition

Infectivity(log LDso or ID~0/ml) in r -~ -~ Mice BHK zI

1o24 < 2 --

I o.2 9"9 9'8

9"5 9"o 9"6

64o < 4

Io-I i o.o

9"8 I o.o

1280 2 --

9'9 Io'5 IO.I

9"7 io.o 9'7

above reduced the titres to ~< 4. In all our experiments this loss of HA1 titre was achieved without loss of infectivity or complement fixing activity (Table 1). In one experiment, the amount of sialic acid remaining on the virus after treatment with neuraminidase was also measured. A preparation of purified Indiana virus, containing I "3 × IOll p.f.u./ml (titration in B H K cells) was incubated with 50 units of enzyme at 37 °C for 4o min and the incubation continued for a further 20 min with 25 additional units of enzyme. The sialic acid content was reduced from o'343 .umol/ml to o.oi 7/~mol/ml, whereas the titre after treatment was virtually unchanged at I'4 x lo 1~ p.f.u./ml. To test whether the cells in which the virus was grown and assayed was a crucial factor, Indiana virus was grown in B H K 2I and L cells and assayed in each cell before and after treatment with neuraminidase. The results in Table 2 demonstrate that, despite the fall in H A I titre, treatment with neuraminidase did not cause any significant loss of infectivity of each virus. If the degree of sialylation of the surface glycoprotein has a role in determining the ability o f the virus to attach to the host cell, it could be argued that the infectivity titre of the virus grown in L cells is lo-fold lower than that grown in BHK 2~ cells because of a lower level of sialic acid. However, the ratio of infectivity to H A I units is the same for the viruses grown in the two cells (Table 2). In a further attempt to ascertain whether sialic acid has a role in the infectivity of VSV, the Indiana serotype was grown in Aedes albopictus cells in the expectation that it would contain low levels of sialic acid. Infectivity titres of I,~s LDs0/ml (measured in unweaned mice) were obtained when the virus was grown in the presence of 2 ~ foetal calf serum. However, purified virus grown in B H K 2[ cells and A. albopictus cells with H A I titres of IO24 and complement fixing activities of 8o and 74 units respectively contained z.i ¢tg and 3.z ttg sialic acid/mg protein as estimated by the thiobarbituric acid method (Warren, I959). Investigation of this high level of sialic acid in the A. albopictus cell-grown virus has shown that cells grown in the presence of serum contain high levels of sialic acid. Even after washing several times with phosphate buffered saline the cells contained I4/zg sialic acid/mg protein compared with 9/zg sialic acid/mg protein for B H K 21 cells. Stollar et al. (I976) also found appreciable quantities of sialic acid in A. albopictus cells grown in the presence o f foetal calf serum and they attribute it to the glycoprotein in the calf serum. Our results

Short communications

I99

Table z. Comparative infectivity o f V S V - Indiana grown in B H K or L celia" in

different hosts

Cell for virus growth

Reciprocal titre of haemagglutinin inhibition

Infectivity (log IDs0 or LDs0/ml) in : :BHK cells L cells

Mice

B H K - Control Neuraminidase

2560 < 4

9"5 9"4

9"3 9"0

Io.o 9"8

L - Control Neuraminidase

2oo < 2

8.6 8"4

8.I 8.I

8"7 8.8

a n d those of Stoll ar et al. (1976) differ from those of Schloemer & W a g n e r (1975 b) who f o u n d very low levels of sialic acid in A. albopictus cells grown in the presence of foetal calf serum a n d also in the virus prepared in these cells. Clearly there is a discrepancy between our results with VSV a n d those described by Schloemer & W a g n e r (1974, I975a, b). If sialic acid has a role in the life cycle of VSV it would seem unlikely that the virus is a r t h r o p o d - b o r n e . It would be surprising for a virus to depend on a host that c a n n o t provide it with a constituent, sialic acid, which is essential for its efficient replication. If, o n the other hand, our results with VSV represent the m o r e correct situation, then the question would still remain open whether VSV is disseminated b y insects or whether vertebrates are the ' p i v o t a l a n d perhaps only host involved in the n a t u r a l life cycle of the v i r u s ' (Stollar et aL I976). Clearly this question is i m p o r t a n t in studies on the dissemination a n d evolution of VSV and requires further clarification. We wish to t h a n k Miss M. Jennings for providing the A. albopictus cells.

The Animal Virus Research Institute, Pirbright, Woking, Surrey, England

B. CARTWRIGHT

F. BROWN REFERENCES

COMPANS,R. W. (I974). Haemagglutination-inhibition:rapid assay for neuraminic acid-containing viruses. Journal of Virology x4, 13o7-I 3o9. FEOERER, K. r., BURROWS,R. ~ aROOKSaV,J. a. 0967). Vesicular stomatitis virus - the relationship between some strains of the Indiana serotype. Research in Veterinary Science 8, ~o3-I I7. KENNEDY,S. I. T. (1974). The effect of enzymes on structural and biological properties of Semliki Forest virus. Journal of General Virology z3, 129-I 43SCHLOEMER,R. H. & WAGNER,R. R. 0974)- Sialoglycoprotein of vesicular stomatitis virus: role of the neuraminic acid in infection. Journal of Virology x4, 270-281. SCHLOEMER,R. H. & WAGNER,R. R. (i 975 a). Cellular adsorption function of the sialoglycoprotein of vesicular stomatitis virus and its neuraminic acid. Journal of Virology xS, 882-893. SCHLOEMER, R.H. & WAONER,R. R. (I975b). Mosquito cells infected with vesicular stomatitis virus yield unsialylated virions of low infectivity. Journal of Virology xS, Io29-Io32. STOLLAR, V., STOLLAR, B.D., KOO, R., HARRAP, K.A. & SCHLESINGER, R.W. 0976). Sialic acid contents of Sindbis virus from vertebrate and mosquito cells. Virology 69, I O4-~15. WARREN,L. (1959). The thiobarbituric acid assay of sialic acids. Journalof Biological Chemistry z34, I97I1975.

(Received ~ October I976)

Role of sialic acid in infection with vesicular stomatitis virus.

I97 J. gen. Virol. (1977), 35, ~97-I99 Printed in Great Britain Role of Sialic Acid in Infection with Vesicular Stomatitis Virus (Accepted 20 Decemb...
200KB Sizes 0 Downloads 0 Views