0300-9629/92 $5.00+ 0.00

Camp. Biochem.Physiol. Vol. lOlA, No. 3, pp. 471-476, 1992 Printed in Great Britain

Pcrpmon Pressplc

THE DETECTION OF INTRACELLULAR BLUETONGUE VIRUS PARTICLES WITHIN OVINE ERYTHROCYTES R. A. NUNAMAKER,* J. A. ELLrs,t J. G. WIGINGTON*and N. J. MACLAIXLANS *U.S. Department of Agriculture, Agricultural Research Service, Arthropod-borne Animal Diseases Research Laboratorv. P.O. Box 3965. Universitv Station, Laramie, WY 82071-3965, U.S.A. Telephone: (307) 72i-b308; tDtpartment of Veterinary Sciences, University of Wyoming, Laramie, WY 82071, U.S.A.; SDepartment of Veterinary Pathology, University of California, Davis, CA 95616, U.S.A. (Received 4 July 1991) Abstract-l. We report here a simplified method for detecting viruses and other antigenic agents in red blood cells (RBCs). Using a nonionic detergent to prepare cytoskeletons, the interior of RBCs can be scanned rapidly using immunoelectron microscopy. 2. In this study, RBCs from bluetongue (BLU) virus-infected sheep were adsorbed directly onto Formvar-coated, gold electron microscope grids. 3. Cytoskeletons were prepared and then probed using a monoclonal antibody to VP 7, a structural BLU-virus protein and Protein-A gold. 4. Of the cc 32,000 RBCs that were examined from BLU virus-infected sheep, 34 (0.106%) contained labelled BLU virus particles. 5. No labelled particles were observed in any of ca 8000 RBCs taken prior to BLU virus inoculation of sheep. 6. If the antigenic BLU virus particles (which may be viral cores) are in fact infectious, this method of sequestration of virus within RBCs could contribute to the prolonged viremia typical of this arboviral disease, which is known to occur concurrently with circulating neutralizing antibody.

INTRODUCI'ION (BLU) virus with by Luedke (1970) who noted that BLU viremia in sheep, cattle and goats was associated primarily with cells rather than plasma. Quantitative assay of virus in the plasma, huffy coat and washed RBC fractions showed that during the acute stages of the disease there was lo-100 times the concentration of virus in the washed RBCs than in the huffy coat fractions and lOOO-1,000,000 times the concentration as compared with the plasma fractions. He found that on day 28 postinfection, when blood contained specific BLU antibodies, the whole blood samples in anticoagulant were negative for isolation of BLU virus, whereas washed erythrocyte fractions yielded viral titers as high as l x 104”. He concluded that BLU virus appears to be very closely associated with the RBCs during the early acute and convalescent stages of BLU virus infection. Foster and Alders (1979) observed that BLU virus is a membranophilic virus that acquires a membrane from the host cell by budding exocytosis or cell lysis. They suggested that the host cell membrane could protect the virion from BLU virus-specific antibody and permit the long viremic periods of weeks in sheep and months in cattle. In their study of the association of BLU virus with blood cells in cattle, Collisson and Barber (1983) noted a high afhnity of the virus for cell membranes. Recently, BLU virus was shown to agglutinate human, ovine and porcine RBCs (Eaton and The association of bluetongue blood cells was first reported

1989). This suggests that the association of BLU virus with RBCs may play a role in the dissemination of virus throughout the body of the vertebrate host and also may help maintain the virus within the gut of the Cuficoih vector following a blood meal. In a study of the infection of the midgut of Culicoides variipennis with BLU vies, Sieburth et al. (1991) noted that virus particles were associated with sheep RBCs as early as 6 hr after ingestion of the blood meal. Furthermore, immunogold labelling revealed the presence of BLU virions within the RBCs as early as 2 days post-ingestion. This indicated that although the initial infection of insect midgut cells occurs within 2 hr post-ingestion of the blood meal, further infection of midgut cells can take place later, after breaks occur in the peritrophic membrane and BLU virus in the blood meal is released from the sheep RBCs. The adsorption of virus to erythrocyte stroma, or some other undetined association with erythrocytes, has been demonstrated or suggested for numerous diseases, including Rift Valley fever virus (Mims, 1956), African swine fever virus (Sirbu et al., 1964), Newcastle disease virus (Overman, 1958) and BLU virus (Luedke, 1970). However, reports of the intraerythrocytic location of virus have been relatively few. Emmons et al. (1972) demonstrated the intraerythrocytic location of Colorado tick fever virus by electron microscopy (EM) and phase microscopy and concluded that the prolonged duration of viremia was apparently due to the intracellular location of virus in erythrocytes, where it was protected from antibody or other host defense mechanisms. Crameri,

471

R. A. NUNAMAKER et al.

472

In the present study, immunogold labelling and transmission EM were used to determine whether BLU virus can be detected intracellularly in RBCs from sheep that are infected MATERIALS

with this virus.

AND METHODS

Sheep

Four yearling Suffolk sheep (021, 022, 023 and 024) of both genders were obtained locally in Wyoming. These sheep were clinically normal and tested negative for group specific anti-BLU virus antibodies. Each animal was inoculated intravenously with 2 ml of BLU virus serotype 10 (1 x 106.6TCID,) propagated in baby hamster kidney (BHK-21) cells. Blood

Blood samples were collected from each sheep at 0, 3, 7, 14 and 21 days post-infection (DPI). Two milliliter heparinized aliquots were centrifuged at 3400g for 8 min. The serum and bulfy coat were removed and the RBCs were diluted 1: 10 (RBC: sterile PBS buffer). The diluted samples were maintained at 4°C until the following day, when they were processed according to the procedure described below in “Preparation and immunolabelling of cytoskeletons.” Preparation

of EM grids

Two hundred mesh gold EM grids were coated with 0.5% Formvar. They received a carbon film in an Illini vacuum evaporator (Tolono, Illinois) for added stability and electron conductivity, and were placed in 1% Triton X-100 for 5 min to render them hydrophilic. The grids were air dried, then sterilized under short wavelength ultraviolet light for 30 min. They were then placed in 35 x 10 mm sterile

petri dishes that contained 2ml sterile PBS and stored overnight at 4°C. Monoclonal

antibody

Monoclonal antibody (1 AA4E4) produced to BLU virus serotype 11 (which crossreacts with BLU virus serotype 10) and specific for the VP7 structural protein (Mecham et al., 1990) was provided as ascitic fluid by J. 0. Mecham. Preparation

and immunolabelling

of cytoskeletons

One milliliter aliquots of the previously diluted blood samples were added to the Petri dishes that contained the sterile PBS and EM grids. The Petri dishes were again placed at 4°C overnight to allow the RBCs to settle and adhere to the grids. Cytoskeletons were prepared essentially according to the procedure of Hyatt et al. (1987) with modifications. The grids with adsorbed RBCs were fixed in 0.1% glutaraldehyde in 1% Nonidet P-40 (NP-40) for 2 min. Next, they were washed three times (5min each) in PBS, pH 7.2, washed 10 min in 5% lish gelatin in PBS and washed three times (3 min each) in 1% fish gelatin in PBS. The grids were then incubated with monoclonal antibody that had been diluted 1: 500 in phosphate buffered gelatin (PBS, 0.1% gelatin, 0.5% bovine serum albumin) for I hr at 37°C. They were washed six times (3 min each) in 1% fish gelatin in PBS and put into 5 nm Protein-A gold for 1 hr at 37°C. After three washes (3 min each) in 1% fish gelatin in PBS they were washed three times (3 min each) in PBS. Cells were fixed in 2.5% glutaraldehyde in PBS for 10 min and washed three times (5 min each) in PBS. They were post-fixed in 1% 0~0, in PBS for 10 rain and washed three times (1 min each) in double distilled water. The cells were dehydrated for 5 min each in 70,95 and 100% EtOH and critical point dried with CO, in a Balzers CPD 620 critical point drier (Balzers,

Fig. 1. Sheep erythrocytes adsorbed to gold EM grid. The high electron density does not allow for the study of internal structure. Bar = 100 nm.

Detection of viruses in RBC Liechtenstein). “Controls” consisted of RBC samples that

were collected from the sheep just prior to BLU virus inoculation and subjected to immunolabelling as described, As a check for nonspecific labelling, grids with RBCs were incubated in abnormal peritoneal fluid in place of the monoclonal antibody. As a further check for nonspecific labelling, some grids were incubated in Protein-A gold but not in primary antibody. For added stability, the grids received an additional carbon coat. They were viewed at 60 kV in a Philips LS 410 transmission EM. Double labehg To reduce the possibility that any viral particles attached to the outside of the RBC membrane might become dislodged during the detergent extraction procedure and end up inside the cell prior to labelling, a double labelling procedure was utilized for several grids. The RBCs were incubated (pre- and post-infection) with primary antibody, then labelled with 15 nm Protein-A gold prior to the detergent extraction. After a series of washes and blocking steps, the cells were subjected to the described immunolabelling protocol utilizing 5 nm gold. In this way, any viral particles that were initially on the outside of the erythrocytes, but “internalized” during the detergent extraction would be labelled with 15 nm gold. Likewise, particles that were initially (prior to RBC membrane dissolution) intracellular would be labelled with 5 nm gold. Virus isolation Virus isolation from whole blood on days 0, 3, 7, 14 and 21 post-inoculation from sheep 021, 022, 023 and 024 was carried out using a cell culture technique (MacLachlan and Thompson, 1985; MacLachlan et al., 1990). Briefly, whole blood (unseparated) or the separated cell fractions were lysed by resuspension in 2 mM Tris buffer. Diluted aliquots

413

Table 1. Bluctonnuc virus titers in whole blood from experimentally i&ted sheep (021, 022, 023, 024) at O-21 days post-infection (DPI) DPI 0

3 I 14 21

021

022

023

024

Neg. > lti’f > Iti.1 10’” Nag

Nee

Nag

> lti,’ > IV’ IO” ;$A

> 104.’ > le.1

Nes

Ie-9

Gg

=-IO” > 104.’

IOZ.6 $1

*Virus not isolated from 0.1 ml washed and lysed blood cells. tTiters reported as TCIDdml washed and resuspended, unseparated cells.

were added to confluent BHK-21 cells and the cultures examined daily (for 10 days) for cytopathic effects. Titers are reported as log,,, TCID,/ml of washed and resuspended blood cells (resuspended at approximately physiological concentration). RESULTS Virus isolation

Virus titers in whole blood were highest on days 3 and 7 post-infection (Table 1). At 7 DPI, the highest titers of virus were obtained from the red cell fraction, followed by the buffy coat and the plasma (Table 2). Immunoelectron microscopy

Although intact RBCs are too electron dense to be viewed by conventional transmission EM (Fig. I), a detergent extraction procedure produces RBC

Fig. 2. Cytoskeleton of sheep erythrocyte. The electron dense areas (arrows) are presumably rich in iron. Bar = 100 nm.

R. A. NUUAMAKER et al.

474

Table 2. Bluetongue virus titers in blood fractions from experimentally infected sheep (021, 022, 023,024) on day 7 oost-infwtion r

021 101.4* IO’” > IO’.’

Fraction Plasma Red cells Buffv coat

022 IO’.9 > lo”1 > 103,’

*Titers reported as TCID,/ml unseparated calls. tVirus not detected.

023

024

Net?t I@9 > IO’.’

Net? > 10s.’ > lo’.’

wash& and resuspended

skeletons that can be easily studied (Fig. 2). Immunolabelled virus particles were observed within the erythrocytes of infected sheep at 3,7, 14 and 21 DPI. Although relatively few erythrocytes contained labelled particles, immunogold labelling with a monoclonal antibody reactive with VP 7 confirmed that BLU virus particles were present within some RBCs from the infected sheep (Table 3, Fig. 3). The ultrastructure of these viral particles was altered during cytoskeleton preparation, yet the appearance of the particles was indistinguishable from those obtained from BLU virus serotype lo-infected BHK-21 cell cultures that had been processed in the same manner (Fig. 4). The labelled particles had a mean diameter of 59 nm, nearly the same size as BLU virus serotype 1 cores purified by Mertens et al. (1987). We examined cu 32,000 cells from the four infected sheep (ca 2000 cells each at 3, 7, 14 and 21 DPI). Seventy-one well-labelled virus particles were observed in a total of 34 RBCs. Consequently, only

Fig.

3.

Erythrocyte

0.106% of the total cells examined contained labelled BLU virus particles. No labelled particles were observed in any of the cu 8000 RBCs examined from samples taken just prior to inoculation of the sheep with virus. No particles that were labelled with 15 nm gold prior to detergent extraction were found within the cell, yet numerous particles labelled with 5 nm gold after detergent extraction were observed to be intracellular. DISCUSSION

We have described detecting viruses within microscopy. By preparing adsorbed to EM grids,

a simplified method for RBCs by immunoelectron cytoskeletons from RBCs one can view an entire

Table 3. Immuno1abe11edbluetongue virus particks observed within erythrocytes of experimentally infected sheep (021, 022, 023, 024) at O-21 days post-infection (DPI) DPI

021

022

023

024

0

Neg. 4t 9 4 2

Neg

Neg

Net?

5 I 3 2

4 I 3 2

3 1 14 21

6 8 2 3

*No labellad virus particles were observed within ca 2000 erythrocytes. tNumbcr of labelled virus particles observed within ca 2000 erythrccytes.

skeleton from sheep infected with bluetongue (BLU) virus Note immunolabelled BLU virus particle (arrow). Bar = 100 nm.

serotype

10.

Detection of viruses in RBC erythrocyte rapidly since the cutting of ultrathin sections is not required. Our study demonstrated that BLU virus particles can be found within RBCs of BLU virus-infected sheep. Evidence that these particles were indeed intracellular was provided by the double-labelling experiment. This indicates that the intracellular goldlabelled viral particles were not simply displaced particles previously attached to the outside of the erythrocyte. The particles were easily labelled by using a Mab to VP7 [a predominantly internal, virus structural protein (Mertens et al., 1987)] and Protein-A gold. Additionally, the mean diameter of the particles (59 nm) was approximately the same as virus cores of BLU virus serotype 1 (58 nm) described by Mertens et al. (1987). Sieburth et al. (1991) observed that BLU virus particles were taken up into ovine RBCs in the insect infective blood meal as early as 6 hr after ingestion. The viral particles inside the RBCs were presumed to be sheltered by the digestive process because the small fraction in the anterior central part of blood cells remained unchanged until the end of digestion (Edman, 1970; Hirumi et al., 1971). BLU virus may be released during RBC digestion and enter midgut cells not previously infected. In the present study, BLU virus particles were detected within sheep erythrocytes on days 3, 7, 14 and 21 post-infection. Although the mode(s) of entry of the viral particles into the erythrocyte is not known, the possibility exists that RBC precursor cells

475

are infected in the bone marrow and are subsequently released into the blood stream. If the BLU virus particles that we observed within the RBCs are indeed infectious BLU cores (Mertens et al., 1987), they could presumably persist in the bloodstream of the infected host for prolonged periods despite the presence of serum antibody. This virus could be stabilized and then released upon degradation of the RBC, which may circulate for 175 days in cattle (Vacha, 1983). In an EM study of blood cells from calves experimentally infected with BLU virus, Morrill and McConnell (1985) noted the presence of BLU virus-like particles within the cytoplasmic vacuoles of infected monocytes and lymphocytes. Interestingly, the particles were 60nm in diameter, virtually the same size as the labelled intraerythrocytic particles that we observed. They did not find virus-like particles associated with either erythrocytes or platelets. However, since their study utilized ultrathin sections of blood components and comparatively few cells can be studied, it is likely that virus-like particles within the erythrocytes would be missed if the particles were present in the low numbers indicated by our study. Moreover, erythrocytes are quite electron dense, consequently, the relatively rare occurrence of virus-like particles could go unnoticed in TEM preparations. Although the infection of erythrocyte precursor cells could account for the presence of the immunolabelled BLU virus particles observed in mature, non-nucleated erythrocytes, there must be alternative

Fig. 4. Four densely labelled bluetongue (BLU) serotype 10 virus particles in baby hamster kidney cell cultures. Bar = 100 nm.

416

R. A. NUNAMAKER et al.

or additional explanations (labelled particles were observed as early as 3 days post-infection, and presumably, erythrocyte precursor cells could not be infected this early during the infection cycle). An alternative explanation is that the particles are viral cores that penetrate the outer membrane of the circulating erythrocyte following release from the primary site of virus replication within the host. According to Hyatt (personal communication) there is evidence that BLU virus cores are capable of direct cell penetration. Admittedly, the procedure used for the preparation of cytoskeletons may result in the loss of many constituents of the cell, including viral particles, since most of the cytoplasm is removed. This could result in a significant underestimate of the number of RBCs that contain BLU virus particles in viva. The sequestration of BLU virus particles within RBCs may be of significance to the course of BLU disease. Sequestered viral particles could contribute to prolonged infections, since viremias could occur concurrently with neutralizing antibody. Acknowledgements-We thank Drs B. M. Gorman, J. E. Jewell. T. D. G’Toole and T. E. Walton for useful discussion, C. E. Nunamaker for photography and N. C. Bozek and D. M. Heiser for help with preparation of the manuscript. REFERENCES

Collisson E. W. and Barber T. L. (1983) Blood cells associated with bluetongue virus infection in cattle. Proc. A. Meeting Am. Ass. Vet. Lab. Diagn. 26,287-300. Eaton B. T. and Crameri G. S. (1989) The site of bluetongue virus attachment to glycophorins from a number of animal erythrocytes. J. gen. Viral. 70, 3341-3353. Edman J. D. (1970) Rate of digestion of three human blood fractions in Aedes aegypti (Diptera: Cuhcidae). Ann. Entornol. Sot. Am. 63, 1778-1779. Emmons R. W., Gshiro L. S., Johnson H. N. and Lennette E. H. (1972) Intra-erythrocytic location of Colorado tick fever virus, J. gen. Virof. 17, 185-195. Foster N. M. and Alders M. A. (1979) Bluetongue virus:

a membraned structure. 37rh A. Proc. Elecrron Microscopy Sot. Am. 37, 48-49. Hirumi H., Burton G. J. and Maramorosch K. (1971) Electron microscopy of friend murine leukemia virus in the midgut of experimentally infected mosquitoes. J. Viral. 8, 801-804. Hyatt A. D., Eaton B. T. and Lunt R. (1987) The grid-cellculture technique: the direct examination of virus-infected cells and progeny viruses. J. Microsc. 145, 97-106. Luedke A. J. (1970) Distribution of virus in blood components during viremia of bluetongue. Proc. 74rh A. Meeting US. Animal Health Ass. 74, 9-21. MacLachlan N. J., Jagels G., Rossitto P. V., Moore P. F. and Heidner H. W. (1990) The pathogen&s of experimental bluetongue infection of calves. Vet. Path. 27, 223-229. MacLachlan N. J. and Thompson J. (1985) Bluetongue virus-induced interferon in cattle. Am. J. Vet Res. 46, 1238-1241. Mecham J. O., Dean V. C., Wigington J. G. and Nunamaker R. A. (1990) Detection of bluetongue virus in Culicoides variipennis (Diptera: Ceratopogonidae) by an antigen capture enzyme-linked immunosorbent assay. J. med. Entomol. 27, 602-606. Mertens P. P. C., Burroughs J. N. and Anderson J. (1987) Purification and oronerties of virus oarticles. infectious subviral particles and cores of bluetongue virus serotypes 1 and 4. Virology 157, 375-386. Mims C. A. (1956) Rift Valley fever virus in mice. I. General features of the infection. Br. J. exp. Path. 37, 99-109. Merrill J. C. and McConnell S. (1985) An electron microscopic study of blood cells from calves experimentally infected with bluetongue vies. In Bfuetongue and Related Orbiviruses (Edited by Barber T. L. and &him M. M.), DO.279-287. Alan R. Liss. New York. O&man J. R. (1958) Vim&a in human mumps virus infections. Arch. Internal Med. 102, 354-362. Sieburth P. J., Nunamaker C. E., Ellis J. A. and Nunamaker R. A. (1991) Infection of the midgut of Culicoides varii&mnis (Diptera: Ceratopogonidae) with bluetongue virus. J. med. Entomol. 28, 74-85. Sirbu Z., Ieremia D. and Bona C. (1964) Immunofluorescent microscopy in the diagnosis of swine fever. Br. Vet. J. 120, 587-591. Vacha J. (1983) Red cell life span. In Red Bfood Cells of Domestic Mammals (Edited by N. S. Agar and P. G. Board). Elsevier, N.Y.

The detection of intracellular bluetongue virus particles within ovine erythrocytes.

1. We report here a simplified method for detecting viruses and other antigenic agents in red blood cells (RBCs). Using a nonionic detergent to prepar...
884KB Sizes 0 Downloads 0 Views