- Email: [email protected]
A simple method for obtaining highly viable virus from culture supernatant
z r’
Iii
*
GP I ..J
LSEVIER
7
Journal
of Virological
Methods
46 (1994) 349-352
Journal of \~irolo@al wlotls
Short Communication
A simple method
for obtaining highly viable virus from culture supernatant
(Accepted
14 September
1993)
Abstract density-gradient methods are used to purify viruses. However, these procedures are not only time consuming and cumbersom, recovery of viable viruses are often quite low. In this report, a single-step concentration technique was used to concentrate a mutant of Moloney murine leukemia virus (~1) virus from culture supernatants by ultrafiltration. A special ultatiltration unit with a 100000 mol wt cut-off was able to concentrate viruses about 30-fold without losing any infectivity. In comparison, traditional sucrose density gradient purified viruses lost a significant portion of their infectivity. This technique could be used for concentrating other viruses for many useful purposes where more viable viruses are needed, e.g., study of virus-cell binding.
Traditionally
Kc~ NYW~Y:Retrovirus;
Ultrafiltration;
Viable
1. Introduction A prerequisite for better understanding virus-cell interaction has been the availability of large quantites of viable and infective virus particles. This is particularly true for the more fragile enveloped viruses. Infective viruses are especially needed for studying virus binding and penetration of cellular membranes. Although some viruses are purified by ion-exchange chromatography (Downing et al, 1992), traditionally most of the viruses are purified by centrifugation over a density gradient which can be quite cumbersome and time consuming (Lambert et al., 1980, Levine et
*Corresponding
author.
Fax: +
I 512 2372444
0166~0934:94/$07.00 (‘ 1994 Elsevier Science B.V All rights SSDI 0166.0934(93)EOl29-P
reserved
al., 1989, Todd et al., 1990). Sucrose was first used as a density gradient in the early 1950’s to study a plant virus (Brakke, 19.51, 1953), Since then, several other substances including percoll, renograffin, and metrizamide have been used as density gradient to purify different viruses with varying degrees of success. Whiie some studies found sucrose as a superior gradient over percoll and renografin (Mbignino and Menezes, 1991), others have found percoll as a better gradient for virus purification (Bustos et al., 1991). Nonetheless, a significant loss in infectivity as much as 99% is associated during the density gradient purification process (Wunner and Pringle, 1976). We have been studying a murine retrovirus rsl, a mutant of Moloney murine leukemia virus, for the past several years (Wang, 1990, Wong et al., 1991, Saha and Wong, 1992a, 1993). We have used the conventional sucrose density gradient to purify 1s1 for molecular biological and immunological studies (Wong, 1990). We have also observed significant loss in infectivity during this process (unpublished observation). We have recently used an ultrafiltration technique to concentrate monoclonal antibodies from culture supernatant (Saha et al., 1992). In this technique, we used a special ~ltration unit, Centricon(Amicoll, Beverly MA) which has a 100000 mol wt cut-off, to separate the heavy-weight antibody molecules from lighter weight serum and media. Here we used a similar approach to concentrate rsl virus from culture supernatant. tsl was produced from a chronically infected thymus-bone marrow (TB) cell line as previously described (Wong et al., 1991, Saha and Wong, 1992a). The culture supernatant was concentrated using Centrico~l-100 according to the manufacturer.s directions. The supernatant was concentrated >90% with each spin since most of the medium with serum albumin passed through the filter membrane leaving behind a concentrated virus. We were able to concentrate the original culture supernatant between 2530 fold since from a starting volume of about 25 ml, we filtered the final virus solution to less than I ml. We then compared the titers of the original virus. the concentrated virus, and virus purified through conventional sucrose density gradient on 15 F cells as previously described (Wong et al., 1951). As shown in Table 1. the ultrafiltered virus titer was almost 30 times more than that of the original viral sup, while sucrose density purified virus (resuspended in PBS to give a final concentration of about 30 times over the original viral sup) had titers only about 22.3 fold more than that of the original sup inspite of being 30 times more concentrated. We repeated these experiments on three times using original sup from various TB cell
Table I Virus
titers”
obtained
from
diffcrcnt
sources tsl (Infectious
Source Culture
supernatant
Culture
supcrnatant
after
Culture
supernatant
after
“Virus
units:ml)
3.6 2 2.4 x
10”
sucrose gradient
6.8 2 3.1 x
IO”
ultrafiltration
1.7 *
IOX
titers were determined
in Centricon-
on 15F ceils as described
(Wang
et :11., 1981).
.6l
x
K. Saha et al./Journal of Virological Methods 46
(19941
349-352
351
lines and obtained similar results. These results clearly show that virus concentrated by this new ultrafiltration technique did not lose any infectivity while sucrose gradient purified virus lost significant infectivity. We also compared the binding abilities of the viruses unpurified or purified by different ways since viable viruses are normally needed for virus-cell binding. FACS analyses showed that virus concentrated by ultrafiltration could bind much better when compared to sucrose gradient purified virus or unpurified virus (data presented elsewhere). This supports the results of our infectivity assay and reconfirms that the virus obtained by ultrafiltration were more viable than density gradient purified virus. Enveloped viruses like tsl has their primary receptor binding (env) protein (gp70 in case tsl) protrudes out as a knob on the surface of the virus particle (Wong, 1990). It is suggested that harsh treatment at high speed centrifugation during sucrose gradient purification process may tear off many of these surface ‘env’ proteins resulting in the loss of infectivity. We believe that a much gentler speed during ultrafiltration (only about 500 x g compared to about 100 000 x g during sucrose gradient) was primarily responsible for the recovery of this highly viable virus. There are limitations in the use of ultrafiltration for virus purification, for example, very small viruses like the chicken anaemia agent (CAA) may not be purified in this way because of their smaller sizes, and the concentrated virus may be associated with other high molecular weight cellular proteins. Nonetheless, we believe that this single step, simple purification process could be very useful with a broad range of viruses, especially whenever a highly viable concentrated virus sample is required.
2. References Brakke, M.K. (1951) Density gradient centrifugation: a new separation technique. J. Am. Chem. Sot. 73, 1847. Brakke, M.K. (1953) Zonal separations by density-gradient centrifugation Arch. Biochem. Biophys. 45, 275. Bustos, J.. Zamora, P., Mejia, E., Varek, Y. and Gomez, B. (1991) Purification of rubella virus by isoprenic gradients: continuous percell versus discontinuous sucrose. Arch. Virol. 118, 285. Downing, L.A.. Bernstein, J.M. and Walter, A. (1992) Active respiratory syncytial virus purified by ionexchange chromatography: characterization of binding and elution requirements. J. Viral. Methods. 38. 215. Lambert, D.M.. Pons, M.W., MGuy, G.N. and Dorsch-Hasler, K. (1980) Nucleic acids of respiratory syncytial virus. Virology 36, 837. Levine, S. and Hamilton, R. (1969) Kinetics of respiratory syncytial virus growth cycle in Hela cells. Arch. gesamte virusforschung. 28, 122. Mbignino, A. and Menezes, J. (1991) Purification of human respiratory syncytial virus: superiority of sucrose gradient over percoll, renograftin and metrizamide gradients. J. Viral. Methods 31, 161. Saha, K., Case. R. and Wong, P.K.Y. (1992) A simple method of concentrating monoclonal antibodies from culture supernatant by ultrafiltration. J. Immunol. Methods 151, 307. Saha. K. and Wong, P.K.Y. (1992a) rsl, a temperature sensitive mutant of Moloney murine leukemia virus TB, can infect both CD4+ and CD8+ T cells but requires CD4+ T cells in order to cause paralysis and immunodeticiency. J. Virol. 676. 2639. Saha, K. and Wong. P.K.Y. (1993) Rudimentary thymus of SCID mouse plays an important role in the development of retrovirus-induced neurologic disorders. Virology 195, 211. Todd, I>.. Creelan, J.L., Mackie, D.P.. Rixon. F. and McNulty, MS. (1990) Puritication and biochemical
characterization of chicken anaemia agent. J. Gen. Viral. 71. 819. Wang. P.K.Y (1990) Moloney murine leukemia bitus temperature-sensitive mutants: a model for retrovirus-induced neurologic disorders. Curr. Top. Microbial. Immunol. 160. 29. Wang, P.K.Y.. Soong. M.M. and Yuen. P.H. (1981) Replication of murine leukemia virus in hrterologous cells:Interaction between ecotropic and xenotropic viruses. Virology 109, 366. Wang. P.K.Y.. Srurek, P.F.. Floyd. E.. Saha. K. and Brooks, B.R. (1991) Alteration from T- to B-cell tropism reduces cytocidal effects in thymocytes but not nemovirulence induced by /.?I. a mutant 01 Moloncy murine leukemia virus TB. Proc. Natl. Acad. Sci. USA XX. XY9l. Wunner. W.H. and Pringle, C.R. (1976) Resptratory syncyttal virus protems. Virology 73. 22X.
Iii
*
GP I ..J
LSEVIER
7
Journal
of Virological
Methods
46 (1994) 349-352
Journal of \~irolo@al wlotls
Short Communication
A simple method
for obtaining highly viable virus from culture supernatant
(Accepted
14 September
1993)
Abstract density-gradient methods are used to purify viruses. However, these procedures are not only time consuming and cumbersom, recovery of viable viruses are often quite low. In this report, a single-step concentration technique was used to concentrate a mutant of Moloney murine leukemia virus (~1) virus from culture supernatants by ultrafiltration. A special ultatiltration unit with a 100000 mol wt cut-off was able to concentrate viruses about 30-fold without losing any infectivity. In comparison, traditional sucrose density gradient purified viruses lost a significant portion of their infectivity. This technique could be used for concentrating other viruses for many useful purposes where more viable viruses are needed, e.g., study of virus-cell binding.
Traditionally
Kc~ NYW~Y:Retrovirus;
Ultrafiltration;
Viable
1. Introduction A prerequisite for better understanding virus-cell interaction has been the availability of large quantites of viable and infective virus particles. This is particularly true for the more fragile enveloped viruses. Infective viruses are especially needed for studying virus binding and penetration of cellular membranes. Although some viruses are purified by ion-exchange chromatography (Downing et al, 1992), traditionally most of the viruses are purified by centrifugation over a density gradient which can be quite cumbersome and time consuming (Lambert et al., 1980, Levine et
*Corresponding
author.
Fax: +
I 512 2372444
0166~0934:94/$07.00 (‘ 1994 Elsevier Science B.V All rights SSDI 0166.0934(93)EOl29-P
reserved
al., 1989, Todd et al., 1990). Sucrose was first used as a density gradient in the early 1950’s to study a plant virus (Brakke, 19.51, 1953), Since then, several other substances including percoll, renograffin, and metrizamide have been used as density gradient to purify different viruses with varying degrees of success. Whiie some studies found sucrose as a superior gradient over percoll and renografin (Mbignino and Menezes, 1991), others have found percoll as a better gradient for virus purification (Bustos et al., 1991). Nonetheless, a significant loss in infectivity as much as 99% is associated during the density gradient purification process (Wunner and Pringle, 1976). We have been studying a murine retrovirus rsl, a mutant of Moloney murine leukemia virus, for the past several years (Wang, 1990, Wong et al., 1991, Saha and Wong, 1992a, 1993). We have used the conventional sucrose density gradient to purify 1s1 for molecular biological and immunological studies (Wong, 1990). We have also observed significant loss in infectivity during this process (unpublished observation). We have recently used an ultrafiltration technique to concentrate monoclonal antibodies from culture supernatant (Saha et al., 1992). In this technique, we used a special ~ltration unit, Centricon(Amicoll, Beverly MA) which has a 100000 mol wt cut-off, to separate the heavy-weight antibody molecules from lighter weight serum and media. Here we used a similar approach to concentrate rsl virus from culture supernatant. tsl was produced from a chronically infected thymus-bone marrow (TB) cell line as previously described (Wong et al., 1991, Saha and Wong, 1992a). The culture supernatant was concentrated using Centrico~l-100 according to the manufacturer.s directions. The supernatant was concentrated >90% with each spin since most of the medium with serum albumin passed through the filter membrane leaving behind a concentrated virus. We were able to concentrate the original culture supernatant between 2530 fold since from a starting volume of about 25 ml, we filtered the final virus solution to less than I ml. We then compared the titers of the original virus. the concentrated virus, and virus purified through conventional sucrose density gradient on 15 F cells as previously described (Wong et al., 1951). As shown in Table 1. the ultrafiltered virus titer was almost 30 times more than that of the original viral sup, while sucrose density purified virus (resuspended in PBS to give a final concentration of about 30 times over the original viral sup) had titers only about 22.3 fold more than that of the original sup inspite of being 30 times more concentrated. We repeated these experiments on three times using original sup from various TB cell
Table I Virus
titers”
obtained
from
diffcrcnt
sources tsl (Infectious
Source Culture
supernatant
Culture
supcrnatant
after
Culture
supernatant
after
“Virus
units:ml)
3.6 2 2.4 x
10”
sucrose gradient
6.8 2 3.1 x
IO”
ultrafiltration
1.7 *
IOX
titers were determined
in Centricon-
on 15F ceils as described
(Wang
et :11., 1981).
.6l
x
K. Saha et al./Journal of Virological Methods 46
(19941
349-352
351
lines and obtained similar results. These results clearly show that virus concentrated by this new ultrafiltration technique did not lose any infectivity while sucrose gradient purified virus lost significant infectivity. We also compared the binding abilities of the viruses unpurified or purified by different ways since viable viruses are normally needed for virus-cell binding. FACS analyses showed that virus concentrated by ultrafiltration could bind much better when compared to sucrose gradient purified virus or unpurified virus (data presented elsewhere). This supports the results of our infectivity assay and reconfirms that the virus obtained by ultrafiltration were more viable than density gradient purified virus. Enveloped viruses like tsl has their primary receptor binding (env) protein (gp70 in case tsl) protrudes out as a knob on the surface of the virus particle (Wong, 1990). It is suggested that harsh treatment at high speed centrifugation during sucrose gradient purification process may tear off many of these surface ‘env’ proteins resulting in the loss of infectivity. We believe that a much gentler speed during ultrafiltration (only about 500 x g compared to about 100 000 x g during sucrose gradient) was primarily responsible for the recovery of this highly viable virus. There are limitations in the use of ultrafiltration for virus purification, for example, very small viruses like the chicken anaemia agent (CAA) may not be purified in this way because of their smaller sizes, and the concentrated virus may be associated with other high molecular weight cellular proteins. Nonetheless, we believe that this single step, simple purification process could be very useful with a broad range of viruses, especially whenever a highly viable concentrated virus sample is required.
2. References Brakke, M.K. (1951) Density gradient centrifugation: a new separation technique. J. Am. Chem. Sot. 73, 1847. Brakke, M.K. (1953) Zonal separations by density-gradient centrifugation Arch. Biochem. Biophys. 45, 275. Bustos, J.. Zamora, P., Mejia, E., Varek, Y. and Gomez, B. (1991) Purification of rubella virus by isoprenic gradients: continuous percell versus discontinuous sucrose. Arch. Virol. 118, 285. Downing, L.A.. Bernstein, J.M. and Walter, A. (1992) Active respiratory syncytial virus purified by ionexchange chromatography: characterization of binding and elution requirements. J. Viral. Methods. 38. 215. Lambert, D.M.. Pons, M.W., MGuy, G.N. and Dorsch-Hasler, K. (1980) Nucleic acids of respiratory syncytial virus. Virology 36, 837. Levine, S. and Hamilton, R. (1969) Kinetics of respiratory syncytial virus growth cycle in Hela cells. Arch. gesamte virusforschung. 28, 122. Mbignino, A. and Menezes, J. (1991) Purification of human respiratory syncytial virus: superiority of sucrose gradient over percoll, renograftin and metrizamide gradients. J. Viral. Methods 31, 161. Saha, K., Case. R. and Wong, P.K.Y. (1992) A simple method of concentrating monoclonal antibodies from culture supernatant by ultrafiltration. J. Immunol. Methods 151, 307. Saha. K. and Wong, P.K.Y. (1992a) rsl, a temperature sensitive mutant of Moloney murine leukemia virus TB, can infect both CD4+ and CD8+ T cells but requires CD4+ T cells in order to cause paralysis and immunodeticiency. J. Virol. 676. 2639. Saha, K. and Wong. P.K.Y. (1993) Rudimentary thymus of SCID mouse plays an important role in the development of retrovirus-induced neurologic disorders. Virology 195, 211. Todd, I>.. Creelan, J.L., Mackie, D.P.. Rixon. F. and McNulty, MS. (1990) Puritication and biochemical
characterization of chicken anaemia agent. J. Gen. Viral. 71. 819. Wang. P.K.Y (1990) Moloney murine leukemia bitus temperature-sensitive mutants: a model for retrovirus-induced neurologic disorders. Curr. Top. Microbial. Immunol. 160. 29. Wang, P.K.Y.. Soong. M.M. and Yuen. P.H. (1981) Replication of murine leukemia virus in hrterologous cells:Interaction between ecotropic and xenotropic viruses. Virology 109, 366. Wang. P.K.Y.. Srurek, P.F.. Floyd. E.. Saha. K. and Brooks, B.R. (1991) Alteration from T- to B-cell tropism reduces cytocidal effects in thymocytes but not nemovirulence induced by /.?I. a mutant 01 Moloncy murine leukemia virus TB. Proc. Natl. Acad. Sci. USA XX. XY9l. Wunner. W.H. and Pringle, C.R. (1976) Resptratory syncyttal virus protems. Virology 73. 22X.