- Email: [email protected]
A Rapid and Simple Procedure to Detect the Presence of MVM in Conditioned Cell Fluids or Culture Media
D.R.
841664—BIOL 25/4 (ISSUE)————MS 0108
Biologicals (1997) 25, 415–419
A Rapid and Simple Procedure to Detect the Presence of MVM in Conditioned Cell Fluids or Culture Media Audrey Chang,* Steve Havas, Flavia Borellini, Jeffrey M. Ostrove and Robert E. Bird Scientific Development, MA BioServices, Inc., 9900 Blackwell Road, Rockville MD 20850, U.S.A.
Abstract. During the manufacture of biopharmaceuticals, numerous adventitious agents have been detected in Master Cell Banks, end-of-production cells as well as bulk harvest fluids. Recently, a number of large-scale production bioreactors have become infected with Minute Virus of Mice (MVM) during cGMP (current good manufacturing practices) operations, and this has resulted in both the loss of product and the need for major cleaning validation procedures to be put in place. We have developed a simple DNA extraction/PCR assay to detect the presence of MVM in cell culture supernatant (conditioned cell fluids). This highly specific assay can detect 10 or fewer genome equivalents (copies) of MVM following PCR and gel electrophoresis visualization. For routine high-throughput detection, 30–100 copies could be consistently detected. The extraction procedure was shown to reliably detect MVM at a concentration of 1 TCID50/ml. The combination of the extraction/PCR procedures establishes a powerful, sensitive, specific assay that can detect the presence of MVM sequences with a 1-day turnaround time. = 1997 The International Association of Biological Standardization
Introduction The safety testing of cell substrates for the presence of viruses in biotechnology products is an important issue and is required by regulatory agencies worldwide. At the FDA Viral Safety meeting held in June 1995, the presence of Minute Virus of Mice (MVM) in Chinese Hamster Ovary cell cultures grown in large-scale fermenters was brought to the attention of the biotechnology industry.1 MVM belongs to the Parvoviridae family, a group of small, non-enveloped virus containing single stranded DNA, whose members can infect a wide variety of hosts ranging from vertebrates (including man) to insects. The natural host for the MVM is the house mouse (Mus musculus) and the highly contagious virus is prevalent in both laboratory and wild mice populations. Although MVM infection in the adult mouse is usually asymptomatic, infection in neonatal mice and hamsters can result in deformity and death. Numerous studies have shown MVM’s involvement with biological processes including immunosuppression, cellular differentiation, and cellular proliferation. From a safety and regulatory *To whom correspondence should be addressed: Audrey Chang, MA BioServices, 9900 Blackwell Rd, Rockville MD USA 20850 1045–1056/97/040415 + 05 $25.00/0/bg970110
standpoint, it is important to insure that biopharmaceuticals produced are free from contaminating viruses such as MVM. The family is divided into two genus. MVM belongs to the genus Parvovirus, often referred to as ‘‘autonomous’’ parvovirus to distinguish it from the genus Dependovirus (e.g. adeno-associated virus, AAV) which requires a co-infection with a helper virus in order to complete its replication cycle.2 The DNA replication of the viruses from both genera occurs in the nuclei of dividing infected cells. Parvovirus viral replication involves the cellular replicative functions and requires that the infected cell pass through S phase. It is this close relationship between cell division and MVM viral replication that predisposes rapidly dividing murine cells, commonly grown in bioreactors, to MVM virus contamination. Several in vitro laboratory assays can detect the presence of MVM.3 Autonomous parvovirus has been shown to agglutinate erythrocytes and a standard haemagglutination assay (HA) using guinea-pig erythrocytes has been used since the early 1970s. More recently, infectivity assays measuring the cytopathic effect (CPE) of MVM on susceptible cell culture monolayers has been used for diagnostic purposes. In addition to the HA and CPE assays, the detection of 7 1997 The International Association of Biological Standardization
D.R.
841664—BIOL 25/4 (ISSUE)————MS 0108
416
A. Chang et al .
infected cells by fluorescent antibody staining (FA) and plaque assays have been developed to detect the presence of MVM. All these assays have been instrumental in determining strain and serotype differences for MVM. However, from an industry perspective, issues of duration (for instance, the CPE assay requires 13 days), complexity, and low sensitivity make these assays impractical for routine screening. Determining the presence of an MVM-infection in a bioreactor of several thousand litres during a production run requires an assay with a rapid turnaround time. Polymerase chain reaction (PCR)-based assays can detect the presence of MVM viral DNA sequences specifically and rapidly. Furthermore, PCR is the most sensitive method of detection presently available. It represents an increase of 100-fold in sensitivity over the most sensitive plaque assay.1 A rapid and straightforward DNA sample preparation procedure is equally important in the development of a quick and reliable MVM PCR assay. We present a MVM-PCR assay which includes a ‘‘mini-prep’’ sample preparation from conditioned cell fluids, a PCR amplification specific for MVM sequences, and a diagnostic restriction endonuclease digest that confirms that the PCR product amplified originated from MVM sequences.
Materials and methods Reagents All chemicals and reagents were of the highest molecular biology grade commercially available. Stock solutions wre made in nuclease free sterile water (Baxter, Deerfield, IL). Oligonucleotides were purchased from Gibco/BRL, Gaithersburg, MI). The plasmid pML-N (7660 bp) containing the complete MVM genome (0–5149 nt) was a generous gift from Dr David Pintel of the University of Missouri. The plasmid DNA was used to transform Escherichia coli DH5a cells and the transformed cells were screened for the presence of pML-N. A plasmid DNA preparation of pML-N was prepared and quantitated by Lofstrand Labs Limited, Gaithersburg, MD. The plasmid copy number was calculated using the molecular weight of the plasmid and Avogadro’s constant. One microgram of the plasmid pML-N preparation is approximately equivalent to 1⋅2 × 1011 copies. Concentrated plasmid DNA stock was stored at −20°C and stocks of the bacterial culture in 50% glycerol were stored at −80°C.
Viruses and cells The Crawford strain of MVM (ATCC VR-663) (lot (022096P) with a titre of 1⋅3 × 109 TCID50/ml in 324 K cells was used in the studies. Chinese Hamster Ovary (CHO) cells were grown to confluence in HAM F-12 media, supplemented with 10% fetal bovine serum (FBS), 0⋅1% gentamycin and 0⋅1% fungizone. 324 K and Mus dunni were grown to confluency in Dulbecco’s modified Eagle’s medium (DMEM) media, supplemented with 5% FBS and 1% non-essential amino acids. All cell lines and MVM virus were grown and maintained at MA BioServices. Sample preparation Conditioned cell fluids (CCF) Cell culture supernatants from either CHO, 324 K, or Mus dunni cells were harvested after a minimum of 72 h in culture, and clarified with a low speed centrifugation (1000 g for 10 min). MVM virus The MVM viral stock was freshly diluted in the clarified CCF medium before each experiment. Extraction Different dilutions of the MVM virus in clarified CCF were extracted along with clarified cell fluid (NEG), and dH2O (H2O) as negative controls. Additional CCF samples were extracted and used as the background for positive control plasmid spikes in the subsequent PCR amplification step (NEG + PC). The miniprep extraction of MVM DNA from conditioned cell fluids was adapted from a plasmid miniprep procedure4 with several modifications. Briefly, 225 ml of each sample was placed in a 1⋅5-ml microcentrifuge tube with 150 ml of 2 × TENS solution (20 mm Tris–HCl pH 8⋅0, 0⋅2 mm EDTA pH 8⋅0, 0⋅2 m NaOH, 1% sodium dodecyl sulfate), 40 ml of 3 m NaOAc pH 8⋅2, and 2 ml of mussel glycogen (20 mg/ml). The tubes were vortexed for 2–5 s to mix. Cold absolute ethanol (1 ml) was added, and the tubes were briefly vortexed and centrifuged in an Eppendorf 541C5 at 14 000 rpm for 5 min at room temperature. The ethanol was gently decanted and the pellets were rinsed with 1 ml of room temperature 70% ethanol. The pellets were air-dried for 5 min at room temperature and resuspended in 20 ml of sterile nuclease-free water. The NEG + PC samples were resuspended in 10 ml to accommodate for the positive control plasmid spike volume of 10 ml. The entire resuspension volume was used in the PCR amplification step.
D.R.
841664—BIOL 25/4 (ISSUE)————MS 0108
A. Chang et al .
PCR Amplification A synthetic oligonucleotide [MVM-1391: 5'-d(TTGCCA-TGC-TAT-TTG-CTG-TG)-3'] that anneals to the MVM nucleotide 1391–1411 sequence (accession ( J02275) serves as the forward primer in the PCR amplification. Oligonucleotide MVM-1716 [5'd(TTG-TGG-TCA-TGA-TGA-CTG-GT)-3'] that anneals to the MVM nucleotide 1697–1716 was the reverse primer. A 325-bp region from the non-structural gene of MVM (nt 1391 to nt 1716) was amplified using primers MVM1391 and MVM 1716. PCR conditions were optimized with regard to primer, dNTPs and Mg2 + concentrations, and cycling conditions in a Perkin Elmer 2400 thermal cycler. Amplification was performed using the hot start technique5 in the presence of 50 mm KCl, 10 mm Tris–HCl pH 8⋅3, 0⋅2 mm dNTPs, 1⋅75 mm MgCl2, 0⋅2 mm of each primer, and 2⋅5 U Taq polymerase in a total volume of 100 ml. Cycling conditions were: 94°C for 3 min (1 cycle); 94°C for 40 s, 56°C for 40 s, 72°C for 40 s (40 cycles), 72°C for 2 min (1 cycle). Serial dilutions of the positive control plasmid, pML-N, were freshly prepared and mixed with NEG + PC extraction samples to serve as positive controls. PCR products wre analysed by electrophoresis on a 2⋅5% MetaPhor (FMC) agarose gel containing 0⋅25 mg/ml ethidium bromide. The products of amplification were visualized with a UV transilluminator.
Results Sensitivity of PCR amplification reaction Primers were chosen from the MVM non-structural coding region to specifically amplify a 325-bp region. Dilutions of the pML-N plasmid ranging from 1000 copies to 1 copy were used to evaluate the sensitivity of the PCR assay in viral genome equivalents (copy number). As Figure 1(a) illustrates, amplification of pML-N produced a single band of 325 bp, clearly visualized by ethidium bromide staining in the lanes containing 1000, 100, 30, and 10 copies pML-N. The 325-bp band was not detectable at the 3, 1 and 0⋅1 copy level. To investigate the reproducibillity of the PCR assay, the same panel of template copy levels was tested with multiple operators on different dates. The results are presented in Figure 1(b) and show that using these amplification conditions, 30 to 100 copies of the MVM template are routinely detected.
417
Sensitivity of PCR Amplification for detection of MVM viral particles in conditioned cell fluids (CCF) A ‘‘ten-minute’’ plasmid mini-prep DNA isolation procedure4 was modified and adapted to extract MVM viral particle DNA from conditioned cell fluids. Conditioned cell fluids were harvested from Chinese Hamster Ovary cells after 72 h in culture, and clarified with a low-speed centrifugation. Serial 10-fold dilutions of purified MVM particles (having a titre of 1⋅3 × 109 TCID50/ml) were prepared using clarified Chinese Hamster Ovary CCF as the diluent and extracted as described in Materials and methods and PCR amplified. A ‘‘water only’’ tube was included as a negative control for the extraction and PCR procedure to ensure no contaminating MVM DNA was introduced during the sample manipulations. As Figure 2(a) illustrates, no PCR products are generated in the ‘‘water only’’ or the CHO CCF extracted material (lanes 1 and 2). The absence of the MVM-specific band in the CHO CCF was not due to inhibition of the PCR since the addition of 100 copies of the positive control plasmid to the CHO CCF (lane 3) produced a 325-bp band with an intensity similar to the band produced by the 100 copies of pML-N in the absence of CHO CCF (a)
1
2
3
4
5
6
7
8
(b) Copy No.
100
30
10
3
1
0
Positives/ total runs
6/6
6/6
5/6
2/6
1/6
0/6
Figure 1. Amplification of MVM sequences with the positive control pML-N template. (a) Serial dilutions of pML-N ranging from 0⋅1 to 103 copies/ml were freshly prepared from concentrated stocks and tested with using optimized MVM PCR conditions. The 325-bp PCR product is indicated by the arrow. Lane (1) 1000 copies; (2) 100 copies; (3) 30 copies; (4) 10 copies; (5) 3 copies; (6) 1 copy; (7) no DNA negative control; (8) 123 bp ladder. (b) Summary of several repeat experiments performed using 0, 1, 3, 10, 30 and 100 template copies/reaction.
D.R.
841664—BIOL 25/4 (ISSUE)————MS 0108
A. Chang et al .
418
(a)
1
2
3
4
5
6
7
8
MVM-derived PCR product with RsaI generated a restriction pattern consisting of two fragments of 171 bp and 154 bp. As shown in Figure 3, the RsaI restriction pattern generated by both the pML-N plasmid and the viral template was consistent with the MVM Crawford strain, confirming that the 325-bp PCR product originated from MVM sequences. No other non-specific bands were generated.
(b) Concentration TCID50/ml
100
10
1
Positive/ total runs
6/6
6/6
3/4
Figure 2. Amplification of DNA extracted from CHO CCF spiked with MVM virus. Shown are the results of the PCR on DNA extracted from serial dilutions of the MVM virus using uninfected CHO CCF as the diluent. Lane (1) H2O: extraction and PCR negative control; (2) NEG: CHO CCF; (3) NEG + PC: CHO CCF + 100 copies of pML-N DNA spike; (4) 10 − 7 dilution MVM virus; (5) 10 − 8 dilution; (6) 10 − 9 dilution; (7) pML-N, 100 copies; (8) 123 bp ladder. The 10 − 9 dilution is equivalent to 1 TCID50/ml virus concentration (0⋅225 TCID50 total tested). (b) Summary of several repeat experiments performed using 100, 10, and 1 TCID50/ml per reaction.
Extraction of MVM DNA from other CCFs In addition to CHO CCF, we tested the MVM DNA extraction/PCR amplification procedure using CCF from two additional cell lines: Mus dunni (normal mouse fibroblasts) and 324 K (human embryo kidney cells). MVM virus particles were diluted in each CCF using the same dilution as for CHO CCF and DNA was extracted as described above. The results of the PCR amplification are shown in Figure 4(a) and 4(b). Each of the CCF tested negative for the 325-bp MVM PCR product. However, in the presence of the MVM virus, the 325-bp MVM PCR product was generated. This demonstrates that the assay is not cell-line specific and that various cell lines may be tested successfully using this procedure. 1
2
3
4
5
6
(lane 7). When the various MVM dilutions were added to the CCF for DNA extraction and PCR analysis, the 325-bp band was obtained at the 10 − 7, 10 − 8 and 10 − 9 dilutions (lanes 4, 5, and 6). No other non-specific bands were seen. In terms of viral titre endpoint, the 10 − 9 dilution corresponds to one TCID50/ml MVM virus. The same virus dilutions were tested with multiple operators on different dates. As shown in Figure 2(b), 10 to 1 TCID50/ml MVM virus concentration could be routinely detected by this methodology. Restriction analysis of PCR products To confirm PCR specificity, amplified products produced from both the plasmid template and the virus DNA were digested with the restriction endonuclease RsaI and analysed by high-resolution agarose gel electrophoresis. As predicted from the reported MVM sequences,6 digestion of the 325-bp
Figure 3. Diagnostic restriction endonuclease digestion: The 325-bp PCR products generated for Figure 2 were digested with 10 units of the restriction endonuclease RsaI (Lanes 2–5). Lane (1) 100 copies pML-N undigested; (2) 100 copies pML-N digested with RsaI; (3) MVM virus dilution 10 − 7 + RsaI; (4) 10 − 8 + RsaI; (5) 10 − 9 + RsaI; (6) 50 bp ladder. The large arrow indicates the 325-bp undigested band. The two small arrow heads indicate the position of the 171 bp and 154 bp RsaI digestion products.
D.R.
841664—BIOL 25/4 (ISSUE)————MS 0108
A. Chang et al .
(a)
1
2
3
4
5
6
7
(b)
1
2
3
4
5
6
7
419
the PCR protocol often resulted in 10- to 100-fold differences in assay sensitivity. The limit of detection for the optimized PCR reaction when using purified DNA as a template was shown to be as low as 10 viral genome equivalents. On a routine basis, the assay could detect between 30 and 100 viral genome equivalents. The powerful level of sensitivity offered by PCR, paired with a rapid and simple DNA extraction procedure establishes a reproducible assay that is specific, sensitive, and has a turnaround time of less than 1 day. In addition, this method offers the significant advantage of identifying the amplified DNA sequence by its electrophoretic mobility and its restriction cleavage pattern. From an industry perspective, an inexpensive, rapid, simple method for detection of MVM in CCF allows decontamination and containment procedures to be initiated at an early stage to minimize loss. This is especially important for large-scale bioreactors. The assay we have developed can be used as a routine screening tool for the early detection of MVM contamination in culture supernatants.
References
Figure 4. Amplification of DNA extracted from other CCF spiked with MVM virus: Shown are the results of the PCR on DNA extracted from serial dilutions of the MVM virus using clarified cell fluids from confluent. Mus dunni (mouse fibroblasts) cells (a) or 324 K (human embryo kidney) cells (b) as the diluent. Lane (1) H2O: extraction and PCR negative control; (2) NEG: CCF; (3) NEG + PC: CCF + 100 copies of pML-N DNA spike; (4) 10 − 7 dilution MVM virus; (5) 10 − 8 dilution; (6) 10 − 9 dilution; (7) 100 bp ladder. The 10 − 9 dilution is equivalent to 1 TCID50/ml virus concentration (0⋅225 TCID50 total tested).
Conclusion In this study, we present an in vitro assay that can specifically detect MVM contamination in conditioned cell fluids. A high sensitivity of detection was achieved by careful choice of primers and optimization of PCR parameters. Minor changes in
1. Garnick RL. Experience with Viral Contamination in Cell Culture. In: Brown F, Lubiniecki AS (eds). Viral Safety and Evaluation of Viral Clearance from Biopharmaceutical Products. Dev Biol Stand 88: 49–57. 2. Berns KI. Parvoviridae and their replication. In: Fields BN, Knipe DM, Howley PM (eds). Virology. Lippincott-Raven Press, Ltd. 1996: 2173–2198. 3. Ward DC, Tattersall PJ. Minute virus of mice. In: The Mouse in Biomedical Research, Vol. II. Academic Press. 1982: 313–334. 4. Zhou C, Yujun Y, Jong AY. Mini-Prep in ten minutes. Biotechniques 1990; 8: 172–173. 5. Newton CR, Graham A, Hepinstall LE, Powell SJ, Summers C, Kalsheker N, Smith JC, Markham AF. Analysis of any point mutatioin in DNA; the amplication refractory mutation systems (ARMS). Hot start. Nucleic Acids Res 1989, 17: 2503–2519. 6. Astell CF, Thomson M, Merchlinsky M, Ward DC. The complete DNA sequence of minute virus of mice, an autonomous parvovirus. Nucleic Acids Res 1983; 11: 999–1018.
Received for publication 3 March 1997; accepted 14 July 1997
841664—BIOL 25/4 (ISSUE)————MS 0108
Biologicals (1997) 25, 415–419
A Rapid and Simple Procedure to Detect the Presence of MVM in Conditioned Cell Fluids or Culture Media Audrey Chang,* Steve Havas, Flavia Borellini, Jeffrey M. Ostrove and Robert E. Bird Scientific Development, MA BioServices, Inc., 9900 Blackwell Road, Rockville MD 20850, U.S.A.
Abstract. During the manufacture of biopharmaceuticals, numerous adventitious agents have been detected in Master Cell Banks, end-of-production cells as well as bulk harvest fluids. Recently, a number of large-scale production bioreactors have become infected with Minute Virus of Mice (MVM) during cGMP (current good manufacturing practices) operations, and this has resulted in both the loss of product and the need for major cleaning validation procedures to be put in place. We have developed a simple DNA extraction/PCR assay to detect the presence of MVM in cell culture supernatant (conditioned cell fluids). This highly specific assay can detect 10 or fewer genome equivalents (copies) of MVM following PCR and gel electrophoresis visualization. For routine high-throughput detection, 30–100 copies could be consistently detected. The extraction procedure was shown to reliably detect MVM at a concentration of 1 TCID50/ml. The combination of the extraction/PCR procedures establishes a powerful, sensitive, specific assay that can detect the presence of MVM sequences with a 1-day turnaround time. = 1997 The International Association of Biological Standardization
Introduction The safety testing of cell substrates for the presence of viruses in biotechnology products is an important issue and is required by regulatory agencies worldwide. At the FDA Viral Safety meeting held in June 1995, the presence of Minute Virus of Mice (MVM) in Chinese Hamster Ovary cell cultures grown in large-scale fermenters was brought to the attention of the biotechnology industry.1 MVM belongs to the Parvoviridae family, a group of small, non-enveloped virus containing single stranded DNA, whose members can infect a wide variety of hosts ranging from vertebrates (including man) to insects. The natural host for the MVM is the house mouse (Mus musculus) and the highly contagious virus is prevalent in both laboratory and wild mice populations. Although MVM infection in the adult mouse is usually asymptomatic, infection in neonatal mice and hamsters can result in deformity and death. Numerous studies have shown MVM’s involvement with biological processes including immunosuppression, cellular differentiation, and cellular proliferation. From a safety and regulatory *To whom correspondence should be addressed: Audrey Chang, MA BioServices, 9900 Blackwell Rd, Rockville MD USA 20850 1045–1056/97/040415 + 05 $25.00/0/bg970110
standpoint, it is important to insure that biopharmaceuticals produced are free from contaminating viruses such as MVM. The family is divided into two genus. MVM belongs to the genus Parvovirus, often referred to as ‘‘autonomous’’ parvovirus to distinguish it from the genus Dependovirus (e.g. adeno-associated virus, AAV) which requires a co-infection with a helper virus in order to complete its replication cycle.2 The DNA replication of the viruses from both genera occurs in the nuclei of dividing infected cells. Parvovirus viral replication involves the cellular replicative functions and requires that the infected cell pass through S phase. It is this close relationship between cell division and MVM viral replication that predisposes rapidly dividing murine cells, commonly grown in bioreactors, to MVM virus contamination. Several in vitro laboratory assays can detect the presence of MVM.3 Autonomous parvovirus has been shown to agglutinate erythrocytes and a standard haemagglutination assay (HA) using guinea-pig erythrocytes has been used since the early 1970s. More recently, infectivity assays measuring the cytopathic effect (CPE) of MVM on susceptible cell culture monolayers has been used for diagnostic purposes. In addition to the HA and CPE assays, the detection of 7 1997 The International Association of Biological Standardization
D.R.
841664—BIOL 25/4 (ISSUE)————MS 0108
416
A. Chang et al .
infected cells by fluorescent antibody staining (FA) and plaque assays have been developed to detect the presence of MVM. All these assays have been instrumental in determining strain and serotype differences for MVM. However, from an industry perspective, issues of duration (for instance, the CPE assay requires 13 days), complexity, and low sensitivity make these assays impractical for routine screening. Determining the presence of an MVM-infection in a bioreactor of several thousand litres during a production run requires an assay with a rapid turnaround time. Polymerase chain reaction (PCR)-based assays can detect the presence of MVM viral DNA sequences specifically and rapidly. Furthermore, PCR is the most sensitive method of detection presently available. It represents an increase of 100-fold in sensitivity over the most sensitive plaque assay.1 A rapid and straightforward DNA sample preparation procedure is equally important in the development of a quick and reliable MVM PCR assay. We present a MVM-PCR assay which includes a ‘‘mini-prep’’ sample preparation from conditioned cell fluids, a PCR amplification specific for MVM sequences, and a diagnostic restriction endonuclease digest that confirms that the PCR product amplified originated from MVM sequences.
Materials and methods Reagents All chemicals and reagents were of the highest molecular biology grade commercially available. Stock solutions wre made in nuclease free sterile water (Baxter, Deerfield, IL). Oligonucleotides were purchased from Gibco/BRL, Gaithersburg, MI). The plasmid pML-N (7660 bp) containing the complete MVM genome (0–5149 nt) was a generous gift from Dr David Pintel of the University of Missouri. The plasmid DNA was used to transform Escherichia coli DH5a cells and the transformed cells were screened for the presence of pML-N. A plasmid DNA preparation of pML-N was prepared and quantitated by Lofstrand Labs Limited, Gaithersburg, MD. The plasmid copy number was calculated using the molecular weight of the plasmid and Avogadro’s constant. One microgram of the plasmid pML-N preparation is approximately equivalent to 1⋅2 × 1011 copies. Concentrated plasmid DNA stock was stored at −20°C and stocks of the bacterial culture in 50% glycerol were stored at −80°C.
Viruses and cells The Crawford strain of MVM (ATCC VR-663) (lot (022096P) with a titre of 1⋅3 × 109 TCID50/ml in 324 K cells was used in the studies. Chinese Hamster Ovary (CHO) cells were grown to confluence in HAM F-12 media, supplemented with 10% fetal bovine serum (FBS), 0⋅1% gentamycin and 0⋅1% fungizone. 324 K and Mus dunni were grown to confluency in Dulbecco’s modified Eagle’s medium (DMEM) media, supplemented with 5% FBS and 1% non-essential amino acids. All cell lines and MVM virus were grown and maintained at MA BioServices. Sample preparation Conditioned cell fluids (CCF) Cell culture supernatants from either CHO, 324 K, or Mus dunni cells were harvested after a minimum of 72 h in culture, and clarified with a low speed centrifugation (1000 g for 10 min). MVM virus The MVM viral stock was freshly diluted in the clarified CCF medium before each experiment. Extraction Different dilutions of the MVM virus in clarified CCF were extracted along with clarified cell fluid (NEG), and dH2O (H2O) as negative controls. Additional CCF samples were extracted and used as the background for positive control plasmid spikes in the subsequent PCR amplification step (NEG + PC). The miniprep extraction of MVM DNA from conditioned cell fluids was adapted from a plasmid miniprep procedure4 with several modifications. Briefly, 225 ml of each sample was placed in a 1⋅5-ml microcentrifuge tube with 150 ml of 2 × TENS solution (20 mm Tris–HCl pH 8⋅0, 0⋅2 mm EDTA pH 8⋅0, 0⋅2 m NaOH, 1% sodium dodecyl sulfate), 40 ml of 3 m NaOAc pH 8⋅2, and 2 ml of mussel glycogen (20 mg/ml). The tubes were vortexed for 2–5 s to mix. Cold absolute ethanol (1 ml) was added, and the tubes were briefly vortexed and centrifuged in an Eppendorf 541C5 at 14 000 rpm for 5 min at room temperature. The ethanol was gently decanted and the pellets were rinsed with 1 ml of room temperature 70% ethanol. The pellets were air-dried for 5 min at room temperature and resuspended in 20 ml of sterile nuclease-free water. The NEG + PC samples were resuspended in 10 ml to accommodate for the positive control plasmid spike volume of 10 ml. The entire resuspension volume was used in the PCR amplification step.
D.R.
841664—BIOL 25/4 (ISSUE)————MS 0108
A. Chang et al .
PCR Amplification A synthetic oligonucleotide [MVM-1391: 5'-d(TTGCCA-TGC-TAT-TTG-CTG-TG)-3'] that anneals to the MVM nucleotide 1391–1411 sequence (accession ( J02275) serves as the forward primer in the PCR amplification. Oligonucleotide MVM-1716 [5'd(TTG-TGG-TCA-TGA-TGA-CTG-GT)-3'] that anneals to the MVM nucleotide 1697–1716 was the reverse primer. A 325-bp region from the non-structural gene of MVM (nt 1391 to nt 1716) was amplified using primers MVM1391 and MVM 1716. PCR conditions were optimized with regard to primer, dNTPs and Mg2 + concentrations, and cycling conditions in a Perkin Elmer 2400 thermal cycler. Amplification was performed using the hot start technique5 in the presence of 50 mm KCl, 10 mm Tris–HCl pH 8⋅3, 0⋅2 mm dNTPs, 1⋅75 mm MgCl2, 0⋅2 mm of each primer, and 2⋅5 U Taq polymerase in a total volume of 100 ml. Cycling conditions were: 94°C for 3 min (1 cycle); 94°C for 40 s, 56°C for 40 s, 72°C for 40 s (40 cycles), 72°C for 2 min (1 cycle). Serial dilutions of the positive control plasmid, pML-N, were freshly prepared and mixed with NEG + PC extraction samples to serve as positive controls. PCR products wre analysed by electrophoresis on a 2⋅5% MetaPhor (FMC) agarose gel containing 0⋅25 mg/ml ethidium bromide. The products of amplification were visualized with a UV transilluminator.
Results Sensitivity of PCR amplification reaction Primers were chosen from the MVM non-structural coding region to specifically amplify a 325-bp region. Dilutions of the pML-N plasmid ranging from 1000 copies to 1 copy were used to evaluate the sensitivity of the PCR assay in viral genome equivalents (copy number). As Figure 1(a) illustrates, amplification of pML-N produced a single band of 325 bp, clearly visualized by ethidium bromide staining in the lanes containing 1000, 100, 30, and 10 copies pML-N. The 325-bp band was not detectable at the 3, 1 and 0⋅1 copy level. To investigate the reproducibillity of the PCR assay, the same panel of template copy levels was tested with multiple operators on different dates. The results are presented in Figure 1(b) and show that using these amplification conditions, 30 to 100 copies of the MVM template are routinely detected.
417
Sensitivity of PCR Amplification for detection of MVM viral particles in conditioned cell fluids (CCF) A ‘‘ten-minute’’ plasmid mini-prep DNA isolation procedure4 was modified and adapted to extract MVM viral particle DNA from conditioned cell fluids. Conditioned cell fluids were harvested from Chinese Hamster Ovary cells after 72 h in culture, and clarified with a low-speed centrifugation. Serial 10-fold dilutions of purified MVM particles (having a titre of 1⋅3 × 109 TCID50/ml) were prepared using clarified Chinese Hamster Ovary CCF as the diluent and extracted as described in Materials and methods and PCR amplified. A ‘‘water only’’ tube was included as a negative control for the extraction and PCR procedure to ensure no contaminating MVM DNA was introduced during the sample manipulations. As Figure 2(a) illustrates, no PCR products are generated in the ‘‘water only’’ or the CHO CCF extracted material (lanes 1 and 2). The absence of the MVM-specific band in the CHO CCF was not due to inhibition of the PCR since the addition of 100 copies of the positive control plasmid to the CHO CCF (lane 3) produced a 325-bp band with an intensity similar to the band produced by the 100 copies of pML-N in the absence of CHO CCF (a)
1
2
3
4
5
6
7
8
(b) Copy No.
100
30
10
3
1
0
Positives/ total runs
6/6
6/6
5/6
2/6
1/6
0/6
Figure 1. Amplification of MVM sequences with the positive control pML-N template. (a) Serial dilutions of pML-N ranging from 0⋅1 to 103 copies/ml were freshly prepared from concentrated stocks and tested with using optimized MVM PCR conditions. The 325-bp PCR product is indicated by the arrow. Lane (1) 1000 copies; (2) 100 copies; (3) 30 copies; (4) 10 copies; (5) 3 copies; (6) 1 copy; (7) no DNA negative control; (8) 123 bp ladder. (b) Summary of several repeat experiments performed using 0, 1, 3, 10, 30 and 100 template copies/reaction.
D.R.
841664—BIOL 25/4 (ISSUE)————MS 0108
A. Chang et al .
418
(a)
1
2
3
4
5
6
7
8
MVM-derived PCR product with RsaI generated a restriction pattern consisting of two fragments of 171 bp and 154 bp. As shown in Figure 3, the RsaI restriction pattern generated by both the pML-N plasmid and the viral template was consistent with the MVM Crawford strain, confirming that the 325-bp PCR product originated from MVM sequences. No other non-specific bands were generated.
(b) Concentration TCID50/ml
100
10
1
Positive/ total runs
6/6
6/6
3/4
Figure 2. Amplification of DNA extracted from CHO CCF spiked with MVM virus. Shown are the results of the PCR on DNA extracted from serial dilutions of the MVM virus using uninfected CHO CCF as the diluent. Lane (1) H2O: extraction and PCR negative control; (2) NEG: CHO CCF; (3) NEG + PC: CHO CCF + 100 copies of pML-N DNA spike; (4) 10 − 7 dilution MVM virus; (5) 10 − 8 dilution; (6) 10 − 9 dilution; (7) pML-N, 100 copies; (8) 123 bp ladder. The 10 − 9 dilution is equivalent to 1 TCID50/ml virus concentration (0⋅225 TCID50 total tested). (b) Summary of several repeat experiments performed using 100, 10, and 1 TCID50/ml per reaction.
Extraction of MVM DNA from other CCFs In addition to CHO CCF, we tested the MVM DNA extraction/PCR amplification procedure using CCF from two additional cell lines: Mus dunni (normal mouse fibroblasts) and 324 K (human embryo kidney cells). MVM virus particles were diluted in each CCF using the same dilution as for CHO CCF and DNA was extracted as described above. The results of the PCR amplification are shown in Figure 4(a) and 4(b). Each of the CCF tested negative for the 325-bp MVM PCR product. However, in the presence of the MVM virus, the 325-bp MVM PCR product was generated. This demonstrates that the assay is not cell-line specific and that various cell lines may be tested successfully using this procedure. 1
2
3
4
5
6
(lane 7). When the various MVM dilutions were added to the CCF for DNA extraction and PCR analysis, the 325-bp band was obtained at the 10 − 7, 10 − 8 and 10 − 9 dilutions (lanes 4, 5, and 6). No other non-specific bands were seen. In terms of viral titre endpoint, the 10 − 9 dilution corresponds to one TCID50/ml MVM virus. The same virus dilutions were tested with multiple operators on different dates. As shown in Figure 2(b), 10 to 1 TCID50/ml MVM virus concentration could be routinely detected by this methodology. Restriction analysis of PCR products To confirm PCR specificity, amplified products produced from both the plasmid template and the virus DNA were digested with the restriction endonuclease RsaI and analysed by high-resolution agarose gel electrophoresis. As predicted from the reported MVM sequences,6 digestion of the 325-bp
Figure 3. Diagnostic restriction endonuclease digestion: The 325-bp PCR products generated for Figure 2 were digested with 10 units of the restriction endonuclease RsaI (Lanes 2–5). Lane (1) 100 copies pML-N undigested; (2) 100 copies pML-N digested with RsaI; (3) MVM virus dilution 10 − 7 + RsaI; (4) 10 − 8 + RsaI; (5) 10 − 9 + RsaI; (6) 50 bp ladder. The large arrow indicates the 325-bp undigested band. The two small arrow heads indicate the position of the 171 bp and 154 bp RsaI digestion products.
D.R.
841664—BIOL 25/4 (ISSUE)————MS 0108
A. Chang et al .
(a)
1
2
3
4
5
6
7
(b)
1
2
3
4
5
6
7
419
the PCR protocol often resulted in 10- to 100-fold differences in assay sensitivity. The limit of detection for the optimized PCR reaction when using purified DNA as a template was shown to be as low as 10 viral genome equivalents. On a routine basis, the assay could detect between 30 and 100 viral genome equivalents. The powerful level of sensitivity offered by PCR, paired with a rapid and simple DNA extraction procedure establishes a reproducible assay that is specific, sensitive, and has a turnaround time of less than 1 day. In addition, this method offers the significant advantage of identifying the amplified DNA sequence by its electrophoretic mobility and its restriction cleavage pattern. From an industry perspective, an inexpensive, rapid, simple method for detection of MVM in CCF allows decontamination and containment procedures to be initiated at an early stage to minimize loss. This is especially important for large-scale bioreactors. The assay we have developed can be used as a routine screening tool for the early detection of MVM contamination in culture supernatants.
References
Figure 4. Amplification of DNA extracted from other CCF spiked with MVM virus: Shown are the results of the PCR on DNA extracted from serial dilutions of the MVM virus using clarified cell fluids from confluent. Mus dunni (mouse fibroblasts) cells (a) or 324 K (human embryo kidney) cells (b) as the diluent. Lane (1) H2O: extraction and PCR negative control; (2) NEG: CCF; (3) NEG + PC: CCF + 100 copies of pML-N DNA spike; (4) 10 − 7 dilution MVM virus; (5) 10 − 8 dilution; (6) 10 − 9 dilution; (7) 100 bp ladder. The 10 − 9 dilution is equivalent to 1 TCID50/ml virus concentration (0⋅225 TCID50 total tested).
Conclusion In this study, we present an in vitro assay that can specifically detect MVM contamination in conditioned cell fluids. A high sensitivity of detection was achieved by careful choice of primers and optimization of PCR parameters. Minor changes in
1. Garnick RL. Experience with Viral Contamination in Cell Culture. In: Brown F, Lubiniecki AS (eds). Viral Safety and Evaluation of Viral Clearance from Biopharmaceutical Products. Dev Biol Stand 88: 49–57. 2. Berns KI. Parvoviridae and their replication. In: Fields BN, Knipe DM, Howley PM (eds). Virology. Lippincott-Raven Press, Ltd. 1996: 2173–2198. 3. Ward DC, Tattersall PJ. Minute virus of mice. In: The Mouse in Biomedical Research, Vol. II. Academic Press. 1982: 313–334. 4. Zhou C, Yujun Y, Jong AY. Mini-Prep in ten minutes. Biotechniques 1990; 8: 172–173. 5. Newton CR, Graham A, Hepinstall LE, Powell SJ, Summers C, Kalsheker N, Smith JC, Markham AF. Analysis of any point mutatioin in DNA; the amplication refractory mutation systems (ARMS). Hot start. Nucleic Acids Res 1989, 17: 2503–2519. 6. Astell CF, Thomson M, Merchlinsky M, Ward DC. The complete DNA sequence of minute virus of mice, an autonomous parvovirus. Nucleic Acids Res 1983; 11: 999–1018.
Received for publication 3 March 1997; accepted 14 July 1997