Detección de siete virus y de Mycoplasma en suero fetal bovino por PCR en tiempo real

Detecção de sete vírus e de Mycoplasma em soro fetal bovino por PCR em tempo real.

Claudia P Cordero Camacho¹, QF, DrSc; Linamaría Escobar Mármol¹, Biol, MSc (C); Edward F Carrillo Borda¹, Biol, DrSc; Sandra J Morantes Medina¹, Biol, DrSc (C); Fabio A Aristizábal Gutiérrez¹*, QF, DrSc.

¹ Grupo de Farmacogenética del Cáncer, Departamento de Farmacia, Facultad de Ciencias, Universidad Nacional de Colombia, sede Bogotá.

(Recibido: 2 febrero,2011; aceptado: 23 mayo, 2011)

Summary

Objective: real  time  PCR  analysis for the  detection  of seven bovine  pathogenic viruses: Bovine Adenovirus  (BAdV), Bovine  Viral Diarrhea  Virus  Types  1  and  2  (BVDV-1  and  BVDV-2),  Bovine Respiratory Syncytial Virus  (BRSV),  Vesicular Stomatitis Virus (VSV), Bovine  Parainfluenza  Virus  3 (BPIV-3), Bovine Herpes Virus-1 (BoHV-1), and Mycoplasma was conducted using fetal bovine serum (FBS, MICROGEN^®) obtained in Colombia,  aiming  to  include it as part of the serum  quality control. Methods: bovine derived MDBK and human derived HEp-2 cell lines were cultured with the test serum for 21 days, collecting supernatant and cellular samples every 7-days. Once DNA and RNA were extracted, the later was converted into cDNA and both samples were subjected to real time PCR using specific primers and Resolight^® (DNA-binding fluorescent dye). Standard curves were generated using serial dilutions of cloned  specific  viral sequences. Accurate amplification  and  high  efficiency was demonstrated  in these reactions. Results: realtime PCR amplification did not show a persistent increase of viral counts in cultures during the 21-day follow-up. However, for vesicular stomatitis virus, a transient increase was observed at 7 and 14 days in both cell lines, but considered as not conclusive for viral presence. Conclusions: real time PCR analysis showed to be a suitable method for viral detection in fetal bovine serum samples and through this method no consistent viral or mycoplasma presence was detected in the MICROGEN^® fetal bovine serum.

Objetivos:  se  empleó  el método  de PCR  en tiempo  real  para  detectar los virus  patógenos bovinos: Adenovirus  bovino  (BAdV), virus  de la  diarrea  Viral Bovina  tipos  1  y  2  (BVDV-1  y  BVDV-2), Virus Respiratorio Sincitial Bovino (BRSV), Virus de la Estomatitis Vesicular (VSV), Virus de la Parainfluenza Bovina tipo  3  (BPIV-3),  Herpesvirus Bovino-1  (BoHV-1)  y Mycoplasma, en suero  fetal bovino  (FBS, MICROGEN ®)obtenido en Colombia, con el objetivo de incluir estos análisis en el control de calidad del FBS. Metodos: las líneas celulares MDBK de origen  bovino  y HEp-2 de origen humano se cultivaron con el FBS MICROGEN^® por 21 días, tomando muestras de cultivos y sobrenadantes cada 7 días. Una ves extraido el DNA y RNA, a partir de este último se sintetizó cDNA, y en los dos tipos de muestras se analizó la presencia de los agentes patógenos mencionados por PCR en tiempo real empleando iniciadores específicos para cada  uno y Resolight^® (colorante  fluorescente  de unión a DNA). Se generaron curvas estándar con diluciones seriadas de secuencias virales específicas clonadas en plásmidos, que mostraron amplificación específica y altas eficiencias. Resultados: el análisis de los cultivos mantenidos con el FBS en estudio no mostró aumento del número de copias virales detectadas a lo largo del periodo de 21 días de seguimiento, excepto para el virus de la estomatitis vesicular, que mostró un incremento transitorio en los sobrenadantes de los cultivos de las dos líneas celulares a los 7 y 14 días de cultivo, que no se consideró concluyente para la presencia del virus. Conclusiones: el método de PCR en tiempo real mostró utilidad para la detección de virus patógenos y mycoplasma en FBS, y mediante este método  no  se  obtuvieron resultados  que permitan  concluir que  los  patógenos  virales  o  los  mycoplasmas  están  presentes  en  los cultivos mantenidos con el suero fetal bovino MICROGEN^®.

Palabras clave: detección viral, PCR en tiempo real, suero fetal bovino, virus bovinos.

Resumo

Objetivo: Foi utilizado o método de PCR em tempo real para detectar os vírus patogénicos bovinos: Adenovírus Bovino (BAdV), Vírus da Diarreia Viral (BVDV-1 e BVDV-2), Vírus Respiratório Sincicial Bovino (BRSV  ), Vírus  da  Estomatite  Vesicular (VSV),  Vírus  da  Parainfluenza  Bovina  tipo  3  (BPIV- 3),  Herpesvírus  Bovino-1  (BoHV-1) e Mycoplasma, em  soro  fetal  bovino  (FBS,  Microgen^®) obtido  na Colômbia,  de modo a incluir esta  análise no controle de qualidade FBS.  Métodos: as linhas celulares de MDBK de origem bovina e HEp-2 de origem humana foram cultivadas com FBS Microgen^® durante 21 dias, tomando amostras de cultura e sobrenadantes cada 7 dias. Uma vez retirado o DNA e RNA, foi sintetizado  o  cDNA  a  partir do  RNA.  Nos dois  tipos  de amostras  foram  analisadas  para  determinar a presença de patógenos mencionados por PCR em tempo real usando primers específicos para cada um e Resolight^®(corante fluorescente de união à DNA). As curvas padrão foram geradas com diluições em série  de sequências virais específicas, clonadas em  Resultados: plasmídeos, que mostram  amplificação específica e altas eficiências.  a análise das culturas mantidas em FBS em estudo, não mostraram aumento no número de cópias virais detectadas ao longo do período de 21 dias de seguimento, exceto para o vírus da  estomatite  vesicular,  que  mostrou  um  aumento  transitório  nos  sobrenadantes  das  culturas de  duas linhas celulares aos 7 e 14 dias de cultura, que não foi considerado conclusivo para a presença do vírus. Conclusões: O método de PCR em tempo real foi útil para a detecção de vírus patogênicos e mycoplasma em FBS, e por este método foram obtidos resultados que demonstram que patógenos virais ou mycoplasmas estão presentes nas culturas mantidas com soro fetal bovino Microgen^®.

Palavras chave: soro fetal bovino, vírus bovinos.

¤  To cite this article: Cordero CP, Escobar L, Carrillo EF, Morantes SJ, Aristizábal FA. Detection of seven viruses and Mycoplasma in fetal bovine serum by real time PCR. Rev Colomb Cienc Pecu 2011; 24:585-597

*  Corresponding author: Fabio Aristizábal Gutiérrez. Departamento de Farmacia, Universidad Nacional de Colombia, sede Bogotá. Ciudad Universitaria. Cra 30 # 45-03.Edificio 450. Oficina 206. Bogotá D.C. Colombia, Suramérica. E-mail:cpcorderoc@unal.edu.co, faaristizabalg@bt.unal.edu.co

Introduction

Fetal  bovine  serum  (FBS)  is  an  essential supplement  for in  vitro  cell  culture medium,  as  a source  of  growth  factors,  vitamins,  minerals  and hormones  that  stimulate cell  proliferation.  Major concerns  on  FBS are related  with  high  cost  and presence of bovine  viruses  and  Mycoplasma,  non desired  agents that  may alter normal  course of cell cultures.  An  alternative to  this  potential  problem has  been  the  use of  serum-free media,  however, associated  high  costs  are a major drawback (Freshney, 2000).

In  Colombia  the  FBS requirements  have  been satisfied  by  international  suppliers  as  GIBCO-BRL-Invitrogen^®,  SIGMA^®,  Eurobio^®,  Lonza and  Hyclone^®,  which  accounts  for a  high  price of  the product and  increases  culture maintenance expenses.  As  an  alternative  to  foreign  produced and  imported  fetal  bovine serum the  Colombian company MICROGEN^® produces a Colombian fetal bovine serum and offers this product to the internal market and the national scientific community.

In  addition  to  sterility  and  basic biochemical quality  controls, MICROGEN^®   FBS should  be evaluated  for  culture performance and  virus absence,  in  order to  ascertain  its  suitability  for use in cell cultures.

Viral  detection  in  bovine  serum was  based on  antigen  detection  through  ELISA  and  other immunological  methods;  within  the last  15  years, RNA and DNA detection methods by real time RT- PCR have been proved to be more sensitive as well as  highly  specific  (APHIS-USDA, 1995;  Kosinova et al., 2007; Timsit et al., 2010).

The present work was intended to establish a real time RT-PCR analytical method for the detection of bovine viruses and mycoplasma and its  application in  the analysis  of commercial  batches  of the MICROGEN^® FBS.

Materials and methods

Virus selection

Mycoplasma were included as those intracellular parasites  are potential  contaminants  of animal derived  products  and  some  of them can  be present in cattle's blood (Freshney, 2000; Nishizawa, 2010).

Selection of primers for real time PCR

Specific  primers  for real  time  PCR-detection were selected based on previous reports of real time analysis  for the target  viruses  and  basic primer- design requirements: selection of conserved regions as target sequences, melting temperature (Tm) close to 60 °C and PCR product length between 100 and 250  bp. Primer3  software (Rozen  and  Skaletsky, 2000) and the PrimerQuest tool included in the IDT- SciTools (IDT, 2009), were used for primers design and  selection.  Primers  were synthesized  by  IDT (IDT, Coralville, IA).

Construction of positive controls

Two  oligonucleotides  containing  the  target sequences  of  the  primers  were  designed,  synthetized and  ligated  to  suitable  vectors.  Oligonucleotide A  (OligoA)  containing  target  sequences  for VSV, BoHV-1,  BVDV-I  and  BPIV-3  primers,  was  cloned in  the  pGEM^®-T  Easy  cloning  vector (Promega, Madison, WI). Oligonucleotide B (OligoB), containing target  sequences  for  BAdV, BRSV,  BVDV-2  and mycoplasma  primers  was  cloned  in  the  pCR^®   2.1- TOPO^®  cloning  vector  (Invitrogen,  Carlsbad,CA). Cloning  procedure  was  conducted  following manufacturer's instructions.   Both recombinant constructs  were  transformed  in  DH-5α Escherichia coli  competent  cells,  purified  with  the  Mini-prep Wizard  (Promega,  Madison,  WI)  and  quantified  by Qubit^®(Invitrogen, Carlsbad,CA).

Generation of standard curves

A  standard  curve was  constructed  for each  one of the studied bovine viruses and mycoplasma using eight  serial  10-fold  dilutions  of cloned  OligoA  or OligoB. HRM Master, containing the DNA binding dye  ResoLight^® (Roche,  Mannheim,  Germany) was  used  for real  time  detection  with  excitation wavelength  465  nm and  emission  wavelength 510  nm.  Reactions  were carried  out in  10  µL  final volume, in the LightCycler 480  Instrument II using multiwell 96 plates^® (Roche, Mannheim, Germany).

Real  time  PCR reactions  included  1x  HRM Master Mix,  2.5  mM MgCl₂,  0.2  µM each specific primer,  2.0  µL  of quantified  cloned  control oligonucleotide  and  ultra pure  PCR-grade water  to a final volume of 10 µL. Each reaction was carried out  in  triplicate. The real  time  PCR amplification program included: pre-incubation 95 °C for 10 min, 45  amplification  cycles:  95°C  for  10  s,  60°C  for 15 s, 72°C for 10 s, followed by one melting cycle: 95°C for 1 min, 40 °C for 1 min, 60 °C for 1 s, 95°C and cooling to 40 °C.

Amplification  curves,  Ct  values,  slope values of  the curves  and  derivative melting  curves were  obtained  with  the LightCycler 480^® II data analysis  software  (Roche, Mannheim,  Germany). Efficiency  percentage  for  each  standard  curve  was determined by the equation E= (10  ^(-1/m) – 1) x 100, where m  is  the  slope  value  of  the curve.  Ct  value for  each  reaction,  the amplification  cycle at  which fluorescence crosses  the  threshold  line,  was  used as  the  measurement  parameter  and  plotted  against plasmid copy number in the Microsoft Office Excel application (Microsoft, Redmond, WA).

Cultures  of cell  lines  sensitive  to infection  by  the  studied  viruses  were maintained with  reference or test  fetal  bovine  serum. MDBK (bovine kidney  derived),  and  HEp-2  (human  larynx cancer derived)  cell  lines  (ATCC,  Manassas,  VA) were maintained in Eagle's MEM medium (SIGMA, St  Louis, MO),  supplemented  with  penicillin100 U.mL^-1,  streptomycin  100  µg. mL^-1  (Gibco-BRL  , Carlsbad,CA) and 10% reference fetal bovine serum Certified-FBS  (Gibco-BRL^®,  Carlsbad,CA)  or test FBS  MICROGEN^®   (Microgen, Bogotá,  Colombia). Cultures  were maintained  in  75  cm² culture  flasks (TPP,  Trasadingen,  Switzerland) in  standard  culture conditions  37  °C, 5  % CO₂  in  air, 100% humidity. Media was changed once a week.

Cell  cultures  were maintained  with  reference FBS for 7 days or with test FBS MICROGEN^® for 14  to  21  days.  Cellular and  supernatant  samples were  taken  from each  culture when  subcultured  at 7, 14 and 21 days of maintenance and processed for total RNA and DNA extraction.

This  cell  culture  procedure was  based  on  the procedure  used  for  viral  testing  by  Gibco-BRL^® accoding  to  the  Requirements  for ingredients  of animal origin (APHIS-USDA,1995).

RNA  and  DNA  extraction.  Total  RNA  and DNA  were  extracted  from culture samples  using the  phenol-guanidine  isothiocyanate-chloroform method.  1  x  10⁵  cells  (solid  samples)  were mixed with 1 mL of TRIzol^® reagent (Invitrogen, Carlsbad, CA)  and  1  mL  of supernatant  (liquid  samples) was  mixed  with  3  mL  of  TRIzol^®   LS reagent (Invitrogen,  Carlsbad,  CA).  Extraction  was  made according  to  the  manufacturer's  instructions.  Total RNA  was  resuspended  in  20  µL  of  nuclease-free DEPC-treated  water;  DNA  was  resuspended  in 8  mM NaOH  solution  added  with HEPES  and EDTA  solutions  (Invitrogen, Carlsbad, CA).  RNA and  DNA  samples  were  quantified  with  the  Qubit fluorometric method (Invitrogen, Carlsbad, CA).

Reverse  transcription.  RNA  was  converted into  cDNA  using  the SuperScript  III First-strand system (Invitrogen,  Carlsbad,  CA),  following the manufacturer's instructions. Each   RT reaction  included:  1X  RT  buffer,  5  mM  DTT, RNAse  inhibitor 40  U,  dNTPs  0.5  mM each, random  hexamers  50  ng,  SuperScript  III reverse transcriptase  200  U,  2.5  µg  total  RNA  and  DEPC- treated water for a final volume of 20 µL. Reactions were incubated at 25 °C for 5 min, 50 °C for 60 min and 70 °C for 15 min. RT products were diluted 1:2 with DEPC-treated water. A conventional PCR with -actin  specific  primers  was  made to  verify  cDNA synthesis, using the RT reaction product as template.

Similar PCR reactions were made using genomic DNA  as  template to  verify  the  suitability  of these samples  as  PCR substrates.  Each  conventional PCR reaction contained 1X Taq  polymerase buffer, 2mM MgCl₂, 0.2 mM each dNTPs, 0.1 µM forward primer (GGCACCCAGCACAATGAA  GATCAA), 0.1  µM reverse primer  (ACTCGTCATACTCC TGCTTGCTGA),  1µL  RT  reaction  product or 100  ng  genomic  DNA,  0.5  U  Go-Taq  polymerase (Promega,  Madison,  WI),  and  PCR-grade  water to  25  µL  final  volume.  Amplification  reactions were  performed  in  the My  Cycler^®   thermal cycler (BIORAD,  Hercules, CA);  amplification  program started with a denaturation step of 94 °C for 2 min, followed by 40 cycles of 94 °C for 30 s, 60 °C for 15 s, 72 °C for 10 s. PCR products were evaluated by  electrophoresis  in  2%  agarose  gels,  stained with  SYBR Safe^®   (Invitrogen,  Carlsbad, CA) and visualized in a Digimage^® (Major Science, Saratoga, CA)  gel  documentation  system.  The expected sizes for the PCR products were 121 bp for bovine derived cells and 133 bp for human derived cells.

Real time PCR Assay

Real  time  PCR detection  of the  seven  viruses and  mycoplasma were performed  using  cDNA as  template, BAdV,  BoHV-1  and   Mycoplasma detection  was  made also  in  DNA. Real  time reactions  were  set  up  using  the  same conditions described  for the  construction  of standard  curves, using 2 µL of diluted cDNA product as template or 10 ng of total DNA. In negative controls nuclease- free water was added as template instead of sample. In  positive  controls  dilutions  of the  correspondent control oligo were added as template. Each sample was evaluated in duplicate. For each reaction the Ct value was  determined  and Tm for the  product  was also observed.

To  avoid  cross  contamination  all  technical procedures  were held  under controlled  conditions, under laminar flow  hoods  and  carried  out  in laboratories  where  no  other viruses  were  under investigation.

Linear  regression  for each  standard  curve  was obtained and the concentration of each sample was automatically  calculated  based  on  linear equation, using the LightCycler 480^® II data analysis software (Roche, Mannheim, Germany).

Results

Selection of primers for real time PCR

The sequences of the specific primers selected for the real time PCR reactions are presented in table 1.

Standard curves

Standard curves were obtained using each specific primers pair (Table 1) and the correspondent control oligonucleotide A or B (Figures 1 and 2).

As  serial  dilutions  of  the  cloned  control oligonucleotides  were used  for  standard  curves' construction, the amount of template was expressed as  copies  number of the  recombinant  plasmid containing  the target  sequence:  ranging  from 2 to  2  x  10⁷  target  copies.  Amplification  curves  for each  set  of  serial  dilutions  showed  that  increasing C_t values  were obtained as the copy  number in  the dilution  decreased  (Figure 1).  A  linear trend  was observed  for  each  dilution  series  when  C_t  values were  plotted  against  the copy  number (Figures  2 and  3).  Melting  curves  obtained  for  the products of these reactions showed a unique peak indicating the absence  of  primer-dimmers  and  non-specific amplification (Melting curves not shown).

In  the  standard  curves  obtained  with  control oligoA corresponding to VSV, BVDV-1, BoHV-1 and BPIV-3, the linear pattern was lost, or no amplification was  detected,  for  the lower dilution  (two  copies), therefore 20  copies  dilution  was  considered  as  the lower detection limit for these three viruses (Figure 2).

Small standard deviation values were obtained in the four data sets and regression coefficient (R²) wasabove 0.98, indicating high adjustment to the linear model.

In  the standard  curves  constructed  with  oligoB the linear pattern  was  conserved  for  the  20  copies dilution  but  not  for two  copies  dilution, which showed lower Ct  values than the ones obtained for the 20 copies dilutions.

Based  on  these results  the  detection  limit  for BAdV,  BRSV  and  mycoplasma  was  established at 20  copies.  For  BVDV-2  no  amplification  was detected  for  2  and  20  dilutions  thus  the  detection limit was  set at  200  copies.  Ct  values  decreased proportionally  to  copy  number increase,  including the highest  concentration  (2  x  10⁷  copies).  Values over 0.99  were obtained  for the regression coefficient (R²) of the four series (Figure 3).

In  addition  to  the observed  linear  pattern  in the standard  curves,  efficiency  percentage  for each  real  time PCR  data  set  was  calculated  (Table 2). Amplification  efficiencies  above 78%  were calculated for VSV, BoHV-1, BAdV, BPIV-3, BRSV, BVDV-1, BVDV-2 and mycoplasma standard curves, indicating  a  proper geometrical  increase  of the copy  number along  the PCR processes.  Maximum efficiency  was  not  obtained  for  all  reactions  as  we used  the  same  standardized  primer and  magnesium chloride  concentration  in  the  seven  specific amplification  reactions, looking  for  establishment of homogeneous  amplification  conditions  which facilitate the analysis procedure.

Detection of viruses  and  mycoplasma  in  culture samples

After conversion  of  total RNA  to  single-strand cDNA,  resultant  cDNA  and  total DNA  (DNA) samples  were  analyzed  by  conventional  PCR amplification for -actin to check for the efficiency of  the  reverse  transcription  reaction  and  for the suitability  of these  samples  to  be used  as  PCR templates.

All total DNA samples from solid samples gave a positive result for  Β-actin amplification but liquid samples,  corresponding  to  culture  supernatants  did not  show  amplification, except  for samples  L9  and L10. This  must  be related  with  the  low  amount  of nucleic acids  present  in  the culture supernatant (Figure 4).  PCR products  showed  sizes  between 100-200 bp, which correspond to the expected sizes (121  bp  for bovine derived  cells  and  133  bp  for human derived  cells).  Size differences  between  the products  were  not  determined  as  the  resolution  of agarose gels is around 10 bp.

Amplification  for Β-actin  on  unpurified  cDNA products  was  positive for  all  the tested  samples, indicating  a  successful  result  for  the RT  reactions and  the suitability  of  synthesized  cDNAs  as PCR templates.  Obtained  PCR products  were,  as expected, in the 100-200 bp size range (Figure 5).

Detection  of  bovine  viruses.  Once samples were  confirmed  to  be  useful  as  PCR  templates, viral detection was performed according to  cellular sensitivity  to  viral  infection.  cDNA  samples from MDBK  cell  line  cultures  were analyzed for  detection  of  seven  pathogenic  viruses  and Mycoplasma,  as  this  bovine  derived  cell  line  is sensitive  to  the  infection  of the  seven  analyzed viruses and also to mycoplasma. On cDNA samples from HEp-2  cell  line cultures,  PCR detection  was only conducted for VSV and BAdV viruses, as this cell line  is  not  sensitive to  infection  by  the  other viruses.  Each  set  of  samples  was  analyzed  within the same real  time PCR run,  including  negative controls  and  positive  controls  corresponding  to dilutions of the cloned control oligoA or oligoB as corresponded (Table 4).

Evaluation  for  viral  presence  in  total  DNA samples  was  conducted  only  on  solid  derived samples  and  the  liquid  derived  samples  that showed  positive amplification  for Β-actin  (L9  and L10).  BoHV-1,  BAdV  and  Mycoplasma  were  the only  three  pathogens  analyzed  in  DNA  samples as  BoHV-1,  BAdV  and  Mycoplasma  have DNA genome, while the other viruses have RNA genome.

Positive  controls  included  in  each  sample set gave positive amplification, the Ct value obtained for these standards  were used  by  the  software  to  adjust the data set to the standard curve and calculate copy number  for  each  sample.  For  some  of the negative control  reactions  a  Ct  value was  obtained  and  a copy  number  was  calculated.  These results  were considered as noise in the detection, as Ct values for negative controls corresponded to late cycles and the calculated  copy  number were  under the  detection limit observed in the standard curves (Table 4).

Detection  of  Vesicular  Stomatitis  Virus  (VSV). A  Ct  value of  39.8  ±  1.2  was  obtained  for the negative control and Ct values for the samples were around  this  value.  The calculated  copy  number for the negative control  was  16  viral  copies,  higher copy  number values  were obtained  for  samples derived  from MDBK  cultures  maintained  with the reference FBS and  test  FBS.  Among  MDBK samples  the highest  copy  number values  were obtained  for S8  and  L10  samples,  corresponding to  the  culture maintained  for 21  days  with  the test-FBS. These  values  showed  to  be  higher than copy  number calculated  for cultures  maintained 7  and  14  days  with  the test  FBS and  7  days  with the reference FBS,  nevertheless, 21  days  values were  in  the  same order of magnitude.  If  the  VSV were  actively  present  in  the culture,  a  geometrical increase in  the detected  viral copy  number must have occurred  along  the  21-day  follow-up  and this  viral  increase might  have  been  accompanied by  cytophatic  morphological  changes  in  the  cell culture. As this pattern was not observed for MDBK cultures, data are not conclusive for the presence of the VSV.  Within  the  HEp-2  derived  samples,  high copy numbers were calculated for supernatants from cultures maintained 7 and 14 days with the test FBS. The  copy  number had  a  slight  increase along  thefollow-up  period,  but  not  a geometrical  increase, in  addition  no  cytophatic effect  was  observed  in these  cultures,  thus, these  data cannot  support  the presence of the viral  infection  neither completely discard  it  in  the cultures  maintained  with  the test FBS.

Detection  of  Bovine  Herpes  Virus-1  (BoHV-1). Only  samples  from  MDBK  cultures  were  tested for  the BoHV-1  presence.  A  mean  Ct  value of 35.0  ±  0.3,  corresponding  to  a  mean  copy  number of  227  copies  was  calculated  for  the negative control in this data set. Ct values and copy number values  calculated  for the  MDBK  cell  line cultures maintained  with  reference-FBS and  test-FBS were close to  the values  for the  negative control.  None of  the  calculated  copy  values  was  higher than  the calculated for the negative control. In addition there was  not  and  increasing  pattern  in  the  viral  copy number along the culture follow-up period with the test-FBS. These results cannot support the presence of  the  BoHV-1  in  the MDBK  cultures  treated  with the reference or test fetal bovine serum.

BoHV-1  detection  was  also  conducted  in the genomic DNA  samples  that  gave a positive amplification  for the  Β-actin  gene.  In  this  set of  samples  no  Ct  value was  determined  for the negative control reactions. In the DNA samples, late Ct  values  matching  with  low  copy  numbers  were obtained. In sample S5, from the solid sample of the culture held by  14  days  with  the test-FBS,  a  copy number of 190  was  obtained,  nonetheless,  lower copy  numbers  were obtained for the corresponding S5  cDNA and S8  DNA  and  cDNA samples,  which came from the 21-days culture with test-FBS. Thus, the S5  DNA  sample  result  seems  to  be  a transient increase in  viral  copy  number,  but  cannot  be considered  as  an  indicator of  BoHV-1  presence in the MDBK samples treated with the test-FBS.

Detection  of  the  Bovine  Adenovirus  (BAdV).  In this  detection  a  late  mean  Ct  value  was  obtained for  the  negative  control  41.6  ±  0.6  corresponding to  a  copy  number of 4.  Ct  values  obtained  for the samples were late too. These results indicate that the presence of the bovine adenovirus was not detected in  the cultures  of MDBK  and  HEp-2  cell  lines maintained with the reference-FBS or the test-FBS.

Detection of BAdV in the DNA samples showed amplification  with  late  Ct  values  for  samples obtained  from MDBK  and  HEp-2  cell  lines maintained with the reference and test FBS by 7 and 14 days, but no increment on the copy number was observed  along  the  follow-up  period.  This  results togheter with  the  data obtained  for  cDNA  samples indicate that  BAdV was  not detected in  the treated cultures.

Detection  of  the  Bovine Viral  Diarrhea  Virus type  1  (BVDV-1).  Analysis  for BVDV-1  were carried  out  in  MDBK  samples.  A  mean  Ct  value of 39.2 ± 1.6 was obtained for the negative control and  later Ct  values  were obtained  for  the samples, except  for  the L1  sample,  supernatant  from  the culture maintained  with  the reference-FBS.  The copy number for this sample was 897. As no further studies were done with this FBS, this isolated result is  not  conclusive for the presence of BVDV-1  in the cultures maintained with the reference FBS. No viral  presence  was  detected  for  MDBK  cultured with test-FBS.

Detection  of  Bovine Parainfluenza  Virus  3 (BPIV-3).  MDBK  cell  line  cultures  were used  in this  detection.  Late Ct  value  corresponding  to  low calculated  copy  numbers  were obtained  for the negative control, as  well  as  for  the  samples.  No presence of  this  virus  can  be deduced  from  these results (Table 4).

Detection  of  Bovine Respiratory  Syncytial Virus  (BRSV).  In  the analysis  for  this  virus  no  Ct value was  obtained  for the  negative  control.  Copy numbers  as  high  as  110  copies  were  calculated for  the  supernatant  samples,  but  there was  not progressive  increase in  the copy  number detected for  samples  from 7, 14  and  21-days  culture. These results indicate that the high copy number observed for sample L7, the supernatant of the 14-day culture with  test-FBS does  not  represent  presence of the respiratory  syncytial  virus  in  this cell  cultures treated with the test-FBS.

Detection  of  the  Bovine Viral  Diarrhea  Virus type 2 (BVDV-2). PCR reactions for BVDV-2 were done  on  MDBK  derived  samples.  Copy  number calculated for the negative control  and  the samples were  lower than  10, except  for  the  sample  from the supernatant  of the culture held for 7  days  with the test  FBS (Sample  L4)  and  lower values  were obtained  for the  samples  of the  same culture after 14  and  21  days  follow-up,  indicating  that  the L4 sample  value  is  an  isolated  result  and  cannot  be considered  as  an  indicative  for the  presence of the tested virus in the MDBK cultures maintained with the test FBS. In addition the calculated copy number was below the linear detection limit established with the standard curve, indicating that this result is not a consistent positive.

Discussion

Reports  on  viral  detection  by  means  of real time PCR show  that  this  technique  is  suitable for  the  detection  of  bovine  viruses  in  different biological matrices and  demonstrates  to  be specific and  as  sensitive  as,  or more sensitive than  the immunological  methods  like fluorescent  antibody test and immunohistochemistry (Timsit et al., 2010).

Specificity  of the PCR reaction  depends  on primer design  and  can  be improved  by  the use of probes. Primers used in our analysis were tested for specificity  by  BLAST and  no  specific  probes  were used,  as  detection  was  made  with  the  ResoLight^® DNA-binding  dye.  Real-time RT-PCR employing the non specific fluorescent SYBR Green-I has been extensively used for bovine viral detection yielding satisfactory  results  and  is  also  considered  as  a better option  over probes-based  amplification,  as the result does not rely on the perfect match of the probes,  which  can  be  hindered  by  viral  mutations (Kosinova,   et al.,  2007).  Here we  replaced  the use  of SYBR  Green-I by  ResoLight^®,  a new non  specific  DNA-binding  fluorescent  dye  with  a higher incorporation  frequency, expecting  a higher sensitivity. Despite precaution taken in the technical procedures  a positive amplification  response was obtained for the negative controls. We consider that the highly  DNA-incorporation  frequency  of the Resolight  ,  which  accounts  for its  high  sensitivity, might  also  be  responsible  for the  high  noise level detected in these experiments.

The  use  of cloned  viral  sequences  as  positive controls  is  a resource that  allows  for  generation  of standard  curves  without  the use  of whole active viruses  at  the  laboratory. This  strategy  has  been used previously and proved to be useful (Kosinova et al.,  2007;  Young  et al.,  2006), in  our  work  the constructed  control  oligonucleotides  A  and  B rendered  specific  amplification  and  high  linearity standard  curves,  detecting  as  low  as  2x10E1 pathogen  copies  for  six  viruses  and  mycoplasma and as low as 2x10E2 for BVDV-2. Detection limits for bovine viral diarrhea virus by real-time RT-PCR was  established  at  1  x  10E1  DNA/µL, copies  by other authors, which lies in the same detection level (Young et al., 2006).

For the detection  of  the seven  selected  viruses and  Mycoplasma,  we  applied  here  an  enrichment procedure  in  which  susceptible cell  lines  were cultured  for a  21-day  period  with  the  test  fetal bovine serum MICROGEN^®,  collecting  cellular  and supernatant  samples  each  7  days.  This  procedure was  based  on  the viral  detection  test  applied  to  the reference FBS by  the manufacturer  (APHIS-USA, 1995). The aim of this protocol was the generation of serial data from the same culture, to make a follow- up  of  the  culture.  In  this  way  if  the presence of a target  virus  was  detected  in  a  sample,  this  could  be compared to the previous and following samples. The projected result was that an increasing copy number along the serial samples taken after 7, 14 and 21 days of culture will account for the presence of the virus.

Prior to real time PCR viral specific amplification, all  samples  were subjected  to Β-actin  PCR amplification,  to  check  for  the  presence of template in  samples.  Only  samples  giving  a  positive amplification  in  this  reaction  were  analyzed  for viral  presence,  to  ensure  that  negative  results  will not  account  for absence  of  template  in the  samples. Within  the  results  obtained  from  the  real  time PCR reactions  there were samples  with  a high  copy number, but  these  were isolated  samples  and  no increasing  pattern  was  observed  when  compared within  the  sample  set.  A  persistent  high  viral  copy number was calculated for the supernatants from the 7  and  14-day  follow-up  for the  VSV, in  the HEp- 2  cell  line  culture maintained  with  the  test  FBS, nevertheless  the  copy  number  increase was  not conclusive for viral presence.

VSV is  a  highly  prevalent  virus  in  Colombia, which has gained space since the eradication of the foot  and  mouth  disease  virus  (Cajas  et al.,  2003). This  situation  explains  the  possible viral  presence detected  in  the  cultures  maintained  with  test-FBS (MICROGEN   FBS),  even  after the  filtration  by 0.22  µm  filters  included  in  the  FBS production process.  According  to  this  the presence  VSV  in the Colombian  MICROGEN^®  fetal  bovine  serum might  be  considered  as  one  of the  highest  risks  in this product and special attention should be given to the testing of these virus in future analysis for other batches of the product.

Viral  analysis  carried  out  in  this  study  did not  detect  the  presence of  the pathogenic bovine viruses or Mycoplasma  in the tested  MICROGEN^® fetal  bovine serum, except  for VSV, which  was detected  in  a  low level.  The viral  detection procedure described here should be included in the quality  control  analysis  of  the  MICROGEN^® fetal bovine  serum,  to  ensure  the  viral  and  mycoplasma absence for  each  batch  of  the  product  before its commercialization.

Acknowledgment

This  work  was  supported  by  the Colombian Administrative Department  of Science, Technology and Innovation:  Colciencias; contract No  038-2007 (ACAC-Colciencias 179-206).

This  work  was  partially  developed  at  the facilities  of  the Institute  of Biotechnology  at  the Universidad Nacional de Colombia, Bogotá D.C.

References

1. Animal and Plant Health Inspection Service USDA (US). Code of Federal Regulations, Requirements for ingredients of animal origin. Title 9CFR Part, 113.53. Washington; 1995.         [ Links ]

2. Betancur C, Orrego A, González M. Seroepidemiological study of parainfluenza  3  virus  in  bovines  with  reproductive  failure, from Montería-Colombia. Rev Med Vet 2010; July/ Dec:63-70.         [ Links ]

3. Cajas  JG, Rojas  JA,  Almansa  JE, Wilches  MT,  Sánchez C, Acosta  O.  Comparación  de  la  respuesta  inmunológica  a  las proteínas  del  virus  de  la  estomatitis  vesicular  entre  bovinos vacunados  e  infectados  de  manera  natural  o  inducida.  Rev Corpoica 2003; 4:1-5.        [ Links ]

4. Freshney  RI.  Culture  of  animal  cells:  a  manual  of  basic technique.  4th  ed.  Hoboken,  (NJ):  John  Wiley  and  Sons  Inc; 2000.        [ Links ]

5. GIBCO Invitrogen Cell Culture. Fetal Bovine Serum Brochure B-066802-r1 0407; 2007.         [ Links ]

6. IDT-Integrated DNA Technologies, Inc. PrimerQuest, SciTools. [accesed on July 2009] URL: http://www.idtdna.com/Scitools/ Applications/Primerquest/         [ Links ]

7. Kosinova E, Psikal I, Robesova B, Kovarcik K. Real-time PCR for quantitation of bovine viral diarrhea virus RNA using SYBR Green I fluorimetry. Vet Med 2007; 52:253-261.         [ Links ]

8. Nishizawa  I,  Sato  M,  Fujihara  M, Sato  S,  Harasawa R.Differential  detection  of  hemotropic  Mycoplasma  species  in cattle by melting curve analysis of PCR products. J Vet Med Sci 2010; 72:77-79.         [ Links ]

9. Piedrahita  L, Montoya,  L  Pedraza,  F. Bovine  Herpes  Virus type  1  (BoHV-1)  as  a  possible  cause  of encephalitis  in  cattle of the  Colombian  Magdalena  medio  region:  serological  study and  epidemiological  analysis.  Rev  Colomb  Cienc  Pecu  2010; 23:191-198.         [ Links ]

10. Rozen  S,  Skaletsky  H. Primer3  on  the  www  for  general  users and  for  biologist  programmers.  In:  Krawetz  S, Misener  S editors.  Bioinformatics  Methods  and  Protocols:  Methods  in Molecular Biology. Totowa: Humana Press; 2000. p.365-86.         [ Links ]

11. Timsit  E,  Maingourd  C,  Le  Dréan  E,  Belloc C, Seegers  H, Douart A, Assié S. Evaluation of a commercial real-time reverse transcription polymerase chain reaction kit for the diagnosis of bovine respiratory syncytial infection. J Vet Diagn Invest 2010; 22:218-241.         [ Links ]

11. Vargas,  DS,  Jaime J,  Vera VJ.  Perspectives  to  control  bovine viral  diarrhea  virus  (BVDV).  Rev  Colomb  Cienc  Pecu  2009; 22:677-688.         [ Links ]

12. Young  NJ,  Thomas  CJ,  Collins  ME, Brownlie  J.  Real-time RT-PCR  detection  of Bovine  Viral  Diarrhea  virus  in  whole blood using an external RNA reference. J Virol Methods 2006; 138:218-222.        [ Links ]