anti-thioredoxin antibodies

[Brian Talbot]

I am using the "Thiofusion" expression system from Invitrogen and I would 
like to detect the thioredoxin using antibodies. However, the 600 dollars 
per ml from American Diagnostica Inc seems a little too much. Does anyone 
know of another source of anti-thioredoxin antibodies?

Brian G Talbot
Departement de Biologie
Universite de Sherbrooke
Sherbrooke, Que. Canada
btalbot at courrier.usherb.ca
INTRODUCTION

      Bovine viral diarrhea virus (BVDV) is an enveloped positive strand RNA virus
(Hermodsson et al., 1962). BVDV is currently classified as a member of the genus
pestivirus in the Flaviviridae Family (Frankiet et al.,1991). Other member of the
genus pestivirus are Hog cholera virus (HCV) of pigs and border disease virus
(BDV) of sheep (Horzinek et al.,1981). BVDV occurs in tissue culture as 2 biotypes
cytopathic (cp-BVD) or non-cytopathic (ncp-BVDV) (Lee et al., 1957).

      BVDV is an economically important pathogen of cattle that is distributed
worldwide. BVDV infection in cattle can range from clinically inapparent to the fatal
mucosa disease ( Bolin, 1993).  The predominant clinical disease is acute bovine
diarrhea viral (BVD). Manifestations of BVD are fever, lymphopenia, inappetence,
and diarrhea. Since 1993 in Quebec clinical sever form of acute BVD that is
characterized by: fever, diarrhea, lymphopenia, thrombocytopenia, hight morbidity
and mortality and seldom hemorrhaging was described (Pellerin et al., 1994). This
form was previously reported (Corapi et al., 1989 and 1990).

      The number of BVDV-specific proteins described is varied in dependent of
experimental methodologies and reagents used by different authours and low level
production in infection cells of most BVDV polypeptides.

      Monoclonal antibodies (MAbs) against proteins of BVDV have been produced
by several laboratoires (Peter et al., 1986, Bolin,et al., 1988, Xue et al., 1990,
Shannon, personal communication,1994) and they were used for antigenic
characterization of differents BVDV strains and isolats. XUE et al. (1990)  divided
40  cp and ncp BVDV isolates into six groups by neutralization test (NT) using 5
MAbs gp 48 kDa-specific or gp 53 kDa-specific. Magar et al. (1988) were classified
5 reference strains and 10 Canadian isolates into three groups on the basis of their
reactivity by immunofluorecence and neutralization with a BVDV neutralizing MAb
against gp 53 kDa. In a recently study, Pellerin et al. (1994) divided 15 Canadian
isolates into two groups by NT using six polyclonal antibodies (PAb) and by the
sequances of 5` untranslated region. The detection of BVDV is usually carried out
by peroxidase-linked antibody assay (PLA assay), or by immunofluorecence. Two
moyor method of cell fixation are normally used in PLA assay. Most authors used
aceton for fixation (Afshar et al., 1989, Cay et al., 1989., Edwards et al., 1991).
Neverthles, the formalin have also been used (Haines et al., 1992, Shannon et al.
personal communication, 1994).

      The purpose of present communication was to study the antigenic variation
among 13 Quebec isolates and three reference strains and two US isolates by PLA
assay, NT and radioimmunoprecipitation assay (RIPA) using the PAbs and MAbs. In
this study for the first time we compared two fixation methods on PLA assay for
detection and eventually classification of BVDV.

MATERIAL AND METHODS

Virus and cells

      The cytopathic BVDV , NADL, Singer and C24V-Oregon strains and the non
cytopathic New York-1 (NY-1) strain were obtained from the American Type Culture
(ATCC, Rockville, Maryland). In this study three field isolates before 1993 outbreak
in Quebec and 10 field isolates  during  1993 outbreak, from different geographic
regions, were isolated in our laboratory. All the isolates were cloned by three cycles
of plaques purification. Two isolates 87-2552 and 86-7061 were  obtained from Dr.
H. C. Minocha (Kansas State University, Manhattan, Kansas, 66506)

      Madin-Darby Bovine Kidney (MDBK) cells free of BVDV (ATCC) were grown in
Earle's  minimum essential medium (MEM, Gibco Canada Inc.), supplemented with
2.2.g/lit sodium bicarbonate, 10% of fetal bovine serum (FBS) (Gibco, Grand Island,
New York) and antibiotics. All virus were propagated in MDBK cells in the presenc
of 2% FBS. 

Polyclonal and monoclonal antibodies
      
      Specific bovine BVDV polyclonal antisera (PAb-1) was prepared in calves by
immunization with NADL and Singer strains. PAb-2 was obtained by mixing 70 sera
from clinicals cases during the 1993 outbreak in Quebec. Two polyclonal antibody,
PAb-3, Singer-specific and PAb-4, 87-2552-specific were obtained form Dr. H.C.
Minocha.
      
      The supernatant culture of the MAbs; 2NB2, P1H11, P4G11, P4A11, P3C12, P1D8,
1NB5, N2B12, P3D11, P3F6, O2A1, 40B4 and N4B4 were obtained from Dr. A. D. Shannon
(Elizabeth Macarthur Agricultural Institue, Camden, New South Wales, Australia).
They were produced against a total of 5 BVDV isolates, including the reference
strain Oregon-C24V, and two New South Wales cp and two ncp isolates. The ascite
fluid of the MAbs D89, C17, E15, F15, D11, and B7, were obtained form Dr. H.C.
Minocha. The first four MAbs were produced against the NADL strain and the last 
two were  produced  against the field isolated 87-2552. 

Peroxidase-linked antibody assay (PLA assay)

      The MDBK cells were grown in microtitre plate (96 well plats) and were
infected with each of BVDV strains or isolates. After 3-4 day, two different fixation
methods were used. In the first method, the plates were washed with PBS-Tween
(0.05% Tween 20, Merk), fixed by added 100  l/well of acetone 20% for15 min, and
dryed at 30  C for 3-4 h.  In the second method  the plates were fixed by added 100
 l of formalin to each well ( 1:7 dilution of 37% formaldehyde solution). After 10 min,
the plates were washed by PBS-Tween and dryed at room temperature (RT). The
plates fixed by both methods were blocked by100  l/well 5% skim-milk powder in
PBS and incubated for 30 min at 37  C. Washing steps throughout the test were
performed with PBS-Tween (0.05% Tween 20) and then the plates were dryed at RT.
The plates were incubated with 100  l of each MAb (dilution 1: 100 was used for
ascite fluid and no dilution for supernatent culture)  for 90 min at 37  C
temperature. Each MAB was tested in triplicate. For the detection of bound MAbs,
100  l/well Goat Anti-Mouse IgG (H+L) Horseradish Peroxidase Conjugate ( BIO-RAD)
(dilution1: 200 in PBS-Tween 0.1%) containing 1% gelatin, were added and incubated 
for 90 min. The substrate was 3-amino-9-ethylcarbazole (Sigma) dissoved in 0.3 ml
of dimethyl-formamide (Merck), , and diluted to 5 ml in 0.05 acetate buffer, pH 5
containing 1.67  l hydrogen peroxid (30%). After 10 - 15 min incubation, the tests
were readed by microscopic examination of the plates. All the negatif resultats were
checked at least two times with the same method. 

Neutralization test (NT)

      NT were performed in 96 well tissue culture plate. 50  l of serial fourfold
dilution (prediluted 1:125) of 6 MAbs; D89, C17, E15, F15, D11 et B7 and 2 PAbs (PAb-
1 and PAb-2) were incubated with 50  l of virus suspension containing 50 TCID50 for
1h at 37  C. Then 100  l of a MDBK cell suspention containing 3 x106 cells/ml were
dispensed per well. The assay was done in quatruplicate. The plates were kept at
37  C in CO2 incubator for  4 days. The highest dilution that inhibited the cytopathic
effet in at least 50% of the cultures was considered as the neutralization titer for
each antibody to different cytopathic viruses. For testing of noncytopathic virus,
after 4 days the plate were stained by PLA assay, as described above, except that
PAb-1 (1:50 in PBS-Tween 0.1% and rabbit antibovine IgG horseradish peroxidase
conjugate (diluted1: 400) were  used. The neutralization titer of the antibodies the
against ncp virus was considered as the highest dilution that was negative in PLA
assay at least 50% of the culture.

Radioimmunoprecipitation assay (RIPA)

      Antigen for RIPA was prepared from MDBK cells infected with BVDV strains
and isolats; NADL, 86-7061, 89-E770, 91-Domar, 93-1 and 93-671. RIPA proceduces
was performed as described by  XUE et al.(1990).

RESULTS

      Peroxidase-linked antibody assay - The binding of the 19 MAbs to 19 different
BVDV strains and isolates was determined by an PLA assay to ascertain the pattern
and extent of antigenic variation among viruses. In addition two different fixation
methods were compared. In the basis of aceton-fixation method, BVDV strains and
isolates in this study divided in 6 groups (Table 1). Seven MAbs presented different
reactivity patern, the rest of the MAbs were positives (P1H11, P4G11, P4A11, P3C12,
P3D10, D89, C17, E15, F15 and B7) or negatives (P3F6 and N4B4) for all the strains
and isolates. NADL as NY-1 atrain showed a pattern different from all the other
viruses. C24V-Oregon strain and 87-2552 isolate were the only virus react with
MAbs O2A1 and 40B4. All the 1993 Quebec isolated and Quebec isolates before 1993
without  exception were classsified in two distint groups. 

      However, in formalin-fixation method, viruses showed minor different
reactivity patterns (Table 2). Cytopathic strain C24V-Oregon and field isolate 87-
2552 reacted with all of MAbs used. The rest of BVDV have the same pattern, they
reacted with all of MAbs except O2A1 and 40B4.
      
      Neutralization test - Two PAbs and six MAbs were used in NT against 19
strains and isolates. All BVDV strains and isolates were classifieded into 4 groups
based on there neutralization reactivities with PAbs. We consider a neutralization
titer of 8000 or more as a "strong reaction"  and a neutralization titer of 2000 or less
as "weak reaction". Results of NT with two PAbs (PAb-1 and PAb-2 ) were presented
in Table 3. Group I includes all strains and isolates that reacted strongly with PAb-1
and weakly with PAb-2. Group II includes those with a strong reaction  with both
PAbs. Group III is represented with strains or isolates that have weak reaction with
PAb-1 and react strongly with poly Ab-2. Group IV is represented with Quebec
isolates ; 93-1, 93-3, and 93-619 and one US isolate 87-2552 which reacted weakly
with both PAbs. 
           The neutralization titer of MAbs to each strain and isolate are
      given in Table 3. MAbs D89 and F15 neutralized 77.8% (14/18) of virus
      strains and isolates in this study. Two MAbs C17 and E17 reacted only
      with two strains NADL and Singer (11.1%). The MAbs D11 and B7 both
      reacted only with isolate 87-2552 (5%). None of the MAbs neutralized all
      the virus at the dilution used in this study. Based on the NT with
      MAbs, group I and group IV are subdivided into 3 and 2 groups
      respectively. Ia :  viruses in this group were neutralized by MAbs D89,
      C17, E15 and F15. Ib : are those neutralized by any MAbs. Ic : is
      represented by viruses that were neutralized by D89 and F15. IVa :
      viruses that were neutralized by D89 and F15. All virus in groups Ic,
      II, III, and IVa (66.7%) have the same pattern of reactivity with MAbs.
      IVb : with one isolate, 87-2552 has only reacted with MAbs D11 and B7.

      Radioimmunoprecipitation assay - When extracts from cells labelled with L-
[35S] methionine were analysed by SDS-PAGE (Fig 1) we identified the presence of
the following virus-induced polypeptides: 120 kDa, 80 kDa, 69 kDa, 53 kDa and 25
kDa in NADL strain, 89-7061, 93-1 and 93-671 isolates(Fig 1, lane 2 to 5). The 25 kDa
polypeptide was detected as a diffuse and fainted band. Quebec isolates 89-E770 and
91-Domar gave only the precipited polypeptides120 kDa , 80 kDa  and 53 kDa (Fig
1, lane 6 and 7).  In addition, in NADL strain viral polypeptide of 135 kDa was
precipited. Differences in the mobility of 120 kDa polypeptide was noted between
different strains.

      Efficiency of the different PAbs for precipitating viral polypeptides was
determined. The comparison of the polypeptide profiles of the NADL strain and 93-1
isolate with 4 PAbs was shown in Fig 2.  Viral polypeptide 25 kDa, 53 kDa, 69 kDa,
80 kDa, and 120 kDa precipitated by all of 4 PAbs (Fig 2, lane 5 to 12). The 
polypeptides 53 kDa, 69 kDa, and 80 kDa appeared as the most highly labelled
polypeptides. The polypeptide profiles for the two viruses were not similar for 4
PAbs. Immunoprecipitation of NADL polypeptiedes with PAb-2 and PAb-3 showed the
presence of135 kDa polypeptide (Fig 2, lane 6 and 7). The PAb-3 precipited one 32
kDa polypeptide (Fig 2, lane 7 and 11). 69 kDa polypeptide was observed in both
viruses (Fig 2, lane 5 and 9). 

DISCUSSION

The PLA assay was used for detection or classification of BVDV by many authors.
Two methods of fixation was used, aceton (Afshar et al., 1989, Cay et al., 1989.,
Edwards et al., 1991) or formalin (Haines et al., 1992, Shannon et al. personal
communication,1994). We compaired both methods.

      Fixation affects the activity of antigens, the choice of fixative depends upon
the antigen to be detected. Formaline preserves cell stucture better than aceton
(Junqueira et al. 1989). In PLA assay when we used formalin-fixation method, all of
MAbs except O2A1 and 40B4 reacted with all of BVDV, indicating a high degree of
structural conservation at the level of three different proteins (p8o, gp 53 and gp48
kDa). This result is confirmed the previous found of Shannon et al. (personal
communication, 1994). They described that the supernant culture  MAbs in our
study detected 95 to 100% of 115 to 151 strains and New South Wales isolates. Only
two MAbs, O2A1 and 40B4, were specific of C24V-Oregon strain. Whatever, Haine et
al. (1992) , observed that the detection of BVDV by immunohistochemical stain in
formalin-fixed tissue with diferent MAbs did not bring many success. In their study
only 1 MAb from 33 MAbs reacted positivly. Probably hight formalin concentration
and long processing for deparaffinization, rehydration and inactivation of
endogenous tissue peroxidases, plaed a role in denaturation of virus epitopes.
      
      In the aceton-fixation method 7 MAbs lost their positive reaction with some 
or all of BVDV. Aceton affected some of the epitopes that we attemp to demonstrate.
This may be attributed to the conformation-dependant nature of many BVD viral
epitopes (Moennig et al., 1990) which renders then sensitive to the denaturing
effects of fixation. There is another explication for negative result with 2 MAbs 2NB2
and P1D8. These MAbs react with the soluble non-stractural protein of approximate
molecular weight 125 or 80 kDa (depending on biotype). Edvard et al.(1991)
suggested that this antigen was expressed most strongly shortly after infection of
cell cultures, so some of the negatif results may be due to later fixation the
cultures.

      However the drastic effet of aceton-fixation allowed us to demonstrated
different stability of viruses tested. In spite of same reaction in formalin-fixation
method, indicating that all strains and isolates have the same epitopes but that
these might be differently expressed in each strain or isolate. This property allowed
us to classify BVDV in base of their stabilization in aceton- fixation method. It is
interesting to note that the 1993 Quebec isolates and Quebec isolates before 1993
were  in two different groups, indicating an antigenic variation among them. 

      The use of NT with 2 PAbs to allowed to us to classified strains and isolates
in 4 groups. In addition, these strains and isolates were clasified in 7 subgroups
using 6 MAbs. In the present study, the viruses strains and isolates showed various
degrees of reactivity in NT. Variation in neutralization titer may indicate that
changes in the amino acid sequences of epitopes that result in minor antigenic
drifts of  viruses. The Mabs D89 and F15 appears to react with a highly conserved
epitope. They neutralized all BVDV tested, excepted groups Ib and IVb. 

      Percentage of competive inhibition for two MAbs D89 and E15 is between 70
to 80% (XUE W., et al., 1990) indicating that these two epitopes are highly related.
Although in the PLA assay (both methods) epitope for MAb E15 was physically
present on all viruses, only NADL and Singer strains were neutralized by this MAb.
This indicated that the neuralizing ability of E15 epitopes is not present in Quebec
isolates. This lack of neutralization could be due to  minur change in amino acid
sequences or in conformation. Similar result reporting by Xue et al. (1990), for MAb
F15. This MAb did not neutralize the US isolates or did so with very low neutralizing
titer, but in immunoflurecence it reacted with  the most of the isolates. 
      
      All of Quebec isolates have the same pattern in the basis of MAbs, except that
87-E770 isolates. They resemble to two references strains NADL  and Singer in the
basis of MAbs D89 and F15, and look like the NY-1 and C24V- Oregon strains in the
basis of the rest of MAbs. In NT, Quebec isolates were distributed in different
antigenic groups.
      
      Comparison of reactivity pattern of Quebec isolates with six MAbs which used
simultaneous in PLA assay and NT, presented that,  MAb  binding to virus in PLA
assay is independet to neutralization activity. But for neutralization activity,
binding the MAb to virus in indispensable. There is not any virus  with
neutralization activity which presented negative result in PLA assay. We obtained
the same result with compaire the immunofluresence test and  neutralizaton activity
of MAbs to  40 BVDV isolates which reported by Xue et al. (1990). There is only one
isolates which present negatif result with immunofluresence, but low titer (1: 100)
in NT. Aceton fixation which was used in their study could modify the epitope and
couse negatif result in immunofluresence. All the Quebec isolates had the same
reactivity in PLA assay in formalin-fixation method. They reacted with  six MAbs
common in PLA assay and NT. In aceton-fixation method only result for MAb D11
came negatif. But in NT, Quebec isolates, except 89-E770 isolate, were neutralized by
the MAbs D89 and F15, and 89-E770 isolate was neutralized by any MAbs.
 
      In conclution,  aceton or formalin fixation method are used fo differents
purposes. For the detection of BVDV strains, the fixation method with the formalin
is the method of choice, whereas the aceton method could be useful for the
classification of BVDV strains from antigenic variations.

      Study of peptidic profile NADL strain and different isolates, demonstrated
variability in mouleculare weight of 120 kDa polypeptide. This difference was noted
previously (Magar et al.,1988).

      Absence of 25 kDa and 69 kDa polypeptide in 89-E770 and 91-Domar isolates,
could be due to low titer of the virus which did not permit on  adequate labelling.
We used multiplicity of infection 2 instead of 5 for these isolates.
      
      We could not detected 37 kDa  and 47 kDa polypeptides. This result partially
confirm the result of  Donis et al. (1987). They detected  these two polypeptides only
in some cp BVD isolates upon longer exposurer of the autoradiograms. Ruben et al.
(1987) found the majority of sera collected from naturally BVD virus-infecion
animals (like as PAb-2 in this study) did not precipite 37 kDa polypeptide.

      Probably in our experience 69 kDa polypeptide is analougous to 62 or 65 kDa 
 polypeptide which reported by many authours (Donis et al., 1987, Magar, et al.,1988,
Collett et al, 1988, Deregt et al., 1991).

      Presence of 25 kDa and absence of 75 kDa polypeptide probably is
consequence of hight concentration of 2- mercaptoethanol (10%) in the sample buffer
(Deregt et al.,1991). In addition 75 kDa polypeptide has a short half-life. Its absenec
may have been due to depletioin of L-[35S] methionine during long labelling.

      Donis et al. (1987) labelled the cell with L- [35 S] cysteine or L-[35S] methionine
under a mild hypertonic condition. They additioned 75 mM excess NaCl shortly
before the radiolabelling to  stop translation of cellular messages. They also used
actinomaycin D. They detected the suplemented polypeptides, 35 kDa, 62 kDa, 25 kDa
and 19 kDa. But in RIPA labelled with L-[35S] methionine , they detected 135 kDa, 62
kDa and 25 kDa, but not 32 kDa, after polongated exposure of autoradiogrames.
However, we detected only  32 kDa polypeptide with PAb-3 (Fig 2, line 7 and 11), but
not with the rest of PAbs.

      In addition we detected 135 kDa polypeptide only in NADL strain with 2 PAbs,
2 and 3 (Fig 1, lane 2, Fig 2, lane 6 and 7), and we identified the 32 kDa polypeptide
with PAb-3 (Fig 2, lane 7 and 11). This results indicated that detection of virus
polypeptides is affected not only by different protocols and reagents which were
used but also by the PAb. The PAb may play an important role for detection of
polypeptides which are less available.


 Afshar, A., Dulac, G. C. and Bouffard, A., 1989. Application ofperoxidase labelled
antibody assay for detection of porcine  IgG antibodies to hog cholera and
bovine viral diarrhea  viruses. Journal of Virological Methods, 23: 253-262.
Bolin, S.,  Moenning, V.,  Gourley, N. E. K. and Ridpath, J., 1988.Monoclonal
antibodies whit neutralizing activity sergregateisolates of bovine viral diarrhea
virus into groups. Aarch.    Virol. 99: 117-123.
Bolin, S. R.,1993. Immunogens of bovine viral diarrhea virus. Vet.Microbiol., 37:
263-271.

Cay, B., Chappuis, G., Coulibaly, C., Dinter, Z., Edwards, S., GreiserMcHugh, I.,
Moennig, V., Nettleton, P. and Wensvoort, G.,1989.   Comparative analysis of
monoclonal antibodies : Report of aninternational workshop. Vet. Microbiol., 20:
123-129. 
Corapi, W. V., French, T. W.  and Dubovi, E. J., 1989. Severethrombocytopenia 
in young calves experimentally infected  whit noncytopathic bovine viral
diarrhea virus. J. Virol., 63:3934-3943.
Corapi, W., Elliott, V., R. D., French, T. W.,   Arthur, D. G.,  Bezek, D. M.and
Dubovi. E. J., 1990. Thrombocytopenia and hemorrhages in   veal calves infected
whit bovine viral diarrhea virus. J. 
      Am.Vet. Med. Assoc., 196:590-596.
Edwards, S., Moennig, V. and Wensvoort, G., 1991. The developmentof an
international reference panel of monoclonal antibodiesfor the differentiation of
hog cholera virus form other pestiviruses. Vet. Microbiol. 29: 101-108.

Francki, R. I. B., Fauquet, C. M.,  Knudson, D. L. and Brown, F., 1991.
Classification and Nomenclature of Viruses. Fifth Report of the  International
Committee on the Taxonomy of Viruses. Arch. 
      Virol., 2(Suppl.): 228-229
Hermodsson, S., and Z. Dinter. 1962. Properties of bovine virus  diarrhea virus.
Nature 19: 893-894.
Haines, D. M., Clark, E. G. and Dubovi, E. J., 1992. Monoclonalantibody-based
immunohistochemical detection of bovine  viral diarrhea virus in formaline-fixed
paraffin-embedded      tissues. Vet. Patho., 29: 27-32.
Hermodsson, S. and Dinter, Z., 1962. Properties of the bovine virusdiarrhea virus.
Nature, 191: 893-894.
Horzinek, M. C., 1981. Non-arthropod-borne togaviruses. Academic Press, London.
Junqueira, L. C., Carneiro, J., Kelley, R. O. (Editors), 1989. BasicHistology, 6.
Appleton & Lange, pp. 1.
Lee, K. M. and Gillespie, J. H., 1957. Propagation of virus diarrheavirus of cattle
in tissue culture. American Journal of   Veterinary Research 18: 952-953.
Moennig, V., 1990. Pestiviruses: a review . Vet . Microbiol. 23: 35-54.
Pellerin, P., Den-Hurk, J. V., Lecomte, J. and Tussen, P., 1994. Identification
of a new group of bovine viral diarrhea virus  strains associated with severe
outbreaks and high mortalities.    Virology, 203: 260-268.
Peters, W., Greiser-Wilke, Moennig, V. and Liess, B., 1986.Preliminary
serological characterization of bovine viral   diarrhea virus strains using
monoclonal antibodies. Vet.  Microbiol. 12: 195-200.
Shannon, A. D., personal communication,1990.
Xue, W., Blecha, F. and Minocha, H. C., 1990. Antigenic variations inbovine viral
diarrhea viruses by monoclonal antibodies. J. ofClinical Microbiol. Aug: 1688-1693.

 
Fig 1- Radioimmunoprcipitation of bovine viral diarrhea polypeptides with L-[35S]
methionine and reacted with PAb-2. The labeled polypeptides are from the fllowing
viruses; NADL strain (lane 2), isolates 86-7061 (lane 3), 93-1 (lane 4), 93-671 (lane
5), 89-E770 (lane 6) and 91-Domar (lane 7). Analysis was done in 7.5% polyacrylamide.
kDa= kilodalton.





Fig 2- Comparison of viral polypeptides from cell infected with NADL strain (lanes
5 to 8), 93-1 isolate (lane 9 to 12) and mock infected with medium only(lane 1 to 40.
Cell lysates were immunoprecipitated with four polyclonal antibodies. PAb-1 (lane
1, 5 and 9), PAb-2 (lane 2, 6 and 10), PAb-3 (lane 3, 7 and 11) and PAb-4 (lane 4, 8
and 12). Analysis was done in 7.5% polyacrylamide gel. kDa= kilodalton. 

 
brian talbot
Biologie