Characterization
of the immune response in wild cougars infected with feline immunodeficiency
virus (FIV)
Jesse Thompson
University
of Montana IBS-CORE Program
Mentor- Dr. Mary Poss
Introduction
Individuals exposed to the same
infectious organism may develop an antibody response to different viral
epitopes. This differential response may reflect features of the organism or
the host. Host factors such as age,
gender, route of exposure, or genetic restriction may influence the antibody
response. On the other hand, viral features can change over time and different
strains may elicit unique antibody responses.
It is well known that in a retrovirus infection, the virus changes
rapidly during the course of the infection and this change may be reflected in
the humoral response of the infected individual. Examples of pathogenesis and changing antibody responses during
the course of infection have been documented in domestic cats infected with
feline immunodeficiency virus (FIV) (Dow, et al., 1990). Wild felids are also infected with FIV, but
infection with this virus causes no apparent symptoms in these animals. No study has been undertaken to examine
antibody responses to FIV infection in cougars or other non-domestic
felids. However, examples of studies on
the antibody responses in wild felids demonstrate that some African cheetahs
and Florida panthers fail to mount a diverse immune response to a viral
pathogen as a result of genetic isolation (Yuhki and O'Brien, 1990, Olmsted et
al., 1992). Thus, understanding the antibody response of a population to an
endemic pathogen may not only provide important information to disease
pathogenesis, but also on population structure.
Many wild cougars living in Yellowstone
National Park and the Snowy Mountain Range in Wyoming are infected with
FIV. Due to migratory patterns and
possibly extended home ranges, these animals do not necessarily constitute
genetically isolated populations (Yuhki and O’Brien, 1990). Preliminary flow
cytometry data of serum samples from these two distinct populations of wild
cougars suggests that 50-80% of these cats are infected with FIV. This assay specifically detects antibodies
to the envelope glycoprotein of the virus.
This data also indicated that some individuals from the Snowy Range who
were seropositive at an earlier date were seronegative when a subsequent sample
was tested. This leads to the hypothesis that the antibody recognition of
specific viral protein in these animals may change over time.
The glycosylated envelope protein of FIV
is comprised of two subunits, a120 kD surface glycoprotein (gp120), and a 40 kD
transmembrane portion (gp40). Other
proteins that elicit an antibody response include the viral capsid proteins p17
and p24, and the polymerase enzyme p55 and p60.
In this study an immunoblot assay was
developed to evaluate the immune response of cougars from the Yellowstone and
Snowy Range populations. Performing immunoblot assays on the serum of these
cougars will not only qualitatively evaluate their individual immune responses,
it will also determine an animal infected with FIV has been overlooked in the
flow cytometry assay, which only detects the antibodies to the viral env protein.
Materials
and Methods
Virus
Preparation and Purification. FIV was pelleted at 100,000xg from the
supernatant of a thymic lymphosarcoma cell line (3201) chronically infected
with cougar lentivirus. Pellets were
resuspended in 500 ml of
Tris-buffered saline (TBS). The protein
pellet was further purified by microultracentrifugation over a 20% sucrose
cushion to eliminate non-viral proteins.
The total protein concentration was determined using Bio-Rad’s coomassie
blue protein assay.
Protein
quantitation. Total protein concentrations of the purified
solutions were determined using a dye that stains proteins; the resulting color
change can be recorded as an optical density reading on a spectrophotometer. By creating a standard curve with known
concentrations of bovine serum albumin (BSA) and then taking the OD readings of
dilutions of unknown samples, the protein concentration of the unknown can be
determined. To be sure that these
readings are reliable, it is optimal to obtain more than one reading for each
sample being quantified in the form of multiple dilutions (Fig. 1).
Fig. 1: Protein
quantitation of a viral preparation from infected cell supernatant. X’s denote the standard curve. o’s denote unknown samples.
Western
Blot Analysis. Virions in the preparation sample were
disrupted in 2% sodium dodecyl sulfate (SDS), 0.1 M dithiothreitol (DTT), 0.5M
Tris (pH 6.8), and resolved on a 10% polyacrylamide gel by SDS-polyacrylamide
gel electrophoresis (SDS-PAGE). Gels
were equilibrated in the appropriate methanol containing transfer buffer and
transferred to a PVDF membrane using Bio Rad’s semi-dry electrophoretic
transfer cell at 15 volts for 15 min.
After transfer, the membranes were blocked with 5% nonfat dry milk and
probed with a dilutional series (1:20 – 1:500) of serum from a PCR+ cougar
using the Bio-Rad multiscreen apparatus.
After washing 3 times in TBS/0.1% Tween for 15 minutes each time, the
membranes were incubated with horse radish peroxidase conjugated goat anti-cat
IgG (1:2000), washed again, and developed with Amersham’s ECL plus western blot
detection kit.
Controls.
PCR was used to confirm that seropositive cougars were actually infected
with FIV. For a positive control, serum
from two FIV+ Yellowstone male cougars was used (YM133 and
YM139). A male cat from the Snowy Range
whose serum tests negative via PCR was used as a negative control (SR635). An additional negative control was serum
from a domestic cat that tests negative via PCR is (UI314).
Results
Viral
preparation. Supernatant from FIV infected cells was
collected and either briefly centrifuged or sterile filtered to remove protein
aggregates and cellular components. The
filtration method was chosen because in addition to being faster, it seemed to
remove more unwanted debris. The
virions in the resulting supernatant were then purified by ultracentrifuge as
described in materials and methods. Gp
120 is the virus envelope and is easily shed.
Retention of this protein is important because this is the antigen
detected by flow cytometry analysis of serum samples, and one of the goals of
this project is to compare immunoblot and flow cytometry results. Two approaches were taken during virus preparation
to retain gp120. First, a spin column
with a filter that allows soluble proteins under 100 kilodaltons to pass
through was used. This results in a
concentrated solution containing intact virions and any disrupted gp120. Second, the resuspended pellet was placed
onto a 20% sucrose cushion and centrifuged at 100,000xg. This resulted in smaller weight proteins
staying in solution while virions were pelleted. These two methods were to observe which gave a cleaner
preparation while retaining gp120; pelleting over a gradient can shear gp120
from virions but may provide a cleaner preparation. Choice of resuspension buffer was also important to consider;
since Tris is present in both the sample buffer and the gel matrix, TBS was
used to resuspend the pellets to provide the proteins with similar
conditions. While both methods resulted
in the appearance of viral proteins on a blot including gp 120, the sucrose
cushion was chosen because all the viral proteins were detectable and the total
protein yield was higher.
Optimizing the Western Blotting.
To obtain good images on the immunoblots, three aspects need to be
considered: the amount of protein being separated in the gel, the amount of
serum being applied to the membrane, and the amount of secondary antibody being
used. It is necessary to first
determine if viral proteins could be detected on a stained SDS-gel. To observe this, the denatured proteins were
separated via SDS-polyacrylamide gel electrophoresis (PAGE) and the gels were
then stained with coomassie blue, a dye that specifically stains proteins. When treated with a destaining solution, the
gels lose the blue color while the proteins retain the dye and their separation
by molecular weight can be observed (Fig. 2).
Fig. 2:
Stained SDS-PAGE gel of separated and denatured viral proteins. Lanes 3 and 5 contain FIV infected cell
preparations. Lanes 4 and 6 contain
preparations from uninfected cells. The
bands in lanes 3 and 5 that do not appear in lanes 4 and 6 correspond to viral
proteins p24 and p55.
Optimal serum and secondary antibody
dilutions were determined by probing lanes of a membrane containing five
dilutions of serum from FIV+ cougars YM 133 and YM139, and three dilutions of
secondary antibody (Fig.3). For both
YM133 and YM139, a dilution of 1:2500 appeared to be too dilute, but at 1:250,
much background was observed. While
continuing to optimize conditions, a dilution of 1:500 of serum was used. A dilution factor of 2000 was determined
appropriate for the secondary antibody.
Fig. 3: Optimized immunoblot conditions. Lanes 1-6, 7-12, and 13-18 were probed with
1:1500, 1:2000, and 1:2500 dilutions of secondary antibody, respectively. Lanes 1, 7, and 13 were probed with secondary
antibody only as a control for non-specific binding. Lanes 2-5, 8-11, and 14-17 were probed with a dilutional series
of YM139 serum (1:50, 1:250, 1:1250, and 1:2500) and secondary antibody. Lanes 6, 12, and 18 were probed with
uninfected cat serum (1:100) and secondary antibody as a negative control. The molecular size marker is in the
leftmost lane.
To determine the optimal protein
concentration to use in the assay, two duplicate gels were loaded with
dilutions of the virus prep. The entire
membrane was probed with a 1:500 dilution of YM133 serum and a 1:2000 dilution
of secondary antibody. The second
membrane was probed with a 1:50 dilution of FIV- UI314 serum and a 1:2000
dilution of secondary antibody (fig.4).
When a membrane was probed with FIV+ cougar serum, bands corresponding
to gp120, gp55, and gp24 appeared; proteins weighing 17 kilodaltons and less
were run off the gel and therefore can not be observed on this image. The
appearance of background bands on the membrane exposed to uninfected cat serum
suggests that this is non-specific recognition of non-viral proteins.
Fig. 4:
A comparison of four varying concentrations of a viral prep. (2mg, 1mg, 0.5mg, and 0.25mg).
The membrane on the left was probed with a 1:500 dilution of YM133 serum
and secondary antibody. The membrane
on the right was probed with a 1:50 dilution of UI314 serum and secondary
antibody. The molecular size marker is
the leftmost strip. Note the
recognition of viral proteins gp120, p55, and p24 by YM133.
Discussion
Chronically
infected individuals like YM133 and YM139 seem to react with all of the viral
epitopes; however, this is not necessarily true of all cougars infected with
FIV.
It is expected that recognition of certain viral proteins
can be detected using immunoblot techniques and that trends in serum reactivity
may be noticed. For example, an
individual’s response to FIV may change over time. A newly infected kitten may generate a different antibody
response than an adult. While not all
of the available serum samples from the Yellowstone National Park and Snowy
Mountain Range wild cougar populations have been assayed by immunoblot, the
conditions are optimized such that six to twelve animals can be tested a
week. Three dilutions of each cougar
will be used to probe the PVDF membrane to be sure to detect serum
reactivity. These will not be as dilute
as the positive controls because if the serum from these animals contains fewer
antibodies, then there may not be as strong a response in them as there is in
YM133 and YM139. Once data analysis
begins trends in serum reactivity of animals of different ages and genders in
the two populations will be analyzed.
Changes in antibody response in a single individual for whom there are
multiple serum samples taken over time will be evaluated. In addition, the antibody response of age
and gender matched individuals will be compared. In the end, a table will be
generated containing immunoblot data on these two populations and illustrating
possible trends.
Acknowledgements
This work
was funded by an IBS-CORE undergraduate research fellowship to Jesse Thompson
through a grant from the Howard Hughes Medical Institute to the University of
Montana. Additional funding was
provided by grants to Dr. Mary Poss at the University of Montana.
Sally
Painter conducted preliminary flow cytometry analysis and Roman Biek conducted
PCR in the Poss lab at the University of Montana. Additional thanks go to the Dr. Stephen Lodmell/ Dr. Walter Hill
lab at the University of Montana for the use of ultracentrifuges.
Additionally, Dr. Mary Poss,
Sally Painter, Roman Biek, David Holley, Rita Luther, and everyone else in the
Poss lab often aided in this research with constructive criticism, thoughts and
ideas, and helpful advice.
Literature
Cited
Dow,
S.W., M. Poss, and E.A. Hoover. Feline Immunodeficiency Virus: A Neurotropic
Lentivirus. J. Acq. Immun. Def. Synd. 1990. 3(7):658-668.
Olmsted,
R.A., R. Langley, M.E. Roelke, R.M. Goeken, D. Adger-Johnson, J.P. Goff, J.P.
Albert, C. Packer, M.K. Laurenson, T.M. Caro, L. Scheepers, D.E. Wildt, M.
Bush, J.S. Martenson, and S.J. O’Brien. Worldwide Prevalence of Lentivirus Infection
in Wild Feline Species: Epidemiologic
and Phylogenetic Aspects. J. Virology 1992. 66(10):
60008-6018.
Yuhki,
N., and S.J. O’Brien. DNA Variation of the Mammalian Major
Histocompatibility Complex Reflects Genomic Diversity and Population
History. Proc. Natl. Acad. Sci. 1990. 87:836-840.