Departments of
* Molecular Microbiology and Immunology and
Veterinary Pathobiology, University of Missouri, Columbia; and
The International Livestock Research Institute, Nairobi, Kenya
Correspondence: D. Mark Estes, Department of Veterinary Pathobiology, University of Missouri, Columbia, MO 65211. E-mail: EstesD{at}missouri.edu
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: CD40L • trypansome • humoral immunity • VSG • KLH • PBS-T
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Because of its extracellular location, the humoral branch of the immune system dominates the response against T. congolense. Polyclonal B-cell activation is a hallmark of infection, as is elevated immunoglobulin (Ig)M, in bovine and murine trypanosomosis [4 5 6 7 ]. Consistent with this idea, studies performed in BALB/c, nude, and severe combined immunodeficiency (SCID) mice suggest that athymic mice (lacking T cells) can respond to the variable surface glycoprotein (VSG) and under certain conditions, control infection [8 ]. SCID mice lacking B cells and T cells show a relatively high degree of susceptibility versus normal or athymic mice [8 ]. Despite the production of trypanosome-specific antibodies, a large fraction of the IgM that is produced in response to infection is polyspecific and/or autoreactive [4 , 10 11 12 ]. One of the major epitopes to which antibodies are formed during trypanosome infection is the VSG. The importance of VSG-specific antibodies in parasite clearance and protection against infection are supported by a variety of independent studies [13 14 15 16 17 18 ]. Differences have been found to exist between N’Dama and Boran cattle regarding the serum antibody titers and the frequency of cells that secrete antibody reactive against VSG and other trypanosome antigens [2 , 12 , 19 , 20 ]. Although trypanosusceptible Boran animals have been shown to produce more IgM specific for buried and linear epitopes of VSG, the cysteine protease, congopain, and the heat shock protein, (hsp)70, as well as nontrypanosome protein, ovalbumin than do N’Dama cattle, N’Dama cattle are found to produce more IgG1 [2 , 3 , 12 , 21 , 22 ]. This suggests that Boran animals infected with T. congolense are not as effective as trypanoresistant N’Dama cattle at isotype switching to IgG1. This finding may be potentially attributed to the higher level of interleukin (IL)-4 produced by infected N’Dama animals compared with that produced by Boran animals [23 ]. Thus, IL-4, which promotes B-cell switching to IgG1 in cattle [24 ], may play a protective role in the immune response to T. congolense through its ability to stimulate the production of higher specificity antigen-reactive IgG1 by B cells in the presence of CD40L stimulation.
It has been shown previously that in response to T. congolense infection, the percentage of CD5+ B cells in the spleen and blood of cattle becomes increased dramatically [25 ]. The expansion of the CD5+ B-cell subset correlates with the increase in low affinity (often nonspecific), self-reactive serum IgM antibodies, and increased IL-10 transcription that is observed in N’Dama and Boran cattle [26 , 27 ]. IL-10, regardless of what cell population produces it, may play a role in causing the broad immunosuppression that is seen during trypanosome infection. A recent study has demonstrated that compared with preinfection levels of peripheral blood mononuclear cells (PBMCs), CD40L mRNA expression becomes decreased during T. congolense infection of trypanosusceptible and trypanoresistant breeds [23 ]. This broad degree of immunosuppression and the potential lack of CD40L expression as a consequence of the general suppression of T-cell responsiveness during T. congolense infection potentially hinder T-cell-dependent isotype switching by B cells.
We sought to investigate the effect of CD40 stimulation on the bovine immune response generated against T. congolense. Given the difficulty of conducting this type of experiment in cattle, a model of T. congolense infection in SCID-bovine chimeric (SCID-bo) mice was used. The data presented herein characterize the immune response generated during T. congolense infection in the SCID-bo mouse and suggest CD40 stimulation provides protection during trypanosome infection.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
II/IIIR (CD16/CD32; Fc Block, Pharmingen). Mice with
percentages of bovine cells at 2% or greater than negative
isotype-control staining were considered reconstituted and used for
experimental purposes.
T. congolense preparation
A frozen African isolate stock of T. congolense
(IL3000) was generously supplied by Dr. John Donelson (University of
Iowa, Iowa City, IA). The stabilate was injected intraperitoneally
(i.p.) into SCID mice to propagate parasites. Upon the development of
high-level parasitemia, blood was collected to harvest parasites. Mice
were exsanguinated from the retrobulbar sinus under anaesthesia
(Metophane, Schering-Plough, Union, NJ). Blood samples containing
parasites were diluted 1:1 with phosphate saline glucose (PSG) freezing
solution containing 20% glycerol (5.4 g/L
Na2HPO4, 0.24 g/L NaH2PO4, 1.7 g/L
NaCl, 10 g/L glucose) and were frozen and stored in liquid nitrogen.
Parasite infection of SCID-bo mice
To determine the optimal dose required for T.
congolense infection of SCID-bo mice, they were injected i.p. with
5, 50, or 200 parasites in 100 µl phosphate-buffered saline (PBS) and
monitored for parasitemia and survival. For all infections that
followed, mice were infected with 
80 parasites (diluted in 100 µl
sterile PBS) via i.p. administration. Of note, the viability of
parasites used in SCID-bo infections was assessed prior to each
experiment by microscopy. SCID-bo mice were injected with 50–75 µg
protein G-purified (Pierce, Rockford, IL) agonistic mouse anti-bovine
CD40 [1LA156 (IgG1), ILRI, Nairobi, Kenya] or mouse IgG1 (MOPC 21,
Sigma Chemical Co., St. Louis, MO) on days -1, 0, and 1 post-parasite
inoculation. Seven days postinfection, mice were treated with 200 µl
1.12 mg/ml Berenil (diaminazene aceturate; Sigma) i.p. This represents
a subcurative dose, because parasitemia eventually redevelops in
SCID-bo mice. Fourteen days postinfection, mice were reinfected with
80 parasites and administered the same anti-CD40 or mIgG1 treatment
as described above (on days -1, 0, and 1 of second infection). Seven
days post second infection, mice were again treated with Berenil. Blood
samples were acquired at weekly intervals to monitor parasitemia.
In all experiments, blood samples collected in heparin obtained by retro-orbital bleeding were centrifuged at 900 g for 20 min, and plasma was removed and stored frozen at -20°C. In some cases, the remaining blood fraction was diluted in 400 µl Hanks’ balanced saline solution (HBSS), overlaid on top of 500 µl Accupaque solution, and centrifuged at 900 g for 20 min. Buffy coats were removed and washed to obtain PBMCs, which were further analyzed by flow cytometric analysis or were used for RNA isolation. A small fraction of blood was also used for determining parasitemia, in which case blood was typically diluted 100-fold in PBS, and parasites were counted using a hemocytometer under 40x ocular lense magnification. Mice exhibiting high levels of parasitemia (107–108) were anaesthesized and sacrificed. Spleens were harvested, ground through stainless steel mesh, passed through 70 µm nylon mesh, overlaid on an Accupaque gradient, and centrifuged to obtain a buffy coat. Spleen buffy coats were harvested, washed in HBSS, pelleted, snap-frozen in liquid N2, and stored at -80°C until subjected to RNA extraction.
Flow cytometric analysis
SCID-bo mice were bled under anaesthesia as previously
indicated. PBMCs were isolated by diluting blood 1:1 in room
temperature-warmed HBSS and overlaying 400 µl diluted blood upon
500 µl Accupaque in a 1.5 ml microcentrifuge tube. Samples were
centrifuged at 900 g 20 min at room temperature. Buffy coats
containing PBMCs were harvested and washed. Prior to staining,
nonspecific binding of antibodies to Fc receptors was blocked by
preincubation of cells with a blocking antibody reactive against
FcII/IIIR (CD16/CD32; Fc Block). Cells were stained with the
following antibodies in the following order: anti-CD45 (CACTB51A, WSU
mAb Center), anti-mouse IgG2a-FITC (Sigma Chemical Co.), mouse
anti-bovine CD5 (CC17, generously supplied by Dr. Chris Howard, IAH,
Compton, UK), anti-mouse IgG1-PE (Becton-Dickinson, San Jose, CA),
biotinylated anti-bovine IgM (BM23, Sigma Chemical Co.), and
streptavidin-allophycocyanin (SA-APC; Pharmingen). Bovine PBMCs were
used as a positive triple-stain control, and single stains of bovine
PBMCs were used to adjust compensation during analysis of triple
stains. Negative triple-stain controls of SCID-bo PBMCs were performed
using a combination of mouse IgG2a (Sigma Chemical Co.), anti-mouse
IgG2a-FITC, mouse IgG1 (Sigma Chemical Co.), anti-mouse IgG1-PE, and
SA-APC. The triple-negative staining control was used to set cursors
for quadrant plot analysis. Cells were stained and washed in
fluorescein-activated cell sorter (FACS) staining buffer containing 1%
bovine serum albumin (BSA) and 0.1% sodium azide azide and were mixed
in 2% PBS-buffered paraformaldehyde. Cells were analyzed using a FACS
Vantage flow cytometer and Cellquest acquisition and analysis programs
(Becton-Dickinson). To obtain fetal lymphocyte populations, fetal
tissues (spleen, liver, and mesenteric lymph node) were macerated, and
cells were isolated over an Accupaque gradient. Isolated cells were
washed in PBS and stained using mouse anti-bovine CD5 (CC17, generously
supplied by Dr. Chris Howard, IAH), rat anti-mouse IgG1-PE
(Becton-Dickinson), and goat anti-bovine IgM-FITC (Kirkegaard and Perry
Labs, Gaithersburg, MD).
Reverse transcriptase-polymerase chain reaction (RT-PCR) of SCID-bo
tissues
RNA was extracted according to the manufacturer’s instructions
(RNeasy RNA isolation kit, Qiagen, Chatsworth, CA), and DNase was
treated with RQ1 DNase (Promega, Madison, WI). DNase-treated RNA was
phenol:chloroform-extracted, ethanol-precipitated, and quantitated
using a UV spectrophotometer. RNA was then used in RT reactions (1
µg/reaction), according to the manufacturer’s instructions (Perkin
Elmer, Foster City, CA). PCR was then conducted on cDNA samples using
primers specific for bovine glyceraldehyde 3-phosphate dehydrogenase
(G3PDH; Genbank accession #AF077815): forward (nt 1–20),
5'-GGAAGCTCACTGGCATGGCC-3'; reverse (nt 274–294),
5'-CCCTGTTGCTGTAGCCAAA-3', and IL-10 (Genbank accession #U00799):
forward (nt 81–102), 5'-AGCTGTATCCACTTGCCAACC-3'; reverse (nt
434–454), 5'-TCTCTTGGAGCTCACTGAAG-3' (40 cycles/reaction). Because of
the high degree of homology between the murine and bovine housekeeping
gene, G3PDH, identification of a restriction-length fragment
polymorphism was necessary to differentiate between bovine and murine
G3PDH cDNA. An MspI (i.e., HpaII) restriction
site was identified in the murine G3PDH sequence that was absent in the
corresponding bovine G3PDH cDNA. Digestion of primed G3PDH RT-PCR cDNA
product yields a murine cDNA product of 225 bp compared with an
undigested bovine G3PDH fragment of 294 bp. MspI-restriction
digestion of cDNA generated from G3PDH-primed RT-PCR reactions was
performed for 1.5–2 h at 37°C. IL-10 RT-PCR samples as well as
digested G3PDH RT-PCR samples were analyzed by resolution on an
ethidium bromide-stained 2% Tris-acetate EDTA (TAE) agarose
gel, followed by Southern hybridization. Southern hybridization was
performed by capillary transfer of nucleic acids to nylon membrane
(GeneScreen Plus II, NEN, Boston, MA), followed by UV cross-linking.
Membranes were prehybridized in PerfectHybPlus containing
heat-denatured herring sperm DNA (100 µg/ml final; Sigma Chemical
Co.) for 4 h at 47°C, followed by addition of
32P-labeled probes overnight, generated using cloned bovine
fragments of bovine G3PDH (nt 1–294), IL-10 (nt 81–454), and a
random-primed DNA labeling kit (Boehringer Mannheim, Indianapolis, IA).
Blots were washed three times in 2 x saline sodium citrate (SSC)
at room temperature and exposed to film (X-OMAT AR, Eastman Kodak Co.,
Rochester, NY) in the presence of intensifying screens. Where
indicated, densitometric analysis was performed on autoradiographs
using Kodak digital imaging/analysis software (Eastman Kodak Co.).
Histology
Spleens fixed in 10% buffered formalin (Fisher Scientific, St.
Louis, MO) were embedded in paraffin, and 5 µm sections were cut. A
portion of sections were stained with hematoxylin and eosin, and others
were individually and immunohistochemically stained. Sections to be
used for immunostaining were pretreated with citrate buffer, pH =
6 (DAKO, Carpinteria, CA) in a steamer. Spleen sections were stained to
detect CD3 (rabbit anti-human CD3; 1:600, DAKO) and bovine IgM/IgG/IgA
(1:200; mouse anti-bovine IgM, IgG, IgA, Serotec, Raleigh, NC) using an
avidin-biotin-peroxidase system (LSAB-2 kit, DAKO) and
peroxidase-3,3'-diaminobenzidine (DAB) development.
Dot blot antibody detection
Plasma from SCID-bo mice was diluted (1:25 in PBS) and blotted
onto a pre-wetted nitrocellulose membrane using a dot blot manifold to
detect total antibody present in plasma. Membranes were then pre-wet in
PBS and blocked for 1–2 h at room temperature in Superblock (Pierce,
Rockford, IL). Following brief washes in PBS + 1% Tween (PBS-T),
membranes were incubated with goat anti-bovine IgM-horseradish
peroxidase (HRP; diluted 1:1000 in PBS-T; Kirkegaard and Perry Labs) or
goat anti-bovine IgG-HRP (diluted 1:2000 in PBS-T; Jackson
Immunoresearch, West Grove, PA) for 30 min at room temperature.
Following 45 min of vigorous washing in PBS-T, blots were developed
using enhanced chemiluminescence (ECL+kit, Amersham, Piscataway, NJ),
and signals were detected using Kodak X-OMAT AR film.
Antigen-specific antibody was detected in SCID-bo plasma in a similar manner. Antigen [100 ng IL3000 VSG, ovalbumin (Sigma Chemical Co.), or keyhole limpet hemocyanin (KLH; Pierce)] was blotted onto pre-wetted nitrocellose. Membranes were dried, blocked, and incubated with diluted SCID-bo plasma (1:25–1:50) for 1 h at room temperature. Following 45 min of vigorous washing, blots were treated as above (i.e., incubation with anti-bovine IgM- or IgG-HRP followed by washing and addition of substrate).
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
View this table: [in a new window] |
Table 1. Representative Assessment of the Percentage of
boCD45+ Cells by Flow Cytometry in SCID-bo Mice for Various
Treatment Groups from Fetal Tissue Donors 1 and 2 at 8 Weeks
Post-Surgery
|
 View larger version (17K): [in a new window] |
Figure 1. Infection of SCID-bo mice with T. congolense. SCID-bo
chimeric mice (two per group) were infected with 200, 50, or 5
parasites i.p. and monitored for the development of parasitemia over
time. Parasitemia was determined by taking a small blood sample via
retro-orbital bleed under anaesthetic and counting parasites under
400x magnification by phase contrast microscopy.
|
80 T. congolense parasites, and on days -1, 0,
and 1 postinfection, 50 µg anti-CD40 (ILA-156), an agonistic
antibody, or control mouse IgG1 was administered i.p. One week
following the start of infection, mice were treated i.p. with Berenil,
a drug that exhibits trypanocidal activity. Mice were treated with a
subcurative dose of Berenil at 7 days postinfection to prolong survival
such that the immune response generated against the parasite infection
would have sufficient time to develop. Seven days post-Berenil
treatment, mice were reinfected with 80 parasites and administered
antibody as performed during the primary infection. Seven days
following this, mice were again Berenil-treated and monitored for the
development of parasitemia and survival. As shown in Figure 2
, 27 days following the start of the experiment in which SCID-bo
mice from donor 1 was used (14 days following the second infection),
mice began to die as a result of infection. It is interesting that by
30 days post-primary infection, 5/6 anti-CD40-treated mice had
survived, and only 2/6 mice receiving the control isotype (mIgG1)
remained, suggesting that anti-CD40 treatment afforded mice some degree
of protection. At this timepoint (30 days), however, remaining SCID-bo
mice in both treatment groups began to succumb to parasitemia and were
thus sacrificed for analysis. In a second experiment (fetal tissues
from SCID-bo mouse donor 2), a similar infection and treatment schedule
was used in which 75 µg anti-CD40 or mouse IgG1 control antibody was
administered to mice i.p. Mice were monitored for survival or
parasitemia levels (unpublished results) greater than 1 x
108 parasites/ml, at which time mice were sacrificed. A
survival plot for this experiment (SCID-bo mice with fetal tissue
implants from donor 2) is shown in Figure 2
. Similar to the results
obtained for the infected SCID-bo mice constructed using donor 1, mice
constructed from donor 2 displayed enhanced survival (or maintained
lower levels of parasitemia) when anti-CD40 treatment was administered.
The difference in survival kinetics observed between the first and
second experiment may be attributable to such factors as donor
variability, reconstitution efficiency, and/or parasite viability. Of
note, in a third experiment, the effect of withholding Berenil
treatment was demonstrated to result in the earlier development of
parasitemia. In this particular experiment, a difference in survival
among antibody treatments was not observed (unpublished results). Thus,
in the absence of the trypanocidal drug, infection ensues at a more
rapid pace, potentially obscuring any protective effect antibody
treatment may have in the SCID-bo mouse. However, in combination with
Berenil treatment that effectively reduces parasitemia, CD40
stimulation elicits a protective effect during trypanosome infection in
SCID-bo mice.
 View larger version (26K): [in a new window] |
Figure 2. Survival curve of SCID-bo mice infected with 80 T.
congolense parasites i.p. Six SCID-bo mice reconstituted with
tissues from fetal donor 1 or donor 2 were treated with anti-CD40 (50
µg in donor 1, 75 µg in donor 2) or the isotype control mouse
IgG1(mIgG1) i.p. on days -1, 0, and 1 postinfection. Seven days
postinfection, mice were treated with Berenil to reduce parasite load.
Seven days following treatment (14 days postinfection), mice were again
infected with 80 parasites i.p. Antibody treatment was administered
at days -1, 0, and 1 post second infection (i.e., days 13, 14, and 15
post initial infection). Seven days following the second infection
(i.e., day 21), mice were again treated with Berenil to reduce parasite
load. Mice were monitored for development of parasitemia and survival.
At day 30 (donor 1) or day 28 (donor 2), remaining mice were sacrificed
for analysis. In the experiment shown in the donor 1 graph, six mice
were used per treatment group, and in the experiment shown in the donor
2 graph, the anti-CD40 group consisted of seven mice, and the mIgG1
isotype control group consisted of eight mice.
|
2%). Lymph node (mesenteric) and liver
(tissues used for SCID-bo implantation), however, are populated by
larger percentages of B cells (10%). It is interesting that the
majority of IgM+ cells found in the lymph node and liver
expresses CD5. T cells are found in the
CD5hiIgM- population, constituting
approximately 20% of the cell population found in the fetal lymph node
and liver.
 View larger version (20K): [in a new window] |
Figure 3. B-cell populations of second trimester fetal tissues. Lymphocyte
populations from spleen, lymph node, and liver tissues from a second
trimester bovine fetus were analyzed for CD5 and IgM expression by flow
cytometry. Fetal tissues were macerated, and a single-cell
suspension was isolated using an Accupaque gradient. Cells were then
stained and analyzed by flow cytometry as described in Materials and
Methods.
|
II/IIIR (CD16/CD32). The level of CD5 and IgM expression of bovine
cells was conducted by limiting the analysis of CD5 and IgM expression
to the bovine CD45+-gated cell population. The
triple-negative staining control was used to set cursors for quadrant
plot analysis. The percentage of total bovine (boCD45+)
IgM+ cells and percentage of bovine
IgM+CD5+ cells are shown for two experiments in
Figure 4 A
. As indicated by the graphs, the majority of bovine
IgM+ B cells in the SCID-bo mouse expresses CD5. This is
similar to what we observed in this study for the bovine fetal liver
and lymph node IgM+ populations (Fig. 3)
as well as the
neonatal B-cell populations of mouse and human [31
,
32
]. No significant differences were observed in the
total percentage of bovine B cells, or the PBMC populations that were
also analyzed in a similar manner at 14 days post initial infection for
tissue donor 1- and tissue donor 2-recipient SCID-bo mice. As
demonstrated in Figure 4A
, similar to the observation made for
pre-infection populations, nearly all (>90%) IgM+ B cells
coexpressed CD5. However, differences were observed in the total
percentage of IgM+CD5+ cells found in
peripheral blood among treatment groups. As indicated by the graph in
Figure 4A
, anti-CD40-treated mice had a higher mean percentage of total
IgM+ bovine cells and IgM+CD5+
cells in peripheral blood than the mIgG1 control group 14 days
postinfection. Because of the number of mice used in the first
experiment shown (anti-CD40-treated, n=4; isotype control
group, n=5), this difference is not statistically
significant using a Student’s t-test with a 95% confidence
interval. However, the difference among treatment groups in the mean
total percentage of IgM+ cells as well as the mean
CD5+ IgM+ cell percentages 14 days
postinfection in the experiment shown for SCID-bo mice from donor
2 were found to be statistically significant (mean percentage
IgM+ cells, P<0.01; mean percentage
CD5+IgM+ cells, P<0.05; Student’s
t-test). Increases in the percentage of bovine B cells in
the peripheral blood of infected SCID-bo mice treatment were also
observed in a third experiment in which SCID-bo PBMCs were analyzed at
22 days postinfection (unpublished results). The percentage of
CD5+IgM+ cells in the SCID-bo mouse among
treatment groups prior to infection was consistently around 40% the
total CD45+ cells in the circulating peripheral blood
pool. The percentage of B cells in individual mice was
increased, but variable at day 22 postinfection in the one experiment
in which this parameter was evaluated (36+18% vs. 24+12%) (donor 1).
 View larger version (32K): [in a new window] |
Figure 4. Peripheral blood B-cell populations of SCID-bo mice on day 0 and days
14 and 15 postinfection. PBMCs were isolated from SCID-bo mice and
analyzed for the expression of bovine CD45, CD5, and IgM. The
percentage of bovine CD45+-gated cells expressing IgM and
the percentage coexpressing IgM and CD5 were determined for individual
mice and were averaged to determine the mean percentage of
IgM+ cells (total B cells) as well as the mean total
percentage of IgM+CD5+ cells (CD5+
B cells). (A) Analysis of B-cell populations from trypanosome-infected
SCID-bo mice at day 0 (pre-infection) and day 14 following anti-CD40
treatment or mouse IgG1 control antibody. The experimental data shown
for SCID-bo mice constructed from fetal donor 1 consist of five mice
receiving 50 µg mIgG1 isotype control antibody and four mice
receiving 50 µg anti-CD40 treatment on days -1, 0, and 1 of
infection and Berenil treatment seven days postinfection. The
experimental data shown for SCID-bo mice constructed from fetal donor 2
consist of six mice receiving 75 µg mIgG1 isotype control antibody
and six mice receiving 75 µg anti-CD40 treatment on days -1, 0, and
1 of infection and Berenil treatment 7 days postinfection. (B) Flow
cytometric analysis of B-cell population PBMCs from noninfected SCID-bo
mice (n=3) prior to anti-CD40 administration and 15 days
following anti-CD40 treatment (75 µg administered i.p. on days 1, 2,
and 3). (A and B) Error bars represent the standard deviation of
calculated means. Statistical analysis was performed using a Student’s
t-test.
|
Immunohistochemical analysis of T. congolense-infected
SCID-bo mice
To determine the effect of anti-CD40 treatment on splenic
architecture and bovine lymphocyte populations within the spleen of
infected SCID-bo mice, formalin-fixed spleen sections were embedded in
paraffin, sectioned, and visually analyzed by hematoxylin and eosin or
immunohistochemical staining. Hematoxylin/eosin stains indicated that
lymphoid cells were indeed localized in the white pulp areas of SCID-bo
spleens of mouse IgG1 control and anti-CD40-treated mice (unpublished
results). To determine the contribution of bovine T cells to the
splenic white pulp, spleen sections were stained with an antibody
specific for CD3. Staining results are shown for the
isotype-control-treated group in Figure 5
as indicated. CD3+ cells were detected in defined
areas surrounding central arterioles throughout the spleen. T cells
were mainly localized around arterioles in T-cell zones, although a few
T cells were found outside these clusters. Subjective analysis of
stained sections indicated no discernible differences in T-cell
numbers. Finally, an antibody specific for bovine IgM, IgG, and IgA was
used to detect the presence of B cells within spleens of infected
SCID-bo mice. The staining results are shown in Figure 5
(
Ig). Fewer
numbers of B cells, relative to CD3+ T cells, were detected
in spleens. However, B cells appeared to be more abundant in
anti-CD40-treated mice compared with the mIgG1 control group. This
finding correlates well with the increase in B cells observed in the
peripheral blood of anti-CD40-treated mice, as detected by flow
cytometry. The paucity of B cells in the spleens of SCID-bo mice,
relative to T cells, has been demonstrated previously in a SCID-hu
model of lymphocyte reconstitution. In this particular study, human B
cells (expressing CD5) were lacking in spleen but were found localized
in the thymus [33
]. An attempt was made to analyze
remnants of thymus but was unsuccessful as a result of the inability to
consistently identify thymic tissue in SCID-bo mice. Alternative
locations of B cells, such as the peritoneal cavity, remain to be
investigated.
 View larger version (119K): [in a new window] |
Figure 5. Detection of T and B cells in spleens of T.
congolense-infected SCID-bo mice from fetal tissue donor 2.
Sections of paraffin-embedded spleens from control (mIgG1) or
anti-CD40-treated SCID-bo mice were stained with a species
cross-reactive rabbit anti-human CD3 polyclonal antibody (DAKO) or a
mouse monoclonal antibody specific for bovine IgM, IgG, and IgA
(Serotec) and developed using an avidin-biotin-peroxidase system and
DAB as a substrate. Representative stains are shown at 40x original
magnification.
|
 View larger version (34K): [in a new window] |
Figure 6. Detection of IL-10 mRNA in T. congolense-infected SCID-bo
mice. (A) RNA was extracted from the spleens or PBMCs of infected
SCID-bo mice at the time of sacrifice and analyzed by RT-PCR/Southern
blotting. Following DNase treatment, extraction, and precipitation, 1
µg mRNA was used in a reverse transcription reaction. Bovine-specific
IL-10 primers and G3PDH primers were then used to prime and amplify
reverse-transcribed cDNA. G3PDH cDNA was subjected to MspI
digestion to cleave mouse but not bovine G3PDH cDNA products. PCR
products were then resolved on a 2% agarose gel and blotted onto nylon
membrane. Membranes were probed with 32[P]-labeled bovine
IL-10 and G3PDH DNA probes and developed by autoradiography. Results
are shown for splenic mRNA isolated from individual SCID-bo mice
reconstituted from fetal donor 1 and donor 2. The RT-PCR Southern blot
for pooled PBMC mRNA for SCID-bo mice reconstituted from donor 2 is
also shown. (B) Quantitation of the ratio of IL-10 signal to G3PDH
signal for experiments 1 and 2. Densitometric analysis was performed on
the autoradiographs for SCID-bo spleen and PBMC samples of experiments
1 and 2 shown in (A). Using a Kodak digital analysis system, the net
intensities of IL-10 and G3PDH bands were determined. The ratio of the
IL-10 to G3PDH signal was determined for each individual mouse spleen
sample, averaged for each treatment group in experiments 1 and 2, and
graphed. The ratio of the IL-10-to-G3PDH signal was also determined for
the pooled PBMCs and plotted. Error bars represent the standard
deviation of the mean.
|
 View larger version (28K): [in a new window] |
Figure 7. Detection of SCID-bo antibody by dot blot analysis. (A) Total
antibody present in SCID-bo plasma at 14 days postinfection and at the
time of sacrifice. Plasma from SCID-bo mice was pooled and diluted
(1:25 in PBS) and 10 µl blotted onto pre-wetted nitrocellulose.
Membranes were blocked in Superblock, washed, and incubated with goat
anti-bovine IgM-HRP or goat anti-bovine IgG-HRP. Following vigorous
washing in PBS-T, blots were developed using ECL + kit, and
signals were detected using Kodak X-OMAT AR film. Results are shown for
two experiments in which 50 µg (exp. 1)/75 µg (exp. 2) of anti-CD40
(ILA156) or mouse IgG1 was administered to mice as described in
Materials and Methods. (B) Detection of VSG and nontrypanosome
antigen-specific antibody responses in SCID-bo mice. VSG (100 ng),
ovalbumin (OVA), or KLH was blotted onto nitrocellulose. Membranes were
blocked (Superblock) and incubated with pooled plasma diluted 1:50
(upper blot) or 1:25 (lower blot). Bound antibody was detected as
described in (A). Data are shown for plasma taken from anti-CD40 or
mIgG1-treated SCID-bo mice 14 days following parasite infection (7 days
following Berenil treatment). Data are also shown for plasma from
noninfected SCID-bo mice receiving anti-CD40 treatment (d14). Normal,
noninfected C.B-17 serum was used as a negative control. Results are
representative of independent experiments that used mice reconstituted
with fetal tissues from two individual donors.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To begin to study the B-cell responses that occur in this model during trypanosome infection, we sought to first analyze the B-cell populations (CD5+ or CD5-) present in the SCID-bo mouse. An important finding shown here is that the overwhelming majority of bovine IgM+ B cells in the SCID-bo mouse expresses CD5. Analysis of bovine fetal tissues indicated that a significant portion of the IgM+ B-cell populations in the second trimester bovine fetal liver and lymph node express CD5. This is similar to what has been demonstrated for human fetal liver and lymph node [38 , 39 ]. Fetal spleen tissue, however, was virtually devoid of B cells during the second trimester of gestation. Given that fetal thymus, liver, and lymph node tissues are used to reconstitute the immune repertoire of SCID mice, it is not surprising to discover that the majority of bovine B cells reconstituting the peripheral blood lymphocyte population of the SCID mouse expresses CD5. Thus, this reconstituted B-cell population may resemble that of the neonate. The process and rate of maturation of the immune system within the reconstituted SCID-bo mouse in comparison to that which occurs in the neonatal calf remain to be established.
Initially we had intended to use the SCID-bo model to study the conversion of a bovine CD5- B-cell phenotype to a CD5+ phenotype as a consequence of T-cell-independent type 2 (TI-2)-antigenic B-cell activation that may potentially occur during trypanosome infection [8 , 40 ]. However, given the finding that the majority of SCID-bo B cells expresses CD5 prior to infection, we were unable to explore this process of B-cell differentiation using the SCID-bo model. However, the findings obtained using this model to study trypanosome infection were nonetheless exciting. Although no difference among treatment groups was observed with respect to the percent of IgM+ cells expressing CD5, increases in the total percentage of IgM+ B cells and CD5+IgM+ B cells in the peripheral blood of SCID-bo mice treated with anti-CD40 antibody were consistently observed during T. congolense infection. Importantly, in the absence of infection, anti-CD40 treatment was not observed to increase the percentage of B cells in the peripheral blood of SCID-bo mice. Thus, CD40 stimulation did not decrease the relative percentage of B cells expressing CD5 but actually served to expand the overall percentage of CD5+IgM+ B cells as well as the overall percentage of CD5-IgM+ B cells, given that the percent of IgM+ cells, which expresses CD5, remained relatively consistent. This finding suggests that anti-CD40 treatment, although unable to reverse the development of CD5+ B cells, did indeed stimulate B cells (CD5+ and CD5-) to expand in vivo concomitant with parasite infection. It is interesting that similar to anti-CD40-treated SCID-bo mice, trypanoresistant N’Dama cattle are shown to have a higher proportion of B cells during infection than similarly infected susceptible Boran cattle [2 ].
Consistent with studies of high levels of IL-10 production in murine and bovine trypanosomosis [26 , 30 ], IL-10 mRNA expression in the spleen and PBMCs of infected SCID-bo mice was elevated. Previously, IL-10 has been suggested to play a role in the immunosuppression observed during trypanosomosis [30 , 37 ]. Macrophages have been implicated in the production of IL-10 during trypanosome infection [41 ], and T cells and CD5+ B cells may potentially produce IL-10 as well. However, the cell type(s) responsible for producing IL-10 during trypanosome infection in SCID-bo mice remains to be determined.
Importantly, this study demonstrates that IgG antibody specific for VSG is produced by T. congolense-infected SCID-bo mice. Although IgM antibodies reactive against VSG, as well as ovalbumin and KLH, were present in the plasma of uninfected and infected SCID-bo mice, IgG-reactive antibody was only present in T. congolense-infected animals. The presence of VSG-reactive IgM antibody in SCID-bo mice that have not been exposed to T. congolense is not an unexpected finding, because self-reactive antibodies that recognize VSG have been shown previously to exist in the serum of uninfected hosts [34 ]. CD40 stimulation (via ILA156 administration), as expected, was observed to increase the level of IgG detected in the plasma of infected mice compared with mice receiving the isotype control. As stated earlier, this increase in VSG-specific IgG may have afforded mice some degree of protection during parasite infection. Importantly, VSG-reactive antibody was only detected in the plasma of mice 14 days postinfection (7 days post initial Berenil treatment). During high levels of parasitemia, antibody was difficult to detect in the plasma of SCID-bo mice. This is likely a result of the fact that bovine antibody is rather limiting in reconstituted mice and that under conditions of high parasitemia, antibody is rapidly depleted because of high antigen load and clearance.
The results presented for the experiments involving SCID-bo T. congolense infection have interesting implications regarding the SCID-bo model as well as for T. congolense infection. The data indicate that survival of trypanosome-infected SCID-bo mice is prolonged when an agonistic antibody against bovine CD40 (ILA156) is administered. Thus, stimulation of B cells and/or other cell types through CD40 afforded SCID-bo mice a slight degree of protection during T. congolense infection. Although the reasons for this observed effect remain to be determined, it can be speculated that by increasing the level of isotype switching to VSG-specific IgG and by increasing the total number of B cells, CD40 stimulation may have enhanced parasite clearance via overall increased IgG antibody production. Indeed, although trypanosusceptible Boran cattle have been shown to produce more IgM specific for buried and linear epitopes of VSG the cysteine protease, congopain, and hsp70, as well as nontrypanosome protein, ovalbumin than do N’Dama cattle, N’Dama cattle are found to produce more IgG1 [2 , 3 , 12 , 21 , 22 ]. This suggests that Boran animals infected with T. congolense are not as effective at isotype switching to IgG1 as trypanoresistant N’Dama cattle. Thus, similar to what is observed in cattle, isotype selection may play an important role in providing protection during T. congolense infection in the SCID-bo mouse. Additional studies aimed at addressing the role of isotype switching, as well as the factors influencing switching during trypanosome infection are required to develop an understanding as to how cattle may achieve protection via antibody during infection. The SCID-bo model provides a useful tool for studying the development of trypanosomosis in cattle. Through the ability to manipulate the immune response of these heterochimeric mice via antibody or cytokine administration, the factors involved in providing protection during T. congolense infection may be more clearly elucidated.
Received April 7, 2001; revised July 23, 2001; accepted August 3, 2001.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  N. Antoine-Moussiaux, P. Buscher, and D. Desmecht Host-Parasite Interactions in Trypanosomiasis: on the Way to an Antidisease Strategy Infect. Immun., April 1, 2009; 77(4): 1276 - 1284. [Full Text] [PDF]
|
|
|
|
 |
|
 K. C. Lambert, E. M. Curran, B. M. Judy, G. N. Milligan, D. B. Lubahn, and D. M. Estes Estrogen Receptor {alpha} (ER{alpha}) Deficiency in Macrophages Results in Increased Stimulation of CD4+ T Cells while 17{beta}-Estradiol Acts through ER{alpha} to Increase IL-4 and GATA-3 Expression in CD4+ T Cells Independent of Antigen Presentation J. Immunol., November 1, 2005; 175(9): 5716 - 5723. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
