| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Microbiology and Molecular Genetics, Medical College of Wisconsin, Milwaukee, WI 53226
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
As it has been hypothesized that determinant spreading contributes to relapsing disease, there is considerable interest in the T cell repertoire in the CNS throughout the course of EAE. Increasing heterogeneity of the CNS T cell population would be predicted as it is during the relapsing phases that determinant spreading has been detected.
There are multiple ways in which CNS T cell heterogeneity may arise during the course of EAE. Following active immunization using CFA both neuroantigen-specific T cells and mycobacteria-specific T cells are raised in response to immunization. There is consensus that activated T cells, regardless of specificity, are able to readily access the CNS (4, 5, 6, 7, 8, 9). Thus, both neuroantigen-specific and mycobacteria-specific T cells will be found in the CNS. A second source of heterogeneity is the neuroantigen used as immunogen. If a protein is used, there will be a response against multiple epitopes. This is also true, although to a lesser degree, when peptides are used for immunization. Also, the emulsion containing the Ag and mycobacteria persists at the site of injection for weeks or longer, providing a continuing source of stimulation of the immune system. Finally, inflammation at the site of the lesions may render the blood-brain barrier permeable to an influx of host cells, which increases T cell heterogeneity within the CNS.
The use of adjuvant may be avoided by adoptive transfer of encephalitogenic T cell lines to syngeneic recipients. With a few exceptions, the T cell lines used for adoptive transfer were heterogeneous. Expansion of minor populations of T cells in the line after transfer to the recipient may lead to increased donor T cell heterogeneity in the CNS. Because nonspecific T cells from the recipient cross into the CNS during the course of disease, another layer of complexity is added to the CNS T cell population. The use of a well-characterized encephalitogenic T cell clone to induce passive EAE would eliminate heterogeneity contributed by the donor T cell population and allow visualization of the host T cell CNS component.
Recently, we generated several encephalitogenic PLP peptide 139–151-specific T cell clones, and found that by using the CDR3 region of one of these T cell clones, 3-19, as an idiotypic marker, the clone could be tracked in the CNS during the acute and relapse phases of EAE (3, 10). It also allowed ready differentiation of the clone from host-derived T cells, as the marker was unique to the clone.
To assess the degree of heterogeneity within the CNS at different stages of disease following adoptive transfer of an encephalitogenic T cell clone, cDNA prepared from the spinal cord of recipients of T cell clone 3-19 at different stages of disease was amplified using primers for VB gene segments present in the SJL mouse and a CB primer labeled with fluorescein. The amplified products were analyzed by spectratyping (11, 12, 13, 14) to assess CDR3 size heterogeneity.
The spectratyping technique provides a minimal measure of TCR heterogeneity, as each band may be composed of multiple CDR3 regions of equal size but differing compositions. To assess heterogeneity at the single-cell level, it is necessary to clone and sequence the PCR products that form each band. As it would be a very large task to sequence every band in the spectratype for each VB family, we restricted this part of the study to the VB2 band that corresponded to the size of the CDR3 of clone 3-19 for two reasons. First, as the CDR3 sequence of clone 3-19 was known, we could readily differentiate the encephalitogenic CDR3 from other TCR CDR3 regions of this size, and second, it was likely that this band was representative of other CDR3 regions. The band was isolated from the CNS spectratype at different stages of disease and cloned, and the nucleotide sequence was determined. At the acute stage of disease 100% of the sequences obtained were identical with T cell clone 3-19. At remission very few CDR3 sequences were that of the clone, while at relapse, the number of 3-19 sequences obtained was increased relative to the number of heterogeneous sequences. Detailed findings of these studies are reported below.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Female SJL/J (H-2s) were purchased from The Jackson Laboratory (Bar Harbor, ME) and maintained in the Animal Resource Center of the Medical College of Wisconsin (Milwaukee, WI). Mice were used between 8 and 12 wk of age.
Antigens
PLP peptide 139–151 was synthesized by the Nucleic Acid and Protein Core Facility of the Cancer Center of the Medical College of Wisconsin. The amino acid sequence of this peptide is H-S-L-G-K-W-L-G-H-P-D-K-F, where S is substituted for the C found at position 140 in the authentic PLP sequence.
Generation of T cell lines and clones
Female SJL/J mice were immunized with PLP peptide 139–151 (100
µg/mouse) emulsified in IFA supplemented with Mycobacterium
tuberculosis H37Ra (50 µg/mouse). A total of 0.1 ml of emulsion
was injected into four sites on the flanks. At 24 and 72 h
following the initial injections, 400 ng of Bordetella
pertussis toxin (Sigma) was administered i.v. to mice injected
with PLP peptide 139–151. Ten days following initial immunization the
draining lymph nodes were removed, and a single-cell suspension was
prepared and cultured without further separation at 3 x
106/ml, 2 ml/well, in 24-well tissue culture
plates. The culture medium was RPMI 1640 supplemented with 10% FCS,
5 x 10-5 M 2-ME, 100
U/ml penicillin, 100 µg/ml streptomycin, 2 mM
L-glutamine, 10 mM HEPES buffer, and 5 µg/ml
peptide 139–151. After 4 days at 37 C blast cells were isolated on a
Ficoll-Hypaque gradient and resuspended at 2 x
105/ml with irradiated (3000 rad) syngeneic
spleen cells at 1 x 106/ml. Erythrocytes in
spleen cell preparations were lysed with ammonium chloride. Cells were
cultured in the absence of nominal Ag in tissue culture flasks for a
10-day rest period followed by a 4-day restimulation period during
which the surviving cells were cultured at 1 x
105/ml with peptide 139–151 (5 µg/ml) and
fresh irradiated syngeneic spleen cells at 2.5 x
106/ml in 24-well culture plates. Ag-specific T
cells were cloned by limiting dilution in 96-well round-bottom culture
plates from primary cultures of lymph node cells. Irradiated syngeneic
spleen cells were added at 5 x 105/well
along with Ag and 10% Con A supernatant from a rat spleen cell culture
as a source of lymphokines. Positive clones or subclones were
transferred at 
2 wk to 24-well plates for expansion. They were
maintained throughout by subculture with Ag, fresh irradiated spleen
cells (5 x 106/ml), and 10% Con A
supernatant. T cell clones were assessed for monoclonality by PCR
analysis using a set of primers for each VB gene expressed in the SJL
mouse and a B chain constant region primer. Only those clones that
reacted with a single VB primer were used for further study. Clone 3-19
was found to be a VB2, CD4+, Th1 T cell and was
highly encephalitogenic following adoptive transfer to syngeneic
naive mice.
Adoptive transfer
Activated 3-19 T cells (8 x 105 to 1 x 106) were injected i.v. into irradiated (500 rad) female SJL/J mice. Mice were observed daily for clinical signs of EAE throughout the experimental period. Clinical signs of EAE were scored on a scale of 1–4 as previously described (15).
RT-PCR
Total RNA was isolated from spinal cords of recipients of T clone 3-19 and unmanipulated control mice (16). Quantification of RNA was performed by spectrophotometry at 260 nm. For preparation of cDNA 1 µg of total RNA was incubated with 10 µl of 5x buffer, 5 µl of poly(T) (20 µM) 2.5 µl of dNTPs (10 mM), and water to a final volume of 42 µl. The mixture was heated at 68°C for 2 min, then cooled in ice, and 0.5 µl of RNAguard (Pharmacia, Piscataway, NJ), 5 µl of DTT (0.1 M), and 2.5 µl Moloney murine leukemia virus reverse transcriptase were added. The mixture was incubated at 42°C for 45 min. PCR amplification of cDNA was performed as follows. One microliter of cDNA was incubated with 2.5 µl of 10x buffer, 0.5 µl of dNTPs (10 mM), 1 µl of 50 mM MgCl2, 1 µl of primers (20 µM), and water to a volume of 24 µl. The mixture was heated to 95°C, and 1 µl (1 U) Taq polymerase was added. The amplification program was conducted at 94°C for 1 min, the appropriate annealing temperature (55–61°C depending on the Tm of the primers) for 1 min, and 72°C for 1 min. At the end of the program there was a final extension at 72°C for 7 min.
The amplified products were analyzed by agarose gel electrophoresis. To
verify that the products were authentic, they were transferred to a
nylon membrane and hybridized with a 32P-labeled
oligonucleotide with a sequence internal to that of the primers used
for amplification. The oligonucleotides (5'-3') used in this study are
as follows: VB2, ATG AGC CAG GGC AGA ACC TTG TAC; VB3, GAA ATT CAG TCC
TCT GAG GCA GGA; VB4, CTA AAG CCT GAT GAC TCG GCC ACA; VB6, GCC CAG AAG
AAC GAG ATG GCC GTT; VB10, CTT CGA ATC AAG TCT GTA GAG CCG G; VB16, CTC
TGA AAA TCC AAC CCA CAG CAC TGG; and VB17, GAA ATC CTA TCC TCT GAA GAA
GAC. The B chain CDR3 probe for the peptide 139–151-specific 3-19 T
cell clone was TGCAGTGCAAACAGG. The B chain constant region internal
probe was GGCTCAAACAAGGAGACCTTGGGTGGA. The B chain constant region
external primer was CCAAGCACACGAGGGTAGCCT. The CD3
primers were GGA
ACA CAG CGG GAT TCT GG (sense) and CAC CAG CCA TGG TGC CCG AG
(antisense). The hypoxanthine phosphoribosyltransferase (HPRT) primers
were GTT GGA TAC AGG CCA GAC TTT GTT G (sense) and GAT TCA ACT TGC GCT
CAT CTT AGC C (antisense).
In vivo analysis of CNS T cells
At various times after adoptive transfer of the 3-19 T cell clone, mice were sacrificed, and the spinal cord was removed. Total RNA was prepared by the procedure of Chomzynski and Sacchi (16) and was reverse transcribed to prepare cDNA. To assess total CNS T cells, cDNA prepared from mice at different stages of disease was amplified using CD3 and HPRT primers. The CD3 signal was normalized to the HPRT signal from the same sample at each time point. Oligonucleotide primers specific for VB2 and the external B chain constant region were used in PCR amplification of the cDNA obtained from the tissues. cDNA from clone 3-19 was used as a positive control. The integrity of the cDNA prepared was assessed by PCR amplification with oligonucleotide primers for (HPRT). Following amplification with the VB2 and the external B chain constant region primers, the amplification products were transferred to a Nytran membrane (Schleicher & Schuell, Keene, NH) and probed with either a 32P-labeled internal B chain constant region probe or with the 32P-labeled 3-19 CDR3 probe. The blots were then analyzed by a phosphorimager and by autoradiography. cDNA prepared from normal unmanipulated mice served as the negative control.
CDR3 spectratypic analysis
cDNA prepared from the spinal cords of normal mice or recipients of T cell clone 3-19 was amplified using a VB2 primer and a CB primer labeled with fluorescein. The amplified products were then diluted 1/50, added to an equal volume of formamide, and heated at 98°C for 2 min. Seven microliters of the samples were applied to a prewarmed 5% acrylamide sequencing gel. The gels were run at 30 W for 3 h on a Pharmacia ALF automated DNA sequencing apparatus. Analysis of the separated products was performed using Fragment Manager software (Pharmacia). cDNA prepared from normal mouse spleen or thymus served as a positive control. In some instances 20 µl of the undiluted amplified products were separated at 60 W on a 5% denaturing polyacrylamide sequencing gel until the region corresponding to 120–160 bp was near the bottom of the gel. The fluorescein-labeled products were analyzed on a Storm II apparatus (Molecular Dynamics, Sunnyvale, CA), and the data were analyzed with ImageQuant software (Becton Dickinson, Mountain View, CA). Spectratypic analyses were performed on at least three mice at each stage of disease. Representative data are shown in the figures.
Cloning and sequencing of PCR products
Amplified products from the band corresponding to the 3-19 T cell clone were isolated from the acrylamide gel, ligated into the pTAdv vector, and cloned using the AdvanTage PCR cloning kit (Clontech, Palo Alto, CA) according to the manufacturer’s instructions. The plasmid DNA was purified using Qiagen Spin Plasmid Miniprep using the manufacturer’s instructions (Qiagen, Valencia, CA), and sequenced using the Autoread Sequencing kit (Pharmacia) on the ALF DNA sequencer.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
chain. TCR spectratyping of several cDNA
VB families was performed to assess the overall heterogeneity of the
CNS T cell population. Heterogeneity was assessed at the single-cell
level by cloning and sequencing of the TCR CDR3 regions found in cDNA
from mice at different stages of EAE.
The first question to be answered was whether there was a reduction in
total T cell number in the spinal cord during remission. To address
this question cDNA prepared from spinal cord at the acute phase,
remission, and relapse was coamplified using oligonucleotide primers
for CD3 and HPRT. The products were separated on an agarose gel and
analyzed on an Alpha Imager 2000 (Alpha Innotech, San Leandro, CA). The
ratio of the density of the CD3 band at 325 bp to the density of the
HPRT band (165 bp) at each time point was determined. cDNA from thymus
served as a positive control, and cDNA from normal SJL spinal cord
was the negative control. The conditions of the assay were such that
the HPRT signal was not amplified to saturation. Tabulation of the
ratios of CD3 signal to HPRT signal showed that the CD3 signal was
diminished at the recovery stage with respect to the HPRT control,
indicating a reduction in the total T cell population in the CNS (Table I
). The experiment was repeated twice
with similar results. Similar results were found when cDNA from several
other mice at the recovery stage was amplified in this manner.
|
Fig. 1 shows the results of a
spectratypic analysis of spleen T cells from a normal SJL mouse
performed by amplification across the CDR3 region of the B chain of the
TCR using VB family-specific primers and a fluoresceinated CB primer.
The amplified products were then separated on an automated DNA
sequencing apparatus (A.L.F., Pharmacia) and analyzed using Fragment
Manager software (Pharmacia). As the CDR3 regions of different TCR vary
in size, a family of products was generated for each VB family. Each
band represents CDR3 regions of the same size amplified from the cDNA
of one or more T cell clones within the spleen. The bands differ by 3
bp as determined by using standards of known base pair number.
Typically, the spectratypic analysis of normal spleen or thymus T cells
results in a series of six to eight bands of varying intensity, with
the most intense band located centrally, and bands of decreasing
intensity on either side. The distribution of intensities normally
approximates a Gaussian curve. Skewing from a normal distribution is
presumptive evidence for expansion or reduction of a particular subset
of T cells in the immune response.
|
. Each panel represents a different
mouse in which disease was initiated by adoptive transfer of
encephalitogenic T cell clone, 3-19. Upon amplification with a VB2 and
a CB primer, the product of clone 3-19 was 184 bp, as verified by
reactivity with a 32P-labeled 3-19 clonotypic
probe. The CDR3 nucleotide and protein sequence of clone 3-19 was known
(3), making it possible to differentiate donor CDR3
sequences from recipient CDR3 sequences in the 184-bp band. All other
size bands as well as bands generated by amplification with non-VB2
primers must be derived from recipient T cells.
|
A. In each analysis the
same quantity of material was used. Size heterogeneity in each family
is indicated by multiple peaks. It is of interest that there was a
single predominant peak with VB4, -6, and -17 at this stage of disease.
As the spectratype of normal splenic T cells approximated a normal
curve with respect to CDR3 size (Fig. 1
), this implied that there was
selectivity of host T cell entry or expansion in the CNS during this
stage of disease. If migration were a random event, the pattern should
approximate a normal distribution.
|
B. It is
apparent from this panel that there was an overall reduction in TCR
expression during this phase of disease, in agreement with the data
presented in Table I. Only VB17+ TCR appeared to
be undiminished, and, in fact, this family is more heterogeneous than
during the acute phase.
Fig. 3, C and D, shows the results of analyses of
first and second relapses, respectively. It is noteworthy that VB3 TCR
were not present in either mouse at this stage of disease. Each of the
other VB families was more heterogeneous than at the acute phase. In
each instance two or three bands became prominent, in sharp contrast to
the predominance of a single band during the acute phase. As each band
probably contained several different CDR3 regions, the heterogeneity of
the CNS T cell population increased dramatically during the relapsing
stage of disease. Again, the data indicated either selectivity of entry
of certain VB families into the CNS, or selective expansion of certain
families after entry. These findings are representative of several mice
tested in this manner.
Several things are apparent from the spectratypic patterns shown in
Fig. 2. First, during acute disease, the predominant band size (184 bp)
corresponded to that of the donor T cell clone, 3-19. However, one or
more bands of larger size appeared at this stage of disease, which
indicated that a small number of host-derived
VB2+ T cells was present at the initial stage of
disease. Table II shows the percentage of
the total area of each peak in the spectratype. During the acute phase
the 184-bp peak was by far the major peak in each case. Interestingly,
during remission, this peak was reduced from 83 to 26% and from 62 to
12% of the total in Fig. 3, A and B,
respectively. Other bands at 187, 190, 193, and 196 bp increased from
the low levels seen in the acute phase. At relapse the 184-bp band
again increased relative to the larger bands in the spectratype from 26
to 35% in Fig. 3A and from 12 to 32% in Fig. 3B.
|
. Sequence 1
was that of the donor T cell clone, 3-19. Twenty-three of 23 sequences
cloned from PCR-amplified cDNA taken at acute disease were that of the
encephalitogenic T cell clone. None of the other sequences of this size
were found at the acute stage of disease, although a small number of
bands of larger size were noted. During remission the same sequence was
present in 7 of 24 clones, while sequence 2 comprised 6 of 11 clones.
At relapse, 10 of 21 clones were of the original sequence, 1 of 11
comprised sequence 2, and the remaining 9 sequences were unique,
indicating that numerous host-derived T cells of new specificities were
present in the spinal cord at this later time.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Initially, the relative T cell content in the spinal cord at different disease states was determined by PCR analysis using CD3 primers for amplification. The CD3 signal was normalized to the HPRT signal and was found to be stronger during disease stages and reduced at remission. This finding is consistent with our recent report that an encephalitogenic T cell clone was readily detected in the spinal cord during the acute and relapse stages of disease, but was not found during remission (10). Others have reported a decrease in CNS T cell number following acute EAE (17, 18). The mechanism of T cell decline in the spinal cord may be apoptosis (19, 20, 21, 22).
TCR spectratyping has been used previously to analyze T cell populations in the spinal cord of rats (23). T cells were very heterogeneous during both the acute phase and the recovery stage. All VB families were present, and each family consisted of two to seven bands. Since EAE was induced by active immunization with myelin basic protein in complete adjuvant, the number of activated T cells able to enter the CNS would be expected to be quite large and would not necessarily represent only neuroantigen-specific T cells. As it has been reported that encephalitogenic VB8.2+ T cells in the rat contain short TCR B chain CDR3 regions (24), the smallest band of this family was selected for sequencing. Nucleotide sequence analysis revealed that a single short sequence predominated during the early and acute stages of disease (15 of 21), but that this sequence was less predominant at recovery. Unfortunately, neither cDNA prepared from the spinal cords of normal control rats nor that from rats injected with CFA alone was included in the study, so the contribution of T cells activated by mycobacteria in the adjuvant to overall T cell heterogeneity in the spinal cords could not be determined.
In the Lewis rat model of EAE it has been speculated that the encephalitogenic T cells were of early ontogenic origin, as the CDR3 regions of the TCR B chains were very short (24). Interestingly, the SJL T cell clone used in the present experiments also had a very short CDR3 region; however, in the mouse model, the CDR3 regions of the TCR B chains of encephalitogenic T cells are more diverse with respect to size and VB family (10).
It has been hypothesized that epitope spreading is a primary cause of relapsing disease (25, 26, 27, 28, 29, 30, 31, 32). In contrast, there are reports of failure to find convincing experimental evidence for epitope spreading (18, 33). It was hypothesized by the latter that that relapses are due to waxing and waning of the response against the immunodominant epitope, and that apparent epitope spreading is the result of the in vitro stimulation of neuroantigen-specific T cell found in the lymphoid organs. These differences have been ascribed to the different experimental techniques used for disease induction (34).
The use of active and passive EAE models in the investigation of epitope spreading has resulted in confusion, as the Ag depot that remains after active immunization provides a source of continued antigenic stimulation by the neuroantigen as well as mycobacteria over the course of months. In passive disease only the donor effector T cells serve to initiate disease; therefore, it was assumed that new specificities would have to come from sensitization of host T cells. A caveat is that many of the studies of epitope spread to this point have used heterogeneous neuroantigen-specific T cell lines to transfer disease. As these lines may contain minor populations of T cells of specificity other than for the immunodominant peptide, it is possible that increased diversity of T cell specificity resulted from expansion of these minor populations in vivo. In the present study EAE was induced by adoptive transfer of an encephalitogenic peptide-specific T cell clone to eliminate the potential contribution of responses to multiple epitopes associated with the encephalitogenic protein and mycobacteria to heterogeneity of the T cell repertoire in the spinal cord.
The results of the spectratypic analyses are relevant to the issue of epitope spread. The spectratype of normal spleen or other lymphoid tissues normally is seen as a series of several bands in a Gaussian-like distribution around a central prominent band. Although the reason for this distribution is not known, it is assumed that for each VB family a certain size CDR3 region is optimal for the binding pocket in the TCR, and that larger or smaller size CDR3 regions are less frequent. During an immune response, selective expansion of specific T cell clones with different size CDR3 regions may result in skewing of the spectratype. The findings with the VB2 family are of interest. During the acute phase of disease the donor clone was predominant in the spectratype. Later during remission and relapse, multiple bands were found, but they were not typical of the normal distribution. This implies selective expansion of host T cell clones. If there were simply a nonspecific influx of peripheral T cells into the spinal cord due to inflammation, a normal distribution would be predicted.
When other VB families were examined in this manner, the findings were similar. Interestingly, VB3 and VB10 spectratypes were not seen in all mice, again implying a specific response rather than a nonspecific accumulation of T cells in the spinal cord. These findings are in accord with a short previous report of spectratypic analyses of spinal cord T cells in SJL and B10.RIII mice (11). Based on the evidence provided by the spinal cord spectratypes, the possibility of epitope spread as being responsible for the expansion of specific T cell clones is likely. It is unfortunate that due to technical limitations serial samplings of individual mice cannot be analyzed in this manner, as this would be more informative.
As each band may consist of multiple CDR3 regions of identical size, the 184-bp band that corresponded to the VB2 CDR3 of the T cell clone used as a model for study of intraband heterogeneity as the nucleotide sequence of the CDR3 region of the clone was known and could be differentiated from host CDR3 regions. One hundred percent of the VB2 TCR B chain CDR3 region sequences obtained at the acute stage were from the transferred T cell clone. As disease progressed through recovery and relapse, increased heterogeneity was observed in this band, although the original sequence was still the majority of those cloned. Spectratypic analysis also showed increasing heterogeneity and skewing in terms of CDR3 size as disease progressed.
In summary, spinal cord T cells in mice in which EAE was induced by a single encephalitogenic T cell clone become increasingly heterogeneous as disease progresses due to an influx of recipient T cells. The data indicate that increased complexity is not due to a simple accumulation of host T cells, but is due to expansion of specific T cells clones that may be the result of sensitization of recipient T cells by additional epitopes during the inflammatory process.
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Robert B. Fritz, Department of Microbiology and Molecular Genetics, Medical College of Wisconsin, 8701 Watertown Plank Road, Milwaukee, WI 53226.
3 Abbreviations used in this paper: EAE, experimental autoimmune encephalomyelitis; PLP, proteolipid protein; CDR3, third complementarity-determining region; HPRT, hypoxanthine ribophosphoryl transferase; CB, constant region of TCR B-chain.
Received for publication November 15, 1999. Accepted for publication April 6, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ß T lymphocytes in the nervous system in experimental autoimmune encephalomyelitis: its possible implications for recovery and acquired tolerance. J. Autoimmun. 5:401.[Medline]
This article has been cited by other articles:
|
|
 |
|
  M. Brisebois, S. P. Zehntner, J. Estrada, T. Owens, and S. Fournier A Pathogenic Role for CD8+ T Cells in a Spontaneous Model of Demyelinating Disease J. Immunol., August 15, 2006; 177(4): 2403 - 2411. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. F. Shih, L. Mandik-Nayak, B. T. Wipke, and P. M. Allen Massive Thymic Deletion Results in Systemic Autoimmunity through Elimination of CD4+ CD25+ T Regulatory Cells J. Exp. Med., February 2, 2004; 199(3): 323 - 335. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. T. Madakamutil, I. Maricic, E. Sercarz, and V. Kumar Regulatory T Cells Control Autoimmunity In Vivo by Inducing Apoptotic Depletion of Activated Pathogenic Lymphocytes J. Immunol., March 15, 2003; 170(6): 2985 - 2992. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
