|
|
||||||||
Department of Immunology, Institute for Medical Microbiology and Hygiene, University of Freiburg, Freiburg, Germany
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Adoptive transfer experiments using tracer populations of TCR tg T cells in recipient mice represent one approach to follow Ag-specific T cells directly during an immune response (13, 14, 15). However, the generation of TCR tg mice is tedious and analysis of a single, monoclonal T cell response may not always reflect the physiological, polyclonal situation. Soluble peptide MHC class I tetramers can be used to directly visualize Ag-specific CD8 T cells by flow cytometry (16). However, this powerful technique requires knowledge of the immunodominant T cell epitopes and involves laborious construction of the appropriate reagents. Cytoplasmic staining of cytokines or enzyme-linked immunospot (ELISPOT) represent two other methods to quantify virus-specific T cells. However, these methods cannot be used to isolate viable Ag-specific T cells for further biological assays.
In the present study we describe a simple adoptive transfer system to visualize and isolate polyclonal antiviral CD8 T cells and demonstrate with three virusesLCMV, vesicular stomatitis virus (VSV), and vaccinia virus (VV)that the massive proliferation of T cells after viral infection is mainly due to clonal expansion of Ag-specific CD8 T cells.
| Materials and Methods |
|---|
|
|
|---|
C57BL/6 (Thy1.2) mice (B6) were obtained from Harlan Winkelmann (Borchen, Germany), B6.PL-Thy-1a (B6.Thy1.1) mice were a generous gift of Dr. H. Mossmann (Max Planck Institute of Immunobiology, Freiburg, Germany). LCMV TCR tg mice (line 318) have been described previously (17, 18). (C57BL/6 x B6.PL-Thy-1a)F1 mice were bred locally in a conventional mouse house facility.
Viruses
The LCMV-WE isolate was originally provided by F. Lehmann-Grube (Hamburg, Germany) and was grown on L929 cells with a low multiplicity of infection. Vesicular stomatitis virus Indiana (VSVIND) (Mudd-Summers isolate) was originally obtained from D. Kolakofsky (University of Geneva) and was grown with a low multiplicity of infection on BHK cells. Stocks of VV strain WR were produced by infecting BSC 40 cells with a low multiplicity of infection.
Adoptive transfers
Virus immune B6 mice were generated by i.v. injection of LCMV-WE (200 pfu), VSVIND (2 x 106 pfu), or VV (2 x 106 pfu). Sex-matched spleen cells from these virus-immune mice (46 wk after infection) containing CD8 (2 x 106) and CD4 (24 x 106) T cells were injected i.v., in a volume of 0.5 ml medium without FCS, into normal, nonirradiated (C57BL/6 x B6.PL-Thy-1a)F1 mice. One day after cell transfer the F1 recipient mice were challenged with the corresponding virus: LCMV-WE (200 pfu), VSVIND (2 x 106 pfu), or VV (2 x 106 pfu). LCMV TCR tg memory T cells were generated by transferring spleen cells of naive LCMV TCR tg mice containing 1 x 105 tg CD8 T cells i.v. into (C57BL/6 x B6.PL-Thy-1a)F1 mice and infecting the recipient mice 1 day later with LCMV-WE (200 pfu). Mice were designated as LCMV TCR tg memory mice when the period following infection exceeded 4 wk.
Flow cytometry
For detection of C57BL/6- or B6.PL-Thy-1a-derived donor CD8 T cells in (C57BL/6 x B6.PL-Thy-1a)F1 recipient mice PBL were incubated on ice with biotinylated anti-CD8 (Life Technologies, Gaithersburg, MD), PE-labeled anti-CD90/Thy1.1 (PharMingen, San Diego, CA), and FITC-labeled anti-CD90/Thy1.2 (Caltag, South San Francisco, CA) followed by Tricolor-streptavidin (Caltag). For detection of CD4 T cells, PBL were incubated with FITC-labeled anti-CD4, PE-labeled anti-CD90/Thy1.1, and biotinylated anti-CD90/Thy1.2 (all from PharMingen) followed by Tricolor-streptavidin. Staining of PBL was performed in PBS containing 2% FCS, 0.1% NaN3, and 10 U/ml heparin (Liquemin, Hoffmann-La Roche, Basel, Switzerland). PBL were analyzed after lysis of red blood cells using FACS Lysing Solution (Becton Dickinson, Mountain View, CA) according to the instructions of the manufacturer. Cells were analyzed using a FACSort (Becton Dickinson) and CellQuest software.
| Results |
|---|
|
|
|---|
We have previously shown that adoptively transferred naive LCMV TCR tg CD8 T cells expanded in B6 recipient mice by a factor of 103104 after LCMV infection (5). To examine whether LCMV TCR tg memory T cells proliferate to the same extent, spleen cells containing 105 naive or memory LCMV TCR tg cells (Thy1.2) were adoptively transferred into nonirradiated (C57BL/6 x B6.PL-Thy1a)F1 recipient mice that were infected 1 day later with LCMV. LCMV TCR tg memory T cells were generated in vivo using the same adoptive transfer protocol and were isolated 612 wk after infection.
Clonal expansion of the TCR tg T cells in the recipient mice was
monitored by flow cytometry using the allelic Thy1 marker to
distinguish donor (Thy1.2+) and host (Thy1.1+,
Thy1.2+) T cells. As shown in Fig. 1
, naive and memory LCMV TCR tg T cells
proliferated to the same extent in the recipient mice, and 8 days after
infection >40% of PBL (Fig. 1
, a and c) and
>70% of CD8 T cells (Fig. 1
, b and d) were of
donor T cell origin (Thy 1.2). As a control, CD8 T cells were also
stained with Abs specific for the transgenic LCMV TCR (V
2/Vß8)
which confirmed that >95% of Thy1.2+ cells found in these
mice expressed the tg TCR (data not shown). Even at the earliest time
point after infection (day 6), when sizable numbers of the donor T
cells were first detected in the recipient mice, no significant
difference in clonal expansion of naive or memory T cells was observed.
After the acute phase of the infection the number of TCR tg T cells
derived from either naive or memory cells declined with similar
kinetics. Thus, in vivo-generated bona fide CD8 memory T cells exhibit
the same characteristics in clonal expansion and decline after adoptive
transfer and Ag challenge as naive T cells.
|
Recent studies using MHC/peptide tetramers, cytoplasmic cytokine
stainings, or ELISPOT have demonstrated that precursor frequencies of
memory CD8 T cells after viral infections are about 10-fold higher than
determined previously by limiting dilution analysis. In the LCMV
system, Murali-Krisha et al. (6) found that 510% of memory CD8 T
cells in LCMV immune B6 mice were virus-specific. As shown above in the
LCMV TCR tg model, both naive and memory LCMV-specific CD8 T cells were
able to undergo the same vigorous clonal expansion after Ag
(re)challenge in vivo. Therefore, it should be possible to also
visualize polyclonal LCMV-specific memory CD8 T cells from
nontransgenic but LCMV immune mice after adoptive transfer and LCMV
infection using appropriate allelic markers (Thy1) to distinguish donor
from host T cells. To test this hypothesis, B6 mice (Thy1.2) were
infected with LCMV, VSV, or VV to induce virus-specific memory T cells
(Fig. 2
). Four to 6 wk after infection
tracer spleen cells from these virus-immune mice were adoptively
transferred into nonirradiated (C57BL/6 x
B6.PL-Thy-1a)F1 mice. One day after transfer
the F1-recipient mice were challenged with the
corresponding virus, and the kinetics of the adoptively transferred
tracer T cells (Thy1.2) was followed in the F1-recipient
mice (Thy1.1, Thy1.2) by flow cytometry of PBL.
|
In the first set of experiments we determined the in vivo
proliferation of T cells from B6 mice immune to LCMV (=B6/LCMV). Tracer
spleen cells from B6/LCMV mice containing CD8 (2 x
106) and CD4 (24 x 106) T cells were
adoptively transferred into F1-recipient mice. A small
population of Thy1.2+ Thy1.1- cells (35%)
could be detected in normal (C57BL/6 x
B6.PL-Thy-1a)F1 mice, and this percentage did
not significantly increase after cell transfer (Fig. 3
a). After LCMV infection,
donor Thy1.2+ T cells from B6/LCMV mice proliferated
strongly in the F1-recipient mice, and on day 8
postinfection 3050% of PBL (Figs. 3
b and
4a) and 7080% of total CD8
T cells (Fig. 4
b) were of donor T cell origin. The extent
and the kinetics of the clonal expansion of donor T cells from B6/LCMV
mice (Fig. 4
, a and b) was strikingly similar to
the data obtained with naive or memory donor T cells from LCMV TCR tg
mice (Fig. 1
).
|
|
Visualization of polyclonal VSV- and VV-specific T cells during infection
Once the adoptive transfer protocol with LCMV had been
successfully established, the same basic approach was used to visualize
VSV- and VV-specific T cells. Tracer spleen cells from VV or VSV immune
B6 mice were adoptively transferred into F1-recipient mice,
which were then challenged with the corresponding virus 1 day later.
Flow cytometric analysis revealed that donor T cells from B6/VSV or
B6/VV mice expanded massively in the F1-recipient mice
(Fig. 5
, a, b,
e, and f). Importantly, Thy1.2+ donor
T cells from VSV or VV immune mice did not expand after infection with
the unrelated LCMV (Fig. 5
, d and h). The slight
proportional decrease of these VSV or VV immune T cells in the
recipient mice observed after LCMV infection was probably due to the
massive expansion of host LCMV-specific T cells.
|
Competitive expansion of LCMV- and VSV-specific memory T cells
CD8 memory T cells but not naive cells were shown to be driven to
activation and proliferation by IFNs induced during viral infections
(22). Since memory CD8 T cells secrete higher levels of IFN-
than
naive cells after stimulation (10, 23), it was important to test
bystander expansion of VV-specific memory CD8 T cells in the presence
of vigorously proliferating LCMV-specific memory CD8 T cells and vice
versa. (C57BL/6 x B6.PL-Thy-1a)F1 mice
were given a 1:1 mixture of spleen cells containing 1 x
106 CD8 and 12 x 106 CD4 T cells from
LCMV immune B6.PL-Thy-1a (Thy 1.1) and from VV immune B6
(Thy 1.2) mice (Fig. 6
A).
Afterward, the recipient mice were infected with either LCMV or VV and
PBL were analyzed 8 days after infection. As shown in Fig. 6
B, only donor T cells from LCMV but not from VV immune mice
expanded after LCMV infection and vice versa. Importantly, immune CD8 T
cells being specific for the unrelated virus did not increase above
background levels.
|
| Discussion |
|---|
|
|
|---|
production at the single-cell level. In our experimental approach,
whole virus was used as antigenic challenge whereas the former studies
used the three known antigenic peptides of LCMV in B6 mice to visualize
virus-specific CD8 T cells. The finding that both approaches yielded
similar numbers of reactive T cells during acute infection further
indicates that all the major peptide epitopes of LCMV in B6 mice are
now known. Following the acute phase of the infection the frequency of
LCMV-specific T cells per CD8 cell decreased roughly by a factor of 2
in our system (Fig. 4
The small degree of virus-specific CD4 T cell expansion found here
might be due to 1) a lower frequency of virus-specific T helper cells
when compared with CD8 T cells, 2) differences in viral Ag
presentation, and/or 3) a lower potential of CD4 memory T cells to
proliferate in vivo. Recently, the frequency of CD4 memory T cells has
been determined in the LCMV system by two groups using limiting
dilution analysis, ELISPOT, and intracellular IFN-
expression
(24, 25, 26). These studies revealed frequencies of LCMV-specific CD4 T
cells ranging from 1/10 to 1/600 CD4 T cells during the acute infection
and from 1/100 to 1/1200 CD4 T cells in long-term memory. Thus, in
comparison to CD8 T cells, 10- to 100-fold lower numbers of
LCMV-specific CD4 T cells are induced in the course of a LCMV
infection. Therefore, the selective expansion of CD8 T cells in the
transfer system described here using LCMV is readily explained by the
higher frequency of LCMV-specific T cells in the CD8 subset when
compared with CD4 T cells.
Frequencies of VSV or VV-specific CD4 or CD8 T cells have not yet been determined precisely. Our data revealed vigorous CD8 T cell expansion but only minimal Ag-specific proliferation of CD4 T cells after VSV or VV infection. This finding may also indicate that in these viral infections the numbers of the induced VSV- or VV-specific CD8 T cells are significantly higher than of the corresponding CD4 T cells.
The result that adoptively transferred T cells from virus-immune mice did not expand after infection with the unrelated viruses indicated that bystander proliferation induced by IFNs was not sufficient to induce detectable T cell expansion in this system. This finding appears to be in contrast with a previous study by Tough et al. (22), who reported induction of bystander proliferation of CD8 memory T cells by viral infections. However, it is important to note that bystander proliferation of memory CD8 T cells was examined by Tough et al. using the 5-bromo-2'-deoxyuridine (BrdU) incorporation technique. Doubling of cells is easily detected by this method, whereas in our system doubling of cells would not result in significant clonal expansion in the context of the vigorous proliferation of the Ag-specific CD8 T cells.
In initial experiments we directly transferred LCMV immune spleen cells from B6 (Thy1.2) into B6.PL-Thy-1a (Thy1.1) mice. Clonal expansion of the transferred T cells after infection was similar to that described above; however, donor T cells disappeared rapidly in some recipient mice, suggesting rejection of donor T cells by the host immune system due to allelic Thy1- and/or other unknown differences of the two "congenic" mouse strains used.
In conclusion, we describe a simple transfer system which allows us to visualize a polyclonal virus-specific T cell response in vivo. This experimental approach may also be useful to directly follow T cell responses against other infectious agents or autoantigens where the epitopes recognized by T cells are not yet well characterized. By using different Thy1 alleles of donor and recipient mice, Ag-specific memory T cells can easily be isolated for further characterization.
| Acknowledgments |
|---|
| Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Hanspeter Pircher, Department of Immunology, Institute for Medical Microbiology and Hygiene, Hermann-Herder-Strasse 11, University of Freiburg, D-79104 Freiburg, Germany. E-mail address: ![]()
3 Abbreviations used in this paper: tg, transgenic; LCMV, lymphocytic choriomeningitis virus; ELISPOT, enzyme-linked immunospot; B6, C57BL/6 mice; pfu, plaque- forming units; VSV, vesicular stomatitis virus; VV, vaccinia virus. ![]()
Received for publication December 17, 1998. Accepted for publication February 11, 1999.
| References |
|---|
|
|
|---|
-herpesvirus 68. Am. J. Pathol. 145:818.[Abstract]
This article has been cited by other articles:
![]() |
M. Koschella, D. Voehringer, and H. Pircher CD40 Ligation In Vivo Induces Bystander Proliferation of Memory Phenotype CD8 T Cells J. Immunol., April 15, 2004; 172(8): 4804 - 4811. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. R. Fulton, J. Smith, C. Cunningham, and C. F. Cuff Influence of the Route of Infection on Development of T-Cell Receptor {beta}-Chain Repertoires of Reovirus-Specific Cytotoxic T Lymphocytes J. Virol., February 1, 2004; 78(3): 1582 - 1590. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Voehringer, M. Koschella, and H. Pircher Lack of proliferative capacity of human effector and memory T cells expressing killer cell lectinlike receptor G1 (KLRG1) Blood, November 15, 2002; 100(10): 3698 - 3702. [Abstract] [Full Text] [PDF] |
||||
![]() |
S.-K. Kim, M. A. Brehm, R. M. Welsh, and L. K. Selin Dynamics of Memory T Cell Proliferation Under Conditions of Heterologous Immunity and Bystander Stimulation J. Immunol., July 1, 2002; 169(1): 90 - 98. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |