CD4 T cells inhibit the CD8 T cell response during low-dose virus infection

(1)

CD4 T cells inhibit the CD8 T cell response during

low-dose virus infection

Stephen Cose

1

, Clair Brammer, David J. Zammit, D. A. Blair

and

Leo Lefranc

x

ois

Division of Immunology, Department of Medicine, University of Connecticut Health Center, Farmington, CT 06030, USA

1_{Present address: School of Clinical Veterinary Science, University of Bristol, Langford, BS40 5DU, UK}

Keywords: CD4 T cells, cellular immunology, immune responses, vesicular stomatitis virus

Abstract

CD4 T cells are not thought to play a significant role in generating an effective primary CD8 T cell response to most viral infections. We have challenged this view by demonstrating that antigen-specific CD4 T cells can indeed suppress the proliferation of antigen-antigen-specific naive CD8 T cells in response to low doses of vesicular stomatitis virus. This finding is in contrast to the established observations that at high antigen loads CD4 T cells play little role in generating CD8 T cell responses, and that in non-infectious model systems CD4 T cells actually help the CD8 T cell response. Our results suggest that at low infectious doses, CD4 T cells play a much larger role in controlling infections than previously appreciated.

Introduction

Several studies have recently defined the role of CD4 help in activating CD8 T cells through the interaction of CD40 and its ligand (1–3). These important studies were performed in non-infectious systems, however, and little evidence exists to show a similar role for CD4 T cells during a live infection. Indeed, CD4 T cells have been shown to play little or no role in gen-erating a CD8 T cell response to viral infection (4, 5), with the exception of herpes simplex virus-1 infection of the footpad (6). This lack of requirement for CD4 T cell help in resolving viral infections may be due to the fact that many viruses can bypass CD4 help by providing their own inflammatory signals (7, 8). However, CD8 T cells generated in the absence of CD4 T cells are defective in the memory phase of the immune response (9–11), indicating that although CD4 T cells may be superfluous for the activation phase of a viral challenge, they are required for the memory phase. It therefore remains uncertain as to what role CD4 T cells play during the primary phase of a viral challenge.

Part of the explanation may lie with viral dose, since large viral loads contribute to the massive expansion of CD8 T cells (12, 13), and may thus diminish the role of CD4 T cells during such infections. Such a mechanism may certainly explain the discrepancy that exists between infectious and non-infectious systems regarding CD4 T cell help. The question remains though as to whether the two systems can be reconciled. Under low-dose infection conditions, the relative paucity of

initial viral proteins may mean that direct Toll-like receptor (TLR) signaling is somewhat diminished, and thus CD4 T cells may well play a more active role in generating the CD8 T cell response. Since the initial antigen encounter is small, one might expect that CD4 T cells would serve to help the CD8 T cell response in a manner similar to that seen in non-infectious systems. We have examined this issue by looking at the role that antigen-specific CD4 T cells play in resolving a low-dose vesicular stomatitis virus (VSV) infection. In contrast to previous studies which showed that CD4 T cells play little role in generating an effective primary CD8 T cell response, even to VSV (4, 5, 9–11, 14, 15), we show that, during a low-dose infection, CD4 T cells can directly affect the outcome of the CD8 T cell response. This is not through conventional CD4 help since addition of antigen-specific CD4 T cells actually inhibited the generation of CD8 T cells. These results are surprising, and suggest that CD4 T cells play a much larger role in the early stages of natural viral infection (where viral antigen may be initially limiting) than previously realised.

Methods

Mice

All mice were housed on-site under specific pathogen-free conditions. C57BL/6-Ly5.2 mice (CD45.1) were purchased

Correspondence to: S. Cose; E-mail: s.cose@bristol.ac.uk Received1 February 2006,accepted13 May 2006

Transmitting editor: S. Hedrick Advance Access publication 12 June 2006

doi:10.1093/intimm/dxl061 For permissions, please e-mail: journals.permissions@oxfordjournals.org

at Pennsylvania State University on February 27, 2014

http://intimm.oxfordjournals.org/

(2)

from CR-NCI. OT-I, OT-II and TEatransgenic mice (CD45.2) on

a Ragÿ/ÿbackground were bred in-house. The TEamice are

a transgenic line whose CD4 T cells are specific for the Ea52–68

peptide from the MHC class II protein I-E, presented in the

context of MHC class II I-Ab(16), and were a kind gift of Prof.

Alexander Rudensky, Department of Immunology, University of Washington, Seattle, WA, USA.

Adoptive transfer and immunization

Single-cell suspensions of pooled spleen and lymph nodes

from OT-I and/or OT-II or TEatransgenic mice were lysed of

RBCs, washed, counted and passed through nylon mesh (40

lm) before injection intravenously (IV) into recipient

(C57BL/6-Ly5.2) mice, in the numbers stated. Before injection, the cells were labeled with 5-(and-6)-carboxyfluoroscein succinimidyl ester (CFSE) (Invitrogen, Carlsbad, CA, USA) as previously described (17). One day later, the mice were immunized IV with recombinant VSV–ovalbumin peptide (OVA) or VSV–OED

(O = SIINFEKL peptide, E = Ea52–68peptide, D = DS-RED II)

with the indicated amount of virus, or immunized with OVA-loaded splenocytes. OVA-OVA-loaded splenocytes were prepared

as previously described (18), and 25 3 106 OVA-loaded

splenocytes injected IV into recipient mice. Recombinant VSV constructs were produced as previously described (19, 20). VSV–OVA (21) is a recombinant VSV expressing the ovalbumin gene, and VSV–OED was produced using the same protocols as above.

Flow cytometry

Single-cell suspensions were prepared from the spleens of mice. After lysis of the RBCs, splenocytes were stained with

combinations of the following antibodies: PE-labeledaCD11a,

PerCP-labeled aCD8 or aCD4 and Cy5-labeled aCD45.2

(eBioscience, San Diego, CA, USA). To detect OVA-specific

or VSV nucleoprotein-specific CD8+_{T cells, splenocytes were}

stained for 1 h with allophycoerythrin (APC)-labeled H-2Kb

tetramers containing either the OVA-derived peptide SIIN-FEKL or the N protein-derived peptide RGYVYQGL and washed prior to addition of the above antibodies. After washing, the relative fluorescence intensities were then measured using a FACSCalibur (BD Biosciences, San Jose, CA, USA) and data analysed using FlowJo software (Tree Star, Ashland, OR, USA). For some experiments, splenocytes

were first stained with PE-labeled H-2Kb _tetramers

contain-ing the OVA-derived peptide SIINFEKL and APC-labeled

H-2Kb tetramers containing the N protein-derived peptide

RGYVYQGL. After washing, the cells were then stained

with PECy7-labeled aCD11a, Alexa405-labeled aCD4,

PerCPCy5.5-labeledaCD8 and APCCy7-labeledaCD45.2

(eBio-science). After washing, the relative fluorescence intensities were measured using an LSRII flow cytometer (BD Bio-sciences) and data analysed using FlowJo software (Tree Star).

Neutralizing antibody assay

Serum from mice was serially diluted, in duplicate, in a 96-well flat-bottomed plate and incubated with 100 plaque-forming

units (p.f.u.) of VSV–OVA for 1 h. Following addition of 53104

baby hamster kidney (BHK) cells per well, the plates were incubated for a further 2 days. Antibody titers were then

determined by the lowest dilution of serum where the BHK cells showed no viral cytopathic effect.

Results

Recruitment of CD8 T cells depends on the viral dose

High doses of virus can generate primary CD8 T cell responses in the absence of CD4 T cell help, probably through engagement of TLRs (7, 8). To test the role of CD4 T cells in response to lower doses, we immunized mice with titrated doses of VSV–OVA and measured the frequency of respond-ing CD8 T cells at the height of the CD8 T cell response (day 6) using MHC class I tetramers (Fig. 1). CD11a was used in conjunction with the tetramer staining because both activated and memory CD8 T cells express high levels of CD11a (22), and staining with this molecule in conjunction with tetramer identifies bona fide antigen-specific CD8 T cells (23). Lower doses of VSV–OVA resulted in lower numbers of CD8 T cells being recruited into the primary immune response, for both the N-specific (Fig. 1A) and OVA-specific (Fig. 1B) CD8 T cells. Note that the OVA-specific CD8 T cell response was ~3-fold lower than that of the N-specific CD8 T cells at all doses lower

than 13106p.f.u. VSV–OVA (compare Fig. 1A with B). This

difference in CD8 T cell recruitment may reflect differences in expression of viral proteins in VSV infections and/or in naive

CD8 T cell precursor frequency, such that at 13 102 p.f.u.

VSV–OVA, the OVA-specific CD8 T cells could not be distinguished above background staining (Fig. 1B). In all subsequent experiments, we therefore used the lowest dose of VSV–OVA in which we could detect a significant population

of OVA-specific CD8 T cells (13103p.f.u.), and compared our

results at this dose with the infectious dose of 13106p.f.u.

VSV–OVA.

Extra ‘help’ during a low-dose infection decreases the CD8 T cell response

To determine whether the primary CD8 T cell response after low-dose infection was dependent on help from CD4 T cells, we adoptively transferred into recipient mice OVA-specific

CD4+ transgenic (OT-II) T cells prior to infection with either

a high (13106_{) or low (1}

3103_{) dose of VSV–OVA (Fig. 2). In}

non-infectious systems, cognate CD4 help is required to generate a CD8 T cell response (1–3, 24), and addition of exogenous specific CD4 T cells increases the antigen-specific CD8 T cell response (25). We reasoned that the inclusion of additional help prior to a low-dose infection might mirror the results seen in such non-infectious systems. Consistent with earlier reports that the CD8 T cell response to high viral doses is largely independent of CD4 help (4, 5, 9–

11, 14, 15), the addition of 13105_{OT-II CD4 T cells prior to}

a high-dose infection had no effect on the resulting CD8 T cell response (Fig. 2). This was true for both the N-specific and the OVA-specific CD8 T cell response (N specific, 27% without OT-II cells versus 22.9% with; OVA specific, 12.5% without OT-OT-II cells versus 13.4% with; upper panels, Fig. 2A and B, respectively). However, when the OT-II cells were adoptively transferred prior to a low-dose infection, the CD8 T cell response actually decreased when compared with the no-transfer controls. This was the case for both the N-specific and

(3)

OVA-specific CD8 T cells (N specific, 1.9% without OT-II cells to 0.1% with; OVA specific, 1% without OT-II cells to 0.1% with; lower panels, Fig. 2A and B, respectively). This was a surprise finding, and suggested that the role of CD4 T cells in responding to low-dose infection was distinct from that previously described for either non-infectious (1–3) or in-fectious systems (4, 5, 9–11, 14, 15).

To show that such an effect was a general characteristic of CD4 T cells during a low-dose infection and not an artifact of the specific TCR transgenic system used, we repeated the above experiments using different transgenic CD4 T cells. In

this case, the transgenic T cells were isolated from TEamice,

a transgenic line whose CD4 T cells recognize the Ea52–68

peptide from the MHC class II I-E protein (16). Immunization of

recipient mice after adoptive transfer of TEacells with a high

dose of VSV–OED, a recombinant VSV encoding the SIINFEKL

(OVA) and Ea52–68 peptides, again revealed no requirement

for the TEa cells in generating an effective CD8 T cell

response. Both N-specific and OVA-specific CD8 T cells

were generated at similar levels regardless of whether TEa

cells were present or not (N specific, 33.8% without TEacells

and 30.8% with; OVA specific, 9.1% without TEa cells and

7.6% with; upper panels, Fig. 2C and D, respectively).

How-ever, as with the OT-II transfer system, when TEa cells were

adoptively transferred prior to a low-dose infection, a reduc-tion in both the N-specific and OVA-specific CD8 T cells was seen compared with the no-transfer controls (N specific, 6.7%

without TEacells to 2.9% with; OVA specific, 4.2% without TEa

cells to 0.7% with; lower panels, Fig. 2C and D, respectively).

This response is antigen specific since transferred TEacells

do not divide when mice are given VSV–OVA, a virus that does

not express the cognate antigen for TEacells.

To show that the adoptively transferred CD4 T cells were responding to the infection, we measured the fold increase of

transferred OT-II or TEacells following infection compared with

uninfected controls (Fig. 2E; open bars, OT-II cells; closed

bars, TEacells). When mice were infected with a low dose of

virus (VSV–OVA for OT-II transfer, open bars, left-hand side;

VSV–OED for TEatransfer, closed bars, left-hand side), both

sets of transgenic CD4 T cells increased by>10-fold over the

uninfected controls. When high doses of virus were used (right-hand bars), the transgenic CD4 T cells increased by

>100-fold. Examination of the CFSE dilution of the transferred

CD4 T cells showed that >95% of the CD4 T cells had

undergone division, regardless of the viral dose used (data not shown). It is interesting to note that despite the greater expansion of the transgenic CD4 T cells following high-dose infection, inhibition of the CD8 T cell response was only seen following low-dose infection, where the expansion of the CD4 T cells was more limited. These results show that the effect on the CD8 T cell response to infection was a general character-istic of CD4 T cells and occurred only in response to a low-dose infection.

Increasing the number of CD4 T cells does not affect the kinetics of the CD8 T cell or B cell response

Since CD4 T cells are known to help the priming of CD8 T cells, at least in non-infectious systems, it was possible that the CD4 T cells in our system were in fact priming the CD8 T cells to more rapidly resolve the infection. If this was the case, the CD8 T cell response may actually have peaked before day 6, and we were simply looking at the end stages of the resolu-tion of primary infecresolu-tion. To rule this out as a possibility, we Fig. 1. Titrated doses of virus causes a concomitant decrease in responding CD8 T cells. Mice were infected with varying doses of VSV–OVA and splenocytes examined 6 days later for the presence of antigen-specific CD8 T cells. (A) CD8-gated T cells stained for N-tetramer and CD11a. (B) CD8-gated T cells stained for OVA-tetramer and CD11a. Numbers within the plots refer to the percentage of CD8-gated T cells staining double positive. Data are representative plots from two individual mice.

(4)

Fig. 2. High numbers of antigen-specific CD4 T cells inhibit the CD8 T cell response following low-dose infection. Mice received either 0 OT-II cells (A and B; left-hand panels) or 13105OT-II cells (A and B; right-hand panels) or 0 Teacells (C and D; left-hand panels) or 13105Teacells (C and D; right-hand panels). One day later mice received either 13106_{or 1}₃₁₀3_{p.f.u. VSV–OVA (A and B; top and bottom panels, respectively)}

or 13106or 13103p.f.u. VSV–OED (C and D; top and bottom panels, respectively). Numbers inside the panels represent the percentage of gated CD8 T cells falling within the top right-hand quadrant. Data are representative plots from five separate experiments, with two mice per experiment. (E) Contribution of transferred CD4 T cells in the above panels, represented as a fold increase in the transferred transgenic CD4 T cells (open bars, OT-II; closed bars, TEa) following infection, compared with uninfected controls. Error bars represent the standard error of the mean.

(5)

conducted a time course experiment looking every 2 days for the presence of N-specific and OVA-specific CD8 T cells (Fig. 3A and B). Only results for the OVA-specific CD8 popula-tion are shown; however, similar results were found with the N protein-specific CD8 T cells. When mice were given a

high-dose infection (1 3 106 p.f.u.), we saw the standard

bell-shaped curve of the responding OVA-specific CD8 T cells, peaking at day 6 regardless of whether OT-II cells were adoptively transferred or not (Fig. 3A; closed circles, no OT-II;

open circles, 105 _{OT-II; note that the two sets of data are}

virtually superimposed upon each other). When mice that had received no OT-II cells were given a low-dose infection, the peak of the responding OVA-specific CD8 T cells was also on day 6 (Fig. 3B, closed circles), indicating that the CD8 T cell response to a low-dose infection runs a similar course to that of

a high-dose infection. However, when 13105OT-II cells were

transferred into recipient mice prior to a low-dose infection, an earlier peak in the responding OVA-specific CD8 T cells was not seen (Fig. 3B, open circles), suggesting that the extra help afforded by the OT-II cells was not generating a quicker CD8 T cell response. The CD8 T cell response was not delayed past day 8 since increases in either N-specific or OVA-specific CD8 T cells were detected at any time point up to day 14 post-infection (data not shown).

It was also formally possible that rather than helping the CD8 T cell response, the OT-II cells were affecting the B cell response. To examine this, we determined the neutraliz-ing serum antibody titer from the same mice as above. We found no difference in the neutralizing antibody titers in mice

infected with 13106p.f.u. VSV–OVA, regardless of whether

OT-II cells were adoptively transferred or not prior to infection

(Fig. 3C; closed circles, no OT-II; open circles, 105OT-II). In

this case, the titers rose at similar levels between the two groups of mice each day up to day 8. When mice were

infected with 1 3103_{p.f.u. VSV–OVA, neutralizing antibody}

titers could only be detected in mice that did not receive OT-II cells and rose to the level of detection at day 6 (Fig. 3D, closed circles). As with the CD8 T cell response, no neutralizing antibody could be detected at any time point when mice were

given 13105OT-II cells prior to low-dose infection (Fig. 3D,

open circles). No neutralizing antibody titer could be detected as far out as day 14 post-infection (data not shown). These data suggest that during a low-dose infection, high numbers of antigen-specific CD4 T cells can act independently to clear infection without the CD8 T or B cell response. This is in accordance with a recent study which showed that virus-specific CD4 T cells actually impaired the induction of neutralizing antibodies, and consequent deliberate removal of CD4 T cells improved the neutralizing antibody response (26). That the CD4 T cells were working to independently clear virus is evidenced by the fact that no antigen-specific CD8 T or B cell response could be detected during the acute phase of the immune response.

Small numbers of adoptively transferred CD4 T cells affect the CD8 T cell response

We now had a system where the number of adoptively transferred antigen-specific CD4 T cells directly affected the adaptive immune response following infection, and this effect could be tracked by looking at the virus-specific CD8 T cell response using MHC class I tetramers. We next wondered whether the effect of the additional CD4 T cells on the responding CD8 T cells was dose dependent or whether our results were merely an artifact of the high transfer numbers. We reasoned that if low numbers of transferred CD4 T cells were still capable of affecting the CD8 T cell response, then this mechanism of CD4-mediated inhibition may be a natural

Fig. 3. High numbers of antigen-specific CD4 T cells do not generate earlier antigen-specific CD8 T or B cell responses. Mice received either 0 or 13105_{OT-II cells 1 day prior to infection with VSV–OVA. (A) Percentage of OVA-specific CD8}+_{T cells following infection with 1}₃₁₀6_{p.f.u. VSV–OVA}

or (B) 13103_{p.f.u. VSV–OVA. (C) Titer of neutralizing antibodies in serum of mice from (A). (D) Titer of neutralizing antibodies in serum of mice}

from (B). Closed circles, mice that received zero OT-II cells; open circles, mice that received 13105_{OT-II cells. Data represent four individual}

mice.

(6)

phenomenon occurring at low pathogenic doses rather than an artifact of experimental design.

Figure 4 shows the results of titrating into recipient mice graded numbers of OT-II cells prior to low-dose infection with VSV–OVA. Without additional CD4 T cells, the N- and OVA-specific CD8 T cell response was ~5 and 3% of the total CD8 T cell population, respectively (Fig. 4; closed bars, N-specific CD8 T cells; open bars, OVA-specific CD8 T cells). The transfer of up to 750 OT-II cells prior to low-dose infection did not significantly alter the number of responding N- or OVA-specific CD8 T cells compared with the untransferred controls. How-ever, the addition of 1000 OT-II cells significantly decreased

the N-specific CD8 T cell response (P<0.05) compared with

the untransferred control. The presence of even larger numbers of OT-II cells resulted in a significant decrease in both the responding N- and OVA-specific CD8 T cells

com-pared with the untransferred controls (Fig. 4; N specific,P<

0.005; OVA specific,P<0.05 at 104and 105transferred OT-II

cells,P<0.005 at 106_{transferred OT-II cells). Thus, as few as}

1000 transferred OT-II cells were capable of affecting the CD8 T cell response. This number approaches the level of detection for transferred T cells in uninfected controls, and is in line with other studies that have examined the functional capacity of CD8 T cells at low transfer numbers (27, 28).

CD4-mediated CD8 inhibition: competition for antigen?

Having established a system whereby small numbers of CD4 T cells could affect the responding CD8 T cell population fol-lowing primary viral infection, we wanted to assess how the CD4 T cells were mediating this effect. As a positive control for

CD4 help, we adoptively transferred either 0 or 13106OT-II

CD4 T cells along with either 0 or 13103_{OT-I CD8 T cells and}

1 day later primed mice with OVA-loaded splenocytes (Fig. 5A and B). This priming regimen renders the OT-I T cell response CD4 dependent when the OT-I T cells are transferred in sufficiently low numbers (29). We found that, in the absence of any additional CD4 help, the OVA-specific CD8 T cell response was minimal regardless of whether OT-I cells were present or not before priming (0.1%; Fig. 5A, bottom and top left panels,

respectively). The addition of 1 3 106 _{OT-II cells prior to}

priming increased the OVA-specific CD8 T cell response to 0.6% without additional OT-I cells (Fig. 5A, top right-hand

panel) and 0.8% when 13103OT-I cells were co-transferred

with the OT-II cells (Fig. 5A, bottom right-hand panel), showing again the requirement for CD4 help in priming the CD8 T cell response in this system. Additionally, the presence of the OT-II cells also helped the OT-I T cells dominate the OVA-specific response at the expense of the endogenous OVA-specific

T cells. Fig. 5B shows the CD11ahi_{(activated) CD8-gated T cell}

population (right-hand quadrants in Fig. 5A) plotted as the OT-I donor marker against OVA-specific MHC class I tetramer staining. When no OT-I T cells were present, the OVA-specific response was completely dominated by the endogenous CD8 T cells (Fig. 5B, upper panels, lower right-hand quadrants).

When 13103OT-I T cells were transferred without additional

OT-II cells, the resulting OVA-specific CD8 T cell population was equally split between responding OT-I T cells and endogenous OVA-specific CD8 T cells (Fig. 5B, bottom left-hand panel; compare upper right- and lower right-left-hand quadrants, respectively). However, when the small number

of OT-I T cells were co-transferred with 13106OT-II T cells, the

OT-I T cells dominated the OVA-specific CD8 T cell response at the expense of the endogenous OVA-specific CD8 T cells (Fig. 5B, bottom right-hand panel; compare upper right- and lower right-hand quadrants, respectively). Such competition for antigen between CD8 T cells is not a new finding, and has indeed been recorded for CD4 T cells as well (13, 30, 31).

The above experiment clearly showed a role for CD4 T cells in helping to initiate a CD8 T cell response (albeit in a non-infectious system), as well as highlighting the antigenic competition known to occur between CD8 T cells (12, 13, 31, 32). Additionally, the experiment showed that the adoptively transferred OT-II cells were not defective in themselves, since they could clearly help the CD8 T cell response following a specific priming regimen. We then repeated the above transfer experiments using the low-dose VSV–OVA priming regimen in order to determine the effect that additional OT-II cells would have on the responding OT-I T cells. As previously shown, additional OT-II T cells caused a reduction in the responding OVA-specific CD8 T cell population when com-pared with the control (0.4% from 0.8%; Fig. 2C, upper right-and left-hright-and panels, respectively). When no additional OT-I T cells were transferred, the endogenous OVA-specific CD8 T cells responded to infection (Fig. 5D, lower right-hand quadrants in upper panels; left-hand panel, 0 OT-II and 0 OT-I;

right-hand panel, 106OT-II and 0 OT-I). Transfer of 13103OT-I

T cells in the absence of additional OT-II cells resulted in an OVA-specific CD8 T cell population that consisted of both OT-I T cells and responding endogenous CD8 T cells (Fig. 5D, bottom left-hand panel; upper right quadrant, OT-I T cells;

lower right quadrant, endogenous T cells). However, when 13

106OT-II T cells were co-transferred with 13103OT-I T cells

prior to infection, the OVA-specific CD8 T cell response was once again almost completely abrogated (Fig. 5D, bottom right-hand panel). Significantly, the number of donor OT-I T cells could not be distinguished above control staining in

mice that had received 13106_{OT-II cells (Fig. 5D, compare}

right-hand panels; upper right, 0 OT-I cells; lower right, 103

OT-I cells). Since the OT-I T cells are capable of responding Fig. 4. Low numbers of antigen-specific CD4 T cells affect the CD8

T cell response to low-dose infection. Mice were seeded with varying numbers of transgenic OT-II cells, infected with 13103_{p.f.u. VSV–}

OVA 1 day later and the percentage of N-specific (filled bars) or OVA-specific (open bars) CD8 T cells in the spleen determined 6 days following infection. Error bars represent the standard error of the mean, and asterisks represent significance compared with the untransferred control (*P<0.05,**P<0.005).

(7)

to infection in the absence of additional OT-II cells (Fig. 5D, bottom left-hand panel), the absence of the OT-I T cells in the presence of transferred OT-II cells suggests that the trans-ferred CD4 T cells are acting at some level to inhibit antigen-specific CD8 T cell proliferation. This figure therefore highlights that it is not only the priming regimen but also the dose of virus that determines how a CD4 T cell helps the immune system respond to infection.

Discussion

We have described here an infection model that has allowed us to examine the role that CD4 T cells play in priming the CD8

T cell response to primary infection. In contrast to other infection models, including high-dose VSV infections (4, 5, 14, 15), we have found that antigen-specific CD4 T cells play a prominent role in the outcome of the primary CD8 T cell response. This role, however, is not the traditional role of CD4 T cell help described in non-infectious models (1–3), and has allowed us to directly examine the effect of antigen-specific CD4 T cells on the generation of the primary CD8 T cell response. Exactly how the CD4 T cells are affecting the immune response to a low-dose infection remains to be determined, although they are clearly preventing antigen-specific CD8 T cells from expanding. We can envisage three (not necessarily mutually exclusive) events to account for our Fig. 5. Antigen-specific CD4 T cells compete with antigen-specific CD8 T cells during low-dose infection. Mice received either 0 or 13106OT-II cells (A–D; left- and right-hand panels, respectively) and either 0 or 13103_{OT-I cells (A–D; top and bottom panels, respectively) 1 day prior}

to priming. (A and B) Mice were primed using OVA-loaded splenocytes and sacrificed 5 days later. Data show (A) gated CD8 T cells or (B) gated CD11ahi_{CD8 T cells [right-hand quadrants in (A)]. (C and D) Mice were primed using 1}₃₁₀3_{p.f.u. VSV–OVA and sacrificed 6 days later. Data}

show (C) gated CD8 T cells or (D) gated CD11ahiCD8 T cells [right-hand quadrants in (A)]. Numbers within each panel represent the percentage of gated cells falling within that quadrant. Data are representative of three individual mice per group.

(8)

surprising findings. The first and perhaps most likely event is that the CD4 T cells themselves are activated to produce

anti-viral cytokines (such as IFN-c) which rapidly prevent the

spread of virus. In such a scenario, B cells and CD8 T cells may not necessarily be required because other innate cells may effectively clear the infection. Two other possibilities are that the CD4 T cells themselves are acting directly to clear the infection in conjunction with antigen-presenting cells (APCs), and in so doing, prevent the activation and expansion of CD8 T cells. This may be through the deliberate removal of antigen directly from the APC by killing it (33–35), or alternatively, the CD4 T cells may directly down-modulate the level of peptide– MHC complexes expressed by APCs presenting viral pep-tides. The latter form of competition has been shown to occur within the CD8 T cell subset (36, 37), although in these experiments the competing CD8 T cells were unable to modulate other peptide–MHC complexes on the same APC. For CD4 T cells to compete directly with CD8 T cells in such a manner would mean that the CD4 T cell, unlike the CD8 T cell, must induce the loss of numerous peptide–MHC complexes, along with other co-stimulatory molecules. Experiments are now underway to determine the precise nature of such CD4-mediated inhibition. Whatever the mechanism, it is clear that additional antigen-specific CD4 T cells are capable of working to inhibit the CD8 T cell response to a low-dose viral infection. Our results suggest that cell-mediated immune responses to low-dose viral infections may be qualitatively different to that described for both high-dose viral infections and non-infectious model systems. Since we mostly succumb to viral disease initially through a low-dose pathogen encounter, determining the mechanisms and immune response out-comes to low-dose infections will have important implications on our understanding of natural infections, and indeed, how best to combat and/or vaccinate against them.

Acknowledgements

The authors would like to thank Dr Stuart Berzins (Melbourne University, Australia) for critical reading of the manuscript. This work was supported by grants from the National Institutes of Health to L.L.

Abbreviations

APC allophycoerythrin APC antigen-presenting cell

BHK baby hamster kidney

CFSE 5-(and-6)-carboxyfluoroscein succinimidyl ester

IV intravenously

OVA ovalbumin peptide

p.f.u. plaque-forming units TLR Toll-like receptor VSV vesicular stomatitis virus

References

1 Bennett, S. R., Carbone, F. R., Karamalis, F., Flavell, R. A., Miller, J. F. and Heath, W. R. 1998. Help for cytotoxic-T-cell responses is mediated by CD40 signalling.Nature393:478.

2 Ridge, J. P., Di Rosa, F. and Matzinger, P. 1998. A conditioned dendritic cell can be a temporal bridge between a CD4+ T-helper and a T-killer cell.Nature393:474.

3 Schoenberger, S. P., Toes, R. E., van der Voort, E. I., Offringa, R. and Melief, C. J. 1998. T-cell help for cytotoxic T lymphocytes is mediated by CD40-CD40L interactions.Nature393:480. 4 Buller, R. M., Holmes, K. L., Hugin, A., Frederickson, T. N. and

Morse, H. C., 3rd. 1987. Induction of cytotoxic T-cell responses

in vivoin the absence of CD4 helper cells.Nature328:77. 5 Rahemtulla, A., Fung-Leung, W. P., Schilham, M. W.et al.1991.

Normal development and function of CD8+ cells but markedly decreased helper cell activity in mice lacking CD4. Nature

353:180.

6 Smith, C. M., Wilson, N. S., Waithman, J. et al. 2004. Cognate CD4(+) T cell licensing of dendritic cells in CD8(+) T cell immunity.

Nat. Immunol.5:1143.

7 Iwasaki, A. and Medzhitov, R. 2004. Toll-like receptor control of the adaptive immune responses.Nat. Immunol.5:987.

8 Janeway, C. A., Jr and Medzhitov, R. 2002. Innate immune recognition.Annu. Rev. Immunol.20:197.

9 Janssen, E. M., Lemmens, E. E., Wolfe, T., Christen, U., von Herrath, M. G. and Schoenberger, S. P. 2003. CD4+ T cells are required for secondary expansion and memory in CD8+ T lymphocytes.Nature421:852.

10 Shedlock, D. J. and Shen, H. 2003. Requirement for CD4 T cell help in generating functional CD8 T cell memory.Science300:337. 11 Sun, J. C. and Bevan, M. J. 2003. Defective CD8 T cell memory

following acute infection without CD4 T cell help.Science300:339. 12 Murali-Krishna, K., Altman, J. D., Suresh, M.et al.1998. Counting antigen-specific CD8 T cells: a reevaluation of bystander activation during viral infection.Immunity8:177.

13 Butz, E. A. and Bevan, M. J. 1998. Massive expansion of antigen-specific CD8+ T cells during an acute virus infection.Immunity

8:167.

14 Lefrancois, L., Olson, S. and Masopust, D. 1999. A critical role for CD40-CD40 ligand interactions in amplification of the mucosal CD8 T cell response.J. Exp. Med.190:1275.

15 Marzo, A. L., Vezys, V., Klonowski, K. D.et al.2004. Fully functional memory CD8 T cells in the absence of CD4 T cells.J. Immunol.

173:969.

16 Grubin, C. E., Kovats, S., deRoos, P. and Rudensky, A. Y. 1997. Deficient positive selection of CD4 T cells in mice displaying altered repertoires of MHC class II-bound self-peptides.Immunity

7:197.

17 Lyons, A. B. and Parish, C. R. 1994. Determination of lymphocyte division by flow cytometry.J. Immunol. Methods171:131. 18 Carbone, F. R. and Bevan, M. J. 1990. Class I-restricted

processing and presentation of exogenous cell-associated anti-genin vivo. J. Exp. Med.171:377.

19 Lawson, N. D., Stillman, E. A., Whitt, M. A. and Rose, J. K. 1995. Recombinant vesicular stomatitis viruses from DNA. Proc. Natl Acad. Sci. USA92:4477.

20 Schnell, M. J., Buonocore, L., Whitt, M. A. and Rose, J. K. 1996. The minimal conserved transcription stop-start signal promotes stable expression of a foreign gene in vesicular stomatitis virus.

J. Virol.70:2318.

21 Kim, S. K., Reed, D. S., Olson, S. et al. 1998. Generation of mucosal cytotoxic T cells against soluble protein by tissue-specific environmental and costimulatory signals. Proc. Natl Acad. Sci. USA95:10814.

22 Masopust, D., Jiang, J., Shen, H. and Lefrancois, L. 2001. Direct analysis of the dynamics of the intestinal mucosa CD8 T cell response to systemic virus infection.J. Immunol.166:2348. 23 Masopust, D., Vezys, V., Marzo, A. L. and Lefrancois, L. 2001.

Preferential localization of effector memory cells in nonlymphoid tissue.Science291:2413.

24 Bennett, S. R., Carbone, F. R., Karamalis, F., Miller, J. F. and Heath, W. R. 1997. Induction of a CD8+ cytotoxic T lymphocyte response by cross-priming requires cognate CD4+ T cell help.J. Exp. Med.

186:65.

25 Behrens, G. M., Li, M., Davey, G. M. et al. 2004. Helper requirements for generation of effector CTL to islet beta cell antigens.J. Immunol.172:5420.

26 Recher, M., Lang, K. S., Hunziker, L.et al.2004. Deliberate removal of T cell help improves virus-neutralizing antibody production.Nat. Immunol.5:934.

(9)

27 Blattman, J. N., Antia, R., Sourdive, D. J.et al.2002. Estimating the precursor frequency of naive antigen-specific CD8 T cells.J. Exp. Med.195:657.

28 Marzo, A. L., Klonowski, K. D., Bon, A. L., Borrow, P., Tough, D. F. and Lefrancois, L. 2005. Initial T cell frequency dictates memory CD8+ T cell lineage commitment.Nat. Immunol.6:793.

29 Mintern, J. D., Davey, G. M., Belz, G. T., Carbone, F. R. and Heath, W. R. 2002. Cutting edge: precursor frequency affects the helper dependence of cytotoxic T cells.J. Immunol.168:977.

30 Smith, A. L., Wikstrom, M. E. and Fazekas de St Groth, B. 2000. Visualizing T cell competition for peptide/MHC complexes: a specific mechanism to minimize the effect of precursor frequency.Immunity13:783.

31 Kedl, R. M., Kappler, J. W. and Marrack, P. 2003. Epitope dominance, competition and T cell affinity maturation.Curr. Opin. Immunol.15:120.

32 Probst, H. C., Dumrese, T. and van den Broek, M. F. 2002. Cutting edge: competition for APC by CTLs of different specificities is not functionally important during induction of antiviral responses.

J. Immunol.168:5387.

33 Echchakir, H., Bagot, M., Dorothee, G. et al. 2000. Cutaneous T cell lymphoma reactive CD4+ cytotoxic T lymphocyte clones display a Th1 cytokine profile and use a fas-independent pathway for specific tumor cell lysis. J. Investig. Dermatol.

115:74.

34 Eljaafari, A., Farre, A., Duperrier, K. et al. 2001. Generation of helper and cytotoxic CD4+T cell clones specific for the minor histocompatibility antigen H-Y, after in vitro priming of human T cells by HLA-identical monocyte-derived dendritic cells. Trans-plantation71:1449.

35 Grogg, D., Hahn, S. and Erb, P. 1992. CD4+ T cell-mediated killing of major histocompatibility complex class II-positive antigen-presenting cells (APC). III. CD4+ cytotoxic T cells induce apoptosis of APC.Eur. J. Immunol.22:267.

36 Kedl, R. M., Rees, W. A., Hildeman, D. A. et al. 2000. T cells compete for access to antigen-bearing antigen-presenting cells.

J. Exp. Med.192:1105.

37 Kedl, R. M., Schaefer, B. C., Kappler, J. W. and Marrack, P. 2002. T cells down-modulate peptide-MHC complexes on APCsin vivo. Nat. Immunol.3:27.