Reversal of chronic to resolved infection by IL-10 blockade is LCMV strain dependent


Full correspondence: Dr. Kirsten Richter, Institute of Microbiology, ETH Zurich, Wolfgang-Pauli-Str. 10, HCI G407, 8093 Zurich, Switzerland

Fax: +41-44-632-10-98



Chronic viral infections lead to CD8+ T-cell exhaustion, characterized by impaired cytokine secretion. The immune-regulatory cytokine IL-10 promotes chronicity of infection with lymphocytic choriomeningitis virus (LCMV) Clone 13, as absence of IL-10 or blocking of IL-10R during early LCMV Clone 13 infection results in viral clearance. Thus, treatment of humans suffering from chronic viral infections with IL-10 neutralizing or IL-10R blocking antibodies was proposed to boost virus-specific T-cell responses to enhance control or even clear the viral infection. Here we demonstrate that although CD4+ and CD8+ T cells can produce elevated levels of cytokines in IL-10−/− mice early after infection compared with WT mice, IL-10−/− mice cannot clear an infection with the quicker replicating LCMV strain Docile, eventually resulting in T-cell exhaustion. These data suggest that the success of IL-10 blockade to control chronic viral infections may critically depend on the virulence of the infecting strain.


Chronic viral infections such as HIV, HCV, and hepatitis B virus are a major health concern. One hurdle inhibiting the clearance of chronic infections is the functional inactivation of antiviral T cells, also termed T-cell exhaustion. T-cell exhaustion is characterized by a gradual loss of cytokine production of the antiviral CD8+ T cells [1]. High dose lymphocytic choriomeningitis virus (LCMV) Clone 13 and LCMV Docile infection represent well-characterized murine models for chronic viral infections [1].

The host-derived immune inhibitory cytokine IL-10 is crucial in driving T-cell exhaustion and viral chronicity following LCMV infection of mice [2, 3]. Early disruption of IL-10 receptor signaling was shown to prevent the loss of CD8+ T-cell effector functions leading to virus clearance already by day 9 post-infection with LCMV Clone 13 [2, 3]. Also in the case of HIV- or HCV-specific T-cell responses ex vivo blockade of IL-10 evokes enhanced T-cell effector functions [4-6] and is therefore proposed to be used for treatment purposes. Unlike high dose infection of LCMV Clone 13, which can be cleared after about two months by immune-competent hosts, the LCMV strain Docile is known to establish life-long persistence if inoculated at similar doses [7]. In addition, T-cell exhaustion is more severe in LCMV Docile infected mice [7]. To assess whether blockade of IL-10/IL-10R interaction generally results in reversal of chronic to resolved viral infection, we compared LCMV Clone 13 and LCMV Docile infection, both representing LCMV strains leading to chronic infection when inoculated at high doses. Surprisingly, unlike Clone 13, LCMV Docile cannot be cleared by IL-10−/− mice but instead leads to a chronic infection that is accompanied with T-cell exhaustion. This differential susceptibility to IL-10 blockade is likely related to variations in virulence of LCMV Clone 13 and Docile.

Results and discussion

CD4+ and CD8+ T cells exhaust after high dose LCMV Docile infection of IL-10−/− mice

To address whether IL-10 is a crucial host cell factor determining persistence of LCMV not only after Clone 13 infection [2, 3], but also after infection with the LCMV strain Docile, C57BL/6, and IL-10−/− mice were infected with 106 pfu of either strain. As expected from previous reports [2, 3], Clone 13 infected IL-10−/− mice showed higher percentages of degranulating (Fig. 1A and Supporting Information Fig. 1) and cytokine-producing (Fig. 1A and B and Supporting Information Fig. 1) antiviral CD8+ and CD4+ T cells than WT mice. Eventually, IL-10−/− mice were able to reduce Clone 13 titers substantially by day 10 post-infection (Fig. 1C and Supporting Information Fig. 1). In contrast, LCMV Docile-infected IL-10−/− mice exhibited exhausted T cells and developed a chronic infection like that in Docile-infected WT controls (Fig. 1A–C and Supporting Information Fig. 1).

Figure 1.

CD8+ T cells exhaust after high dose LCMV Docile infection. (A–C) Groups of 3–4 C57BL/6 or IL-10−/− mice were i.v. infected with either 106 pfu of LCMV Docile or 106 pfu of LCMV Clone 13. On day 10 post-infection, splenocytes were restimulated with the peptides gp33 or np396 and the percentages of degranulating (CD107) and IFN-γ secreting CD8+ T cells were determined by flow cytometry. Data are shown as (A) representative flow cytometry plots pregated on CD8+ T cells and (B) summary of all mice. (C) The virus titer in the spleen was determined on day 10 post-infection. (D) Groups of 3–4 IL-10−/− and C57BL/6 mice were i.v. infected with 104, 105, or 106 pfu LCMV Docile and the virus titer in spleen was analyzed on day 10 post-infection. One IL-10−/− mouse infected with 105 pfu and one IL-10−/− mouse infected with 106 pfu had to be euthanized during the course of the experiment. Each symbol represents an individual mouse. The solid lines indicate the means, the dashed lines the detection limits. (A, B) Data from one representative experiment of two performed. (C) Data from two pooled experiments. Significance was determined by unpaired two-tailed Student's t-test. *p < 0.05; **p < 0.01; ****p < 0.0001; n.s., not significant.

To evaluate if the inability of the IL-10−/− mice to clear LCMV Docile depends on the titer of the viral inoculum, virus titers were determined after infection with decreasing doses of LCMV Docile. However, none of the virus doses tested resulted in a scenario where C57BL/6 mice displayed a chronic infection while IL-10−/− mice could clear the virus (Fig. 1D, similar in other organs (data not shown)).

To investigate whether LCMV Clone 13 induces a host factor in absence of IL-10 that promotes virus clearance more efficiently than LCMV Docile, IL-10−/− mice were co-infected with 5 × 105 pfu LCMV Docile and 5 × 105 pfu Clone 13. Similar to the LCMV Docile infected IL-10−/– mice and in contrast to the Clone 13 infected IL-10−/− mice, co-infected mice could not clear the infection (Supporting Information Fig. 1J) and exhibited lower CD8+ T-cell responses than Clone 13 infected IL-10−/− mice (data not shown). This could be explained by a more efficient induction of immune inhibitory factors by LCMV Docile or by an increased replication speed and subsequent higher levels of antigen, which are known to efficiently induce the exhaustion of antiviral T cells [8, 9].

Early T-cell responses in IL-10−/− and C57BL/6 mice after LCMV Docile or Clone 13 infection

It is well established that persistent high amounts of antigen drive CD8+ T-cell exhaustion [8, 9]. Furthermore, it was reported that IL-10 acts directly on CD4+ T cells, but not on CD8+ T cells during acute LCMV infection [10] and that CD8+ T-cell exhaustion is more severe after LCMV infection when CD4+ T cells are absent [11]. We therefore hypothesized that increasing virus-specific T-cell help from the beginning of chronic LCMV infection might facilitate control of LCMV Docile infection in IL-10−/− mice. To test this hypothesis, LCMV-specific TCR transgenic Smarta CD4+ T cells were adoptively transferred prior to LCMV Clone 13 or Docile infection and cytokine production by antiviral CD4+ and CD8+ T cells was analyzed on days 4, 5, and 6 post-infection. As depicted in Figure 2A–D, the priming of antiviral cytokine producing CD4+ T cells was not more potent in Clone 13 than in Docile-infected IL-10−/− mice and demonstrated to be more efficient than in C57BL/6 mice in both cases. However, the percentage of degranulating and IFN-γ-producing gp33- and np396-specific CD8+ T cells was higher after the infection of IL-10−/− mice with Clone 13 than with Docile (Fig. 2E–H), likely reflecting the already decreased virus titers in Clone 13 infected IL-10−/− mice (Fig. 2I–L). The data depicted show functional exhaustion on a per cell basis. The differences between the experimental groups were also reflected by the total cell numbers (data not shown).

Figure 2.

Comparison of the T-cell priming in IL-10−/− and C57BL/6 mice after Docile or Clone 13 infection. One day prior to infection with 106 pfu of either LCMV Docile or LCMV Clone 13, groups of three IL-10−/− and C57BL/6 mice were transferred with 5 × 105 MACS purified Ly5.1+ TCR transgenic Smarta CD4+ T cells. Splenocytes were analyzed on days 4, 5, and 6 post-infection for the percentage of (A) Ly5.1+ CD4+ T cells. The percentages of (B) IFN-γ+, (C) TNF-α+, and (D) IL-2+ of Ly5.1+ CD4+ T cells were determined after stimulation with the peptide p13. The percentages of (E and F) CD107+ or (G and H) IFN-γ+ of CD8+ T cells were determined after stimulation with the peptides (E and G) gp33 and (F and H) np396. Mean + SD of three mice per group is depicted. One representative of two experiments is shown. (I–L) 2 × 105 MACS-purified Smarta CD4+ T cells were transferred to C57BL/6 and IL-10−/− mice 1 day prior to infection with 106 pfu of either Clone 13 or Docile. Virus titers were analyzed on day 6 post-infection in (I) spleen, (J) kidney, (K) liver, and (L) lung. Each symbol represents an individual mouse; the solid lines indicate the means and the dashed lines the detection limits. Significance was determined by unpaired two-tailed Student's t-test. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

LCMV Clone 13 infects the same cell types as LCMV Docile

A point mutation in the glycoprotein (F260L) that alters receptor tropism [12] was shown to contribute to the higher ability of Clone 13 to persist compared to its parental strain Armstrong [12]. Therefore, we investigated if Clone 13 and Docile infect different cell types. However, both virus strains exhibited a similar tropism and infected DCs and macrophages most efficiently (Supporting Information Fig. 2).

In addition, it was shown that chronic LCMV infection leads to a reduction in the numbers of APCs that is in turn responsible for the decreased T-cell responses [13]. Therefore, we compared the composition of cellular subsets in the spleens of WT mice infected with LCMV Docile or Clone 13. However, the percentage of CD4+ and CD8+ T cells, B cells, monocytes, macrophages, CD4+ and CD8+ DCs, and NK and NKT cells was not significantly different in mice infected with Clone 13 or Docile on days 2 and 5 post-infection (data not shown).

LCMV Docile replicates quicker than LCMV Clone 13

Interestingly, the percentages of cytokine-producing CD4+ and CD8+ T cells in Clone 13 infected WT mice were elevated compared with those from Docile-infected WT mice (Fig. 1A and B and Supporting Information Fig. 1) already by day 6 post-infection (Fig. 2). The stronger degree of exhaustion in Docile-infected mice might reflect lower levels of antigen in Clone 13 infected mice compared to Docile-infected mice. As it was shown recently that a more active viral polymerase facilitates viral persistence [14], we compared the replication speed of Clone 13 and Docile in vivo. Indeed, Clone 13 infected mice displayed lower virus titers in various organs already early after infection, likely reflecting decreased replication speed compared to Docile (Fig. 3A–D). Interestingly, the differences were bigger in peripheral organs (Fig. 3C and D, similar trends in liver and brain (data not shown)) than in secondary lymphatic organs (Fig. 1A and B). To formally exclude a contribution of a different activation of innate immune cells by Clone 13 and Docile, viral replication speed was compared in MC57G fibroblasts in vitro. In line with our hypothesis, Docile titers increased significantly quicker than Clone 13 titers (p< 0.005, Two-way Anova test; Fig. 3E). The plateau of the maximal virus titers was reached quicker in Docile-infected cells than in Clone 13 infected cells (Fig. 3E). Thus, Docile's elevated replication speed likely causes chronicity in vivo. It cannot be ruled out, however, that a different recognition by cell-intrinsic innate mechanisms or different potency of cell-intrinsic innate mechanisms against both strains contribute to the different velocity of viral spread.

Figure 3.

LCMV Docile replicates to higher titers than Clone 13. C57BL/6 mice were i.v. infected with 106 pfu of LCMV Docile or LCMV Clone 13 and virus titers in (A) spleen, (B) mesenteric lymph node, (C) lung, and (D) kidney were determined on day 10 post-infection. Each symbol represents an individual mouse, the solid lines indicate the means and the dashed lines the detection limits. (E) MC57G cells were infected with a MOI of 0.1. Virus titers in the supernatants were analyzed at the indicated time points. The values represent the ratio between the titers in supernatants from Docile-infected cells to Clone 13 infected cells. Each symbol represents a quadruplicate. The solid lines indicate the means. One representative of two experiments is shown. (F) C57BL/6 and IL-10−/− mice were i.v. infected with 106 pfu of LCMV Docile in addition to anti-PD-L1 or mock treatment and morbidity was assessed by core body temperature measurements. Data are shown as mean + SEM of four mice per group. One representative of three experiments is shown.

First indications that blockade of IL-10/IL-10R might not be a generally successful intervention against chronic infections came from a study reporting that even though Clone 13 infected IL-10−/− mice initially possess more functional antiviral T cells, they cannot clear the infection by day 30 [15]. This obvious discrepancy to the studies observing clearance of Clone 13 in IL-10−/− mice [2, 3] might reflect slight differences in the virulence of the Clone 13 isolates used in the different studies. Our own observations support this hypothesis as inoculation of IL-10−/− mice with Clone 13 at higher titers results in persistent infection (data not shown). Thus, the presented data indicate that the absence of IL-10R signaling can turn an otherwise chronic infection into an acute resolved one in situations when the replicative speed of the persistently infecting virus is below a given threshold and when the inoculum does not exceed a specific threshold. It is therefore important to include these considerations when thinking about targeting IL-10 or IL-10R in the context of persistent viral infections, as it is likely that neutralization of IL-10 or blockade of the IL-10R will not be sufficient to clear quickly replicating viruses. Furthermore, also in case of chronic LCMV Clone 13 infection, IL-10R blockade was only very effective in enhancing viral control when blockade was initiated during the first 5 days of infection. Later initiation of treatment resulted in much more moderate viral control [2]. It is therefore likely that multiple immune regulatory molecules such as PD-L1/PD1, LAG-3, Tim-3, CTLA-4, and TGF-β or immunoregulatory cells such as Treg cells will have to be targeted simultaneously [1] in addition to the treatment with antivirals in case of quickly replicating chronic infections or when initiation of treatment is started during established chronic infection. It should be noted, however, that caution has to be taken concerning the attempts to use IL-10/IL-10R blocking drugs alone or in combination with additional immune-stimulatory interventions during chronic viral infections as potent (re)invigoration of antiviral T-cell responses before establishment or perhaps also during chronic infections bears the risk of potentially fatal immunopathology [8]. Indeed, when we combined simultaneous inhibition of signaling through IL-10R and PD-1 during establishment of chronic LCMV Docile infection this resulted in exacerbated fatal immunopathology (Fig. 3F).

Materials and methods

Virus, viral peptides, lymphocytes stimulation, flow cytometry, and adoptive transfer of lymphocytes

LCMV Docile and Clone 13 was propagated on Madin Darby canine kidney (MDCK) or baby hamster kidney cells (BHK), respectively. Titers were determined by plating of serial dilutions on MC57G cells and subsequent intracellular staining of the viral nucleoprotein as described previously [16]. Lymphocytes were isolated and stained as previously described [8]. Adoptive transfers were done as previously described [8].


C57BL/6 (Janvier Elevage, Le Genest Staint Isle, France), IL-10−/− mice (>10 generations backcrossed to C57BL/6 background) and Ly5.1+ Smarta mice (C57BL/6 background) [17] were kept under specific pathogen-free conditions and were i.v. infected with LCMV Docile or Clone 13. Mice were daily treated i.p. with 200 μg αPD-L1 (clone 10F.9G2, BioXCell, West Lebanon, NH, USA) if indicated. The body core temperature was measured using a rectal temperature probe of a Homeothermic Blanket Control Unit (Harvard Apparatus, Holliston, MA, USA). Animal experiments were performed according to the regulations of the cantonal veterinary office (animal experimentation number 146/2008).


We thank Nathalie Oetiker and Franziska Wagen for excellent technical assistance and the members of the Oxenius group for fruitful discussions. This work was supported by the ETH Zurich and the Swiss National Science Foundation (Grant No. 310030-129751 to AO).

Conflict of interest

The authors declare no financial or commercial conflict of interest.


lymphocytic choriomeningitis virus