SEARCH

SEARCH BY CITATION

Keywords:

  • Killer cell lectin-like receptor G1;
  • NK cells;
  • Rodent;
  • T cells

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information

The killer cell lectin-like receptor G1 (KLRG1) is expressed by NK and T-cell subsets and recognizes members of the classical cadherin family. KLRG1 is widely used as a lymphocyte differentiation marker in both humans and mice but the physiological role of KLRG1 in vivo is still unclear. Here, we generated KLRG1-deficient mice by homologous recombination and used several infection models for their characterization. The results revealed that KLRG1 deficiency did not affect development and function of NK cells examined under various conditions. KLRG1 was also dispensable for normal CD8+ T-cell differentiation and function after viral infections. Thus, KLRG1 is a marker for distinct NK and T-cell differentiation stages but it does not play a deterministic role in the generation and functional characteristics of these lymphocyte subsets. In addition, we demonstrate that E-cadherin expressed by K562 target cells inhibited NK-cell reactivity in transgenic mice over-expressing KLRG1 but not in KLRG1-deficient or WT mice. Hence, the inhibitory potential of KLRG1 in mice is rather weak and strong activation signals during viral infections may override the inhibitory signal in vivo.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information

The killer cell lectin-like receptor G1 (KLRG1) belongs to the C-type lectin family and contains a single ITIM in its cytoplasmic domain. The human gene is part of the NK gene complex, whereas the murine homolog of KLRG1 maps 2 cM distant from the complex 1, 2. KLRG1 was first described in the rat and was originally termed mast cell function-associated antigen, given that antibody ligation inhibited the secretory response in RBL-2H3 mast cells 3, 4. In both humans and mice, KLRG1 is found on subsets of NK cells and T cells but not on other cell types, including mast cells 5–7. Viral, bacterial or parasite infections strongly induce KLRG1 expression in NK cells and T cells and most T cells with effector or effector-memory phenotypes are KLRG1+8–11. T cells expressing KLRG1 have normal effector functions but the proliferative capacity of these cells is impaired 7, 11–14. In addition, KLRG1 was shown to serve as a marker for short-lived effector CD8+ T cells during viral infection 15, 16.

Within the NK-cell population, KLRG1 is predominantly found in the most mature CD11b+CD27 NK-cell subset in mice 17–19 and in CD56dim NK cells in humans 7. Of interest, NK cells from MHC-class-I-deficient mice express lower levels of KLRG1 20. Moreover, KLRG1 expression by NK cells after murine cytomegalovirus (MCMV) infection has been demonstrated to inversely correlate with the ability to produce IFN-γ 21. Thus, similar to T cells, KLRG1 is also a marker for NK cells that are approaching the end of their differentiation stage.

Members of the classical cadherin family have been identified as ligands for both human and mouse KLRG1 22–25. In addition, inhibition of T and NK-cell function by interaction of KLRG1 with E-cadherin has been demonstrated in some but not all experimental settings 22–24, 26. These findings suggested a role for KLRG1 in dampening KLRG1+ lymphocytes in tissues expressing cadherins in order to prevent immunopathology 27. The crystal structure of KLRG1 in complex with E-cadherin has recently been solved 28. It shows that KLRG1 binds to a highly conserved site on cadherins that overlaps with the site involved in homophilic trans interaction but is distinct from the αEβ7 (CD103) binding site. An exceptionally weak affinity of KLRG1 to cadherins has further been noted substantiating the notion that KLRG1–cadherin interaction occurs through multivalent binding and involves the formation of multimeric receptor/ligand complexes 26.

Despite KLRG1 being widely used as a lymphocyte differentiation marker, and the substantial progress made in structural and functional characterization of KLRG1, the role of KLRG1 in vivo is still poorly defined. To address this issue, we generated KLRG1-deficient mice by homologous recombination. The characterization of these mice indicates that KLRG1 is dispensable for normal CD8+ T-cell differentiation and memory cell formation after viral infections. In addition, KLRG1 deficiency did not affect development and function of NK cells in the various assays used in this study.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information

Generation of KLRG1-deficient mice

KLRG1-deficient mice were generated by homologous recombination using a targeting construct that carries a lacZ reporter gene and a neo cassette inserted into the third exon of the mouse Klrg1 gene (Fig. 1A). This exon encodes the neck region and the proximal half of the C-type lectin domain of KLRG1 2. A homologous recombinant HM1 ES cell clone (M31) was injected into B6 blastocysts and resulting 129/B6 chimeric mice were crossed with B6 mice to attain germ line transmission. Homologous recombination was confirmed by Southern blot analysis of the offspring (Fig. 1B) and Klrg1+/− mice were first backcrossed to the B6 background for six generations. The Klrg1 gene locates 2.2 cM outside the NK gene complex (NKC) on chromosome 6 2. To generate KLRG1 KO mice carrying the well-characterized NKC of B6 mice, we used a marker-assisted strategy to identify offspring in the consecutive B6 backcrosses that carried a recombination between the disrupted Klrg1 gene (59.2 cM) and the NKC (62 cM). In the 11th and 12th backcross generation, we identified such Klrg1+/− mice that were intercrossed to generate a KLRG1-deficient mouse line. Northern blot analysis of spleen cells from lymphocytic choriomeningitis virus (LCMV)-infected mice showed that Klrg1−/− mice did not express KLRG1 mRNA in contrast to Klrg1+/− or Klrg1+/+ mice (Fig. 1C). To demonstrate lack of KLRG1 expression at the protein level, lymphocytes from KLRG1 KO and WT mice were stained with KLRG1-specific mAb. The results revealed that that NK cells and LCMV-activated CD8+ T cells from KLRG1 KO mice were not stained by anti-KLRG1 mAb in contrast to cells from WT mice (Fig. 1D). KLRG1 KO mice were born in the expected Mendelian ratio and the mice exhibited no visible alterations in major organs or overt pathology. In addition, primary and secondary lymphoid organs such as thymus, spleen and lymph node did not reveal detectable abnormalities with respect to total cell numbers and lymphocyte subset composition (data not shown).

thumbnail image

Figure 1. Generation of KLRG1 KO mice. (A) Partial restriction map of the KLRG1 locus and the targeting construct. Exons are represented by filled boxes. (B) Southern blot analysis of Kpn I-digested tail DNA from Klrg1+/+ and Klrg1+/− mice using the 5′-flanking probe as indicated in (A). (C) Northern blot analysis of spleen cells from LCMV-infected WT, heterozygotes and KO mice using a KLRG1 cDNA probe. Equivalent sample loading was confirmed by ethidium bromide staining (bottom). (D) KLRG1 expression as determined by staining with KLRG1-specific mAb of NK cells (left) and CD8+ T cells (right) from LCMV-infected WT and KO mice. Shown are representative staining from three independent experiments.

Download figure to PowerPoint

Normal induction of virus-specific CD8+ T-cell response in KLRG1 KO mice

KLRG1 is strongly induced in CD8+ T cells after viral infections 9–11. To determine whether KLRG1 deficiency influences induction of virus-specific CD8+ T cells, KLRG1 KO mice were infected with LCMV or vesicular stomatitis virus (VSV) and virus-specific CD8+ T cells were enumerated with MHC class I tetramers. The experiments revealed that KLRG1 KO mice generated a normal LCMV- or VSV-specific CD8+ T-cell response at the acute and the memory phase of the infection (Fig. 2A). Moreover, effector and memory T-cell subsets analyzed by CD62L and CD127 expression were indistinguishable in KO and WT mice for both types of infections (Fig. 2B and C). Thus, despite being abundantly expressed by anti-viral effector CD8+ T cells, KLRG1 deficiency did not affect induction of antigen-specific CD8+ T cells after LCMV and VSV infection. Similar results were obtained when CD8+ T-cell responses to infections with vaccinia virus (VV) or Listeria monocytogenes were examined (data not shown).

thumbnail image

Figure 2. Analysis of the LCMV- and VSV-specific CD8+ T-cell response in KLRG1 KO mice. (A) WT and KLRG1 KO mice were infected with LCMV or VSV and the percentage of virus-specific CD8+ T cells were determined in the spleen with MHC class I tetramers at the indicated time points after infection. Dots represent values from individual mice. (B) Expression of KLRG1, CD62L and CD127 by LCMV-GP33- and VSV-NP52-specific CD8+ T cells from WT and KO at the acute (day 7/8) and memory phase (week 5) of infection. Data are representative of two to three independent experiments with three animals per time point. (C) Percentage of CD62L+ and CD127+ cells of LCMV-GP33- and VSV-NP52-specific CD8+ T cells from WT (closed bars) and KO mice (open bars) at the acute (day 7/8) and memory phase (week 5) of infection. Mean values including SEM from two to three independent experiments with three animals per time point are shown.

Download figure to PowerPoint

CD8+ T-cell differentiation after LCMV infection occurs normally in the absence of KLRG1

The results shown above were performed with mice with a polyclonal TCR repertoire involving a broad range of different affinities for viral antigens. To extend our analysis to a system with a monoclonal TCR with defined affinity for the nominal antigen, we used the well-characterized P14 TCR transgenic model specific for the LCMV GP33 antigen. First, we determined co-expression of KLRG1 with several T-cell differentiation markers in this system. In effector and memory P14 CD8+ T cells, KLRG1 expression was more frequent in CD5-, CD27-, CD62L- and CD127-negative cell subsets compared with the corresponding marker positive subsets (Fig. 3A). In addition, KLRG1 expression was increased in IFN-γ secreting P14 cells but decreased in cells producing IL-2 after stimulation (Fig. 3B). Thus, KLRG1 was preferentially expressed by CD8+ T cells with a “late” differentiation phenotype.

thumbnail image

Figure 3. Correlation of KLRG1 expression by effector and memory P14 cells with T-cell differentiation markers. (A) KLRG1 expression in effector and memory P14 T cell subsets defined by CD5, CD27, CD62L and CD127 expression. P14 T cells (Thy1.1+) generated by adoptive transfer into B6 mice followed by LCMV infection were analyzed in the spleen at the indicated time points after infection. Representative dot plots gated on Thy1.1+ P14 T cells show co-expression of KLRG1 with the indicated markers (left panels); data (mean+SEM) of KLRG1 expression are summarized as bar graphs (right panels), in various cell subsets. (B) KLRG1 expression by cytokine-positive or cytokine-negative P14 T-cell subsets. Cytokine production was determined by intracellular staining after GP33 peptide stimulation. Representative dot plots are gated on P14 T cells and data are summarized in bar graphs showing mean+SEM. Data are pooled from two independent experiments with two to three mice per group.

Download figure to PowerPoint

To determine whether KLRG1 played a causal role in CD8+ T-cell differentiation, expression of the T-cell differentiation markers used above was compared in P14 T cells from KLRG1 KO and WT mice at the acute and at the memory phase of the LCMV infection. Adoptively transferred P14 T cells from KLRG1 KO and WT mice proliferated to the same extent in recipient mice after LCMV infection and gave rise to similar numbers of memory T cells (Fig. 4, left). In addition, expression of CD5, CD27, CD62L and CD127 on effector and memory P14 T cells and their capacity to secrete IFN-γ and IL-2 after antigen stimulation did not differ between KO and WT cells (Fig. 4, right). Thus, these data indicate that the differentiation pathways of P14 T cells after LCMV infection were not altered in the absence of KLRG1.

thumbnail image

Figure 4. KLRG1 deficiency does not influence LCMV-induced P14 T-cell differentiation and function. P14 T cells (Thy1.1+) from WT (black circles and bars) and KLRG1 KO mice (open circles and bars) were adoptively transferred into B6 mice followed by LCMV infection. At the indicated time points after infection, absolute numbers of P14 T cells were determined in the spleen (left panel), their cell surface phenotype was analyzed by staining with CD5-, CD27-, CD62L- and CD127-specific mAb (middle panel) and their capacity to produce IFN-γ and IL-2 after GP33 peptide stimulation was determined by intracellular cytokine staining (right panel). Dots present values from individual mice and the bars show mean+SEM from two independent experiments with two to three mice per group.

Download figure to PowerPoint

Impaired proliferation capacity of memory P14 T cells is independent of KLRG1 expression

We and others have previously demonstrated that repetitively stimulated P14 memory T cells express high levels of KLRG1 and are impaired in their proliferation capacity after antigen stimulation 11, 29. In addition, recent data in the human system indicate that KLRG1 signaling induces defective Akt phosphorylation and proliferative dysfunction of highly differentiated CD8+ T cells 14. To determine whether KLRG1 is causally linked to impaired proliferation, P14 T cells from KLRG1 KO and WT mice were used in consecutive adoptive T-cell transfer experiment as outlined in Fig. 5A. Confirming previous findings 11, 29, “tertiary” P14 memory T cells from WT mice were mostly KLRG1+ and expanded only marginally after antigen stimulation in vivo when compared with naïve or primary memory P14 cells (Fig. 5B and C). However, “tertiary” P14 memory T cells from KLRG1 KO mice also proliferated poorly, demonstrating that the impaired proliferative capacity of these cells was not due to KLRG1 expression.

thumbnail image

Figure 5. Impaired proliferation of repetitively stimulated P14 memory T cells is not due to KLRG1 expression. (A) Experimental system: P14 T cells (Thy1.1+) from WT and KLRG1 KO mice were adoptively transferred into B6 mice followed by LCMV infection. After 3–4 wk, primary (1°) memory P14 cells were isolated and re-transferred into B6 mice followed by LCMV infection to generate secondary (2°) and subsequently tertiary (3°) P14 memory T cells that were finally transferred into B6 mice followed by LCMV infection. (B) Expansion of naïve P14 and of 1°, 2° and 3° P14 T cells memory after transfer and LCMV infection. Percent values of P14 WT (black circles) and of P14 KLRG1 KO cells (open circles) of total CD8+ T cells in PBL of the recipient mice at day 8 p.i. are shown. (C) KLRG1 expression by naïve P14 and by 1°, 2° and 3° P14 memory T cells used for the adoptive transfers. P14 WT (black circles) and of P14 KLRG1 KO cells (open circles). Dots present values from individual mice and one out of two independent experiments is shown.

Download figure to PowerPoint

KLRG1 expression and memory T-cell inflation after MCMV infection

Infection of mice with MCMV leads to CD8+ T-cell memory inflation whereby the magnitude of the response to some epitopes (i.e. M38 or m139 in B6 mice) increases with time, whereas T-cell reactivity to other epitopes (i.e. M45 in B6 mice) contracts after the peak of the acute phase 30, 31. Interestingly, KLRG1 expression by M38- or m139-specific CD8+ T cells also increased in the course of the infection whereas the portion of KLRG1+ cells within the pool of M45-specific CD8+ T cells decreased (Fig. 6A). This observation prompted us to examine epitope-specific CD8+ T cells in MCMV-infected KLRG1 KO mice. The results, however, revealed that KLRG1 deficiency did neither alter frequency nor distribution of M45-, M38- or m139-specific CD8+ T cells in MCMV-infected mice at the acute or memory phase of the infection (Fig. 6B).

thumbnail image

Figure 6. KLRG1 expression and memory T-cell inflation after MCMV infection. (A) B6 mice were infected with MCMV and KLRG1 expression was determined in MCMV epitope-specific CD8+ T cells from the spleen as identified by intracellular IFN-γ staining. (B) WT and KLRG1 KO mice were infected with MCMV and the CD8+ T-cell response to the M45, M38 and m139 epitopes was determined in the spleen by intracellular IFN-γ staining. Dots present values from individual mice analyzed at the indicated time point after infection.

Download figure to PowerPoint

KLRG1 is dispensable for NK-cell development and function

KLRG1 is expressed by 30–50% of NK cells and NK-cell activation is associated with KLRG1 upregulation 18, 20, 21. KLRG1 KO mice had normal numbers of CD3 NK1.1+ NK cells in spleen, liver and lung and expression of various stimulatory and inhibitory receptors including 2B4, Ly49A, Ly49C, Ly49D, Ly49G2, Ly49I, Ly49F, NKG2A/E/C and NKG2D was also not different (data not shown). Infection of KLRG1 KO mice with viral (VSV, Vaccinia, LCMV, MCMV) or bacterial (L. monocytogenes) pathogens resulted in a decrease of immature CD11bCD27+ NK cells and an increase of more mature CD11b+CD27+ and CD11b+CD27 NK-cell subsets. As depicted in Fig. 7A, the different types of infections induced distinct patterns of these three NK-cell subsets, but KLRG1 deficiency did not influence their proportions. Similarly, IFN-γ production induced by NK1.1 antibody-ligation (Fig. 7B), cell-mediated lysis of RMA-S target cells by poly(I:C)-activated NK cells (Fig. 7C) and NKG2D-triggered IFN-γ responses by virus-activated NK cells (Fig. 7D) did not differ between KLRG1 KO and WT mice. Moreover, the viral elimination kinetics after infection with MCMV was similar in both types of mice (Fig. 8A). To avoid strong NK-cell activation via Ly49H/m157 interaction after MCMV infection 32, 33, we finally used mutant MCMV lacking m157 (△m157) 34. We also failed to observe a difference in viral titers in spleen of KLRG1 KO and WT mice under these conditions (Fig. 8B). MCMV titers in liver and lungs of KO mice were very slightly increased but we consider these differences too small to allow any further conclusion. Taken together, these data indicate that KLRG1 is dispensable for normal development and function of NK cells in the assays used here.

thumbnail image

Figure 7. Normal NK cell responses in KLRG1 KO mice. (A) Proportions of NK cell subsets defined by CD27 and CD11b expression in the spleen 20 h after i.v. injection of the indicated infectious agents. The gates used to define the three NK cell subsets are indicated in the CD27/CD11b dot plot and the histograms show KLRG1 expression of the subsets in non-infected WT mice. The bars indicate the percentage of CD3NK1.1+ cells with a CD27+CD11b (subset I), CD27+CD11b+ (subset II) or a CD27CD11b+ (subset III) cell surface phenotype. WT (black bars) and KO (open bars). (B) NK1.1-triggered IFN-γ production and (C) cytolytic activity against RMA-S target cells of poly(I:C)-activated NK cells from WT (closed circles) and KO mice (open circles). (D) NKG2D-triggered IFN-γ production of LCMV- and VSV-activated NK cells from WT (black bars) and KO mice (open bars). Mean values±SEM from two independent experiments.

Download figure to PowerPoint

thumbnail image

Figure 8. KLRG1 is dispensable for efficacious control of MCMV infection. WT (black circles) and KLRG1 KO (open circles) mice were infected with (A) 5×104 PFU SGV MCMV or (B) 5×105 PFU △m157 MCMV and viral titers were determined by standard plaque assay at the indicated time points after infection. Titers in individual mice (circles) and mean values (horizontal dashed line) are shown. Virus titers were calculated per organ of spleen, lungs and salivary glands (submandibular gland) and per gram of liver. The detection limit of the virus plaque assay is indicated by the horizontal dotted line and circles with vertical arrow indicate values below detection level. Data are derived from two independent experiments.

Download figure to PowerPoint

E-cadherin-mediated inhibition of NK-cell function in mice requires high KLRG1 expression levels

Members of the classical cadherin family were recently identified as ligands for KLRG1 22, 23, 25. In addition, we demonstrated that human E-cadherin expressed by K562 target cells inhibited effector function of freshly isolated human NK cells 24 but we failed to observe an inhibitory effect of E-cadherin when IL-2-activated mouse NK cells and B16 target cells were used 22. To test whether E-cadherin expressed by K562 cells could inhibit NK-cell function in the murine system, IL-12-pre-activated mouse NK cells were co-cultured with E-cadherin- or mock-transduced K562 cells and IFN-γ production was determined by intracellular cytokine staining. As shown in Fig. 9A, the IFN-γ response of NK cells from KLRG1-transgenic (TG) mice that constitutively express KLRG1 was significantly decreased by stimulation with E-cadherin- when compared with mock-transduced K562 cells. In contrast, NK cells from KO mice were not inhibited by E-cadherin and we even observed that K562-E-cadherin stimulator cells triggered NK cells from these mice more efficiently when compared with mock-transduced K562 cells.

thumbnail image

Figure 9. Reactivity of NK cells from KLRG1 TG, KLRG1 KO and WT mice to K562 target cells expressing E-cadherin. (A) IL-12-preactivated NK cells from KLRG1 TG and KLRG1 KO mice were cultured with mock- or E-cadherin-transduced K562 cells. After 5 h, cells were surface stained for CD3 and NK1.1 and intracellular IFN-γ. Representative dot plots gated on CD3 cells (top panels). The numbers in the quadrants indicated the percentages of IFN-γ+ cells of CD3NK.1.1+ cells. Data are pooled from five independent experiments and the ratios of IFN-γ+ cells after stimulation with E-cadherin compared with mock-transduced K562 cells are shown. *p<0.05, **p<0.01 (bottom panel). (B) IL-12-preactivated NK cells from B6 WT mice were cultured with mock- or E-cadherin-transduced K562 cells. After 5 h, cells were surface stained for CD3, NK1.1, KLRG1 and intracellular IFN-γ. Representative dot plots gated on CD3NK.1.1+ cells (top panels). The numbers in the quadrants indicate the percentage of IFN-γ+ cells of KLRG1 and KLRG1+ NK cells. Data are pooled from four independent experiments and the ratios of IFN-γ+ cells after stimulation with E-cadherin- compared with mock-transduced K562 cells for KLRG1 and KLRG1+ NK cells are shown (bottom panel). (C) KLRG1 expression on CD3NK1.1+ cells from KLRG1 TG, KLRG1 KO and WT mice.

Download figure to PowerPoint

Next, it was of interest to determine whether E-cadherin expressed by K562 cells also inhibited KLRG1+ NK cells from normal WT mice. In contrast to KLRG1 transgenic NK cells, IFN-γ production by KLRG1+ NK cells from normal WT mice was not affected by E-cadherin expression (Fig. 9B). Of note, the level of KLRG1 expression by NK cells from KLRG1 TG mice was considerably higher when compared with NK cells from WT mice (MF 186 versus 43) (Fig. 9C). These data indicate that high KLRG1 expression levels by NK cells are required for E-cadherin-mediated inhibition in the murine system. It is noteworthy that functional activity of human KLRG1+ NK cells could be significantly inhibited by E-cadherin in the same assay system used here with K562 target cells 24. Since natural KLRG1 expression by ex vivo isolated NK cells from humans and mice are similar, these data point to a difference in the inhibitory capacity of mouse and human KLRG1.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information

In an attempt to unravel the role of KLRG1 in vivo, we generated and characterized KLRG1-deficient mice. Although a number of different infection models and assays systems were used, we failed to observe an effect of KLRG1 deficiency on NK and T-cell differentiation and function in vivo. How can these “negative” findings be rationalized? In the targeting vector used for homologous recombination, the Klrg1 gene was disrupted by insertion of a LacZ/neomycin expression cassette into the third exon. Appropriate homologous recombination was confirmed by Southern blotting and lack of KLRG1 expression was also verified at the mRNA and at the protein level. Alternatively spliced transcripts of the Klrg1 gene are detectable at a low level but none of these transcripts is predicted to give rise to a protein with residual KLRG1 activity since they either lack a transmembrane region or lead to a frame shift in the extracellular part 2. Thus, the lack of a phenotype in KLRG1 KO mice is very unlikely due to imperfect ablation of the Klrg1 gene. Of note, KLRG1 is present in the genome as a single copy gene 6 and closely related receptors are not known.

Inhibition of T and NK-cell function by antibody- or E-cadherin-mediated ligation of KLRG1 has been documented by several groups 21–23, 25, 26. It is, however, important to stress that all inhibition experiments published so far in the murine system have been performed with retrovirally transduced cell lines or transgenic lymphocytes that over-express KLRG1. We demonstrate here that E-cadherin expressed by K562 target cells could only inhibit NK cells from transgenic mice over-expressing KLRG1 but not from normal mice. This indicates that the inhibitory potential of mouse KLRG1 is rather weak and requires high levels of expression. It is therefore possible that the weak inhibitory signal through KLRG1 was overruled by strong activation stimuli in the infections models used here. Model systems that are accompanied with lower activation of immune cells may therefore be suited better to unravel the function of KLRG1 in vivo.

The reactivity of lymphocytes is regulated by a considerable number of activating and inhibitory receptors. Elimination of only one type of inhibitory receptor with even a weak inhibitory potential may therefore not be sufficient to detectably alter their functional activity. It is also possible that the loss of KLRG1 in NK or T cells is compensated by altered expression of other cell surface recognition structures. The observed increased reactivity of NK cells from KLRG1 KO mice toward E-cadherin-transfected target cells was unexpected. Besides KLRG1, there is only one additional receptor, αEβ7 (CD103), known to be expressed on lymphocytes that can bind E-cadherin 35. However, the NK cells used in our experiments did not express CD103 (data not shown). In addition to its adhesive role, E-cadherin is also involved in the Wnt signaling pathway by sequestering β-catenin and is also known to inhibit the ligand activation of receptor tyrosine kinases 36. Thus, it is possible that ectopic expression of E-cadherin in K562 cells alters the expression of other yet undefined cell surface molecules that may play a role in NK-cell recognition.

KLRG1 expression has been associated with distinct stages during NK and T-cell differentiation and differences between KLRG1+ and KLRG1 lymphocytes subsets have been demonstrated in several instances. This includes the decreased ability of MCMV-activated KLRG1+ NK cells to produce IFN-γ 21, the low level of KLRG1 expression by non-responsive NK cells lacking self-MHC-specific inhibitory receptors 20, 37, the impaired capacity of KLRG1+ effector/memory T cells to proliferate 7, 11, 13, 14, 29, the paucity of KLRG1+ effector/memory cells to produce IL-2 and inability of KLRG1+ effector cells to give raise to long-lived memory T cells 15, 16. Importantly, the experiments performed here revealed that KLRG1 serves as marker for these lymphocyte differentiation stages and their functional characteristics but it does not play a deterministic role. Of note, treatment of B6 mice with anti-KLRG1 mAb did also not affect induction of LCMV-specific CD8+ T cells determined by MHC class I tetramer staining and did also not influence the extent of CD62L- and CD127-downregulation in these cells during the acute phase of the infection (data not shown).

Even though our study did not reveal alterations of immune functions in the absence of KLRG1, we certainly cannot exclude the possibility that KLRG1 regulates T-cell or NK-cell functions that we have not investigated in this first characterization of these mice. We have recently observed that KLRG1-E-cadherin binding can also strengthen the interaction between cells 26. Thus, the effect of KLRG1 deficiency on lymphocyte adhesion in epithelial tissues expressing E-cadherin such as lung, intestine or skin will have to be tested. In addition, autoimmune models in which slightly activated lymphocytes persist in such tissues could now be used together with the KLRG1-deficient mice generated here.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information

Generation of KLRG1 KO mice

The targeting vector was constructed by insertion of an IRES-LacZ-neo cassette into the PvuII site of exon 3 of KLRG1 using a 10 Kb KpnI fragment containing exons 1, 2 and 3. The fragment was derived from a PAC clone (RPCIP711N196Q3) from a mouse 129/Sv genomic library RPCI 21 (RZPD, Berlin, Germany). The NotI-linearized construct was transfected by electroporation into 129/Sv-derived HM1 ES cells, and G418-selected ES cell colonies were screened for homologues recombination by Southern blot analysis. Two out of 120 clones carried the recombined Klrg1 gene and one of them (M31) was injected into B6 blastocysts and resulting chimeric mice were crossed with B6 mice to attain germ line transmission. Germline competent mice were first backcrossed to B6 mice for six generations. To identify offspring that carried a recombination between the targeted KLRG1 allele (59.2 cM on chromosome 6) and the NKC (62.12–63.6 cM), DNA from tails of the offspring during further backcrossing to B6 mice were analyzed by PCR for a B6/129 sequence polymorphisms of the CdKn1b (p27Kip1) gene 38 (62.0 cM) using the following primers: 5′-GTTACTTTTGAGTGCAGGAG-3′ and 5′-TTTCTTAGCCACATCTTTGC-3′. This PCR yielded a product of 165 bp for B6 and 95 bp for 129/Sv mice. The presence of the targeted KLRG1 allele was determined by a PCR specific for NEO (5′-CTTGGGTGGAGAGGCTATTC-3′ and 5′-TCATTTACACTCCCTGGTTGTCCGGAAATG-3′) that resulted in an 800 bp product. Four out of 138 Klrg1+/− mice were identified during the B6 backcrossing that carried a recombination event between the targeted KLRG1 allele and the NKC of B6 mice. These mice were intercrossed to generate homozygous KLRG1 KO mice.

Southern and Northern blot analysis

KpnI-digested DNA was electrophoresed in 0.8% agarose gels, transferred to a nylon membrane (Gene Screen, PerkinElmer Life Sciences, Boston, MA, USA) and hybridized with a 32P-labeled 5′-flanking probes as depicted in Fig. 1A. Total RNA was isolated from spleen cells of LCMV-infected mice (day 8 p.i.) with an RNA Isolation Kit (Fluka Chemie AG, Buchs, Switzerland) and 10 μg of total RNA per lane were run on 1.2% agarose gels containing formaldehyde. RNA was transferred to nylon membrane (Gene Screen) and hybridized with a 32P-labeled KLRG1-specific probe. The probe consisted of 164 bp from exons 1 and 2 of KLRG1 generated by PCR using the following primers: 5′-GCTGACAGCTCTATCT-3′ and 5′-AGGATCCGTTGATACATCAGTAG-3′.

Mice

C57BL/6 (B6) mice were obtained from the BioMed Zentrum of the University Hospital Freiburg or from Harlan Winkelmann (Borchen, Germany). P14 KLRG KO mice were generated by mating Thy1.1+ P14 TCR transgenic mice (B6; D2-Tg(TcrLCMV)318Sdz/JDvsJ) 39 with KLRG1 KO mice. KLRG1 transgenic mice (B6, CBA/J-Tg(Klrg1)1Dhr) 20 were obtained from Thomas Hanke (University of Würzburg, Germany). Mice were bred at the BioMed Zentrum and were kept under specific pathogen-free conditions. Female or male mice were used at 8–20 wk of age and all animal experimental protocols used in this study were approved by the Regierungspräsidium Freiburg.

Viruses

Stocks of lymphocytic choriomeningitis virus (LCMV) WE strain, VSV Indiana (VSVIND), VV strain WR were produced by infecting L929 fibroblast cells, BHK and BSC 40 cells, respectively, with a low multiplicity of infection. Stocks of MCMV, Smith strain and mutant MCMV lacking m157 (△m157) 34 were produced in cell culture using B6 mouse embryo fibroblasts or by serial passage of salivary gland homogenates in BALB/c mice in vivo. Tissue culture-derived MCMV was used for inducing T-cell responses and salivary gland virus (SGV) for NK-cell studies. Mice were infected i.v. with 200 PFU LCMV-WE, 2×106 PFU VSVIND, 2×106 PFU VV, 2×106 PFU tissue culture-derived MCMV (i.p.), 5×105 PFU tissue culture-derived Δm157 MCMV (i.v.) or 5×104 PFU SGV MCMV (i.p.).

Flow cytometry

Cells (105–106 in 50–100 μL) were stained with appropriately diluted mAb (0.1–1 μg in 50–100 μL) in PBS containing 2% FBS and 0.1% NaN3 at 4°C for 30 min. The following fluorescence-labeled mAb were purchased from BD Pharmingen and eBioscience (NatuTec GmbH, Frankfurt, Germany): anti-CD3, -CD5, -CD8, -CD11b, -CD27, -CD62L, -CD127, -NK1.1. Anti-KLRG1 mAb (clone 2F1) 20 was produced in cell culture, purified using protein G and labeled with Alexa488 or Alexa647 (Molecular probes, Invitrogen, Karlsruhe, Germany). LCMV- and VSV-specific CD8+ T cells were detected using PE-labeled H-2Db tetramers complexed with GP33 peptide (KAVYNFATM) and H-2Kb tetramers complexed with NP52 peptide (RGYVYQGL) generated in the laboratory as described 12. Samples were analyzed by a BD FACSCalibur flow cytometer (BD Biosciences) using CellQuest-Pro software (BD Biosciences).

Analysis of T-cell reactivity by intracellular cytokine staining

Spleen cells (105 in 200 μL) were stimulated for 5 h in 10 μg/mL brefeldin A with 10−6 M of the following peptides: LCMV GP33–41 (KAVYNFATM), MCMV M45985–993 (HGIRNASFI), MCMV M38316–323 (SSPPMFRV), MCMV m139419–426 (TVYGFCLL). Intracellular cytokine staining was performed with PE-labeled mAb specific for IFN-γ (XMG1.2, eBioscience) and IL-2 (JES6-5H4, eBioscience) using Cytofix/Cytoperm solution (BD PharMingen). Peptides were purchased from Neosystem (Straßburg, France).

Adoptive T-cell transfers

P14 chimeric mice were generated by adoptive transfer (i.v.) of 105 P14 T cells from P14 KLRG1 KO or P14 WT mice. Repetitive P14 T cell transfers to generate 1°, 2° and 3° memory P14 cells were performed as described 11. Memory P14 T cells used for repetitive adoptive transfers were purified using PE-labeled anti-Thy1.1 mAb and anti-PE MACS-MicroBeads (Milteny, Bergisch Gladbach).

Analysis of NK-cell responses induced by infections

NK cells were activated in vivo by i.v. injection of VSVIND (2×106 PFU), VV (2×106 PFU), L. monocytogenes (106 CFU), LCMV (200 PFU) or 5×104 PFU MCMV (SVG) i.p. After 20 h, spleen cells were analyzed by staining with CD3-, CD11b-, CD27- and NK1.1-specific mAb. The activity of poly(I:C)-activated NK cells (200 μg i.p., 18 h) was determined by intracellular IFN-γ staining using plate-bound stimulation with anti-NK1.1 mAb (10 μg/mL) in the presence of 10 μg/mL brefeldin A or by classical 4 h 51Cr release assays using RMA-S target cells. NK cells were enriched from the spleen by MACS using negative selection (NK cell Isolation Kit, Milteny). NKG2D-triggered responses were determined by intracellular IFN-γ staining of NK cells from LCMV- or VSV-infected mice (day 3 p.i.) using stimulation with RMA-S-H60 cells as described 40. To assess the role of NK cells in MCMV infection, SGV was given i.p. (5×104 PFU) and MCMV titers of homogenized organs were determined on B6 mouse embryo fibroblasts.

NK-cell reactivity to E-cadherin expressing K562 cells

NK cells were enriched from the spleen by MACS using negative selection (NK cell Isolation Kit, Milteny) and cultured in the presence of 5 ng/mL of IL-12 (Preprotech, Hamburg) for 18 h. K562 cells expressing mouse E-cadherin were generated by retroviral transduction as described 24. For stimulation, 105 NK1.1+ cells were co-cultured with 105 mock- or E-cadherin-transduced K562 cells in 96-well round-bottom plates in the presence of 10 μg/mL brefeldin A for 5 h. Afterwards, cells were surface-stained with CD3-, NK1.1- and KLRG1-specific mAb, fixed, permeabilized using Cytofix/Cytoperm solution (BD PharMingen) and stained intracellularly with anti-IFN-γ mAb.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information

The authors thank Nicole Klemm for ES cell work and blastocyst microinjection, Smiljka Vucikuja for technical assistance, Peter Aichele and Andreas Diefenbach for critical comments on the manuscript, Matthias J. Reddehase, Ulrich H. Koszinowski and Lars Doelken for providing initial MCMV stocks, Norma Bethke, Rainer Bronner, Christian Herr, Uwe Griessbaum and Sonja Wagenknecht for animal husbandry, and Juergen Brandel for help with image processing and artwork. This work was supported by the Deutsche Forschungsgemeinschaft DFG (SFB 620, B2 to H. P.).

Conflict of interest: The authors declare no financial or commercial conflict of interest.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information
  • 1
    Butcher, S., Arney, K. L. and Cook, G. P., MAFA-L, an ITIM-containing receptor encoded by the human NK cell gene complex and expressed by basophils and NK cells. Eur. J. Immunol. 1998. 28: 37553762.
  • 2
    Voehringer, D., Kaufmann, M. and Pircher, H., Genomic structure, alternative splicing, and physical mapping of the killer cell lectin-like receptor G1 gene (KLRG1), the mouse homologue of MAFA. Immunogenetics 2001. 52: 206211.
  • 3
    Guthmann, M. D., Tal, M. and Pecht, I., A secretion inhibitory signal transduction molecule on mast cells is another C-type lectin. Proc. Natl. Acad. Sci. USA 1995. 92: 93979401.
  • 4
    Ortega Soto, E. and Pecht, I., A monoclonal antibody that inhibits secretion from rat basophilic leukemia cells and binds to a novel membrane component. J. Immunol. 1988. 141: 43244332.
  • 5
    Blaser, C., Kaufmann, M. and Pircher, H., Virus-activated CD8 T cells and lymphokine-activated NK cells express the mast cell function-associated antigen, an inhibitory C-type lectin. J. Immunol. 1998. 161: 64516454.
  • 6
    Hanke, T., Corral, L., Vance, R. E. and Raulet, D. H., 2F1 antigen, the mouse homolog of the rat “mast cell function- associated antigen”, is a lectin-like type II transmembrane receptor expressed by natural killer cells. Eur. J. Immunol. 1998. 28: 44094417.
  • 7
    Voehringer, D., Koschella, M. and Pircher, H., Lack of proliferative capacity of human effector and memory T cells expressing killer cell lectinlike receptor G1 (KLRG1). Blood 2002. 100: 36983702.
  • 8
    Ibegbu, C. C., Xu, Y. X., Harris, W., Maggio, D., Miller, J. D. and Kourtis, A. P., Expression of killer cell lectin-like receptor G1 on antigen-specific human CD8+ T lymphocytes during active, latent, and resolved infection and its relation with CD57. J. Immunol. 2005. 174: 60886094.
  • 9
    Robbins, S. H., Terrizzi, S. C., Sydora, B. C., Mikayama, T. and Brossay, L., Differential regulation of killer cell lectin-like receptor G1 expression on T cells. J. Immunol. 2003. 170: 58765885.
  • 10
    Thimme, R., Appay, V., Koschella, M., Panther, E., Roth, E., Hislop, A. D., Rickinson, A. et al., Increased expression of the NK cell receptor KLRG1 by virus-specific CD8 T cells during persistent antigen stimulation. J. Virol. 2005. 79: 1211212116.
  • 11
    Voehringer, D., Blaser, C., Brawand, P., Raulet, D. H., Hanke, T. and Pircher, H., Viral infections induce abundant numbers of senescent CD8 T cells. J. Immunol. 2001. 167: 48384843.
  • 12
    Altman, J. D. and Davis, M. M., MHC-peptide tetramers to visualize antigen-specific T cells. Curr. Protoc. Immunol. 2003. Chapter 17, Unit 17.13.
  • 13
    Heffner, M. and Fearon, D. T., Loss of T cell receptor-induced Bmi-1 in the KLRG1(+) senescent CD8(+) T lymphocyte. Proc. Natl. Acad. Sci. USA 2007. 104: 1341413419.
  • 14
    Henson, S. M., Franzese, O., Macaulay, R., Libri, V., Azevedo, R. I., Kiani-Alikhan, S., Plunkett, F. J. et al., KLRG1 signaling induces defective Akt (Ser473) phosphorylation and proliferative dysfunction of highly differentiated CD8+T cells. Blood 2009. 113: 66196628.
  • 15
    Joshi, N. S., Cui, W., Chandele, A., Lee, H. K., Urso, D. R., Hagman, J., Gapin, L. and Kaech, S. M., Inflammation directs memory precursor and short-lived effector CD8(+) T cell fates via the graded expression of T-bet transcription factor. Immunity 2007. 27: 281295.
  • 16
    Sarkar, S., Kalia, V., Haining, W. N., Konieczny, B. T., Subramaniam, S. and Ahmed, R., Functional and genomic profiling of effector CD8 T cell subsets with distinct memory fates. J. Exp. Med. 2008. 205: 625640.
  • 17
    Hayakawa, Y. and Smyth, M. J., CD27 dissects mature NK cells into two subsets with distinct responsiveness and migratory capacity. J. Immunol. 2006. 176: 15171524.
  • 18
    Huntington, N. D., Tabarias, H., Fairfax, K., Brady, J., Hayakawa, Y., Degli-Esposti, M. A., Smyth, M. J. et al., NK cell maturation and peripheral homeostasis is associated with KLRG1 up-regulation. J. Immunol. 2007. 178: 47644770.
  • 19
    Robbins, S. H., Tessmer, M. S., Mikayama, T. and Brossay, L., Expansion and contraction of the NK cell compartment in response to murine cytomegalovirus infection. J. Immunol. 2004. 173: 259266.
  • 20
    Corral, L., Hanke, T., Vance, R. E., Cado, D. and Raulet, D. H., NK cell expression of the killer cell lectin-like receptor G1 (KLRG1), the mouse homolog of MAFA, is modulated by MHC class I molecules. Eur. J. Immunol. 2000. 30: 920930.
  • 21
    Robbins, S. H., Nguyen, K. B., Takahashi, N., Mikayama, T., Biron, C. A. and Brossay, L., Cutting edge: inhibitory functions of the killer cell lectin-like receptor G1 molecule during the activation of mouse NK cells. J. Immunol. 2002. 168: 25852589.
  • 22
    Grundemann, C., Bauer, M., Schweier, O., von Oppen, N., Lassing, U., Saudan, P., Becker, K. F. et al., Cutting edge: identification of E-cadherin as a ligand for the murine killer cell lectin-like receptor G1. J. Immunol. 2006. 176: 13111315.
  • 23
    Ito, M., Maruyama, T., Saito, N., Koganei, S., Yamamoto, K. and Matsumoto, N., Killer cell lectin-like receptor G1 binds three members of the classical cadherin family to inhibit NK cell cytotoxicity. J. Exp. Med. 2006. 203: 289295.
  • 24
    Schwartzkopff, S., Grundemann, C., Schweier, O., Rosshart, S., Karjalainen, K. E., Becker, K. F. and Pircher, H., Tumor-associated E-cadherin mutations affect binding to the killer cell lectin-like receptor G1 in humans. J. Immunol. 2007. 179: 10221029.
  • 25
    Tessmer, M. S., Fugere, C., Stevenaert, F., Naidenko, O. V., Jonathan Chong, H., Leclercq, G. and Brossay, L., KLRG1 binds cadherins and preferentially associates with SHIP-1. Int. Immunol. 2007. 19: 391400.
  • 26
    Rosshart, S., Hofmann, M., Schweier, O., Pfaff, A. K., Yoshimoto, K., Takeuchi, T., Molnar, E. et al., Interaction of KLRG1 with E-cadherin: new functional and structural insights. Eur. J. Immunol. 2008. 38: 33543364.
  • 27
    Colonna, M., Cytolytic responses: cadherins put out the fire. J. Exp. Med. 2006. 203: 261264.
  • 28
    Li, Y., Hofmann, M., Wang, Q., Teng, L., Chlewicki, L. K., Pircher, H. and Mariuzza, R. A., Structure of natural killer cell receptor KLRG1 bound to E-cadherin reveals basis for MHC-independent missing self recognition. Immunity 2009. 31: 3546.
  • 29
    Masopust, D., Ha,S. J., Vezys, V. and Ahmed,R., Stimulation history dictates memory CD8 T cell phenotype: implications for prime-boost vaccination. J. Immunol. 2006. 177: 831839.
  • 30
    Karrer, U., Sierro, S., Wagner, M., Oxenius, A., Hengel, H., Koszinowski, U. H., Phillips, R. E. and Klenerman, P., Memory inflation: continuous accumulation of antiviral CD8+T cells over time. J. Immunol. 2003. 170: 20222029.
  • 31
    Munks, M. W., Cho, K. S., Pinto, A. K., Sierro, S., Klenerman, P. and Hill, A. B., Four distinct patterns of memory CD8 T cell responses to chronic murine cytomegalovirus infection. J. Immunol. 2006. 177: 450458.
  • 32
    Arase, H., Mocarski, E. S., Campbell, A. E., Hill, A. B. and Lanier, L. L., Direct recognition of cytomegalovirus by activating and inhibitory NK cell receptors. Science 2002. 296: 13231326.
  • 33
    Smith, H. R., Heusel, J. W., Mehta, I. K., Kim, S., Dorner, B. G., Naidenko, O. V., Iizuka, K. et al., Recognition of a virus-encoded ligand by a natural killer cell activation receptor. Proc. Natl. Acad. Sci. USA 2002. 99: 88268831.
  • 34
    Bubic, I., Wagner, M., Krmpotic, A., Saulig, T., Kim, S., Yokoyama, W. M., Jonjic, S. and Koszinowski, U. H., Gain of virulence caused by loss of a gene in murine cytomegalovirus. J. Virol. 2004. 78: 75367544.
  • 35
    Cepek, K. L., Shaw, S. K., Parker, C. M., Russell, G. J., Morrow, J. S., Rimm, D. L. and Brenner, M. B., Adhesion between epithelial cells and T lymphocytes mediated by E-cadherin and the alpha E beta 7 integrin. Nature 1994. 372: 190193.
  • 36
    van Roy, F. and Berx, G., The cell-cell adhesion molecule E-cadherin. Cell. Mol. Life Sci. 2008. 65: 37563788.
  • 37
    Fernandez, N. C., Treiner, E., Vance, R. E., Jamieson, A. M., Lemieux, S. and Raulet, D. H., A subset of natural killer cells achieves self-tolerance without expressing inhibitory receptors specific for self-MHC molecules. Blood 2005. 105: 44164423.
  • 38
    Kemp, C. J., Kim, K. H. and Philipp, J., The murine gene Cdkn1b (p27(Kip1)) maps to distal chromosome 6 and is excluded as Pas1. Mamm. Genome 2000. 11: 402404.
  • 39
    Pircher, H., Burki, K., Lang, R., Hengartner, H. and Zinkernagel, R. M., Tolerance induction in double specific T-cell receptor transgenic mice varies with antigen. Nature 1989. 342: 559561.
  • 40
    Lucas, M., Schachterle, W., Oberle, K., Aichele, P. and Diefenbach, A., Dendritic cells prime natural killer cells by trans-presenting interleukin 15. Immunity 2007. 26: 503517.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  9. Supporting Information

See accompanying Commentary: http://dx.doi.org/10.1002/eji.201040506

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.