Histological analysis of CD11c-DTR/GFP mice after in vivo depletion of dendritic cells

Abstract

To investigate the dependence of individual immunological processes on DC, a transgenic mouse system (CD11c-DTR/GFP mice) has been developed that allows conditional depletion of CD11c⁺ DC in vivo through administration of diphtheria toxin. We have performed careful histological analysis of CD11c-DTR/GFP mice at different time points after diphtheria toxin injection and confirmed the transient depletion of CD11c⁺ cells from lymph nodes and spleen. Unexpectedly, the injection of diphtheria toxin completely depleted marginal zone and metallophilic MΦ from the spleen and their sinusoidal counterparts from the lymph nodes. This finding limits the use of CD11c-DTR/GFP mice for the analysis of the role of DC to models and read outs that are proven to be independent of marginal zone and sinusoidal MΦ.

Introduction

Based on their ability to express MHC class II and costimulatory molecules and to take up antigens, B cells, MΦ and DC are thought to have the capacity to stimulate naïve T cells and are thus referred to as professional APC. During the past 20 years however, a large body of evidence accumulated suggesting that DC are the only cells fulfilling all criteria of a professional APC: DC were found to be unchallenged in their potency to prime allogeneic T cell responses both in vivo and in vitro[1–3]. Recently, a novel diphtheria toxin (DT) based system was developed, which allows the short-term inducible ablation of DC in vivo: A single administration of DT to transgenic mice that express the primate DT receptor (DTR) under control of the DC-specific CD11c promoter was found to result in depletion of CD11c⁺ DC [4]. Depletion of the DC compartment resulted in a complete inability to cross-prime CD8⁺ T cells responses against Listeria monocytogenes and Plasmodium yoelii[4] and to prime LCMV-specific CD8⁺ T cell responses [5].

Unexpected results that we obtained when analysing antiviral B cell responses in DT-treated CD11c-DTR/GFP mice, however, prompted us to use immunohistology to carefully analyse the consequences of DT administration to CD11c-DTR/GFP mice to the structure of lymphoid organs and to the individual cell types present in those lymphoid organs. We found that a single injection of DT resulted in efficient and transient depletion of DC but, unexpectedly, also in complete and protracted depletion of marginal zone and metallophilic MΦ.

Materials and methods

Mice

CD11c-DTR/GFP mice [4], H-2D^b−/− and control C57BL/6 mice were obtained from the Biologisches Zentrallabor (University of Zurich) and were kept in specific pathogen free facilities. All mice were backcrossed for at least 10 generations to C57BL/6 mice. Age- and sex-matched mice that were at least 6 weeks of age were used for all experiments. Animal experiments were performed according to institutional guidelines and to the Swiss federal and cantonal laws on animal protection.

For systemic DC depletion, CD11c-DTR/GFP transgenic mice were injected intraperitoneally with 4 ng/g body weight DT (in PBS; Sigma Chemical Co., St. Louis, MO, USA).

Generation of bone marrow chimeras

To generate mixed bone marrow chimeras, H-2D^b−/− mice were lethally irradiated (10 Gy) with a ⁶⁰Cobalt source. Bone marrow was isolated from femurs and tibias of age and sex matched donor mice and a mixture of 2·5 × 10⁶ CD11c-DTR/GFP + 2·5 × 10⁶ H-2D^b−/− bone marrow cells were injected i.v. In these chimeras, the only H-2D^b-positive CD11c⁺ cells will also express the transgenic DTR, and will therefore be susceptible to DT-mediated depletion. Mice received 1 mg/ml sulfadoxin and 0·2 mg/ml trimethoprim in the drinking water during the first 2 weeks, and were rested for 7 weeks before use.

Preparation of and treatment with clodronate liposomes

Liposomes were prepared as described earlier [6]. In brief, soy phosphatidylcholine, DL-a-tocopherol, and cholesterol (all from Sigma-Aldrich, St. Louis, MO, USA), in 1 : 0·01 : 0·3 molar ratio, were dissolved in chloroform. After evaporation of the solvent, dry lipid film was dispersed in 10 ml of clodronate solution (Ostac, Roche, Switzerland) by careful shaking. Suspension was sonicated three times for 5 min and freeze-thawed in three cycles of liquid nitrogen and water at 37°C. Liposomes were washed twice in PBS (50 000 × g) to remove free clodronate. For LCL, final solution was adjusted to an OD600 of 1·2, and 100 µl corresponding to 10 mg were injected i.v. Mice were either treated with empty liposomes or left untreated as controls. To confirm depletion of marginal zone MΦ[6], spleens were isolated 3 days after injection of clodronate- or empty liposomes and 5 µm cryosections were stained with rat anti-marginal zone MΦ (ERTR-9 [7] and with rat anti-marginal metallophilic MΦ (MOMA-1; Biomedicals, Augst, Switzerland [8] (data not shown). Using the same method, we found that both metallophilic and marginal zone MΦ were present at normal numbers and localization 3 weeks after injection of clodronate-liposomes (data not shown).

Immunohistology

At indicated time points after DT injection and LCMV infection, organs were removed and snap frozen in liquid nitrogen. Five µm cryosections were fixed in acetone for 10 min and subsequently incubated with rabbit anti-GFP (RDI Research Diagnostics), rat anti-interdigitating DC DEC-205 (anti-CD205; NLDC-145, Biomedicals), rat anti-marginal zone MΦ (ERTR-9 [7]), rat anti-marginal metallophilic MΦ (MOMA-1; Biomedicals, Augst, Switzerland [8]), rat anti-red pulp MΦ (F4/80; HB-198, American Type Culture Collection), rat anti-CD4 (YTS 191 [9]), rat anti-CD8 (YTS169 [9]), or rat anti-CD19 (1D3). CD11c on DC was stained with the hamster mAb N418 (HB-224, American Type Culture Collection). Goat anti-rat Ig (Caltag Laboratories, Burlingame, CA, USA) in 5% normal mouse serum was used as a secondary reagent and alkaline phosphatase-labelled donkey anti-goat Ig (Jackson ImmunoResearch Europe Ltd, Soham, UK) in 5% normal mouse serum as tertiary reagent. Alternatively, alkaline phosphatase-labelled rabbit anti-hamster Ig or goat anti-rabbit IgG was used as secondary reagent. The substrate for the red colour reaction was naphtol As-Bi phosphate/New Fuchsine. Endogenous alkaline phosphatase activity was quenched by levamisole. Sections were counterstained with haemalum.

Immunofluorescence

Five µm cryosections of spleens from DT-treated or control H-2D^b−/− + CD11c-DTR/GFP → H-2D^b−/− chimeras were fixed for 10 min in acetone and subsequently incubated with PBS containing 10% FCS and 10 µg/ml rat anti-FcγRI/II (2·4G2). Sections were then incubated with mouse anti-D^b-FITC (B22-249 [10]) plus hamster anti-CD11c-biotin (clone HL-3, PharMingen, BD Biosciences, San Jose, CA, USA) or rat anti-MOMA-1-biotin (Biomedicals, Geneva, Switzerland). After washing in PBS, sections were incubated with streptavidin-TRITC (Southern Biotechnology Associates, Birmingham, AL, USA). All incubations were carried out at room temperature for 1 h. Finally, sections were washed in PBS and were counterstained with DAPI. Sections were mounted by using DAKO fluorescent mounting medium (DakoCytomation, Glostrup, Denmark) containing antibleach and were examined using a fluorescence microscope.

Results and discussion

Diphtheria toxin transiently depletes CD11c⁺ DC in CD11c-DTR/GFP mice

DC from transgenic mice that express the primate DTR (CD11c-DTR/GFP) as a CD11c promoter-driven transgene (CD11c-DTR/GFP mice) are susceptible to death induced by DT [4,11]. The toxicity of DT is strictly dependent on receptor-mediated endocytosis, thus enabling conditional depletion in vivo of those cell types that express the primate DTR [4,12]. Using FACS analysis, it was previously shown that a single injection of DT efficiently depletes CD11c⁺ cells from the spleen CD11c-DTR/GFP mice [4]. We extend this finding here to histological analysis. We injected CD11c-DTR/GFP, transgene-negative littermates or C57BL/6 mice i.p. with 4 ng/g body weight of DT and removed spleen, inguinal lymph nodes and mesenteric lymph nodes for immunohistology at different time points thereafter. Eighteen hours after a single DT injection, virtually all DC were depleted from the spleen as visualized by the absence of CD11c⁺ and of transgene expressing GFP⁺ cells (Fig. 1). CD11c⁺ and GFP⁺ cells were detectable again at 42 h after DT injection, and reached comparable numbers to not-depleted mice after 66 h (Fig. 1). At this time point, however, the inderdigitating DC (DEC205⁺) only started to be detectable again. Histological analysis at later time points after depletion revealed that the DEC205⁺ cells reached normal numbers and localization after 5d (data not shown). Neither injection of PBS in CD11c-DTR/GFP mice nor injection of DT in transgene-negative littermates or B6 mice depleted CD11c⁺ cells (data not shown), thus confirming and refining previously published findings [4]. Depletion of GFP⁺ cells and DC in inguinal and mesenteric lymph nodes was comparable to that in spleen (data not shown).

Fig. 1.

Injection of Diphtheria Toxin depletes dendritic cells and marginal zone MΦ from CD11c-DTR mice. Spleen sections of CD11c-DTR mice were stained with Abs of the indicated specificity at indicated time points after i.p. injection of 100 ng DT. Ab staining resulted in a red precipitate. GFP, indicating transgene (DTR)-expressing cells; CD11c, indicating DC; MOMA-1, indicating metallophilic macrophages; ERTR-9, indicating marginal zone macrophages. Original magnification: ×100 for all panels.

Diphtheria toxin depletes marginal zone MΦ in CD11c-DTR/GFP mice

In experiments designed to explore the role of DC in antiviral B cell responses we found an unexpectedly high susceptibility of DT-treated CD11c-DTR/GFP mice to VSV infection, revealing itself as 100% lethality at 6–8 days after infection with a usually nonlethal dose of 2 × 10⁶ pfu VSV-Indiana (IND). This prompted us to carefully analyse whether the architecture of secondary lymphoid organs or cells other than DC were affected by DT treatment. To our surprise, we found that, besides the expected depletion of DC, also marginal zone MΦ (MZM, ERTR-9⁺[7]) and metallophilic MΦ (MM, MOMA-1⁺[8]) were completely depleted in spleen (Fig. 1) as were their sinusoidal counterparts in lymph nodes (data not shown) for at least 66 h after a single DT injection. This is in contrast with previously published finding where no effects on splenic F4/80⁺ MΦ were seen [4]; this discrepancy may be explained by the relative insensitivity of FACS analysis compared to immunohistology, as MM and MZM make up a small contribution to the total number of splenic MΦ. The number and localization of CD4⁺, CD8⁺ and CD19⁺ cells was not affected by administration of DT in CD11c-DTR/GFP mice in lymph nodes or spleen (data not shown). Immunohistological analysis of macrophage subsets at later time points after a single DT injection showed few MOMA-1⁺ cells after 5 days and considerable, although still diminished numbers after 10 d. In addition, ERTR-9⁺ cells stayed depleted for at least 10 days (Fig. 2). We found transiently reduced staining of red pulp MΦ (F4/80⁺) in the spleen (Fig. 2), but comparison of F4/80 and CD11b staining showed no effect of DT treatment on the CD11b⁺ population, suggesting that the transient loss of F4/80 staining might be due to marker down-regulation rather than to depletion of F4/80⁺ MΦ (Fig. 3).

Fig. 2.

Injection of Diphtheria Toxin results in a long lasting depletion of marginal zone macrophages in CD11c-DTR/GFP mice. Spleen sections of CD11c-DTR mice were stained with Abs of the indicated specificity at indicated time points after i.p. injection of 100 ng DT. Ab staing resulted in a red precipitate. MOMA-1, indicating metallophilic macrophages; ERTR-9, indicating marginal zone macrophages; F4/80, indicating red pulp macrophages. Original magnification: ×100 for all panels.

Fig. 3.

Injection of Diphtheria Toxin does not affect red pulp macrophages in CD11c-DTR/GFP mice. Spleen sections of CD11c-DTR mice were stained with Abs of the indicated specificity at indicated time points after i.p. injection of 100 ng DT. Ab staining resulted in a red precipitate. F4/80, indicating red pulp macrophages; CD11b, indicating pan-macrophages. Original magnification: ×100 for all panels.

Mechanism of DT-dependent depletion of marginal zone and metallophilic MΦ from CD11c-DTR/GFP mice

The unexpected depletion of MΦ subsets from spleen and lymph nodes may result from:

• phagocytosis by MΦ of dying DC that had taken up DT;

• an indirect structural defect due to DC absence;

• ectopic expression of the transgene in MΦ.

To address the first possibility, we generated mixed bone marrow chimeras in which we reconstituted lethally irradiated transgene-negative mice with a 1 : 1 mixture of CD11c-DTR/GFP transgenic and transgene-negative bone marrow. The spleens of these mice were analysed immunohistologically 18 h after DT injection. Approximately half of the bone marrow-derived cells in the spleen will express the DTR and will thus die upon DT administration. We didn't see any effect of DT-treatment on ERTR-9⁺ or on MOMA-1⁺ MΦ in the spleen of mixed chimeras, suggesting that phagocytosis of dying DC including DT is not the reason for the unexpected MΦ depletion in CD11c-DTR/GFP mice [5]. To distinguish between the remaining two possibilities, we generated H-2D^b−/− + CD11c-DTR/GFP → H-2D^b−/− mixed bone marrow chimeras. All H-2D^b-positive cells in these chimeras are exclusively derived from CD11c-DTR/GFP bone marrow. If depletion of mariginal zone MΦ in CD11c-DTR/GFP mice was due to an indirect effect of DC depletion, we would expect similar depletion of D^b-positive and D^b-negative marginal zone MΦ. However, if depletion of marginal zone MΦ was due to ectopic expression of the DTR transgene, we would expect that only CD11c-DTR/GFP bone marrow-derived, that is H-2D^b-positive, MΦ were affected in the marginal zone. We performed immunofluorescence using anti-H-2D^b-FITC plus anti-CD11c-TRITC or anti-MOMA-1-TRITC: H-2D^b-positive, CD11c-DTR/GFP-derived CD11c⁺ or MOMA-1⁺ cells will appear orange, whereas H-2K^b-positive and therefore H-2D^b−/− derived CD11c⁺ or MOMA-1⁺ cells will appear red. In a first set of experiments we observed that most MOMA-1⁺ cells in mixed bone-marrow chimeras were H-2D^b−/− and therefore host-derived (data not shown), which is probably due to the relatively high radioresistance and low turn-over rate of these cells, and which makes the abovementioned analysis very difficult. To circumvent this problem, we have depleted marginal zone MΦ from H-2D^b−/− + CD11c-DTR/GFP → H-2D^b−/− mixed bone marrow chimeras 8 weeks after reconstitution using clodronate-containing liposomes [6]. We confirmed depletion of marginal zone MΦ by histology (data not shown) and waited another 6 weeks before injection DT, which was found to be amply sufficient for replenishment of the marginal zone by bone marrow-derived MΦ (data not shown). In this situation, marginal zone MΦ were equally derived from CD11c-DTR/GFP (D^b-positive) and from H-2D^b−/− (D^b-negative) bone marrow. We found that DT injection depleted CD11c-DTR/GFP-derived, but not H-2D^b−/− derived CD11c⁺ as well as MOMA-1⁺ cells (Fig. 4), suggesting that depletion of marginal zone MΦ is due to ectopic expression of the transgene by marginal zone and metallophilic MΦ.

Fig. 4.

Injection of Diphtheria Toxin depletes transgene-expressing DC and marginal zone macrophages from H-2D^b−/−+ CD11c-DTR/GFP → H-2D^b−/− mixed bone marrow chimeras. Bone marrow chimeras were injected i.v. with 10 mg clodronate liposomes to deplete marginal zone macrophages 8 weeks after transplantation. Six weeks thereafter, mice were injected with 100 ng diphtheria toxin or were untreated, and spleens were isolated 18 h later. Cryosections were stained with anti-H-2D^b FITC plus anti-MOMA-1 biotin/streptavidin TRITC or anti-CD11c biotin/streptavidin TRITC and were analysed for immunofluorescence. Double positive (FITC/TRITC) cells appear orange, whereas H-2D^b negative cells appear red. Original magnification was ×200.

Consequences for the use of CD11c-DTR/GFP transgenic mice

We have shown that the administration of DT to CD11c-DTR/GFP transgenic mice results in the desired depletion of CD11c⁺ DC, but also in complete depletion of marginal zone MΦ in the spleen and of sinusoidal MΦ in the lymph nodes. Therefore, conclusions on the role of DC in immunological processes from experiments using CD11c-DTR/GFP mice have to be interpreted with great care for those experimental systems that involve marginal zone MΦ. These include responses where marginal zone MΦ are important in antigen sampling and subsequent induction of B cell responses [13] as has been described for Ficoll-coupled antigens [14], VSV [15,16] lymphocytic choriomeningitis virus [16] and Listeria monocytogenes[6]. To illustrate the caveats associated with CD11c-DTR/GFP mice, we infected CD11c-DTR/GFP mice with VSV-IND 18 h after DT treatment (DC and marginal zone MΦ depleted, Figs 1 and 2), 72(h) after DT treatment (DC returned, marginal zone still MΦ depleted, Figs 1 and 2) or clodronate-containing liposomes (marginal zone depleted of MΦ, all DC intact [6]) or without DT injection and measured survival of the animals. All treated animals died after 6–8 days after infection, whereas the control group survived. There was no difference between the two DT-treated groups and the group treated with clodronate-containing liposomes, suggesting that the absence of marginal zone MΦ rather than the absence of DC is responsible for the inability to cope with VSV-infection as illustrated by early death [15–17].