Lipocalin 2 Is Not Essential for Apoptosis.

Lipocalin 2 has been strongly implicated as a proapoptotic factor. For example, in FL5.12 cells, an IL-3-dependent pro-B cell line, IL-3 withdrawal induced lipocalin 2 expression and secretion leading to apoptosis induction (10). To further investigate the involvement of lipocalin 2 in apoptosis, we isolated IL-3-dependent primary bone marrow mast cells from the femurs of Lcn2^−/− and WT mice and cultured them in complete medium containing 3 ng/ml recombinant IL-3. Upon IL-3 withdrawal, both Lcn2^−/− and WT cells underwent apoptosis in a time-dependent manner at a comparable rate (Fig. 1A). After 96 h, 29% and 39% of cells remained viable in Lcn2^−/− and WT cultures, respectively (Fig. 1A). These data suggest that lipocalin 2 is not essential for apoptosis induced by IL-3 withdrawal.

Open in a separate window

Fig. 1.

Lipocalin 2-deficient cells respond normally to IL-3 withdrawal and a variety of apoptotic stimuli. (A) Cultured bone marrow cells from WT and Lcn2^−/− littermates were subjected to IL-3 withdrawal, and the percentage of viable cells remaining was measured by annexin–propidium iodide staining at the indicated time points. Results shown are the mean percentage of viable cells (±SD) of three independent experiments involving at least two mice for each genotype. (B) Thymocytes from WT and Lcn2^−/− littermate mice were treated with the indicated chemical and physical stimuli, and cell viability was assessed by annexin–propidium iodide staining after 24 h.

We next examined a range of chemical and physical agents to determine whether lipocalin 2 might be involved in programmed cell death only in response to certain apoptotic stimuli. Thymocytes treated with dexamethasone, etoposide, staurosporine, sorbitol, PMA, or ionomycin underwent normal apoptosis (Fig. 1B). Apoptosis induced by UV-irradiation, γ-irradiation, heat shock, or serum starvation was also indistinguishable between Lcn2^−/− and WT thymocytes (Fig. 1B). Thus, despite the reported involvement of lipocalin 2 in programmed cell death, loss of lipocalin 2 has no effect on the apoptosis of these cell types under these conditions.

Lipocalin 2 Has an Important Role in Innate Immunity in Vitro and in Vivo.

To test the function of lipocalin 2 in the innate immune response in vivo, we used a murine model of an acute lethal bacterial infection. Lcn2^−/− and WT littermates were i.p. challenged with a lethal dose of Escherichia coli [ATCC 25922; 1.5 × 10⁸ colony-forming units (cfu)/30 g], and the survival rates of the mice were compared (Fig. 2A). Lcn2^−/− mice showed a significant acceleration of lethality such that 74% (14 of 19) of the knockout mice had died after 16 h compared with 21% (5 of 24) of the WT controls. Only 21% of the Lcn2^−/− mice survived and recovered completely from the infection, whereas 54% of the WT mice recovered (P < 0.005; log-rank test).

Open in a separate window

Fig. 2.

Antibacterial effects of lipocalin 2 in vivo and in vitro. (A) Lipocalin 2 protects against bacterial infection in vivo. Shown are survival curves of Lcn2^−/− and WT mice that were i.p. injected with 1.5 × 10⁸ cfu/30 g of E. coli (ATCC 25922). Results shown are the percentage survival from three independent experiments of a total of 19 Lcn2^−/− mice and 24 WT mice (P > 0.005; log-rank test). (B) Lcn2^−/− neutrophils exhibit impaired bacteriostatic activity in vitro. Neutrophils isolated from Lcn2^−/− mice (black bars) and WT littermates (white bars) were incubated with E. coli at a multiplicity of infection of 1. Gray bars indicate cultures of bacteria and WT neutrophils to which 50 μM iron was added. Bacterial growth was determined at the indicated time points as described in Materials and Methods. Data shown are the mean cfu of bacteria (±SD) of triplicate measurements from three independent experiments. Significant differences in bacterial growth between WT and Lcn2^−/− cultures were found at 30 min (, P < 0.001) and at 60 and 120 min (, P < 0.0002).

Several reports have indicated that lipocalin 2 mediates a bacteriostatic effect by sequestering iron that has associated with bacterial siderophores (6, 15). To further analyze this mechanism, we isolated neutrophils from Lcn2^−/− and WT littermates and incubated them in cultures to which E. coli was added at a multiplicity of infection of 1. The in vitro bacteriostatic effects of Lcn2^−/− and WT neutrophils were then compared. As shown in Fig. 2B, the number of viable bacteria remaining in the cultures of Lcn2^−/− neutrophils was significantly higher than the number of viable bacteria remaining in WT cultures (P < 0.001 at 30 min, and P < 0.0002 at 60 min and 120 min). After 120 min of incubation, we measured a 2.5-fold increase in the number of bacteria in the Lcn2^−/− neutrophil cultures compared with the WT. This striking difference in bacterial growth was completely abolished by the addition of 50 μM iron to the cultures containing WT neutrophils and E. coli (Fig. 2B).

Lipopolysaccharides are characteristic components of the cell walls of Gram-negative bacteria and are the major causative agent of sepsis. We therefore examined whether the difference in survival after bacterial infection is due to an altered response to LPS. Although the reaction of Lcn2^−/− mice to septic shock induced by i.p. injection of 50 mg/kg LPS was delayed, there was no significant difference in overall mortality in Lcn2^−/− mice compared with WT littermates (data not shown). We also analyzed the levels of TNF-α, IL-6, IL-12, and IFN-γ in the sera of LPS-treated WT and Lcn2^−/− mice and could not detect any differences in the levels of any of these cytokines (data not shown).

Taken together, these data clearly indicate that lipocalin 2 has an essential role in the antibacterial innate immune response, and that this function can be abrogated by an increase in the local iron concentration. Our results confirm that the sequestration of iron is likely the primary function of lipocalin 2 during bacterial infections.

Generation of Lcn2^−/− Mice.

The targeting strategy used is outlined in Fig. 4A. A genomic DNA clone containing exons 1–6 of the murine Lcn2 gene was isolated from a 129/J mouse genomic library. A targeting vector was designed to replace a 2.5-kb genomic fragment containing Lcn2 exons 1–5 with the PGK-neo cassette. The diphtheria toxin A gene (DT-A) driven by the pMC1 promoter was incorporated into the 3′ end of the vector to allow for negative selection (25). The targeting vector was linearized with NotI and electroporated into 129/Ola ES cells (Bio-Rad Gene Pulser; 0.34 kV, 250 μF). The transfected cells were then cultured in the presence of 300 μg/ml G418 (Sigma-Aldrich) for 10 days. Homologous recombinants were identified by PCR and verified by Southern blot with a PCR-generated 5′ flanking probe (primers: 5′-ATAGCCCTGGCTGTCCTGAA-3′ and 5′-TAAGGTCCCCCTCTAAACC-3′). Correctly targeted clones were injected into C57BL/6J blastocysts, and the generation of chimeras was performed as described in ref. 26. Two independently targeted ES cell clones transmitted the Lcn2 mutation into the germ line. Male chimeric mice were mated to C57BL/6 female mice, and the heterozygous F₁ progeny were intercrossed to generate lipocalin 2-deficient mice. The WT littermates from these crosses were used as controls for all experiments.

All mice were maintained in the animal facility of the Ontario Cancer Institute in accordance with established ethical care regulations of the Canadian Council in Animal Care.

Analysis of Lcn2 Genotypes.

Genotypes of mutant mice and embryos were determined by PCR and confirmed by Southern blot analysis of genomic DNA from tail biopsy samples as described in ref. 27 with the external 5′ probe as depicted in Fig. 4A. For PCR, primers a and b (see Fig. 4A), which are specific for an Lcn2 intron sequence located 5′ and 3′ of the deleted exon 2, were used to detect the WT Lcn2 allele. Primer c, specific for an Lcn2 5′ intron sequence, and primer d, specific for Neo, were used to detect the mutant Lcn2 allele. Primer sequences were as follows: a, 5′-GGT TTG GTG GCA GGC TAT TA-3′; b, 5′-CAG AGT GGC TTT CCC CAT AA-3′; c, 5′-TGC TCC TGC CGA GAA AGT AT-3′; d, 5′-CTC TTC CTC CTC CAG CAC AC-3′.

Immune System Cell Populations.

Populations of thymocytes, splenocytes, lymph node cells, and bone marrow cells in mice were identified by staining with fluorochrome-labeled primary antibodies and quantitated by flow cytometry using a FACSCalibur FACS machine and cellquest software (Becton Dickinson).

Apoptosis in Primary Bone Marrow Cells and Thymocytes.

Bone marrow cells were prepared from femurs of WT or Lcn2^−/− mice and cultured for 14 days in OptiMEM medium containing 10% FCS and 3 ng/ml recombinant IL-3 (R & D Systems). The nonadherent cells were transferred into a new culture flask every 2 days for the first 6 days. To assay apoptosis, the cells were washed twice in PBS to remove any residual IL-3, resuspended at 0.5 × 10⁶ cells per ml in fresh IL-3-free medium, and cultured for 24–96 h in the presence or absence of exogenous IL-3 (3 ng/ml). At each time point, the cells were stained with Annexin V (BD Bioscience) plus propidium iodide and analyzed by flow cytometry.

Thymocytes were prepared from WT or Lcn2^−/− mice by standard procedures (28), plated at 1 × 10⁶ cells per ml in 24-well plates in RPMI medium 1640 containing 10% FCS, and treated as shown in Fig. 1B. Cell viability was determined after 24 h as for bone marrow cells. All chemicals were purchased from Sigma-Aldrich.

Isolation and Purification of Neutrophils.

Isolation of peritoneal exudate cells (PECs) after casein injection was carried out as described in ref. 29. Neutrophils were purified by discontinuous Percoll gradient centrifugation as described in ref. 30. Neutrophil purity was typically >93%, as judged by flow cytometric analysis of CD11b- and Gr-1-positive cells (data not shown).

Neutrophil Bacteriostatic Activity.

The bacteriostatic activity of isolated neutrophils was determined by an in vitro killing assay as described in ref. 31. Briefly, purified neutrophils (5 × 10⁵) were resuspended in 1 ml of warmed PBS containing 10% fresh serum obtained from syngeneic mice and 0.1% gelatin. These cells were mixed with log phase E. coli (ATCC 25922) in culture tubes coated with silicon (Sigma-Aldrich). The ratio of bacteria to neutrophils was adjusted to 1:1 (multiplicity of infection of 1). The culture tubes were incubated in a shaking water bath at 37°C for 2 h. Triplicate samples were removed for analysis at 0, 30, 60, and 120 min of incubation. The number of viable bacteria left in each sample was determined by plating 0.1 ml of an appropriate dilution of the sample on LB plates. The results were expressed as the mean cfu/ml (±SD) of triplicate cultures.

Bacterial Infection.

Log-phase E. coli (ATCC 25922) were washed, resuspended in 0.5 ml of PBS to the desired concentration, and i.p. injected into Lcn2^−/− and WT littermate mice. Each mouse received 1.5 × 10⁸ cfu E. coli per 30 g of body weight. Injected mice were observed twice daily and were not left alone for >3–4 h during survival experiments. The log-rank test was used for statistical analysis of the survival curve.

Embryonic Kidney Histology.

Embryos and tissues were dissected on ice, fixed with 4% paraformaldehyde (PFA), paraffin-embedded, and sectioned according to standard procedures. Hematoxylin/eosin staining was performed as described in ref. 32.

Mouse Model of Renal Ischemia-Reperfusion Injury.

Male lipocalin 2-deficient mice and WT control littermates (25–30 g) were anesthetized with i.p. pentobarbital (50 mg/kg) and placed on a warming table to maintain 37°C core body temperature. The kidneys were exposed through an abdominal section, and the left kidney was removed after ligation of its vascular pedicle and ureter. The vascular pedicle of the right kidney was clamped with a nontraumatic vascular clamp (Roboz Surgical Instrument, Rockville, MD) for 30 min, during which time the kidney was kept warm and moist. The clamp was then removed, the kidney was observed for return of blood flow, and the incision was sutured with 5-0 nylon. After 24 h of reperfusion, heparinized blood was obtained for measurement of serum creatinine and urea (Vita-Tech, Markham, ON, Canada). The kidney was harvested, and sagittal sections were fixed in 4% formalin. Paraffin-embedded sections (4 μm) were stained with hematoxylin/eosin or by in situ end-labeling of fragmented DNA for detection of apoptotic nuclei and examined histologically.

Histopathology Scoring.

Hematoxylin/eosin-stained kidney sections were scored for histopathologic damage to the tubules in a blinded manner. The area of proximal convoluted tubular necrosis was evaluated by the Jablonski scale (33) ranging from 0 to 4, with no necrosis (0), isolated necrotic cells (1), focal necrosis in inner cortex (2), diffuse necrosis in inner cortex (3), and necrosis involving the whole cortex (4).

We thank M. Nilsen-Hamilton (Iowa State University, Ames) for providing the Lcn 2 antibody; J. Dembowy for technical assistance; and M. Saunders for scientific editing. T.B. is the recipient of a postdoctoral fellowship from the Cancer Research Institute, New York. C.C.C. is supported by Canadian Institutes of Health Research.