Published online before print April 23, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
/
T cells
Department of Microbiology and Immunology, Emory University School of Medicine, Atlanta, Georgia
1Correspondence: Department of Microbiology and Immunology, Emory University School of Medicine, 1510 Clifton Road, Atlanta, GA 30322. E-mail: ziegler{at}microbio.emory.edu
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
/
T cells play a broad, immunoregulatory role has been accumulating steadily. We show here that myeloid cells are disregulated after peritoneal infection with Listeria monocytogenes in mice lacking / T cells. Inflammatory populations of neutrophils and monocytes recruited to the site of infection remained longer. Intracellular cytokine analysis showed that frequencies of myeloid cells producing interleukin-12 and tumor necrosis factor 
were higher and remained elevated longer after infection in mice genetically deficient in / T cells. In vivo dye-tracking studies indicated that the majority of inflammatory monocytes differentiated into resident tissue macrophages in situ. In vitro experiments confirmed that monocytes harvested from mice lacking / T cells were defective in their maturation process. This evidence suggests that / T cells promote differentiation in the monocyte/macrophage lineage. These cells are important for bactericidal activity, inflammatory cytokine production, clearance of inflammatory neutrophils, and ultimately, antigen presentation to T cells. Regulation of monocyte/macrophage differentiation may underlie a broad segment of the phenotypic alterations that have been reported in mice lacking / T cells.
Key Words: macrophages • monocytes • neutrophils • TNF • IL-12
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
/ T cells have been shown to participate in the host response to a variety of infectious agents (reviewed in ref. [1
]). Although they tend to be important for survival only after infection with large doses of pathogens, the immune response to moderate levels of infection is altered quantitatively and qualitatively in their absence. The complexity of the role of / T cells in the immune response is illustrated by the paradox that they have the ability to produce interferon- (IFN-) [2
], yet systemic levels of proinflammatory cytokines are elevated in response to infection with Listeria monocytogenes when / T cells are absent [3
]. Systemic IFN-, interleukin (IL)-12, and IL-6 reached higher levels in / T cell receptor (TCR)–/– mice and remained elevated longer after infection. Similarly, hepatic tissue inflammation was exaggerated when mice deficient in / T cells were infected with Listeria [4
]. / T cells are more prevalent in tissue sites than in the circulation or lymphoid organs. Their function my vary depending on the tissue in which they reside, the activating antigen or pathogen, or their TCR expression. Some subsets may promote the inflammatory response, and others may be anti-inflammatory [5
6
7
].
A hallmark of the early portion of the host response to infection with Listeria is the trafficking of myeloid cells to the site of infection. The peritoneal cavity provides an anatomically defined, model site to monitor changes in neutrophil and macrophage populations following infection. Myeloid cell populations fluctuate dramatically after infection and function in a variety of roles in the host immune response. They not only eliminate pathogenic bacteria directly via phagocytosis but can also contribute to the general expansion of the innate and adaptive responses by their secretion of a variety of cytokines. After up-regulation of class II expression in response to infection, macrophages can also serve as antigen-presenting cells (APCs) to CD4+ and CD8+ /ß T cells. These dynamic changes in myeloid populations occur during the same time-frame that cytokine levels fluctuate following infection.
We reported previously that peritoneal myeloid populations in mice genetically deficient in / T cells (/ TCR–/– mice) differed from those in C57BL/6 (B6) control mice before and 2 days after Listeria infection [3
]. As cytokine production and myeloid population changes were altered in the absence of / T cells, we examined the functional relationships within this system by following myeloid populations and their production of cytokines throughout the course of a developing immune response to Listeria. We used GR1 (LY6-G) as a marker for neutrophils [8
, 9
] and F4/80 as a macrophage marker [10
]. It is interesting that a population of cells expressing both of these markers predominated during the initial 2–4 days of infection. This was followed by a gradual return to a peritoneal myeloid population composed primarily of cells expressing F4/80 but not GR1. Injection of tracking dye provided a means to monitor myeloid populations to determine whether cells were trafficking through the site or were altering their expression of F4/80 and GR1 during the course of the infection. Production of proinflammatory cytokines by myeloid population subsets was monitored by intracellular flow cytometry, and secretion was monitored by enzyme-linked immunosorbent assay (ELISA). The differences that we found in population dynamics and cytokine production in the presence or absence of / T cells suggest that regulation of myeloid populations by / T cells may underlie the exaggerated, proinflammatory response that occurs in their absence. The findings are consistent with the concept that / T cells promote differentiation of inflammatory monocytes to tissue macrophages as an infection resolves.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
/ TCR–/–) [11
] were obtained initially from Jackson Laboratories (Bar Harbor, ME) and then maintained as a breeding colony at Emory University (Atlanta, GA). Mice were housed in filter-topped microisolator cages in a specific, pathogen-free facility and used at 8–16 weeks of age. The Institutional Animal Care and Use Committee approved all experimental procedures. L. monocytogenes wild-type strain #43251 (American Type Culture Collection, Manassas, VA) was grown overnight in brain-heart infusion broth (Difco Laboratories, Detroit, MI) at 37°C with aeration and then washed three times in phosphate-buffered saline (PBS) before intraperitoneal (i.p.) injection. Concentrations were determined by optical density with confirmation by colony counts on brain-heart infusion agar plates.
Cell preparation and culture
Peritoneal exudate cells (PEC) were harvested by lavage with cold Hanks’ balanced saline solution (HBSS) containing 0.06% bovine serum albumin (BSA) and 10 units heparin/ml. Cells were washed, counted, and resuspended for culture in RPMI 1640 supplemented with 10% fetal calf serum (FCS), 5 x 10–5 M 2-mercaptoethanol, 0.5 mM sodium pyruvate, 10 mM Hepes buffer, 50 U/ml penicillin, 50 µg/ml streptomycin, and 2 mM L-glutamine. Cells were used without further separation, or nonadherent cells were separated by vigorous washing with warm HBSS/BSA from cells that had adhered to plastic during incubation for 2 h at 37°C.
Antibodies for flow cytometry
The following reagents were purchased from PharMingen (San Diego, CA): RB6-8C5-phycoerythrin (PE) or -APC (anti-GR1 or Ly-6G), C15.6-APC (anti-IL-12), and MP6-XT22-APC [anti-tumor necrosis factor (anti-TNF-)], AF6-120.1-PE (anti-I-Ab), M1/70-PE (anti-CD11b, Mac-1), HL3-fluorescein isothiocyanate (FITC; anti-CD11c), 3/23-PE (anti-CD40), 16-10A1-PE (anti-CD80, B7.1), and GL1-PE (anti-CD86, B7.2). F4/80-FITC was purchased from Caltag Laboratories (Burlingame, CA). Isotype controls for these antibodies were purchased from PharMingen: rat immunoglobulin G (IgG)-FITC, hamster IgG-FITC, rat IgG-PE, hamster IgG-PE, and rat IgG-APC. An antibody to Fc receptors (FcRs; 2.4G2) was purified from tissue-culture supernatants.
Analysis of cell-surface markers by flow cytometry
PEC (1x106/sample) were pretreated with an antibody to FcRs to reduce nonspecific binding. Optimal concentrations of fluorochrome-conjugated antibodies to proteins expressed on the cell surface were then added for 30 min at 4°C. Cells were washed twice with wash buffer (PBS with 3% FCS and 0.1% sodium azide) and fixed with 1% paraformaldehyde. Data from a minimum of 20,000 PEC were collected using a FACSCalibur flow cytometer (Becton Dickinson, San Jose, CA) and analyzed using CellQuest software. Unstained cells and cells stained with isotype controls were used throughout to distinguish between autofluorescence or nonspecific binding and genuine marker-specific staining.
Detection of intracellular cytokines in individual myeloid populations by flow cytometry
PEC (3x106/ml) were incubated in ultra-low adherence, 24-well tissue-culture plates (Corning/Costar, Corning, NY) for 6.5 h. Cells we cultured in medium alone or stimulated with lipopolysaccharide (LPS; from Salmonella minnesota, Sigma Chemical Co., St. Louis, MO) or lipoteichoic acid (LTA; from Streptococcus pyogenes, Sigma Chemical Co.). Monensin (Sigma Chemical Co.) was added at 1.5 µM to inhibit cytokine secretion and thereby permit cytokine detection in the intracellular compartment [12
].
For analysis by flow cytometry, 1 x 106 cells were incubated 30 min at 4°C with fluorochrome-conjugated monoclonal antibodies (mAb) to cell-surface markers to identify individual cell populations and were then washed twice with wash buffer (PBS with 3% FCS and 0.1% sodium azide). Cells were incubated 15 min at room temperature with 50 µl fixation medium (Fix and Perm kit, Caltag Laboratories) and then washed once. Predetermined optimal concentrations of fluorochrome-conjugated anti-cytokine antibodies diluted in permeabilization buffer (Fix and Perm kit) were added for 15 min at room temperature, followed by two washes. If cells were not permeabilized, staining with anti-cytokine antibodies was reduced to background levels (data not shown). Background fluorescence was less than 0.5% after incubation with Ig isoptype controls conjugated to appropriate fluorochromes. Cells were examined on a FACSCalibur flow cytometer, and the data were analyzed using CellQuest software (Becton Dickinson). Gating on populations of cells based on their differential expression of cell-surface markers allowed for analysis of frequencies of cytokine-producing cells within individual subpopulations.
Analysis of cytokines secreted in vitro by ELISA
Cytokines secreted in response to in vitro restimulation were measured in tissue-culture supernatants from plastic-adherent PEC. Whole PEC (2x106/well) were cultured for 2 h in 24-well plastic tissue-culture plates. Nonadherent cells were removed by vigorous washing with warm culture medium. Adherent populations were then incubated for 24 h with medium alone or stimulated with LTA (10 µg/ml). Culture supernatants were analyzed by sandwich ELISA for secreted TNF- or IL-12 p40. For analysis of TNF-, a matched pair of antibodies was purchased from R & D Systems (Minneapolis, MN). For the IL-12p40 ELISA, antibodies were purified from culture supernatants from the C17.8.20 and C15.6.7 hybridomas (a generous gift from Dr. Giorgio Trinchieri, Wistar Institute, Philadelphia, PA) by Protein G chromatography or by ammonium sulfate precipitation from serum-free supernatants. C15.6.7 was biotinylated using standard techniques. Both assays were developed using extravidin-alkaline phosphatase (Sigma Chemical Co.) and p-nitrophenylphosphate as substrate (BioRad, Hercules, CA). Absorbance was read at 405 nM using a microplate reader (Bio-Tek Instruments, Inc., Winooski, VT). Standard curves were constructed using known amounts of recombinant murine cytokines. Sensitivity limits for the ELISAs were 
25 pg/ml.
Analysis of cell trafficking in the peritoneal cavity using tracking dye and flow cytometry
PEC were labeled in vivo by injection of PKH26 fluorescent dye (PKH26 red fluorescent cell-linker mini-kit, Sigma-Aldrich, St. Louis, MO). PKH26 (1x10–6 M) was injected i.p. in 1 ml manufacturer-supplied diluent. The injection dose was optimized empirically. This dye is incorporated into lipid layers of cell membranes, and serum proteins inhibited the reaction. As serum proteins are plentiful in the peritoneal cavity, the cell labeling is localized and terminated rapidly. Preliminary experiments confirmed that cell labeling was localized at the injection site such that splenocytes, bone marrow cells, or peripheral blood cells were not labeled after i.p. injection (data not shown). Cell subsets were identified using mAb tagged with other fluorochromes, and their fate was then tracked in vivo [13
].
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (55K): [in a new window] |
Figure 1. Changes in peritoneal myeloid cell populations in B6 control and / TCR–/– mice after i.p. infection with L. monocytogenes. PEC were harvested from uninfected mice or at the indicated times after i.p. injection of 5 x 103 Listeria per mouse. PEC were pooled from three mice per group. Surface expression of F4/80 and GR1 was analyzed by flow cytometry. Percentages of cells exhibiting F4/80 bright/GR1–, F4/80+/GR1+, or F4/80–/GR1 bright phenotypes are indicated in the appropriate sections of each dot plot. F4/80 was conjugated to FITC. Most of the autofluorescence and nonspecific binding was observed in the intermediate fluorescence-intensity range for FITC binding (x-axis). CD11b [membrane-activated complex-1 (Mac-1)] was expressed on all three of these populations and did not discriminate among them (see Table 1
). Population changes were confirmed in multiple experiments (see Fig. 2
).
|
/ T cells, this pattern of myeloid population dynamics was similar, but the return to a preinfection pattern was delayed in / TCR–/– mice. The greatest differences were apparent between 6 and 14 days after the infection (Fig. 1) . Neutrophils (bright GR1+ cells) and cells expressing F4/80 and GR1 remained elevated longer in / TCR–/– mice. Conversely, the "mature" tissue macrophage population (bright F4/80+GR1–) reappeared more slowly in the absence of / T cells.
This kinetic pattern was confirmed in multiple experiments (Fig. 2
). A sharp decrease in bright F4/80+ cells occurred after infection in control and / TCR–/– mice, but the return of these cells to the peritoneal cavity was consistently delayed in mice deficient in / T cells (Fig. 2A)
. Conversely, an increase in cells expressing F4/80 and GR1 followed infection in both groups but remained elevated longer in the / TCR–/– mice (Fig. 2B)
. Similarly, the influx of neutrophils that followed infection was not resolved as quickly in the / TCR–/– mice as in B6 controls (Fig. 2C)
. Differences were statistically significant (P<0.05) at Day 0 (Fig. 2A)
and at Day 6 (Fig. 2B) . Data from Day 10 were not analyzed statistically, as only three observations were available for that time-point. In two experiments, changes in myeloid cell populations were followed for longer times. By 3–4 weeks postinfection, there were no longer any apparent differences in these populations between normal and / TCR–/– mice, and the peritoneal myeloid population was phenotypically similar to that of a naïve mouse (data not shown).
 View larger version (13K): [in a new window] |
Figure 2. Changes in peritoneal myeloid cell populations after infection with Listeria are consistent in multiple experiments. Using procedures described in Figure 1
, myeloid populations were followed kinetically in B6 control and / TCR–/– mice. Data shown are means ± SEM. The number of observations for individual days was as follows: Day 0, n = 8; Day 2, n = 12; Day 6, n = 5; and Day 10, n = 3. Each observation was based on a pool of cells from three mice per group. Differences were statistically significant (P<0.05) at Day 0 (A) and at Day 6 (B), as determined by ANOVA with a Bonferroni post-test using Prism software. Data from Day 10 were not analyzed statistically, as only three observations were available for that time-point.
|
/ T cells [3
]. Increased cytokine production was detectable by ELISA in vivo and in vitro in response to nonspecific and antigen-specific stimuli in / TCR–/– mice. To examine the potential contribution of myeloid cells to those differences, we examined intracellular cytokine expression in each of the myeloid populations that was defined by the expression of F4/80 and GR1. Using flow cytometry and electronically selecting cells expressing these markers alone or cells that coexpressed them, it was possible to analyze the percentage of cells in each subset that was producing IL-12 or TNF- under defined conditions. These two cytokines were selected as representative of macrophage proinflammatory cytokines that are known to play a role in host defense against Listeria and other pathogens for which a T helper cell type 1 (Th1) response is required for protection. PEC were examined in the absence of infection or on Day 2 or Day 6 after an i.p. infection with Listeria. Cells were incubated in medium alone or with LPS or LTA, nonspecific stimuli from gram-negative or gram-positive bacteria, respectively. Monensin was added to all cultures to inhibit cytokine secretion and optimize detection by flow cytometry.
IL-12 and TNF- expression in peritoneal myeloid cells from uninfected mice is shown in Figure 3
. By gating on the typical population of bright F4/80+GR1– cells found in uninfected mice, we found that these cells expressed TNF- but not IL-12 ex vivo in the absence of further stimulation in vitro. Stimulation with LPS increased the frequency of TNF--expressing cells and induced detectable expression of IL-12. The frequencies of cells from / TCR–/– mice that expressed these proinflammatory cytokines were consistently higher than frequencies seen in B6 control mice. For example, 12.6% of resident macrophages from / TCR–/– mice produced TNF- in response to LPS, and only 7.2% of these cells from B6 control mice expressed TNF-.
 View larger version (61K): [in a new window] |
Figure 3. Proinflammatory cytokine production in peritoneal myeloid cells from uninfected mice. PEC pooled from groups of four naive B6 control or / TCR–/– mice were cultured in the presence of 1.5 µM monensin for 6.5 h with medium or with 1 µg/ml LPS. Cells were examined by flow cytometry for expression of F4/80 and GR1. The predominant myeloid population in these uninfected mice (F4/80 bright/GR1–) was examined further after fixation and permeabilization for cytokine-producing cells. Numbers indicate the percentages of gated F4/80 bright/GR1– populations that were producing IL-12 or TNF- as indicated. In a second experiment, the frequencies of cytokine-producing cells were higher for both mouse strains, but the pattern of increased numbers of cells producing cytokines in / TCR–/– mice was confirmed.
|
in the absence of further stimulation in vitro. Stimulation with LPS induced very small frequencies of IL-12+ cells, but the number of cells expressing TNF- increased dramatically. Nearly half of the neutrophils from mice lacking / T cells expressed TNF-, and only about one-third of neutrophils from B6 control mice were TNF-+. The population expressing F4/80 and GR1 (Fig. 4 , right side) contained low but detectable frequencies of cytokine-positive cells after incubation in medium alone. Stimulation with LPS increased the frequencies of cells expressing IL-12 or TNF-. The percentage of TNF-+ cells was again higher in / TCR–/– mice. Cells from both strains showed higher frequencies of cytokine-positive cells after in vitro stimulation at this interval than on Day 0 or Day 6, consistent with their status as recruited inflammatory macrophages.
 View larger version (58K): [in a new window] |
Figure 4. Myeloid populations harvested on Day 2 after Listeria infection and their production of proinflammatory cytokines after restimulation in vitro. PEC were harvested and pooled from groups of four B6 control or / TCR–/– mice 48 h after i.p. injection of 6.7 x 103 Listeria/mouse and cultured in vitro as described for Figure 2
. Intracellular cytokine expression is shown after gating on the myeloid populations, which predominated at this time after infection: bright GR1+ cells on the left and cells expressing F4/80 and GR1 on the right. Numbers indicate the percentages of gated populations that were producing IL-12 or TNF as indicated.
|
/ TCR–/– mice (Fig. 5
). This population of bright F4/80+ cells contained higher frequencies of IL-12- and TNF--producing cells than the corresponding population found in uninfected mice (see Fig. 3
), suggesting a more activated condition. Similarly, the expression of class II molecules was higher on the bright F4/80+ cells on Day 6 than on Day 0 (see Table 1
). Cells expressing F4/80 and GR1 were still present in both strains but were once again more predominant in / TCR–/– mice (32.5% vs. 8.2% in B6 controls). As on Day 2, significant percentages of these cells expressed proinflammatory cytokines, especially after restimulation in vitro. Overall frequencies of restimulated, cytokine-positive cells were lower on Day 6 (Fig. 5)
than on Day 2 (Fig. 4)
, but PEC from / TCR–/– mice again exhibited a higher frequency of TNF-+ cells. Consistent with the experiments shown in Figures 1
and 2
, the bright GR1+ neutrophils remained elevated in / TCR–/– mice compared with B6 controls. As on Day 2, these bright GR1+ cells were expressing TNF- with higher frequencies in populations from mice lacking / T cells. It is interesting that the frequencies of TNF--expressing cells in the absence of in vitro restimulation in cultures from both strains were higher on Day 6 than on Day 2, suggesting a prolonged state of in vivo activation. Restimulation in vitro had much less impact on Day 6 than on Day 2 with respect to increasing the percentages of cells expressing TNF-.
 View larger version (71K): [in a new window] |
Figure 5. Myeloid populations harvested on Day 6 after Listeria infection and their production of proinflammatory cytokines after restimulation in vitro. PEC were harvested and pooled from groups of seven B6 control or / TCR–/– mice 6 days after i.p. injection of 1.5 x 104 Listeria/mouse and were cultured in vitro as described for Figure 2
. In this experiment, LTA (10 µg/ml) was substituted for LPS. Intracellular cytokine expression is shown after gating on the myeloid populations which predominated at this time after infection: bright GR1+ cells on the left and cells expressing F4/80 and GR1 on the right. In B6 controls, there were sufficient numbers of F4/80 bright/GR1– cells to evaluate for cytokine production. Numbers indicate the percentages of gated populations that were producing IL-12 or TNF as indicated.
|
|
View this table: [in a new window] |
Table 1. Activation Marker Expression on Subsets of Peritoneal Myeloid Cells
|
/ TCR–/– mice, we normalized the data presented in the previous three figures to reflect the combined differences in population dynamics and in frequencies of cells from those populations that expressed IL-12 or TNF-. Data shown in Figure 6
were derived by multiplying the percentage of cells exhibiting the indicated combination of F4/80 and GR1 markers by the percentage of that population expressing the cytokine intracellularly. This number divided by 100 is the normalized frequency presented on the y-axes in Figure 6
. Presence or absence of a graph representing each subset of the myeloid populations reflects the presence or absence of that subset relative to the time of infection. For example, bright F4/80+ cells are present in both strains preinfection, in neither strain on Day 2, and in only the B6 controls on Day 6. The general pattern of lower percentages of bright F4/80+ cells and higher percentages of bright GR1+ neutrophils and double-positive (F4/80+,GR1+) cells in / TCR–/– mice (see Fig. 2
) is superimposed on the differences in frequencies of cytokine-expressing cells in the two strains. As a result, the differences in cytokine-producing cells in uninfected mice are dampened by this analysis, but differences observed at Days 2 and 6 are amplified. There are clearly and consistently more myeloid cells producing proinflammatory cytokines for a longer time post-infection in / TCR–/– mice.
 View larger version (33K): [in a new window] |
Figure 6. Frequencies of cytokine-producing cells adjusted for population size. Peritoneal myeloid populations undergo major changes in response to Listeria infection. To account for this and for differences between B6 control (solid bars) and / TCR–/– (striped bars) mice, frequencies of cytokine-producing cells shown in Figures 3
4
5
were "normalized" using the following formula: (% of population expressing the myeloid marker combinationx% of that population expressing the cytokine)/100. For example, if only one cell type were present and all of the cells were producing cytokine, the normalized frequency would equal (100x100)/100 = 100. Because of population dynamics after infection, not all populations were present in sufficient numbers to analyze at each time-point.
|
/ TCR–/– mice by PEC from / TCR–/– mice harvested 6 days after Listeria infection. We have previously reported higher levels of IL-12p40 and IFN- production by PEC obtained from / TCR–/– mice between 2 and 6 days after i.p. Listeria infection [3
].
 View larger version (26K): [in a new window] |
Figure 7. Secretion of proinflammatory cytokines was also exaggerated in / TCR–/– mice. Plastic-adherent PEC harvested and pooled from seven mice per group on Day 6 after Listeria infection were incubated at 37°C in medium alone or with 10 µg/ml LTA. Supernatants were harvested after 24 h and analyzed by ELISA for IL-12 p40 or TNF-.
|
/ TCR–/– mice/ T cells (Table 1)
. Although some of these populations were present in low frequencies (e.g., bright F4/80+,GR1– cells on Day 2), there were usually sufficient cells available to assess surface-marker expression. Bracketed values in Table 1
indicate low frequencies of that population at the indicated time-point. Changes were evident in some markers after infection, but there were no notable differences between B6 and / TCR–/– mice. In general, peritoneal cells are more highly activated than cells found in the spleen. Class II expression was up-regulated in both groups on bright F4/80+ cells by Day 2 after Listeria infection and remained elevated on Day 6. All other activation markers that we evaluated were expressed on essentially all of the bright F4/80+ cells, even in uninfected mice. Activation markers were present on lower percentages of myeloid cells that expressed F4/80 and GR1, but / TCR–/– mice were similar to B6 controls. Considerable increases in the percentages of neutrophils (bright GR1+F4/80– cells) expressing activation markers were apparent between Days 2 and 6 after infection, but again, there was no major difference in the absence of / T cells. CD11b (Mac-1) was expressed at a high level on all three populations and did not discriminate among them. In addition to analyzing the frequency of expression of activation markers, we evaluated the mean fluorescent intensity for each marker. Levels of expression were similar on cells from B6 and / TCR–/– mice (data not shown).
Cell trafficking in the peritoneal cavity after infection with Listeria
The changes that we have described in F4/80 and GR1 marker expression on PEC following Listeria infection might result from several possible scenarios. These markers might be up-regulated and down-regulated on a resident population of peritoneal cells. Alternatively, cell trafficking might result in one or more turnover events in peritoneal cell populations during the course of infection. Third, a combination of marker regulation and cell trafficking might be operational. To discriminate among these possibilities, peritoneal cells in B6 control mice were labeled in vivo with PKH26 dye (PKH). As dye incorporation is inhibited by serum proteins, only the cells present in the peritoneum at the time the dye is injected were labeled. These cells were then followed kinetically by flow cytometry. PKH was first injected into uninfected B6 mice (Fig. 8
, top row). The dye was incorporated into the resident F4/80+GR1– population (Day 0) and diminished only slightly in fluorescence intensity during the 20-day observation period (see Day 2 and Day 20). This suggests that this population is truly resident and does not traffic significantly in this site if mice are not infected. There was no detectable influx of unlabeled, bright F4/80+ cells during this time-period, consistent with a previous study in which resident macrophage populations were stable for up to 49 days in the absence of peritoneal inflammation [15
]. In contrast to macrophages, T cell populations were dynamic. Fifty percent of the peritoneal T cells had been replaced within 2 days, and the entire T cell population had essentially turned over within 6 days postinfection (data not shown). Infection with Listeria subsequent to PKH labeling of PEC resulted in a population of cells on Day 2 that expressed F4/80 and GR1 as expected but were not labeled with PKH (Fig. 8
, middle row). This indicated that this double-positive population had been recruited to the site as a result of the infection. Cells labeled intensely with PKH were no longer detectable on Day 2, suggesting that this population had been destroyed by the infection or had trafficked elsewhere.
 View larger version (55K): [in a new window] |
Figure 8. Myeloid cell trafficking and maturation in the peritoneal cavity after infection with Listeria. Control B6 mice were injected i.p. with 1 x 10–6 M PKH26 dye to label cells in situ. Data in the top row are from mice that received no further injections. In the middle row, mice were infected i.p. with 3.2 x 104 L. monocytogenes 1 h after injection of PKH26 dye. In the bottom row, mice were infected with Listeria 2 days before injection of PKH26 dye. PEC were harvested on Days 0, 2, or 20 as indicated, stained with F4/80-FITC and GR1-APC to identify myeloid populations, and analyzed by flow cytometry. After electronic selection of the predominant myeloid population present at each time-point as indicated, the level of PKH expression was analyzed by histogram overlay (far-right column). Data presented from individual mice are representative of three to four mice per group.
|
To determine whether the cells that express F4/80 and GR1 that predominate on Day 2 are precursors to the bright F4/80+GR1– cells present on Day 20, we injected PKH on Day 2. This resulted in a bright PKH+ population of double-positive cells on Day 2 (Fig. 8
, bottom row). If these cells remained at the site and changed marker-expression levels, then on Day 20, there should be a population of F4/80+GR1– cells with a level of PKH fluorescence only slightly diminished from the brightness observed on Day 0. However, if cell trafficking brought yet another population of myeloid cells into the peritoneal cavity, then only moderate PKH staining, resulting from phagocytosis of labeled membrane fragments, would be expected. In fact, both levels of PKH fluorescence were found in the F4/80+GR1– population (Fig. 8
, bottom row). The more brightly labeled PKH population predominated, however, indicating that maturation of the myeloid population in situ was the primary pattern. In the three mice tested, 64.1% (±9.9%) of the cells exhibited the brighter fluorescence signal. The presence of a smaller population of moderately bright PKH+ cells suggests that a population of myeloid cells entered the peritoneal cavity subsequent to PKH labeling on Day 2 and acquired PKH fluorescence as a result of phagocytosis. These data strongly support the concept that cells, which express F4/80 and GR1, remain in the peritoneal cavity and mature into the resident population of cells that express F4/80 at high intensity in the absence of GR1. In the absence of / T cells, this change in marker expression is delayed, suggesting that / T cells influence this process. Note also that the proportion of myeloid cells in the PEC remains stable, indicating that although expression of cell-surface markers changes, there is neither a major gain nor loss of myeloid cells during the response to Listeria infection.
In vitro maturation of myeloid cells is deficient in the absence of / T cells
To examine this myeloid maturation process in vitro, we collected PEC on Day 2 following i.p. Listeria infection when cells expressing both markers predominated. After 48 h of culture at 37°C, the percentage of bright F4/80+GR1– cells had increased dramatically in populations from Listeria-infected, B6 control mice (Fig. 9
). This transition was far less pronounced in cultures from / TCR–/– mice. The ratio of bright F4/80+GR1– cells to cells expressing F4/80 and GR1 shown in Figure 9
highlights the differences between control and / TCR–/– mice. To determine whether myeloid growth/maturation factors might influence this process, GM-CSF or IL-3 was added to the cultures. Neither of these factors significantly influenced the expression of F4/80 or GR1 on cells from B6 control mice (Fig. 9)
. A much different pattern was seen with cells from / TCR–/– mice, in that the expression of F4/80 in relation to GR1 was increased considerably, suggesting that these growth/maturation factors might be deficient in the absence of / T cells. Even with the addition of these factors, the maturation of cells from / TCR–/– mice was not equivalent to that found in B6 control mice.
 View larger version (21K): [in a new window] |
Figure 9. Maturation of myeloid cells in vitro is altered in / TCR–/– mice. PEC harvested and pooled from three mice per group on Day 2 after Listeria infection (1.1x104 i.p.) were cultured in medium alone or with the addition of 10 ng/ml recombinant granulocyte macrophage-colony stimulating factor (rGM-CSF) or 10 ng/ml rIL-3. Expression of F4/80 and GR1 was analyzed by flow cytometry on freshly harvested cells (ex vivo) and after 48 h in vitro. The ratio between F4/80 bright/GR1– cells and F4/80+/GR1+ cells is indicated for each strain. Data are representative of two experiments.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
/ T cells play a critical role in this process. In their absence, the number and frequency of cells in the monocyte/macrophage lineage which expresses proinflammatory cytokines were strikingly elevated, resulting in increased secretion of these cytokines. TNF--producing neutrophils also persisted longer in the peritoneum in the absence of / T cells. Neutrophils have been shown to persist in hepatic lesions in Listeria-infected mice that lack / T cells [4
], possibly as a result of a deficit in monocyte chemoattractant protein-1 (MCP-1) [16
]. Reduced MCP-1 could lead to decreased recruitment of monocytes that were found necessary to clear the neutrophilia. We have reported elsewhere that the antigen-specific /ß T cell-mediated immune response is also elevated in / TCR–/– mice [3
], possibly as a result of the prolonged inflammation.
Paradoxically, monocyte/macrophage-lineage cells that express GR1 have been characterized as "suppressor macrophages" in several other systems because of their ability to inhibit the proliferative responses of naïve T cells to polyclonal stimuli [17
18
19
] or to enhance cell death in activated T cells [18
]. This pattern at first seems inconsistent with our findings of enhanced, proinflammatory and Th1 responses in / TCR–/– mice. The functional outcome may be highly dependent however, on the nature of the stimulus and the duration of exposure to these suppressor macrophages. Although we have reported an increased Th1 response to polyclonal or complex antigen stimulation [3
], more recent experiments have shown that if T cells were stimulated with an immunodominant Listeria peptide, the response was suppressed in / TCR–/– mice (manuscript in preparation). In a tumor model, increases in immature myeloid cells that accumulated in response to tumor growth were associated with impairment of immune responses [20
]. The dichotomy between immune suppression in the tumor system and enhancement in the bacterial infection model may be a result of the more prolonged inhibition of the differentiation of immature myeloid cells in the tumor model [21
]. The complexities of the interactions among / T cells, monocyte/macrophage-lineage subsets, and development of antigen-specific T cells are under continuing investigation in our laboratory.
There are several possible ways in which / T cells might exert their influence on the monocyte/macrophage lineage. Under normal circumstances, the fate of inflammatory macrophages recruited to the peritoneal cavity in response to infection might involve emigration to other sites, death, or differentiation in situ into mature tissue macrophages as the infection is resolved. Our data most strongly support the third possibility. Under conditions of sterile inflammation (thioglycollate injection), recruited macrophages have been shown to emigrate to draining lymph nodes as the inflammation subsides [22
]. After Listeria infection, however, inflammatory macrophages labeled with PKH were still found in the peritoneal cavity 18 days later (see Fig. 8
). The persistence of these labeled cells coupled with the observation that the fraction of PEC represented by cells of the monocyte/macrophage lineage does not differ appreciably between Day 2 and Day 20 (see Fig. 8
) suggest that emigration does not play a major role. As the number of cells recruited to the peritoneal cavity in response to thioglycollate is much greater than after Listeria infection (data not shown), it is not surprising that different mechanisms may modulate the inflammatory macrophage population.
Cell death may play some role as indicated by the presence on Day 20 of a subpopulation of cells with an intermediate level of PKH intensity (Fig. 8)
. Unlabeled, inflammatory macrophages that may have entered the peritoneum subsequent to dye-labeling on Day 2 appear to have acquired PKH label via phagocytosis of dead cells. Some cells may have died as a direct result of the bacterial infection. Alternatively, a population of activated macrophages found in the peritoneal cavity 6 days after Listeria infection may be selectively killed by / T cells [23
]. In the absence of this killing mechanism, persistence of activated macrophages could contribute to the prolonged inflammation seen in / TCR–/– mice. Cells susceptible to killing by / T cells were characterized by Percoll density gradient separation [23
]. Susceptible cells were in a low-density fraction, expressed relatively high levels of F4/80, and were described as mature macrophages [23
], suggesting that they likely expressed only F4/80. Earlier studies showed that cells of lower density expressed F4/80 more brightly and were more differentiated than cells of higher density [24
]. However, if these cells expressed F4/80 and GR1, then this killing mechanism might at least partially underlie our observations. We thus examined F4/80 and GR1 expression on PEC separated on Percoll gradients in our laboratory. We found dim F4/80+GR1+ cells not only in the low-density fraction but in other fractions as well, indicating the heterogeneity of the maturing population (data not shown). One interpretation is that as these cells mature, they become susceptible to killing by / T cells. Thus, in the absence of / T cells, a delay in maturation combined with absence of the killer cell population results in a relative increase in this population. The data from PKH-labeled cells in our experiments strongly suggest that the majority of the inflammatory macrophages remains in the peritoneal cavity and differentiates into mature tissue macrophages. The myeloid population differences we observed between control and / TCR–/– mice suggest that / T cells induce myeloid-lineage maturation, similar to the effect reported for human / T cells on DC maturation [25
]. There was no significant decrease in cell numbers in the peritoneal cavity during the 3 weeks following initial infection (not shown). Moreover, cells of the monocyte/macrophage lineage comprise 25% of the PEC on Day 2 and Day 20 (Fig. 8)
, indicating there is not a major loss of the monocyte/macrophage component in the peritoneal cavity. Approximately two-thirds of the inflammatory macrophages labeled with PKH 2 days after infection remained brightly labeled in the peritoneal cavity on Day 20. Two lines of evidence indicate that cytokine production is lower in the more mature tissue macrophages. First, the frequency of cytokine-producing cells is lower in resident macrophages (see Fig. 3
), and second, the level of secreted cytokines gradually declines as a Listeria infection is cleared [3
]. Thus, a delay in this maturation process in the absence of / T cells would prolong inflammation.
/ T cells may influence macrophage maturation via several potential mechanisms. Studies in the M-CSF-deficient osteopetrotic mouse have shown that peritoneal macrophages are highly dependent on M-CSF for their maturation [26
]. Macrophage maturation mechanisms are site-dependent, but peritoneal macrophage populations can be restored by injection of M-CSF. Restoration is effective only if the growth factor is injected directly into the peritoneal cavity, suggesting local regulation of this process [27
]. Our in vitro experiments showed that addition of M-CSF (not shown) or GM-CSF could partially overcome the deficit in the in vitro maturation of macrophages in PEC harvested from / TCR–/– mice. Some populations of / T cells have been reported to produce these growth factors. Circulating, bovine / T cells have been shown by gene activation analysis (serial analysis of gene expression) to produce GM-CSF after activation [28
], and murine epidermal / T cells appear to stimulate growth and differentiation of dendritic epidermal cells by secreting M-CSF and GM-CSF [29
]. / T cells may also influence the production of CSFs by one of the many other types of cells that can produce these broadly active growth factors [30
]. / T cells may also affect macrophage maturation by regulation of caspase activation. Cells in which maturation is induced by M-CSF rely on activation of caspase-3 or caspase-9 for the maturation process to proceed normally [31
]. In M-CSF-dependent cells, activation of these caspase molecules results in terminal differentiation rather than apoptosis. The underlying molecular pathway and its regulation are not well understood. It is interesting that caspase activation is also required for differentiation of epidermal keratinocytes [32
]. / T cells and keratinocytes are found in close association in the epidermis, and their functional interactions have been investigated extensively [33
, 34
]. Although / T cells have been shown to produce keratinocyte growth factor [35
], they may also initiate caspase activation at tissue sites where caspases are involved in terminal differentiation of specialized tissue subsets.
Another potential mechanism might involve the induction of the synthesis of heat shock proteins (HSP) by macrophages as a necessary step in the maturation process. In the absence of / T cells, HSP65 is not induced in maturing macrophages, resulting in apoptosis rather than differentiation [36
, 37
]. Impaired macrophage maturation in these studies was associated with increased susceptibility to infection with intracellular protozoan parasites.
We have also considered the possibility that macrophage populations may be altered by changes in the clearance of Listeria in / TCR–/– mice. In our laboratory and others, however, there are no clear and consistent differences in clearance of Listeria in the absence of / T cells. Most reported changes are modest, and in fact, some subsets of / T cells enhance clearance, and those expressing V1 significantly inhibit it [38
]. Thus, any modest changes in bacterial numbers in the absence of the entire population of / T cells are unlikely to cause the pronounced differences that we have observed.
By using the combination of F4/80 and GR1 to define subsets of myeloid cells, we have been able to discern alterations in myeloid cell dynamics, maturation, and cytokine production in the absence of / T cells. Given the importance of myeloid-lineage cells in the innate, proinflammatory response and their significant role as APC for the antigen-specific /ß T cell response, the regulation of myeloid populations by / T cells may underlie many of the phenotypic differences found in mice lacking / T cells. These differences range from changes in tissue pathology and resistance to a variety of infectious agents to exacerbations in T cell-mediated autoimmune diseases.
Received November 19, 2003; revised February 20, 2004; accepted March 15, 2004.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
T cells Adv. Immunol. 71,77-144[Medline] T cells for production of IFN- is mediated by bacteria via macrophage-derived cytokines IL-1 and IL-12 J. Immunol. 154,5832-5841[Abstract]/ T cells after infection with Listeria monocytogenes Infect. Immun. 69,7213-7223 T cells J. Immunol. 153,3101-3115[Abstract] T lymphocyte subsets defines unique tissue-specific functions J. Leukoc. Biol. 73,306-314 T cells Nat. Rev. Immunol. 3,233-242[CrossRef][Medline] T cells: functional plasticity and heterogeneity Nat. Rev. Immunol. 2,336-345[CrossRef][Medline] gene mutant mice: independent generation of /ß T cells and programmed rearrangements of / TCR genes Cell 72,337-348[CrossRef][Medline] J. Immunol. 165,6056-6066 and nitric oxide-dependent mechanism J. Immunol. 167,4293-4302 T cells cytotoxic for activated macrophages J. Exp. Med. 191,2145-2158 T cells induce dendritic cell maturation Clin. Immunol. 103,296-302[CrossRef][Medline] T cell subsets defines distinct immunoregulatory phenotypes and unexpected gene expression profiles J. Immunol. 170,356-364 T cells in tissue repair Immunol. Res. 21,63-69[CrossRef][Medline] T cells Science 266,1253-1255 T cells prevents apoptosis of macrophages and contributes to host defense in mice infected with Toxoplasma gondii J. Immunol. 159,2375-2381 T cells play an important role in hsp65 expression and in acquiring protective immune responses against infection with Toxoplasma gondii J. Immunol. 155,244-251[Abstract] T cell subset can increase host resistance to a bacterial infection J. Immunol. 165,6472-6479This article has been cited by other articles:
 |
|
 |
|
  F. Gays, J. G. Aust, D. M. Reid, J. Falconer, N. Toyama-Sorimachi, P. R. Taylor, and C. G. Brooks Ly49B Is Expressed on Multiple Subpopulations of Myeloid Cells J. Immunol., November 1, 2006; 177(9): 5840 - 5851. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Zachariadis, J. P. Cassidy, J. Brady, and B. P. Mahon {gamma}{delta} T Cells Regulate the Early Inflammatory Response to Bordetella pertussis Infection in the Murine Respiratory Tract Infect. Immun., March 1, 2006; 74(3): 1837 - 1845. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. E. Egan, J. E. Dalton, E. M. Andrew, J. E. Smith, M.-J. Gubbels, B. Striepen, and S. R. Carding A Requirement for the V{gamma}1+ Subset of Peripheral {gamma}{delta} T Cells in the Control of the Systemic Growth of Toxoplasma gondii and Infection-Induced Pathology J. Immunol., December 15, 2005; 175(12): 8191 - 8199. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Wands, C. L. Roark, M. K. Aydintug, N. Jin, Y.-S. Hahn, L. Cook, X. Yin, J. Dal Porto, M. Lahn, D. M. Hyde, et al. Distribution and leukocyte contacts of {gamma}{delta} T cells in the lung J. Leukoc. Biol., November 1, 2005; 78(5): 1086 - 1096. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Zakharova and H. K. Ziegler Paradoxical Anti-Inflammatory Actions of TNF-{alpha}: Inhibition of IL-12 and IL-23 via TNF Receptor 1 in Macrophages and Dendritic Cells J. Immunol., October 15, 2005; 175(8): 5024 - 5033. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. M. Freeman and H. K. Ziegler Simultaneous Th1-Type Cytokine Expression Is a Signature of Peritoneal CD4+ Lymphocytes Responding to Infection with Listeria monocytogenes J. Immunol., July 1, 2005; 175(1): 394 - 403. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
