Published online before print June 24, 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1
* Veterans Affairs Hospital and
Division of Infectious Diseases, University of Cincinnati College of Medicine, Cincinnati, Ohio, USA
1 Correspondence: Division of Infectious Diseases, Department of Medicine, University of Cincinnati College of Medicine, Cincinnati, OH 45267-0560, USA. E-mail: george.deepe{at}uc.edu
ABSTRACT
Numerous in vitro studies have demonstrated that Histoplasma capsulatum is engulfed by the diverse populations of phagocytic cells including monocytes/macrophages (M
), immature dendritic cells (DC), and neutrophils. The in vivo distribution of H. capsulatum has yet to be examined following an intrapulmonary challenge. To accomplish this goal, we engineered GFP into two genetically dissimilar strains of H. capsulatum, G217B and186R. C57BL/6 mice were infected with each of these strains, and we analyzed the distribution of this fungus in the three major phagocytic populations on successive days. Yeast cells were found in all three populations of cells from Days 1 through 7. Proportionally, DC dominated at Day 1, whereas the majority of yeast cells was detected in neutrophils thereafter. Yeast cells were present in inflammatory and resident M on Day 3, but on Day 7, they were chiefly in inflammatory M. Yeast cells were predominantly in a CD11c+intermediate/high, F4/80–, CD11b+, Ly-6C+, CD205+ DC population. Neutralization of TNF-
or IFN-
produced a significant redistribution of yeast cells. These results reveal the complex nature of intracellular residence of this fungus. Moreover, the findings demonstrate that there is a skewing in the subpopulations of cells that are infected, especially DC.
Key Words: rodent • fungus • innate immunity
INTRODUCTION
The human pathogen, Histoplasma capsulatum, is a leading cause of respiratory illness among the medically important fungi. The organism exists in soil as a saprobe and transforms into the pathogenic yeast phase upon inhalation into mammalian lungs. A cascade of events is set in motion that results in the reduction, but not sterilization, of yeast cells from tissues rich in mononuclear phagocytes. Following inhalation, there is an initial influx of recruited inflammatory cells including monocytes and neutrophils [1
, 2
]. In vitro, monocytes and macrophages (M) are permissive for the growth of yeast cells until cellular immunity is activated [3
, 4
]. Polymorphonuclear neutrophils (PMN), on the other hand, mediate fungistasis [5
]. More recently, immature human dendritic cells (DC) have been demonstrated to exert fungicidal activity [6
].
Although yeast cells are ingested by phagocytes, little is known about the in vivo distribution of yeast cells within these populations during the course of acute histoplasmosis. Prior work by us [2
, 7
8
9
] has analyzed the inflammatory response to H. capsulatum under several conditions but without regard to whether a phagocyte was infected. We have generated GFP containing yeast cells in two genetically distinct but commonly studied strains, G217B and G186R. Using flow cytometry, we endeavored to identify and quantify the cellular populations that harbor the fungus in vivo. We also sought to determine if neutralization of two critically important endogenous mediators of protective immunity to H. capsulatum, IFN- and TNF-, change the distribution of yeast cells within phagocytes. Our data report the cellular populations that are associated with the fungus within the lung and reveal that cytokine neutralization alters the profile of cells that ingest this fungus. These results provide for the first time an examination of the flux of H. capsulatum within inflammatory cell populations during an active infection.
MATERIALS AND METHODS
Mice
C57BL/6 mice were purchased from The Jackson Laboratory (Bar Harbor, ME, USA). Animals were housed in microisolator cages and maintained by the Department of Laboratory Animal Medicine (University of Cincinnati, OH, USA), which is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care. All animal experiments were performed in accordance with the Animal Welfare Act Guidelines of the National Institutes of Health, and all protocols were approved by the Institutional Animal Care and Use Committee of the University of Cincinnati.
Preparation of H. capsulatum and infection of mice
H. capsulatum yeast strains G217B and G186R were prepared as described previously [7
]. To produce infection in naive mice, animals were inoculated intranasally (i.n.) with 105, 2 x 106, or 1.5 x 107 H. capsulatum yeast cells in a 30-µl vol of HBSS.
Construction of GFP-expressing H. capsulatum
GFP-expressing strains 186R and 217B were prepared by Agrobacterium tumifaciens-mediated transformation of H. capsulatum as described [10
]. Briefly, the GFP gene under control of the H. capsulatum calcium-binding protein promoter was cloned downstream of the bleomycin-resistance cassette with the Aspergillus nidulans tryptophan synthetase terminator at the 3' end.
Preparation of lung leukocytes
Lungs were teased apart with the frosted ends of two glass slides. The solution was filtered through 60 µm nylon mesh (Spectrum Laboratories Inc., Rancho Dominguez, CA, USA) and washed three times with HBSS. Leukocytes were isolated by gradient density centrifugation using Lympholyte-M (Cedarlane Laboratories, Hornby Ontario, Canada).
Reagents and flow cytometry
Recombinant GFP was purchased from Invitrogen (Carlsbad, CA, USA). The following antibodies were purchased from BD Biosciences (San Diego, CA, USA): CD62 ligand (CD62L)-, CD11b-, CD11c-, and Annexin V-allophycocyanin; CD80-, CD86-, I-Ab-, membrane-activated complex 3 (Mac-3)-, and Ly-6G-PE; and Ly-6C-biotin and CD8 conjugated to streptavidin-PerCP. CD205-PE was purchased from Miltenyi Biotec (Auburn, CA, USA). Biotin-conjugated CD68 was purchased from AbDSerotec (Raleigh, NC, USA). FITC-conjugated F4/80 was purchased from Caltag Laboratories (Burlingame, CA, USA); 2 x 106 cells were incubated with 0.5 µg antibody in staining buffer (1% BSA in PBS) for 10 min at 4°C. The cells were washed in staining buffer, and fluorescence was measured using a FACSCaliber flow cytometer (BD Biosciences). Between 50,000 and 100,000 events were counted. Absolute values of GFP+ cells were calculated by the number of cells x the percent GFP+. For cell subpopulations, the aforementioned number was multiplied by the percentage of a given subpopulation. Cells were sorted by a FACSVantage.
Treatment of mice with neutralizing mAb to cytokines
Mice were injected i.p. with mAb on the day of infection. Purified mAb were produced by the National Cell Culture Center (Minneapolis, MN, USA). mAb (1 mg) to TNF- (clone XT-22.1) or mAb to IFN- (clone XMG 1.6) was administered. Control animals received an equal amount of rat IgG concomitantly.
Statistical analyses
ANOVA was used to compare groups. P < 0.05 was considered statistically significant.
RESULTS
GFP-expressing H. capsulatum
GFP+ and wild-type (wt) yeast cells manifested distinct fluorescent profiles (Fig. 1A
1B
1D
and 1E
). The mean fluorescence intensity (MFI) of wt G217B was 112.3, and that of G217B-GFP was 2687.6. The MFI of wt G186R was 282.6, and that of G186R-GFP was 1404.7. A depiction of the GFP+ cells in the lung leukocytes in mice infected for 7 days is illustrated in Figure 1C
and F
. The insertion of the GFP gene did not change the virulence or growth of G217B as compared with wt [10
] or that of G186R. The mean number of CFU log10 (±SEM) at Day 7 of infection in mice infected with 2 x 106 G186R (5.0±0.4; n=5 mice) was similar to that of G186R-GFP (4.8±0.4; n=5 mice). Mice infected 2 x 106 yeast cells of GFP-186R, or wt G186R survived for >45 days.
 View larger version (38K): [in a new window] |
Figure 1. Flow cytometry profile of G217B and G186R engineered to express GFP. Yeast cells were harvested from cultures of actively growing yeast cells and analyzed by flow cytometry (B and E). Controls were cells not expressing GFP (A and D). (C and F) Representative dot-plots of GFP+ cells in mouse lungs from Day 7 of infection.
|
(Mac-3+, CD11c–/lo, Ly-6G–, high autofluorescence), neutrophils (Ly-6G+, CD11c–, Mac-3–), and DC (CD11c+intermediate/high, Mac-3–, I-Abhigh, low autofluorescence), as these are the cells that principally ingest H. capsulatum.
The absolute number of leukocytes at each day was similar for mice infected with G217B and G86R (Fig. 2
). On Days 1 and 3, there were significant differences in the proportion and absolute numbers of M in lungs of mice infected with G217B as compared with G186R. On Days 3, 5, and 7, the proportion and number of PMN from mice infected with G217B exceeded that of G186R-infected mice. The proportion and number of DC in mice infected with either strain were similar.
 View larger version (30K): [in a new window] |
Figure 2. Analysis of M, PMN, and DC from the lungs of mice infected with G217B or G186R. Mice were infected with 2 x 106 yeast cells i.n., and cells were analyzed on Days 1, 3, 5, and 7 of infection. M were identified as Mac-3+, CD11c–/lo cells, high autofluorescence; PMN by Ly-6G+, Mac3–, and CD11c–cells; and DC as Mac-3–, CD11cintermediate/high, I-Abhi, low autofluorescence. Data represent pooled mean ± SEM of n = 7–9 mice in two or three separate experiments. *, P < 0.05; **, P < 0.01.
|
or IFN-, both of which cause a marked increase in fungal burden [7
, 11
]. In a second evaluation, mice were infected with G217B for 7 days, and cells stained with CD45, a common leukocyte antigen. The mean percentage (±SEM) of cells that coexpressed GFP and CD45 was 10.0 ± 1.5%, and the percentage of GFP+ CD45– cells was <0.01%. In addition, we sorted GFP+ cells and examined them by fluorescence microscopy. No extracellular yeast cells were observed in the positive or negative sort. This finding was also true for mice receiving mAb to TNF- or IFN-. Neutralization of either of these cytokines causes a marked increase in fungal burden and is associated with death of mice following infection with a nonlethal challenge [7
, 11
]. Greater than 95% of leukocytes from the positive gate contained at least one yeast cell, whereas <0.1% of cells from the negative gate contained a yeast cell. Extracellular yeast cells were not observed. These results indicate that detection of GFP+ events was associated almost exclusively with phagocytes. We were also concerned about GFP escaping from cells and causing a false-positive detection. Mice were inoculated with 1 µg GFP i.n. or saline as a control, and cells were examined after 24 h. In controls, the background green fluorescence was detected in <0.34% of cells. In mice injected with GFP, green-positive cells constituted 0.32% of total cells.
Analysis of the infected phagocyte populations
We determined the number of GFP+ cells during the first 7 days of infection. GFP+ cells increased serially in mice infected with either strain, but the number of GFP+ cells was greater in mice infected with G217B at each day (Fig. 3
).
 View larger version (13K): [in a new window] |
Figure 3. Distribution of lung phagocytes that harbor yeast cells. Mice were infected with 2 x 106 yeast cells, and at Days 1, 3, 5, and 7 postinfection, lung leukocytes were stained and analyzed. Cells were gated on GFP, and a proportion of surface-positive cells was assessed. Absolute numbers were calculated by multiplying the number of leukocytes x the proportion of a positive cell population that harbored GFP+ yeast cells. Data represent mean ± SEM of seven to nine mice/group performed in two or three separate experiments. *, P < 0.05, compared with controls; **, P < 0.01, compared with controls.
|
, we gated on Mac-3+, CD11c–/lo cells and excluded a population of TNF and inducible NO synthase-producing DC that are CD11c+ and intracellular Mac-3+high [12
]. For DC, we analyzed Mac-3–, CD11c+intermediate/high, and I-Abhigh, which we found in preliminary studies to be >97% negative for CD68, a M marker [13
]. For G217B and G186R, the number of M, PMN, and DC containing yeast cells increased between Days 1 and 7. PMN were the most heavily infected cells (Fig. 3)
. The number of M and PMN bearing G186R was considerably less than that of cells infected with G217B on most days. The numbers of infected DC for the two strains only diverged at Day 7 (Fig. 3)
.
As a corollary, we calculated the relative distribution of phagocytes that harbor yeast cells to assess if one population disproportionately contained fungal elements. We gated on all GFP+ cells and determined the proportion of phagocytes that contained yeast cells. M constituted a relatively stable population of infected cells ranging from 20% to 30% at Days 1–7 postinfection (Table 1
). PMN comprised the largest proportion of infected cells, except on Day 1, when DC were the most frequently infected (Table 1)
. DC represented between 30% and 40% of the total infected population on Days 1 and 3 (Table 1)
. The values for PMN increased slightly after Day 3 and declined modestly for DC. Differences between G217B and 186R were insignificant.
|
View this table: [in a new window] |
Table 1. Relative Distribution of GFP+ Phagocytesa
|
or IFN- alter the distribution of yeast cells in phagocytes? and IFN- are independently requisite for host control of infection with H. capsulatum [7
, 14
15
16
17
18
]. We ascertained whether defective immunity induced by neutralization of these cytokines altered the residence of yeast cells. Mice were infected with 2 x 106 yeast cells and given rat IgG or mAb to TNF- or IFN- on the day of infection. Leukocytes from lungs of mice infected for 3 or 7 days were analyzed. Neutralization of TNF- or IFN- did not modify the proportion or number of GFP+ events at Day 3. However, at Day 7, cytokine neutralization was accompanied by a dramatic increase in the proportion and number of infected cells, thus reflecting the increased burden associated with antagonism of TNF- or IFN- (Table 2
). |
View this table: [in a new window] |
Table 2. Percentage and Number of GFP+ Cells in Controls and Those Receiving mAb to TNF- or IFN-a
|
or IFN- resulted in pronounced increases in the number of infected M, PMN, and DC on Day 7, but not Day 3, for both strains (Fig. 4
).
 View larger version (21K): [in a new window] |
Figure 4. Effect of neutralization of TNF- or IFN- on in vivo residence of GFP+ yeast cells. Mice were infected with 2 x 106 yeast cells and given rat IgG, mAb to TNF-, or IFN- i.p. Preliminary data indicated that there were no differences between mice receiving PBS or rat IgG. At Days 3 and 7 postinfection, lung leukocytes were analyzed for the proportion of a given population that contained GFP+ yeast cells. Data represent mean (±SEM) of n = 6–8 mice performed in two or three individual experiments. *, P < 0.05, compared with controls; **, P < 0.01, compared with controls.
|
, this value was 5.6 ± 1.2, and for cells from those given mAb to IFN-, this value was 4.6 ± 0.3. In mice infected with G186R, the values were: controls, 0.7 ± 0.2; mAb to TNF-, 3.4 ± 0.3; and mAb to IFN-, 2.9 ± 0.3.
Does neutralization of TNF- or IFN- alter the distribution of yeast cells in M or DC subsets?
M and DC manifest multiple subpopulations [19
20
21
22
23
24
25
]. M, for example, can be categorized into inflammatory and resident cells based on a number of surface markers including Ly-6C or CD62L [20
, 26
]. Hence, we examined the distribution of GFP+ yeast cells in M and DC subsets in controls and those given mAb to TNF- or IFN-.
We analyzed the data in two ways. First, we ascertained the relative percentage of infected M and DC subpopulations. In controls, only a small fraction of G217B yeast cells was associated with Ly-6C+ cells on Day 3 (Fig. 5A
), but on Day 7, the majority was in this population (Fig. 5C)
. Yeast cells were predominantly in Mac-3+ CD62L+ on Days 3 and 7, although the percentage increased modestly between Days 3 and 7 (Fig. 5A
and 5C)
. A representative FACS profile of these infected cell subpopulations is illustrated in Figure 6
(top and middle panels). G186R yeast cells were found principally in Ly-6C+ and CD62L+ cells on Days 3 and 7 (Fig. 5B
and 5D)
. In uninfected mice (n=4), the mean proportion (±SEM) of Mac-3+ cells that were Ly-6C+ or CD62+ was 28.3 ± 1.8% and 18.4 ± 1.3%, respectively. For G186R and G217B, <30% of Mac-3+ cells coexpressed CD62L and Ly-6C.
 View larger version (30K): [in a new window] |
Figure 5. Effect of neutralization of TNF- or IFN- on in vivo residence of GFP+ yeast cells on M (A–D) and DC (E–H) subsets. Mice were infected and treated as described in Figure 4
. Data were analyzed as percentage of cells that were infected. Data represent mean (±SEM) of n = 5–7. *, P < 0.05, compared with controls; **, P < 0.01, compared with controls.
|
 View larger version (41K): [in a new window] |
Figure 6. Representative dot-plots of phagocytes from mice infected with G217B at Days 3 and 7. Cells were gated on GFP+ cells and analyzed for expression of Mac-3 and Ly-6C (top panels), Mac-3 and CD62L (middle panels), and CD11c and CD205 (bottom panels).
|
DC subsets were analyzed in conjunction with those of M. We determined initially that only a small fraction (
1%) of yeast cells resided in CD11c+CD8+ cells. We focused attention on CD11c+Ly-6C and CD11c+CD205 cells. On Day 3, only a minority of G217B yeasts was detected in CD11c+ Ly-6C+ cells, but by Day 7, most yeast cells were found in this population (Fig. 5E
and 5G)
. On Days 3 and 7, G217B yeast cells were largely observed in CD205+ cells (Fig. 5E
and 5G)
. A representative FACS profile is shown in Figure 6
(bottom panels). By Day 7, 97.3 ± 1.0% of the CD11c+ Ly-6C+ cells coexpressed CD205. For G186R, the yeasts were found chiefly in Ly-6C+ or CD205+ CD11c+ cells (5, F and H). Like G217B, 98.3 ± 0.4% of CD11c+ Ly-6C+ cells coexpressed CD205. In uninfected mice (n=4), the Ly-6C+ population constituted 14.2 ± 1.2% of CD11c+ cells, and the CD205+ population comprised 58.8 ± 3.9% of these cells.
Among the CD11c+CD205+ DC that contained G217B-GFP, 83.5 ± 4.4% were coexpressed CD11b. For G186R, this value was 77.6 ± 2.3%. This population of CD11c+CD205+ cells that harbored yeasts was largely F4/80–. Thus, 70.9 ± 4.1% of G217B-GFP yeast cells were found in CD11c+CD205+F4/80– cells, and 71.7 ± 3.9% of G186R-GFP were present in CD11c+CD205+F4/80– cells.
In studies of DC, the only substantial alteration associated with neutralization of TNF- or IFN- was an increase in the proportion of Ly-6C+ cells at Day 3 of infection with G217B (Fig. 5E)
. The finding that yeast cells of both strains were found nearly exclusively in CD205+ cells was unchanged by administration of mAb to TNF- or IFN- (Fig. 5E
5F
5G
5H)
.
Does a lower inoculum alter the distribution of GFP+ H. capsulatum in M or DC subsets?
The above data established the profile of infected phagocytes in mice given the nonlethal inoculum of 2 x 106 yeast cells. We asked if a lower inoculum would alter that profile. Mice (n=6) were infected with 105 yeast cells i.n. with GFP-G217B, and at Day 7 of infection, we analyzed the proportion of infected M and DC subsets. Yeast cells were found predominantly in Mac-3+Ly-6C+ (75.2±6.9%) or in Mac-3+CD62L+ (85.7±8.1%) cells. Greater than 80% of Mac-3+cells coexpressed Ly-6C and CD62L. For infected DC, >95% of GFP-G217B was present in CD11c+CD205+ cells. GFP-G217B was detected in 67.4 ± 4.9% CD11c+Ly-6C+ cells. These values are similar to those from mice given 2 x 106 yeast cells (see Fig. 5C
and G
).
Absolute numbers of yeast cells in M and DC subpopulations
The aforementioned experiments assessed the proportions of phagocyte subpopulations that harbored yeast cells in control animals and in recipients of anticytokine mAb. In a separate set of studies, we analyzed the absolute number of subpopulations that harbored GFP+ yeast cells. In contrast to the proportions, neutralization of TNF- or IFN- caused a significant increase in phagocyte subpopulations that harbored these cells (Fig. 7
).
 View larger version (21K): [in a new window] |
Figure 7. Effect of neutralization of TNF- or IFN- on in vivo residence of GFP+ yeast cells in M (A–D) and DC (E–H) subsets. Mice were infected and treated as described in Figure 4
. Absolute numbers of infected cells were calculated. Data represent mean (±SEM) of n = 5–7 in two separate experiments. *, P < 0.05, compared with controls; **, P < 0.01, compared with controls.
|
or IFN- a result of a high fungal burden? or IFN- is that the fungal burden is higher. We infected mice with a large inoculum (1.5x107 G217B or G186R yeast cells) and at Day 7 of infection, assessed the phenotype of infected cells. The vast majority of yeast cells was found in Mac-3+Ly-6C+ or Mac-3+CD62L– cells. This finding differed from that of mice given mAb to cytokines. On the other hand, the results for DC subsets were similar to that of mice given mAb to cytokines (Table 3
). Thus, the data with mice given mAb to TNF- or IFN- cannot be explained simply by burden. |
View this table: [in a new window] |
Table 3. Distribution of M and DC Subsets in Mice Infected with a High Inoculum of H. capsulatuma
|
or IFN- antagonism was associated with a sharp decrease in the number of cells that expressed CD80, CD86, and I-Ab for G217B and G186R.
 View larger version (20K): [in a new window] |
Figure 8. Antagonism of TNF- or IFN- alters the number of infected DC expressing CD80, CD86, or class II MHC (I-Ab). Mice were infected with 2 x 106 yeast cells i.n. and given rat IgG, mAb to TNF-, or IFN-. Cells were stained with CD11c and gated on the GFP+ population. Data represent mean (±SEM) of n = 5–7 performed in two separate experiments. *, P < 0.05; **, P < 0.01.
|
DISCUSSION
GFP+ H. capsulatum has been used to examine gene expression during the conversion from mycelial to yeast phase and to assess RNA inhibition [28
, 29
]. Our studies have extended the use of GFP-expressing H. capsulatum by examining the in vivo residence of two genetically distinct strains of this pathogenic fungus in phagocytic populations. The use of GFP-containing fungus provides a more precise analysis of the complexities of intracellular invasion, as a distinction can be made between infected and uninfected phagocytes. The results demonstrated that H. capsulatum was present in diverse phagocytic populations and subpopulations in vivo and that the profile of residence was perturbed by antagonism of TNF- or IFN-, two cytokines that are critically important in optimal host defenses [7
, 11
, 14
15
16
17
18
]. The presence of yeast cells within particular phagocyte populations or subpopulations was dynamic. We detected a shift over time in the proportion of particular phagocyte populations that contained yeasts in immunocompetent mice. Likewise, yeast cells of Cryptococcus neoformans manifest a shift to residence to DC [30
]. An unanticipated observation was that the yeast cells were found largely in a CD8–CD11c+Ly-6-C+CD11b+F4/80–CD205+ subpopulation.
In several experiments, data were presented as the relative proportion that was infected and the absolute numbers. Both of these calculations provide important information. For example, analysis of proportions did lead us to find that yeast cells were almost exclusively in a CD205+ DC. This would not be so evident if one just examines absolute numbers. On the other hand, evaluation of absolute numbers does account for any differences in cell numbers that may arise from altered inflammatory responses.
We restricted our studies to the first 7 days of infection to concentrate on the innate system and avoid the influence of a robust, adaptive immune response. Activation of adaptive immunity generally develops after 7 days of nonlethal pulmonary infection with H. capsulatum [31
, 32
]. These studies were also limited to 7 days, as neutralization of TNF- and IFN- leads to the demise of animals between Days 8 and 14 of infection [7
, 14
]; thus, comparisons between cytokine-deficient and cytokine-sufficient mice could only take place during the first week to avoid analysis of moribund animals.
H. capsulatum strains G217B and G186R are commonly used for in vitro and in vivo experimentation. Both strains have been sequenced (www.genome.wustl.edu/genome.cgi?Genome=Histoplasma%20capsulatum), and the data suggest differences in their genomic architecture. Particular virulence traits such as -(1,3) glucan are known to be present in G186R but are not identified in G217B [29
, 33
]. Another distinguishing characteristic is that the former induces a higher burden of infection than the latter [34
]. Hence, we opted to investigate both strains to determine if the genetic unrelatedness was associated with markedly dissimilar inflammatory profiles. Despite differences in genetic composition, the character of the inflammatory response to each was similar, although not identical. G217B induced a more vigorous neutrophilic response. This finding is possibly attributable to the increased fungal burden associated with this strain [34
]. We also demonstrated that the relative proportions of the phagocyte populations and subpopulations that harbored the yeast cells were quite similar, although there were differences when absolute numbers were calculated. For example, G186R was associated with inflammatory M and DC subpopulations earlier, and this finding may, in part, explain why it is less virulent in mice.
Limitations to our approach should be noted. Small numbers of yeast cells may be difficult to discern from background, especially in lungs, as phagocytes from this organ exhibit substantial autofluoresence [35
]. Another constraint is the distinction of cell populations. Unlike CD4+ or CD8+ cells, definitive markers for phagocyte populations are more variable. As monocytes convert into DC, given the proper stimuli [36
, 37
], it is possible that during the infectious process, a fraction of the cells is in transition rather than fixed in lineage. Ly-6C expression on DC marks an inflammatory population of these cells [38
]. G186R yeast cells were largely associated with Ly-6C+ DC on Days 3 and 7, and proportionally, this value was unchanged by treatment with mAb to TNF- or IFN-. On the other hand, only a small proportion of yeast cells was found in this DC subpopulation, but the values were increased by administration with mAb to TNF- on Day 3 only and mAb to IFN- on Days 3 and 7. In terms of absolute numbers, treatment with anticytokine mAb increased the presence of yeast cells in both populations. Although DC are reported to kill this fungus [39
], the influence of specific subpopulations is not known. Moreover, the effect of cytokine neutralization on the function of these cells remains to be determined.
One of the most striking findings was the presence of yeast cells from either strain in CD8–CD11c+CD11b+F4/80–CD205+ cells. Residence in these cells was largely unperturbed by neutralization of TNF- or IFN- when the data were analyzed as proportions. However, the number of CD205+ DC containing yeast cells was markedly increased on Days 3 and 7 of infection. CD205 is a member of the mannose receptor family and can act as an antigen uptake receptor on DC, and in mice, it is largely, although not exclusively, found on DC [40
, 41
]. Although DC are a major mediator of antigen presentation, it is unlikely that the CD8–CD11c+CD205+CD11b+ population contributes much to this biological process. CD8+CD11c+ cells are capable of vigorous antigen presentation, especially class I MHC antigen, and ingestion, but their phagocytic activity appears to be limited to inert particles and dying cells [42
, 43
]. More recently, a CD11c+CD11b+ lung DC population has been reported to be an antigen-bearing, migratory population [44
]. Whether our population subserves the same function requires additional testing. Mycobacterium tuberculosis also is largely located within CD205+ lung DC, as demonstrated by immunohistochemistry, thus suggesting that this population is important in handling intracellular pathogens [45
].
The data do not discriminate if CD205 is induced upon infection or if a pre-existing CD205+ population is the target. The finding of a high proportion, from 60% to 99%, of infected CD205+ cells as early as Day 3 would suggest that H. capsulatum preferentially infects this population but does not absolutely exclude induction of this surface receptor. The contribution of CD205 in the uptake and fate of H. capsulatum is unknown. If the fungus preferentially infects this population, this surface antigen may be crucial for ingestion.
We also analyzed the expression of costimulatory molecules on the surface of DC as well as the class II MHC molecule I-Ab. Marked increases in the number of DC expressing each of these molecules were noted between Days 3 and 7. This finding correlates with the evolution of the T cell-mediated response to this fungus and supports the importance of these molecules in the generation of a Histoplasma-specific adaptive response [46
]. Administration of mAb to TNF- or IFN- sharply reduced the numbers of these cells. The failure to control immunity in mice deficient in either cytokine may be in part explained by the perturbation in costimulatory molecule and MHC antigen expression on DC.
Neutralization of endogenous TNF- and IFN- substantially increased the percentage of phagocytes harboring H. capsulatum. We found the increase in infected cells to be generalized rather than skewed to a particular phagocyte population. Moreover, neutralization of TNF- or IFN- did not blunt the absolute number of infected inflammatory M. Thus, inflammatory M from mice given mAb to TNF- or IFN- may be just as permissive for H. capsulatum growth as resting M, as the outcome of infection is death. If Ly-6C+ or CD62L+ M possess enhanced antifungal activity, the absence of TNF- or IFN- must blunt it.
The increase in infectivity of M enhances the likelihood of a poorly resolving infection, as these cells are permissive for intracellular growth unless activated by CSFs or IFN- [16
, 47
]. A compounding factor is that TNF- antagonism inhibits apoptosis of leukocytes from mice infected with H. capsulatum and therefore, may prolong the lifespan of infected M [27
]. Neutralization of either cytokine induced a significant increase in the percentage of infected Ly-6C+ M and DC in mice infected with G217B on Day 3. The effect was different for G186R. The shift was reversed completely by Day 7. These findings indicate that infected cells from cytokine-neutralized mice manifest a common perturbation. Despite the differences in the signaling cascades associated with these cytokines, the net result of their neutralization is a similar alteration in the profile of infected cells.
The presence of CD62L or Ly-6C on murine M denotes a population of inflammatory phagocytes [26
]. In general, yeast cells were found largely but not exclusively in inflammatory M; the exception is with strain G217B at Day 3 of infection. Although residence of the strains was noted in inflammatory M, Ly-6C and CD62L may denote distinct populations, as the changes in infectivity that occurred with time and with treatment with mAb to TNF- or IFN- were not identical. Thus, distinct subsets of inflammatory M most likely exist, and they may have different biological functions.
The use of GFP+ yeast cells has also uncovered the fact that apoptosis of phagocytes, which is a feature of infection with this fungus [17
, 27
, 48
], almost exclusively occurs in uninfected cells, at least as early as Day 3. As apoptosis of DC is pivotal in promoting cross-presentation to Histoplasma antigens to CD8+ cells [48
], it may only take a small number of apoptotic DC to drive this process. Bystander apoptosis of M has been demonstrated to be a feature of infection with M. tuberculosis [49
]. The loss of these phagocytes by a bystander effect may serve to eliminate a site of residence for bacterial or fungal invasion.
In summary, GFP has been engineered into H. capsulatum, and this approach has enhanced the ability to elucidate the inflammatory response in vivo. The analysis reveals that the process of infection of phagocytes is dynamic. Exploiting the knowledge gained from investigating intracellular residence will provide a more thorough understanding of the immunopathogenesis of this fungus.
ACKNOWLEDGEMENTS
This work was supported by Grants AI-703337 and AI-62918 from the National Institutes of Health and by a Merit Review from the Veteran’s Affairs.
Received March 3, 2008; revised May 7, 2008; accepted June 4, 2008.
REFERENCES
exacerbates primary and secondary pulmonary histoplasmosis by differential mechanisms J. Immunol. 160,6072-6082 plays a critical role in maintaining secondary immunity in the absence of IFN- J. Immunol. 160,1359-1368 interferon knockout mice Infect. Immun. 65,2564-2569 plays a role in host defense against Histoplasma capsulatum J. Infect. Dis. 162,1349-1353[Medline] interferon Infect. Immun. 55,1014-1016 is important in determining the outcome of experimental histoplasmosis in mice Infect. Immun. 60,4230-4238 J. Immunol. 155,785-795[Abstract]-(1,3)-glucan in virulence Mol. Microbiol. 53,153-165[CrossRef][Medline]-(1,3)-glucan blocks innate immune recognition by the β-glucan receptor Proc. Natl. Acad. Sci. USA 104,1366-1370 switches monocyte differentiation from dendritic cells to macrophages Blood 101,143-150This article has been cited by other articles:
 |
|
 |
|
  A. J. Guimaraes, S. Frases, F. J. Gomez, R. M. Zancope-Oliveira, and J. D. Nosanchuk Monoclonal Antibodies to Heat Shock Protein 60 Alter the Pathogenesis of Histoplasma capsulatum Infect. Immun., April 1, 2009; 77(4): 1357 - 1367. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
