Effects of Anaplasma phagocytophilum on Host Cell Ferritin mRNA and Protein Levels

Cultivation of A. phagocytophilum and maintenance of infection. A. phagocytophilum strains NCH-1 and HZ were cultivated in HL-60 cells as previously described (14). Strain HZ was kindly provided by Ralph Horowitz of New York Medical College (Valhalla) and Yasuko Rikihisa of Ohio State University (Columbus). A. phagocytophilum infection was initiated in vivo by inoculating inbred C3H-scid mice with A. phagocytophilum-infected HL-60 cells. Infection was maintained in C3H-scid mice by inoculating naïve mice with 100 μl of A. phagocytophilum-infected C3H-scid mouse blood every 4 to 6 weeks.

Isolation of human neutrophils.Human neutrophils were isolated from peripheral blood from healthy donors by centrifugation through an equal volume of Polymorphprep (Axis-Shield; Greiner Bio-One, Frickenhausen, Germany) at 470 × g for 30 min. The resulting neutrophil band was removed by aspiration and mixed with equal volumes of 0.45% NaCl and Iscove's modified Dulbecco's medium (IMDM; Invitrogen, Carlsbad, CA)-0.5 mM EDTA to restore isotonic conditions. The cells were centrifuged at 210 × g for 10 min, and the supernatant was removed. The resulting pellet was resuspended and incubated in 10 ml of red blood cell lysing buffer (Sigma, St. Louis, Mo.) for 10 min followed by two successive washes in IMDM-0.5 mM EDTA and enumeration using a hemacytometer.

In vitro A. phagocytophilum infection.Host-cell-free A. phagocytophilum cells were recovered from infected HL-60 cells and added to 5 × 10⁶ HL-60 cells (0.5 × 10⁵/ml) in IMDM-10% fetal bovine serum. The ratios of A. phagocytophilum organisms per HL-60 cell were determined as described previously (12). For in vitro infection of human neutrophils, 5 × 10⁵ neutrophils (1 × 10⁶/ml) in IMDM were added to individual wells of either 6- or 24-well Ultra Low Attachment plates (Corning, Inc.; Corning, NY). To these were added suspensions of host-cell-free A. phagocytophilum. In some cases, zymosan A particles from Saccharomyces cerevisiae (Sigma) that had been opsonized with autologous serum (OpZ) or Escherichia coli were added to neutrophils as previously described (12). At 1, 3, and 6 h postinfection, either total RNA or protein was isolated as described below.

RNA isolation and reverse transcription-PCR (RT-PCR).Total RNA from uninfected or A. phagocytophilum-infected cells was isolated using Trizol (Invitrogen) according to the manufacturer's instructions followed by treatment with RNase-free DNase (1 U/500 μg; Roche Molecular Biochemicals, Indianapolis, IN) for 2 h at 37°C. DNA-free RNA was reextracted using Trizol, ethanol precipitated, and resuspended in diethyl pyrocarbonate-treated water. cDNA stocks (50 μl) were prepared from 5 μg of total RNA using random hexamers and the ProSTAR First Strand RT-PCR kit (Stratagene, La Jolla, CA), followed by PCR amplification using 1 μl of cDNA template. The thermal cycling conditions used were 94°C for 2 min followed by 35 cycles of 94°C for 60 s, 55°C for 60 s, 72°C for 60 s, and a final extension step of 72°C for 7 min. The reaction mixture contained final concentrations of 1.5 mM MgCl₂, 200 μM deoxynucleoside triphosphates, 0.8 μM of each primer, and 2.5 U of Taq polymerase (Roche Molecular Biochemicals). To ensure that RNA templates were free from contaminating DNA, identical reactions were performed in the absence of reverse transcriptase (data not shown). The primers used are listed in Table 1.

Quantitative RT-PCR.cDNAs generated from uninfected and A. phagocytophilum-infected cells were used as templates for quantitative PCR using the MyiQ Real Time Detection System (Bio-Rad, Richmond, CA). Each cDNA was in a reaction mixture that contained final concentrations of 1.5 U platinum Taq DNA polymerase (Life Technologies, Gaithersburg, MD), 20 mM Tris-HCl (pH 8.4), 50 mM KCl, 3 mM MgCl₂, 200 μM deoxynucleoside triphosphates, 500 nM of each primer, and 300 nM TaqMan probe targeting either fhc or flc. The sequences of the TaqMan probes are listed in Table 1. The probes were labeled at their 5′ ends with 6-carboxyfluorescein and at their 3′ ends with carboxytetramethylrhodamine. Thermal cycling conditions, fluorescence detection, and quantification were performed as previously described (14), except that starting quantities were extrapolated in copy numbers instead of nanograms. β-actin gene expression was determined using the TaqMan β-actin Control kit (Perkin Elmer, Foster City, CA) according to the manufacturer's instructions.

Immunoblot analyses.Whole-cell lysates of uninfected or A. phagocytophilum-infected cells were prepared, quantified, fractionated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and transferred to Immobilon-P membranes (Millipore, Bedford, MA) as described previously (14). Human actin and A. phagocytophilum P44 were detected using mouse anti-human actin MAb (Sigma) and rabbit polyclonal anti-A. phagocytophilum serum, respectively, followed by goat anti-rabbit immunoglobulin G-Alexa Fluor 680 (Molecular Probes, Eugene, OR) and goat anti-mouse immunoglobulin G-IRDYE800CW (Rockland Immunochemicals, Gilbertsville, PA), respectively. Bound antibodies were detected using the Odyssey Infrared Imager (LI-COR Biosciences, Lincoln, NE).

Ferritin enzyme-linked immunosorbent assay (ELISA).Whole-cell lysates were examined for ferritin content using the Human Ferritin ELISA kit (Alpha Diagnostic, San Antonio, TX) according to the manufacturer's instructions. Briefly, 30 μg of whole-cell lysate, 30 μg bovine serum albumin (BSA), or 10 μl of recombinant ferritin standards, ranging from 0 to 800 ng/ml, were added in duplicate to wells of a 96-well plate coated with anti-ferritin MAb. One-hundred microliters of horseradish peroxidase-conjugated anti-ferritin MAb were added, followed by a 30-min incubation at room temperature. The plate was washed three times with 300 μl deionized water, followed by the addition of 200 μl detection substrate per well. After 10 min, 50 μl stop solution was added and the absorbance at 450 nm was recorded using an AD 340 Absorbance Detector (Beckman Coulter, Fullerton, CA).

Assessment of fhc and flc mRNA expression during murine A. phagocytophilum infection.Infected blood from C3H-scid mice was used to initiate infection in immunocompetent C3H/HeN mice (3 per group) as previously reported (13). Nine percent of peripheral blood neutrophils in the inoculum had detectable morulae. Control groups were injected with uninfected C3H-scid mouse blood. On days 2 and 8, the respective groups were sacrificed and assessed for bacterial burden by determining the percentage of morulae-positive neutrophils using light microscopy and by PCR. Neutrophils were isolated from pooled splenocytes via positive selection using fluorescein isothiocyanate-conjugated rat anti-mouse Ly6G (Gr-1; a neutrophil-specific marker) MAb (BD Pharmingen, San Diego, CA) followed by anti-fluorescein isothiocyanate MAb conjugated to magnetic beads (Miltenyi Biotec, Auburn, CA). Total RNA was isolated from Gr-1-positive and -negative cells as described above and used as template for RT-PCR.

Statistical analyses.Statistical analyses were performed using the Prism 4.0 software package (Graphpad, San Diego, CA). If one-way analysis of variance (ANOVA) or two-way ANOVA indicated a group difference (P < 0.05), then Tukey's or Bonferroni's test was used, respectively, to test for a significant difference among groups. In some instances, the means for uninfected and infected samples per time point were examined using the Student t test.

Study approval.The Human Investigation Committees at Yale University and the University of Kentucky and the Yale University Institutional Animal Care and Use Committee approved these studies.

HL-60 cell fhc mRNA levels increase after a minimum of 24 h of A. phagocytophilum infection.We have previously reported that HL-60 and rHL-60 cells with established A. phagocytophilum infections exhibit increased fhc mRNA levels (14). However, neither the kinetics of fhc transcription immediately following A. phagocytophilum infection nor the effects of the bacterium on flc expression are known. We therefore added host-cell-free A. phagocytophilum (∼10 organisms per cell) to HL-60 cells, isolated total RNA, and performed quantitative RT-PCR (QRT-PCR) to assess changes in fhc and flc expression over 96 h of infection. For all samples, flc mRNA levels exceeded those of fhc by as much as 57.0- to 297.1-fold (Fig. 1). For the initial 12 h of infection, fhc mRNA levels of A. phagocytophilum-infected cells were 1.2- to 1.7-fold lower than those of uninfected control cells. Consistent with our previous study (14), however, fhc expression by A. phagocytophilum-infected cells increased significantly above those of uninfected controls by 24 h postinfection (hpi) and remained so throughout the duration of the time course. The largest rise in fhc message, a 2.6-fold increase, was observed at 96 hpi. With the exception of 30 min postinfection, little to no change was observed in flc transcription between infected and uninfected control HL-60 cells. These data demonstrate that the increased fhc mRNA levels associated with A. phagocytophilum infection of HL-60 cells are not observable until at least 24 hpi.

• Open in new tab
• Download powerpoint

FIG. 1.

Ferritin mRNA expression in A. phagocytophilum-infected HL-60 cells. Host-cell-free A. phagocytophilum organisms were added to HL-60 cells, and infection was allowed to proceed for 0, 0.5, 1, 3, 6, 12, 24, 48, or 96 h. Uninfected HL-60 cells served as controls. At the appropriate time postinfection, total RNA was isolated from 5 × 10⁶ infected HL-60 cells and used as template for cDNA synthesis. Quantitative PCR was performed using the cDNA templates to assess the relative expression levels of fhc and flc. Data are presented as the mean copies of either fhc or flc transcript per 10³ β-actin transcript copies. Samples were analyzed in triplicate. The error bars indicate standard deviations. The mean results for uninfected and infected cells were compared per time point using the Student t test. Values are statistically significant (*, P < 0.05; **, P < 0.01; and ***, P < 0.001). The results are representative of three independent experiments.

Ferritin protein levels in HL-60 cells decrease in a dose- and time-dependent manner following A. phagocytophilum infection.Though transcriptional expression does play a role in modulating ferritin levels within eukaryotic cells, ferritin homeostasis is primarily regulated at the level of protein translation (23, 24). We therefore investigated changes in ferritin content in HL-60 cells following A. phagocytophilum infection. Host-cell-free A. phagocytophilum organisms were added to HL-60 cells at approximate ratios of 0, 0.4, 1.1, 3.3, and 10 bacteria per cell. Four days postinfection, whole-cell lysates were extracted and analyzed. Samples exhibited no difference in actin levels upon immunoblot analysis, thereby demonstrating they were effectively normalized (Fig. 2A). As expected, a dose-dependent increase in A. phagocytophilum P44 was observed. When examined by ELISA, ferritin protein levels decreased in a dose-dependent manner in response to A. phagocytophilum burden (Fig. 2B). HL-60 cells that were inoculated with 10 organisms per cell demonstrated a threefold decline in ferritin levels at 96 hpi. This decline was maximal, as cultures that were inoculated with 30 bacteria per cell exhibited nearly identical decreases (data not shown). To assess the effect of A. phagocytophilum on HL-60 cell ferritin protein expression at earlier time points, host-cell-free bacteria were added at approximate ratios of 0, 1, and 10 per cell, and the time course was repeated. At 24, 48, and 96 hpi, whole-cell lysates were assessed for ferritin content. While ferritin levels increased per respective samples over the duration of the time course, A. phagocytophilum-infected cells exhibited a dose-dependent decrease in ferritin relative to uninfected controls (Fig. 2C). Significant decreases in ferritin levels were apparent for cells inoculated with 10 organisms per cell as early as 24 hpi, which demonstrated a 1.7-fold reduction. A significant decrease in ferritin expression for cells inoculated with 1 bacterium per cell was not evident until 96 hpi. The reduction in ferritin levels in infected HL-60 cells relative to uninfected controls is apparent as early as 30 min postinfection and continues throughout the initial hours of A. phagocytophilum infection (Fig. 2D). However, the differences in ferritin levels between infected and uninfected controls are not statistically significant until 24 to 48 hpi. The observed changes in ferritin levels are not in response to fluctuations in whole-cell iron concentrations, which remained at levels that were too low to be detected by the ferrozine binding assay (43) throughout the time courses (data not shown). Thus, these results demonstrate that A. phagocytophilum infection of HL-60 cells results in a bacterial dose- and time-dependent reduction in host cell ferritin levels.

• Open in new tab
• Download powerpoint

FIG.2.

Ferritin protein levels in HL-60 cells decrease in a time- and dose-dependent manner following A. phagocytophilum infection. Host-cell-free A. phagocytophilum (Ap) organisms were added to 5 × 10⁵ HL-60 cells at ratios of 0.4, 1.1, 3.3, and 10 per cell. Uninfected HL-60 cells served as controls. Four days postinfection, (A) whole-cell lysates (10 μg) were fractionated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis followed by Western immunoblot analysis using antibodies directed against A. phagocytophilum P44 and actin, and (B) whole-cell lysates (30 μg) were assayed for ferritin content by ELISA. Bovine serum albumin (BSA; 30 μg) served as a negative control. Statistical significance was determined using one-way ANOVA followed by Tukey's test. (C) Dose-dependent infection time course. A. phagocytophilum was added to 5 × 10⁵ HL-60 cells at approximate ratios of 0 (white bars), 1 (gray bars), and 10 (black bars) per cell. Whole-cell lysates (30 μg) were assessed for ferritin content at 24, 48, and 96 h postinfection. Statistical significance was determined using two-way ANOVA followed by Bonferroni's test. The mean values indicated by different letters are significantly different. (D) Host-cell-free A. phagocytophilum organisms were added to HL-60 cells at a ratio of approximately 10 organisms per cell, and infection was allowed to proceed for 0, 0.25, 0.5, 1, 2, 5.5, 8, 12, 24, 48, 72, or 96 h. Uninfected HL-60 cells served as controls. Whole-cell lysates (30 μg) generated from 5 × 10⁶ cells per time point were assayed for ferritin content. The means ± standard errors of duplicate samples are presented. Statistical significance (*, P < 0.05; **, P < 0.01) was determined using two-way ANOVA followed by Bonferroni's test. Results are representative of four to six independent experiments. Ctrl, control.

Human neutrophil flc levels rapidly increase in response to A. phagocytophilum infection in vitro.The effects of A. phagocytophilum on ferritin transcription in neutrophils are unknown. Furthermore, while the promyelocytic HL-60 cell line is permissive to A. phagocytophilum, it does not accurately mimic the natural mammalian host cell for the bacterium. For instance, HL-60 cells are devoid of specific granules and are unable to generate a robust respiratory burst compared to neutrophils. We and others have previously demonstrated that following binding to the neutrophil surface in vitro, A. phagocytophilum internalization is a prolonged process that lasts for several hours (8, 12, 25). This prolonged association of the bacterium with the neutrophil outer membrane stimulates at least some degree of NADPH oxidase assembly and degranulation (12). Because of ferritin's role as a cytoprotectant against oxidative damage and to better understand the neutrophil expression profile in response to intracellular infection, we investigated changes in fhc and flc mRNA levels in neutrophils during the early hours of A. phagocytophilum infection. Human neutrophils were incubated in the presence and absence of host-cell-free A. phagocytophilum at an approximate ratio of 10 organisms per cell. The kinetics of infection paralleled previous observations (12) and achieved a maximum of 90% infected neutrophils by 6 h postaddition of bacteria (data not shown). At 1, 3, and 6 hpi, total RNA was extracted and subjected to RT-PCR. Samples were normalized to β-actin expression. In contrast to that observed during the initial hours of infection in HL-60 cells, neutrophils incubated with A. phagocytophilum demonstrated successive rises in not only fhc transcription but also flc transcription throughout the course of infection (Fig. 3A).

• Open in new tab
• Download powerpoint

FIG. 3.

Ferritin mRNA expression in A. phagocytophilum-infected neutrophils. (A) RT-PCR analyses. Neutrophils (5 × 10⁵) were incubated in the presence and absence of host-cell-free A. phagocytophilum (Ap; 10 per cell) for 1, 3, or 6 h. At the appropriate time point, total RNA was isolated and RT-PCR was performed targeting human fhc and flc as well as β-actin and A. phagocytophilum 16S rRNA genes and p44 transcripts. Results are representative of three experiments. (B) Quantitative RT-PCR analyses. Neutrophils (5 × 10⁵) were incubated with host-cell-free A. phagocytophilum (10 per cell), E. coli (Ec; 10 per cell), or OpZ (10 per cell) for 1, 3, or 6 h. Total RNA was isolated and converted to cDNA, which was used as template for real-time PCR to quantify the relative expression levels of fhc and flc. Samples were analyzed in triplicate. Data are presented as the mean copies of either fhc or flc transcript per 10³ β-actin transcript copies. Error bars indicate standard deviations. Statistical significance was determined using one-way ANOVA followed by Tukey's test. The mean values indicated by different letters are significantly different. Results are representative of four independent experiments. ctrl, control.

To further define the relevance of the observed changes in ferritin transcription associated with A. phagocytophilum infection, the time course was repeated with neutrophils being incubated with host-cell-free A. phagocytophilum, Escherichia coli, or OpZ. The latter two stimuli served as controls for activating neutrophil NADPH oxidase and degranulation. Changes in fhc and flc expression were quantified using QRT-PCR. Similar to that observed for HL-60 cells, flc levels exceeded fhc levels by as much as 20.6-fold (Fig. 3B). Consistent with the semiquantitative RT-PCR studies, A. phagocytophilum infection stimulated a twofold, statistically insignificant rise in fhc expression. Neutrophils incubated with E. coli, however, demonstrated as much as a 12.8-fold increase in fhc transcript levels. Incubation with OpZ yielded as much as a 3.2-fold rise in fhc message, though this change was not significant. In contrast to that observed for neutrophil fhc mRNA levels, flc transcription significantly increased following incubation with A. phagocytophilum, with increases of 16.2-, 20.0-, and 62.9-fold relative to uninfected controls at 1, 3, and 6 hpi, respectively. Ferritin light-chain expression also significantly increased in response to E. coli and OpZ. Thus, ferritin expression, particularly that of flc, rapidly increases during the initial hours of A. phagocytophilum infection and upon exposure to phagocytic stimuli such as E. coli and OpZ.

Neutrophil ferritin protein levels increase during the initial hours of A. phagocytophilum infection in vitro.Neutrophil fhc transcription exhibits a low level of increase, while flc mRNA expression increases dramatically during the initial hours of A. phagocytophilum infection. Similar transcriptional profiles were also observed upon incubating neutrophils with E. coli and OpZ. We therefore examined whether the observed increases in ferritin transcription in neutrophils translates to rises in ferritin protein levels following incubation with each of these stimuli. Host-cell-free A. phagocytophilum, E. coli, or OpZ was added to neutrophils, after which whole-cell lysates were examined for ferritin content at 1, 3, and 6 hpi. At 1 hpi, neutrophil ferritin levels exhibited a three- to fourfold increase for all stimuli relative to uninfected controls (Fig. 4). Ferritin levels of A. phagocytophilum-infected cells remained high throughout the time course, while those of cells exposed to E. coli had declined to levels that were not significantly different from those of uninfected controls by 3 hpi. Ferritin expression by neutrophils exposed to OpZ had also dropped considerably at 3 and 6 hpi. Thus, exposure of neutrophils to A. phagocytophilum results in a sustained rise in ferritin levels, while exposure to either E. coli or OpZ stimulates an initially pronounced spike followed by a subsidence in ferritin amounts. The kinetics of ferritin expression induced by A. phagocytophilum, E. coli, and OpZ parallel the effects each of these stimuli has on NADPH oxidase assembly and degranulation in vitro (14). These data are consistent with the induction of ferritin expression as a cytoprotective response to ROS.

• Open in new tab
• Download powerpoint

FIG. 4.

Ferritin protein levels in neutrophils following A. phagocytophilum infection. Neutrophils (5 × 10⁵) were incubated with A. phagocytophilum (Ap; 20 per cell), E. coli (20 per cell), or OpZ (20 per cell) for 1, 3, or 6 h. Uninfected cells served as controls. Whole-cell lysates (30 μg) were analyzed for ferritin content. The means ± standard errors of duplicate samples are presented. Statistical significance was determined using two-way ANOVA followed by Bonferroni's test. The mean values indicated by different letters are significantly different. Results are representative of four independent experiments.

Changes in fhc and flc expression during murine A. phagocytophilum infection.To extend our analyses in vivo, C3H/HeJ mice were inoculated with A. phagocytophilum. Sham-inoculated mice served as controls. On days 2 and 8, the respective groups were sacrificed and assessed for bacterial burden by determining the percentage of morulae-positive neutrophils using light microscopy and PCR. On days 2 and 8, the percentages of neutrophils with detectable morulae were 1.5% ± 1.1% and 7.9% ± 2.6%, respectively (data not shown). Neutrophils were isolated from pooled splenocytes via positive selection. Total RNA was isolated from Gr-1-positive (neutrophils) and Gr-1-negative (nonneutrophil splenocytes) cells and used as template for RT-PCR. No differences in fhc or flc mRNA levels were apparent in Gr-1-negative cells from infected and uninfected mice on either day (Fig. 5). Consistent with observations for uninfected HL-60 cells and human neutrophils, murine neutrophils (GR-1-positive splenocytes) exhibited strong flc expression. By days 2 and 8 postinfection, however, flc expression had dropped considerably for Gr-1-positive cells. There was no detectable difference in fhc expression between neutrophils from infected and uninfected mice on day 2. Notably, though, fhc transcription was strongly up-regulated in neutrophils from infected mice on day 8, which typically corresponds to peak infection. These profiles are highly consistent with what would be expected to occur in the context of a strong IFN-γ response (47), which is a common manifestation of A. phagocytophilum infection (2, 20, 33). Thus, in vivo A. phagocytophilum infection results in a pronounced increase in neutrophil fhc expression, a pattern that is consistent with the IFN-γ-dominated cytokine profile associated with infections by obligate intracellular pathogens.

• Open in new tab
• Download powerpoint

FIG. 5.

Ferritin mRNA expression in neutrophils recovered from A. phagocytophilum-infected mice. C3H/HeJ mice were inoculated with A. phagocytophilum (Ap). Sham-inoculated mice served as controls. On days 2 and 8, the mice were sacrificed and Gr-1-positive and -negative cells were isolated. Total RNA was extracted and used as template for RT-PCR analyses targeting murine fhc and flc and A. phagocytophilum p44. Samples were normalized according to murine hypoxanthine phosphoribosyltransferase (HPRT) transcript levels. Transcript levels for fhc, flc, and the HPRT gene for Gr-1-negative cells on day 2 are similar to those observed for day 8 and thus are not shown. Results are representative for four separate experiments. ctrl, control.