The authors have declared that no competing interests exist.
Conceived and designed the experiments: TP CL. Performed the experiments: TP KK PP. Analyzed the data: TP KK PP. Contributed reagents/materials/analysis tools: CL YS SC. Wrote the paper: TP CL PP.
Our earlier genome-wide expression study revealed up-regulation of a tryptophan-catabolizing enzyme, indoleamine 2,3-dioxygenase (IDO1), in patients with scrub typhus. This gene has been previously reported to have anti-microbial activity in a variety of infectious diseases; therefore, we aimed to prove whether it is also involved in host defense against
Using LC-MS, we observed an increased ratio of serum L-kynurenine to serum L-tryptophan in patients with scrub typhus, which suggests an active catalytic function of this enzyme upon the illness. To evaluate the effect of IDO1 activation on OT infection, a human macrophage-like cell line THP-1 was used as a study model. Although transcription of IDO1 was induced by OT infection, its functional activity was not significantly enhanced unless the cells were pretreated with IFN-γ, a potent inducer of IDO1. When the degree of infection was evaluated by quantitative real-time PCR, the relative number of OT 47 kDa gene per host genes, or infection index, was markedly reduced by IFN-γ treatment as compared to the untreated cultures at five days post-infection. Inhibition of IDO1 activity in IFN-γ treated cultures by 1-methyl-L-tryptophan, a competitive inhibitor of IDO1, resulted in partial restoration of infection index; while excessive supplementation of L-tryptophan in IFN-γ treated cultures raised the index to an even higher level than that of the untreated ones. Altogether, these data implied that IDO1 was partly involved in restriction of OT growth caused by IFN-γ through deprivation of tryptophan.
Activation of IDO1 appeared to be a defensive mechanism downstream of IFN-γ that limited intracellular expansion of OT via tryptophan depletion. Our work provided not only the first link of in vivo activation of IDO1 and IFN-γ-mediated protection against OT infection but also highlighted the promise of this multifaceted gene in scrub typhus research.
Scrub typhus is a potentially life-threatening infectious disease that is a major cause of acute undifferentiated fever in Asia-Pacific region. It is caused by
Scrub typhus is a potentially life-threatening infectious disease caused by
Earlier studies have shown that IFN-γ is essential in protection against OT infection in animal and cell-based models
In addition, IDO1 also exerts an indirect antimicrobial activity via production of some certain downstream catabolites of kynurenine pathway
Despite its protective role in a variety of infections, IDO1 activation paradoxically appears to be involved in suppression of the immune responses. This was first revealed by a key study showing that IDO1-mediated tryptophan degradation prevents allogeneic fetal rejection in mice
In rickettsial infection, it was reported that IFN-γ-mediated IDO1 activation inhibited the growth of
We hypothesized that up-regulation of IDO1 upon infection with scrub typhus could lead to rescriction of OT growth in host cells according to the fact that OT lacks an enzyme to generate tryptophan, which is an essential material for its expansion
In the present study, we reported that IDO1 acitivity was increased in patients with scrub typhus. Activation of IDO1 by IFN- γ resulted in a lower number of OT load in THP-1 macrophages, and the degree of infection could be partly restored by an inhibitor of IDO1 enzyme. Supplementation with high-dose tryptophan did not only reverse the suppression of OT growth but markedly increased the number of OT in IDO1 active cultures. Our work not only provided the first link of in vivo activation of IDO1 and IFN-γ-mediated protection against OT infection but also highlighted the promise of this multifaceted gene in scrub typhus research.
The study was conducted after obtaining the approval of the Ethics Committee of Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand. Informed and written consent was derived from all patients and control subjects before their blood samples were collected.
Serum samples were derived from patients with acute undifferentiated fever at the first visit to Siriraj Hospital and were stored at −80°C until being used. As a control group, serum samples from ten healthy blood donors were derived and handled in a similar manner. As detected by an indirect immunofluorescent assay, definite diagnosis of scrub typhus was made according to the following criteria: 1) presence of OT-specific antibody at a titer of ≥1∶400 in a single acute sera; or 2) ≥four-fold rising of OT-specific antibody in paired sera collected two weeks apart
Analysis of L-Trp and L-Kyn was performed using a validated high performance liquid chromatography with tandem mass spectrometry (HPLC-MS/MS) method in accordance with the USFDA guidelines [U.S. Department of Health and Human Services Food and Drug Administration Center for Drug Evaluation and Research 2001]. Briefly, 3-nitro-L-tyrosine was added to each sample as an internal standard, and protein precipitation was performed using 10% trichloroacetic acid. Chromatographic separation of subsequent organic layer was carried out on LC-MS/MS with C18, 2.5 µm (50×3.00 mm i.d.). A mobile phase consisting of acetonitrile and 0.1% formic acid (Gradient condition) was delivered at a flow rate of 0.2 ml/min. Mass spectra were obtained using a Quattro Premier XE mass spectrometer (Micromass. UK), operated in multiple reaction monitoring mode. Sample introduction and ionization was performed by electrospray ionization in the positive ion. The mass transition ion-pair for L-Trp [
OT Standard Kato strain (CSUR R163) was propagated using mouse fibroblast cell line L929 as a host. When more than 90% of the host cells were infected, as determined by Giemsa staining, the media was replaced with fresh media. The infected cells were then dislodged from the flask and broken down using 1.0-mm-diameter glass beads with vigorous vortexing. The disrupted cell suspension ensuing from multiple parallel cultures were pooled together and stored as multiple small aliquots in liquid nitrogen. Before being used, a few aliquots of frozen OT inoculum were thawed in water bath at 37°C. Afterwards, the cell suspension was thoroughly mixed and broken down once again with glass beads and vortex before being centrifuged at 500×g for five minutes to sediment host cell lysate. Subsequent OT-containing supernatant was immediately used to infect THP-1 as well as to determine the infectivity of the inoculum by infected cell counting method with some modifications
Twenty-four hours prior to an in vitro infection experiment, 2×105 cells of THP-1 cell suspension were transferred into each well of 24-well plate and treated with 100 nM pharbol 12-myristate 13-acetate (PMA) (Sigma-Aldrich) to induce cell adherence. When IDO1 expression was required, the cells were be treated with 20 ng/ml IFN-γ (R&D Systems, Minneapolis, MN), along with 1 mM 1-methy-L-tryptophan (1-MT) (Sigma-Aldrich), L-Trp (Sigma-Aldrich) supplement, or neither. On the day of the experiments, the culture media was aspirated; and OT-containing inoculum of 5.9×105 ICU, prepared as described in the previous section, was inoculated onto the THP-1 monolayer. Infection process was facilitated by centrifugation at 1,450×g for five minutes and further incubation in a humidified 5% CO2 atmosphere at 37°C for one hour. The inoculum was then replaced with fresh PMA-containing media, together with additional treatments corresponding to each pre-infectious condition; this time point was designated as zero hour post-infection (p.i.). At specified time points, culture media was collected for determination of IDO1 activity or just discarded. The cell layer was then rinsed three times with phosphate buffer saline to wash out extracellular organisms. Finally, the infected cells were collected and further processed for RNA or DNA study.
Total RNA was extracted from each cell culture that had been lyzed in Trizol reagent (Invitrogen, Calsbad, CA) in accordance to the manufacturer's instruction. cDNA was synthesized from subsequent RNA extract using SuperScript III First-Strand Synthesis System (Invitrogen, Carldbad, CA). Specific amplification of TATA binding protein (TBP) and IDO1 gene transcripts was performed in duplicate using LightCycler FastStart DNA Master SYBR Green I reagents (Roche Applied Science), two µL of an appropriate dilution of a cDNA sample, and a specific primer pair for
DNA was isolated from infected cell cultures using standard phenol/chloroform method
To reflect IDO1 activity, the proportion of L-Kyn to L-Trp was calculated for each sample. Differences in the levels of serum L-Trp, L-Kyn or L-Kyn/L-Trp ratio between any two groups of the subjects were evaluated using Mann-Whitney test. For continuous clinical and laboratory variables, their association with the levels of serum L-Trp, L-Kyn or L-Kyn/L-Trp ratio was assessed using Spearman correlation.
For experimental data, differences of L-Trp level, L-Kyn level, L-Kyn/L-Trp ratio, IDO1 expression, and infection indexes between any two culture conditions were evaluated using unpaired t-test with or without Welch's correction as appropriate.
Twenty patients with confirmed scrub typhus were enrolled into the study. Clinical and laboratory data of seventeen patients were available and were summarized in
Characteristics | Median (IQR) |
Age | 42 (27.5–62.0) |
Sex (% of males) | 82.4 (n = 14) |
Fever day at presentation | 7 (4–11) |
Body temperature | 38.0 (37.7–39.0) |
Hospitalization (%) | 76.5 (n = 13) |
Eschar (%) | 17.6 (n = 3) |
Hepatosplenomegaly (%) | 35.3 (n = 6) |
Acute renal failure (%) | 29.4 (n = 5) |
Meningoencephalitis (%) | 17.6 (n = 3) |
Lymphadenopathy (%) | 11.8 (n = 2) |
Septic shock (%) | 5.9 (n = 1) |
DIC (%) | 5.9 (n = 1) |
WBC | 10.36 (8.08–15.25) |
Platelet | 158 (10.55–20.55) |
AST | 61.5 (17–114.5) |
ALT | 31.5 (14–100) |
IQR = interquantile range, DIC = disseminated intravascular coagulopathy, AST = aspartate aminotansferase, ALT = alanine aminotransferase.
To check whether functional activity of IDO1 enzyme actually increased in concordance to our earlier transcriptional study
Serum concentrations of L-Trp (A) and L-Kyn (B) in patients with scrub typhus (n = 20) were compared with those in healthy donors (n = 10). L-Kyn/L-Trp ratios (C) were calculated to reflect IDO1 activity. Data were derived from triplicate measurement. Median and interquantile range are presented.
Correlation analysis between levels of serum L-Kyn, serum L-Trp, or IDO1 activity and characteristics of the patients revealed no association for most of the clinical and laboratory parameters except for serum AST, whose level significantly correlates with serum L-Kyn (ρ = 0.6412,
To prove our hypothesis on the role of IDO1 in OT infection, we first evaluated temporal profiles of IDO1 gene transcription in an experimental model. As shown in
Before being infected with OT, THP-1 cells were treated with IFN-γ alone, IFN-γ combined with 1-MT, or neither. Mock-infected THP-1 cells were used as a control. Fold induction represents levels of IDO1 expression in each culture condition at indicated time points relative to that in mock-infected cells at 6 hours p.i.. For clearer illustration, data of OT-infected and mock-infected cultures are exclusively presented in
To assess functional activity of IDO1 in a cell model, levels of L-Trp and L-kynurenine in culture supernatant of THP1 cells was determined at corresponding time points. As shown in
Before being infected with OT, THP-1 cells were treated with IFN-γ alone, IFN-γ combined with 1-MT, or neither. Mock-infected THP-1 cells were used as a control. Levels of L-Trp (A) and L-Kyn (B, C) were measured in culture supernatant at 6, 24, 72, and 120 hours p.i.. IDO1 activity was reflected by the ratio of L-Kyn to L-Trp (D, E) at each indicated time point.
Despite the different rate of L-Trp consumption, the dynamic of L-Kyn level in OT-infected cultures was similar to that in mock-infected counterparts (
To investigate the effect of IDO1 activation on OT growth; an infection index, which is a relative copy number of OT 47 kDa gene per a human gene MTHFR, was determined in OT-infected cultures with or without IFN-γ-induced IDO1 activation. As shown in
Before being infected with OT, THP-1 cells were treated with IFN-γ alone, IFN-γ combined with 1-MT, or IFN-γ supplemented with L-Trp at 400 µg/ml (A) or 1 mg/ml (B). The infection index represents the degree of OT growth in each culture condition at indicated time points relative to that in mock-infected cells at 1 day p.i.. Means ± SEM from triplicate cultures are shown.
Activation of IDO1 is a host defensive mechanism downstream to IFN-γ that has been proven to limit the growth of various infectious pathogens in both immune and non-immune cells in vitro
From our earlier analysis of genome-wide expression, we have observed an up-regulation of
When THP-1 macrophages were infected with OT, transcription of IDO1 was directly induced by OT infection early after infection. However, a decreased level of L-Trp without a concomitant elevation of L-Kyn in OT-infected cultures as compared with mock infection may imply that higher rate of tryptophan consumption was contributed by increased utilization of the amino acid by the organism for its intracellular activity rather than enhancement of tryptophan catabolism by IDO1 enzyme. We postulated that the increase of IDO1 activity in vivo was a secondary event following the release of IFN-γ rather than a direct induction by the infection itself. For this reason, THP-1 was treated with IFN-γ prior to the in vitro infection to imitate natural infection in human body, in which other IFN-γ-producing cells such as NK cell, γδT cell, Th-1 cells and cytotoxic T cells, are also present. Such postulate seems sensible considering that the rise of IFN-γ during acute scrub typhus has been consistently observed by a number of earlier studies
After pre-treatment with exogenous IFN-γ, a marked increase in transcription of IDO1 along with its functional activity was detected as early as 6 hours p.i.. Concurrently, the intracellular number of OT was significantly depressed by IFN-γ treatment at 5 days p.i.. Even though IFN-γ is known to cause multiple changes in macrophage biology that could influence the outcome of an intracellular infection, our data suggest that IFN-γ-mediated IDO1 activation is responsible for growth restriction of OT to some extent since the number of OT per host cell was partially restored by a competitive inhibitor of the enzyme. However, it must be noted that induction of IDO1 only limited the rate of OT proliferation, but neither froze the growth nor reduced the number of the intracellular organism. For this reason, development of adaptive immune mechanisms seems to be required for eradication of the infection.
From earlier studies, anti-microbial activity of IDO1 has been explained by two non-mutually exclusive mechanisms: deprivation of tryptophan and formation of its downstream metabolites, collectively known as kynurenines. The first mechanism is based on the fact that tryptophan is the rarest essential amino acid in a human cell; therefore, IDO1-mediated tryptophan degradation, which limits the availability of the amino acid to be exploited by the organism, could lead to restraint of reproduction of tryptophan-sensitive microorganisms. In this study, we demonstrated that L-Trp supplementation at 400 µg/ml caused no change in the infection index of IFN-γ treated culture, whereas replenishment of the amino acid up to 1 mg/ml did not just restore but accelerated the growth of OT since the third day post infection. For this reason, it is not surprising why an earlier study that supplemented the culture media with 100 µg/ml of trytophan failed to rescue the growth of OT in IFN-γ-treated murine embryonic cell lines and refuted the role of tryptophan deprivation in control of OT infection
For the second mechanism involving IDO1-mediated kynurenine formation, it has been proved that some tryptophan metabolites, particularly 3-hydroxy-DL-kynurenine and alpha-picolinic acid but not L-Kyn, can exert anti-microbial activity against several extracellular bacteria in a dose-dependent manner
Despite the key findings that have already been discussed, we are also aware of some factors that could influence the interpretation of our experimental data. Firstly, the infection index, used as an indicator of OT growth in this study, was derived by a relative quantification method. It is, therefore, not only sensitive to a change in the number of OT but also to that of host cells at the time of the assessment. In the cultures supplemented with 1-mg L-Trp, we observed a reduction in the number of host cells by time, which might be partly responsible for a marked increase in the infection index of the condition compared with the others. Such cell loss could be a consequence of overwhelming expansion of OT in either tryptophan-rich environment or excessive accumulation of toxic metabolites in the culture system, or both. Secondly, the use of L-Kyn to L-Trp ratio as an indicator of IDO1 activity in THP-1 cells was unfortunately limited by the drop of L-Trp below the detectable level since the first day post infection. Moreover, some ambiguity in the data interpretation also resulted from the surprising rise of L-Kyn level in the culture supernatant of cells treated with 1-MT. Unlike other cultures whose L-Kyn level similarly reached its equilibrium at 72 hours p.i., the concentration of L-Kyn in 1-MT treated cultures continuously rose beyond 120 hour p.i.. This might reflect a higher rate of L-Kyn formation exceeding the capacity of downstream enzymes to catabolize the molecule into downstream metabolites even in the presence of the IDO1-specific inhibitor. However, it is unlikely that the other two tryptophan-degrading enzymes, IDO-2 and tryptophan 2,3-dioxygenase (TDO), were responsible for such phenomenon because the former appears to be expressed as an inactive form in human dendritic cells
Apart from its anti-microbial activity, IDO1 also appears to be involved in suppression of immune responses as well as development of immunological tolerance. This concept originated from a key finding that demonstrated a role of IDO1-mediated tryptophan degradation in prevention of allogeneic fetal rejection
Recently, a high level of serum L-Kyn/L-Trp ratio was reported to be associated with the development of severe complications, like septic shock and multiple organ failure, in patients with major trauma
In conclusion, this is the first report for IDO1 activation in patients with scrub typhus, which has brought this multifaceted gene with promising therapeutic potential into focus in the field of scrub typhus research. We demonstrated here that IDO1-mediated tryptophan deprivation was a downstream mechanism of IFN-γ that helps restrain intracellular expansion of OT. However, further studies are deserved to investigate other potential effects of IDO1 activation on the outcome of OT infection in a more complex experimental model as well as in more patients with scrub typhus.
(TIF)
(DOC)
(DOC)
(DOC)
We would like to thank Dr.Wiwit Tantipattayangkul, Prof.Dr.Didier Raoult, and personnel for generously providing OT strain Kato, as well as particularly helpful technical advice for propagation of OT. We also would like to thank Dr.Angkana Chaiprasert and Dr.Sontana Siritantikorn for kindly providing THP-1 and L929 cell lines for our work respectively.