Does IL-17 Play a Role in Hepatic Dysfunction of Scrub Typhus Patients?

S crub typhus is a zoonotic disease that is caused by Orientia tsutsugamushi (Tamura et al. 1995) and is accidentally transmitted to humans by the bites of larval-stage trombiculid mites (chiggers). It prevails in the rural South and Southeast Asia and the Western Pacific (Phongmany et al. 2006). Patients present with sudden fever and a generalized lymphadenitis, and more than 50% have an inoculation eschar (Jensenius et al. 2004, Phongmany et al. 2006). Severe complications include pneumonitis, meningoencephalitis, disseminated intravascular coagulation, or renal failure (Jensenius et al. 2004, Kim et al. 2008). The dominant strains of O. tsutsugamushi in Korea are the Boryong, Karp, and Gilliam (Chang 1995).

IL-17 is a potent proinflammatory cytokine produced by activated memory T cells (Aggarwal and Gurney 2002). There are at least six members of the IL-17 family in humans and in mice (Aggarwal and Gurney 2002). They represent a distinct signaling system that appears to have been highly conserved across vertebrate evolution (Aggarwal and Gurney 2002). Although the relationship between hepatic dysfunction and scrub typhus has been given little attention in the literature, hepatic dysfunction is always associated in endemic areas (Yang et al. 1995, Park et al. 2003, Hu et al. 2005, Chanta et al. 2007). Several liver histopathologic studies were reported, such as nonspecific reactive hepatitis (Kanno et al. 1996), granulomatous change (Satoh et al. 1991, Chien et al. 1995), vasculitis (Chi et al. 1998), or direct cytopathic injury (Watanabe et al. 2005). However, the mechanism of hepatic dysfunction is still not clear, and cytokine network studies are rare in scrub typhus. The study by Yasumi et al. (2007) suggested the usefulness of serum IL-17 level in evaluating the severity of an acute hepatic injury, thus emphasizing the necessity for the basic investigation of the pathological role of IL-17 in acute hepatitis. From this study, we had an idea that IL-17 may be involved in hepatic dysfunction in scrub typhus, which has not been investigated yet. Therefore, the objective of this study was to evaluate IL-17 as a useful surrogate marker indicating the hepatic dysfunction in scrub typhus.

We included 43 patients with Boryong-type scrub typhus patients and 40 age- and sex-matched control healthy people in our study, admitted between September 2006 and November 2006 to the 1000-bed Eulji University Hospital, Daejeon, Korea. The objective of this study was to compare IL-17 between scrub typhus patients and a healthy group. Because data like these have not been previously studied, we set a difference of mean as 20, a standard deviation of the scrub typhus group as 30, a standard deviation of the healthy groups as 30, α as 0.05, and β as 0.20; 36 persons were needed for each group. We added 10% of 36, so the number that we calculated was 40. As a result, the scrub typhus group was 43, and the healthy group was 40. We enrolled only Boryong-type scrub typhus patients—the most prevalent type in the central region of South Korea, including Daejeon (Chang 1995, Shim et al. 2005) to exclude the degrees of virulence by unique antigenic characteristics of various serotypes such as Karp, Gilliam, Boryong, and Yonchon serotypes. To analyze the relationships between scrub typhus and hepatic/kidney function and hemograms, we excluded patients with previously known hepatic/kidney disease, or hematologic disease. As the cytokine profiles may change significantly depending on the progression of the disease, venous blood samples were collected from scrub typhus patients on admission day 1 before antibiotic therapy. Scrub typhus was confirmed on the basis of either a single indirect immunofluorescent-specific IgM titer of ≥1:10 against O. tsutsugamushi or a fourfold or greater rise in the IFA IgG titer (World Health Organization 2004). We also performed a nested polymerase chain reaction assay to the target 56 kDa protein gene as described previously (Furuya et al. 1993). The DNA sequences were compared with the nucleotide sequences of O. tsutsugamushi registered in the GenBank for sequence homology analysis. IL-17 was measured using human IL-17 immunoassay (Quantikine^®; R&D Systems, Minneapolis, MN) according to the instructions of the manufacturer, which is the quantitative sandwich enzyme immunoassay technique for measuring IL-17. We added Trasylol to prevent the degradation of IL-17. We gathered the data by chart reviews, which included patient demographics, the presence of fever, headache, myalgia, eschar, skin rash, stupor/coma, mechanical ventilation, intensive care unit admission, and abnormal chest X-ray findings, and laboratory findings on admission day 1 before antibiotic therapy {liver function test (LFT) panels (aspartate aminotransferase [AST], alanine aminotransferase [ALT], total bilirubin, alkaline phosphatase, lactate dehydrogenase, prothrombin time, albumin, platelet); kidney function test (KFT) panels (blood urea nitrogen, serum creatinine), and hemograms (hemoglobin [Hb] and white blood cell count)}. We analyzed this data using the Statistical Package for Social Sciences version 14.0 (SPSS, Chicago, IL). Independent t-test, Kolmogorov–Smirnov test, and correlation analysis were used. All variables with a p < 0.05 were considered significant.

The median age and sex ratio (M/F) of the scrub typhus group and the healthy group was 60.0 ± 12.9 versus 56.2 ± 5.5 years, and 23/20 versus 20/20, respectively. The main symptoms and signs in the scrub typhus group were fever (83.7%), headache (72.1%), skin rash (81.4%), and myalgia (83.7%). Eschars were found in 34 patients (79.1%). The median (25 and 75 percentiles) duration of fever was 5 (3, 7) days, and the median maximal body temperature was 38.4°C (37.8°C, 38.8°C). Sixteen patients (37.2%) had abnormal chest X-ray findings, two patients (4.7%) had intensive care unit admission history, and one (2.3%) had mechanical ventilation and showed stupor/coma (Table 1). The IL-17 level (mean, SD) in patients with headaches (43.5, 39.8 pg/mL) was higher than patients without headaches (18.8, 22.5 pg/mL), showing a borderline significance (p = 0.05) (Table 1). All patients made a complete recovery after doxycycline therapy. The median levels (25 and 75 percentiles) of IL-17, AST, ALT, and Hb of the scrub typhus group and the healthy group were described in Table 2, showing the significant difference between the two groups. To determine which factor correlates with IL-17, we conducted correlation analysis between IL-17 and LFT panels, KFT panels, and hemograms. We did not find any correlation with IL-17.

IL-17 is synthesized and secreted by activated CD4⁺ memory T lymphocytes and induces a wide range of inflammatory and immunological responses such as neutrophil recruitment and activation (Kolls and Linden 2004), and Th1 and Th2 responses (Nakae et al. 2002). The potent inflammatory actions identified for several of these factors, and the emerging associations with major human diseases suggest that these proteins may have significant roles in inflammatory processes (Aggarwal and Gurney 2002). For example, IL-17 plays a proinflammatory role in the airways by recruiting and activating neutrophils (Larsson et al. 2000, Linden et al. 2000). Overproduction of IL-17 has been associated with several chronic disease conditions, suggesting a role in these diseases [transplant rejection (Antonysamy et al. 1999), rheumatoid arthritis (Kotake et al. 1999), systemic sclerosis (Kurasawa et al. 2000), psoriasis (Homey et al. 2000), and promotion of tumor growth (Kato et al. 2001)]. Other studies have shown abundant IL-17 in various conditions affected by microbial pathogens, including the synovial fluid of patients with Lyme arthritis or Chlamydia-induced reactive arthritis (Infante-Duarte et al. 2000), Helicobacter pylori–infected gastric mucosa (Luzza et al. 2000), and listeria infection (Umemura et al. 2004).

Liver involvement is not commonly mentioned in the Western literature. No consistent liver enzyme abnormality is even shown in Western medical textbooks (Alfred and Saah 2005). However, liver involvement is always associated with scrub typhus in Korea, one of the endemic countries, as in this report (90.7% of patients showed an abnormality in their LFT) and other Korean reports (average AST, ALT, and γ-GTP were increased to 92.5, 98.3, and 132.2 IU/L, respectively) (Park et al. 2003). Yang et al. (1995) and Hu et al. (2005) reported that hepatic dysfunction occurred in 77% and 96.7% of the scrub typhus patients, suggesting the differential diagnosis of patients from high-risk groups and endemic areas who present with hepatitis-like symptoms should be examined for scrub typhus. The liver histopathologic study showed nonspecific reactive hepatitis (Kanno et al. 1996) or granulomatous changes (Satoh et al. 1991, Chien et al. 1995). The electron microscope showed rod-shaped organisms within the hepatocytes and sinusoids with variable degree of cytoplasmic organelle damage (Pongponratn et al. 1998). The main pathologic change is focal or disseminated multiorgan vasculitis caused by the destruction of endothelial cells and perivascular infiltration of leukocytes. The extent of infiltrating leukocytes around the small blood vessels is closely related with the clinical manifestation of scrub typhus (Chi et al. 1998). Watanabe et al. (2005) reported that the mechanism of liver dysfunction might be a result of the direct cytopathic effect on hepatocytes rather than immunoreaction as in infections caused by other hepatitis viruses considering liver histology and marked dominant levels of alkaline phosphatase or γ-GTP in Gilliam type-scrub typhus.

Several reports document the roles of cytokines in scrub typhus. Chierakul et al. (2004) reported that elevated plasma concentrations of interferon-γ (IFN-γ) in patients with scrub typhus suggested this disease is associated with activation of a cell-mediated immune reaction. In addition, de Fost et al. (2005) reported elevated plasma concentrations of granzymes A and B, IFN-γ-inducible protein 10, and monokine induced by IFN-γ, suggesting activation of cytotoxic lymphocytes is part of the early host response to scrub typhus. Iwasaki et al. (1997) reported the patients with the most severe illness, for whom the disease was nearly fatal, showed increases in circulating cytokines (M-cerebrospinal fluid [CSF], G-CSF, and TNF-α).

We postulated IL-17 might be a cytokine with a role in hepatic dysfunction in scrub typhus, because of reports that IL-17 is a marker of acute hepatic injury (Yasumi et al. 2007). In these data, the IL-17 level was significantly higher in scrub typhus patients compared with the healthy group (p < 0.01). Also, the patients with scrub typhus showed significantly higher AST and ALT (p < 0.01) and lower Hb levels (p = 0.03) compared with the healthy group. However, in correlation analysis, we could not find any correlation between IL-17 and LFT panels, KFT panels, and hemograms. This suggests that the IL-17 level increases independently of hepatic dysfunction, kidney dysfunction, and abnormal hemograms in scrub typhus patients. The paper by Yasumi et al. (2007), which formulated the basis of our hypothesis of a possible association between scrub typhus–induced hepatic dysfunction and IL-17, the AST and ALT levels were substantially high (approximately 20–30-fold higher than that in our report). Thus, it is possible that the absence of any correlation between IL-17 and LFT panels could be just due to the fact the pathology in scrub typhus is not a true hepatitis and a severe inflammatory reaction that could be associated with a high level of IL-17. These results may also explain that hepatic dysfunction is caused by direct invasion of the organism, resulting in hepatocellular damage as Watanabe et al. (2005) has suggested.

IL-17 might have other roles in the pathophysiology of scrub typhus than hepatic dysfunction. In clinical practice, the patients usually present severe constant headaches, which can aid to the diagnosis of scrub typhus in autumn in Korea with eschar and skin rash. Headache has been found in 54–100% of scrub typhus patients (Chang 1995, Song et al. 2004, Varghese et al. 2006), and our data showed 72.1%. The IL-17 level was higher in patients with headaches with a borderline significance (p = 0.05). The pathologic changes of scrub typhus are characterized by focal or disseminated multiorgan vasculitis such as lung and brain due to inflammatory lesions from infiltrating leukocytes and destruction of the endothelial cells lining small blood vessels (Kasuya et al. 1996, Song et al. 2004), and the extent of vasculitis helps explain the great diversity of clinical manifestations that have been encountered. Pathological findings in studies of the central nervous system include a diffuse or focal mononuclear cell exduate in the leptomeninges, and the presence of typhus nodules (clusters of microglial cells) that are distributed throughout the brain substance (Allen and Spitz 1945, Settle et al. 1945, Levine 1946). Analysis of CSF revealed mild to moderate pleocytosis (mainly mononuclear) similar to those of viral meningitis (Berman and Kundin 1973, Pai et al. 1997). In addition, Pai et al. demonstrated the presence of rickettsiae in CSF. Erythrocytes were also seen in the CSF, which could be explained by the presence of generalized vasculitis, the pathological mechanism of scrub typhus (Pai et al. 1997). Although the precise mechanism of these vascular pathophysiological changes remains unclear, we could make an assumption that one of the causes of the pathophysiological consequences may be a result of increased levels of IL-17. Increased IL-17 during O. tsutsugamushi infections has caused vasculitis in brain vessels through pleitropic biological activities such as destruction of endothelial and epithelial cells (Yao et al. 1995) and recruiting inflammatory cells. The sensation of a headache is then picked up by pain receptors located at the base of the brain in arteries and veins (Welch 2005) due to vasculitis.

From these findings, we know that IL-17 increases significantly in scrub typhus, and it was not related to hepatic dysfunction but could be related to vasculitis-associated headaches, which are the basic pathogenesis of scrub typhus. Despite the small number of patients due to serotype selection, this study is worthwhile as a first to investigate the role of IL-17 in scrub typhus. More prospective, large-scale studies are needed on the relationship between scrub typhus and hepatic dysfunction or headaches and for confirmation the exact pathophysiology of headaches in the future. Also, a detailed study of the activation of the cytokine network in patients with scrub typhus may increase our understanding of why some people become sicker than others when they are infected with the same organism.