Introduction

One of the leading pathogens causing sepsis in immunocompromised hosts are Candida spp. [1, 2]. Medical conditions that lead to an immunocompromised state increase susceptibility to Candida sepsis [3]. In addition to exogenous factors, it is believed that genetic variation also plays an important role in susceptibility to sepsis [4–6]. Caspase-12 is an inflammatory caspase, in which a loss-of-function genetic variant has been fixed in some populations by still undefined evolutionary pressures [7–9]. This loss-of-function is due to the presence of a T/C single nucleotide polymorphism (rs497116) on nucleotide position 125 in the CASPASE-12 gene [10]. Although the ancestral variant is still present in African and African-American populations, of which 20–30% express the active variant of caspase-12, it is absent in Asian and Caucasian populations [9, 10].

Functional studies have suggested that functional caspase-12 is a negative regulator of caspase-1 activation, which might result in less cytokine production in response to recognition through pattern recognition receptors. Thus, based on the proposed inhibitory effect on caspase-1 and, consequently, lower IL-1β and IL-18 production, functional caspase-12 may increase the susceptibility to severe sepsis and/or the clinical outcome of sepsis patients [10]. Therefore, it is compelling to assess whether genetic variation in CASPASE-12 plays a role in the susceptibility to Candida sepsis. The aim of this study was to assess whether genetic variants of CASPASE-12 influence the incidence, severity, and mortality of Candida sepsis in a cohort of African-American patients.

Patients, materials, and methods

Subjects were enrolled between January 2003 and January 2009 after informed consent (or waiver, as approved by the Institutional Review Board) at the Duke University Hospital (DUMC, Durham, NC, USA). Infected subjects had ≥1 positive blood cultures for a Candida species while hospitalized. Non-infected controls were recruited from the same hospital wards as infected patients, with no history or evidence of Candida sepsis/invasive candidiasis or any invasive fungal infection.

Genomic DNA was isolated from whole blood using standard procedures. The region of interest of the CASPASE-12 gene was amplified as described previously [10].

Circulating cytokine concentrations of IL-6, IL-8, and IFNγ in infected patients were measured by Multiplex Fluorescent Bead Immunoassays (xMAP technology, Bio-Rad, Veenendaal, the Netherlands), from day 0 up to day 5 after the initial positive blood culture.

Statistical comparisons of frequencies were made between infected versus non-infected subjects using Chi-square tests. Statistical analysis of the cytokine data was performed by using the Mann–Whitney U-test. Overall, a p-value < 0.05 was considered to be statistically significant.

Results

A total of 93 African-American patients and 88 non-infected African-American controls had genetic and clinical data available for the analysis. The demographic data for the study subjects are presented in Table 1.

Table 1 Baseline patient characteristics of African-American patients with Candida systemic infection or uninfected controls recruited at the Duke University Hospital (DUMC, Durham, NC, USA) (n = 181)
Full size table

No significant differences in the distribution of CASPASE-12 genotypes were seen when comparing infected patients (CC 3.9%, CT 25.3%, TT 72.4%) and non-infected controls (CC 2.9%, CT 30.0%, TT 66.1%) (p > 0.05). No associations between the CASPASE-12 genotypes and disseminated disease, persistent fungemia, or 30-day mortality were observed (data not shown).

Serum samples collected from infected patients during the first 5 days after the initial positive blood culture were measured for concentrations of IL-6, IL-8, and IFNγ. Also, measurements of IL-1β and IL-18 were performed in these samples. However, the concentrations of these cytokines were too low to detect (data not shown). Cytokine concentrations decreased over time. No differences in cytokine concentrations were apparent between individuals bearing different CASPASE-12 genotypes (Fig. 1).

Fig. 1
figure 1

IL-6, IL-8, and IFNγ circulating concentrations in infected patients from day 0 up to day 5 after initial positive blood culture, in relation to the CASPASE-12 genotype. TC heterozygous, TT homozygous mutant. The data are presented as mean ± standard error of the mean (SEM)

Full size image

Discussion

Caspase-12 has been suggested to inhibit caspase-1 processing of proIL-1β and proIL-18 into the active cytokines. Genetic variation of CASPASE-12 in populations of African descent has been previously associated with susceptibility to bacterial sepsis [10]. The present study was performed in order to assess the role of caspase-12 in sepsis caused by Candida spp. The results indicate that the CASPASE-12 genotype has no significant effect on the susceptibility and severity of systemic infections with Candida.

Candida is one of the leading pathogens causing sepsis [2, 11, 12]. Pro-inflammatory cytokines such as IL-1β and IL-18 are a crucial factor in eliciting an effective immune response to eradicate the infection. A modulatory step in the production of these cytokines is exerted at the level of caspase-1, a protease that cleaves the pro-form of these cytokines into shorter bioactive proteins [13, 14]. It has previously been reported that CASPASE-12 knockout mice were better capable of clearing both local and systemic bacterial infections compared to wild-type mice, through an improved inflammatory response [15]. The same authors described a similar effect of caspase-12 in patients with bacterial sepsis, with individuals bearing functional caspase-12 being more susceptible to this condition [10]. However, the role of CASPASE-12 genetic variants in fungal sepsis has not been addressed so far.

Firstly, the comparison of CASPASE-12 genotype frequencies in African-American patients with non-infected controls revealed no statistically significant differences. Secondly, no effects of the CASPASE-12 genotype was observed in relation to the clinical outcome of infection, assessed as disseminated disease, persistent fungemia, and 30-day mortality. Furthermore, serum cytokine concentrations during the first 5 days of infection were shown to be unaffected by the CASPASE-12 genotype.

Our findings on the lack of influence of the CASPASE-12 genotype on fungal sepsis contrast with those of Saleh et al. [10, 15], who suggested an important role of this genetic variant in bacterial sepsis. Moreover, circulating cytokine concentrations in infected patients were also not influenced by the CASPASE-12 genotype. It should be emphasized that this is, in particular, true for IL-6 and IFNγ, cytokines that are induced by IL-1β and IL-18, respectively [16–18]. This provides indirect evidence that functional caspase-12 has no clear effect on the production of IL-1β and IL-18 in the context of Candida sepsis. One possible explanation for the discrepancy between this study and that of Saleh et al. [10] is represented by the different cause of sepsis in the two studies, fungal and bacterial, respectively. However, one has to concede that the pro-inflammatory cytokines, of which production is reportedly regulated by the CASPASE-12 genotype, exert similar protective effects in bacterial and fungal sepsis [18–21]. In this respect, a recent study has also failed to reproduce the inhibitory effects of the CASPASE-12 genotype of lipopolysaccharide and Gram-negative bacteria-induced cytokine production [22], bringing into question the biological activity of caspase-12.

In conclusion, although an effect of the CASPASE-12 genotype on the susceptibility to bacterial sepsis has been previously reported in a small cohort of African-American patients [10], this could not be confirmed in our larger cohort of fungal sepsis patients. Furthermore, clinical outcome and in vivo cytokine responses were not influence by the CASPASE-12 genotype. Therefore, we propose that caspase-12 is redundant for systemic host defense in sepsis.