Sepsis

Polymorphisms in CD14, mannose-binding lectin, and Toll-like receptor-2 are associated with increased prevalence of infection in critically ill adults* Ainsley M. Sutherland, BSc; Keith R. Walley, MD; James A. Russell, MD Objective: To test for the association of single nucleotide polymorphisms of the innate immunity receptors cluster of differentiation (CD)-14, mannose-binding lectin, and Toll-like receptor-2 with clinical phenotype in critically ill patients with systemic inflammatory response syndrome. Design: Genetic association study. Setting: Tertiary care mixed medical-surgery intensive care unit at St. Paul’s Hospital, Vancouver, BC, a teaching hospital associated with the University of British Columbia. Patients: A cohort of 252 critically ill Caucasians with systemic inflammatory response syndrome. Interventions: DNA was extracted from discarded blood. Clinical data were gathered by retrospective chart review. Measurements and Main Results: C-159T CD14, the X/Y and B, C, and D polymorphisms of mannose-binding lectin, and T-16933A Toll-like receptor-2 were genotyped using polymerase chain reaction-restriction fragment length polymorphism. We tested for association of genotype with prevalence of positive bacterial cultures, type of organism (Gram-positive, Gram-negative, other), sepsis and septic shock at admission to the intensive care unit, and 28-day survival. CD14 159TT was associated with increased prevalence of positive bacterial cultures and with Gram-negative bacteria. Mannose-binding lectin haplotype pairs XO/O and O/O were also associated with increased prevalence of positive bacterial cultures but not with a specific organism class. Toll-like receptor-2 16933AA was associated with increased prevalence of sepsis and with Gram-positive bacteria. In contrast, the polymorphisms were not associated with increased prevalence of septic shock or altered 28-day survival. Conclusions: Single nucleotide polymorphisms in CD14, mannose-binding lectin, and Toll-like receptor-2 are associated with increased prevalence of positive bacterial cultures and sepsis but not with altered prevalence of septic shock or decreased 28-day survival. Furthermore, CD14 single nucleotide polymorphisms were associated with Gram-negative bacteria and Toll-like receptor-2 with Gram-positive bacteria, whereas mannose-binding lectin was not associated with a particular organism class. Thus, single nucleotide polymorphisms in innate immunity receptors may alter recognition and clearance of bacteria without changing outcomes of critically ill adults with systemic inflammatory response syndrome. (Crit Care Med 2005; 33:638 –644) KEY WORDS: innate immunity; cluster of differentiation-14; mannose-binding lectin; Toll-like receptor-2; immunologic receptors; infection; sepsis; single nucleotide polymorphism T he preprogrammed innate immune response contributes to rapid clearance of bacteria in an effort to prevent or contain infection. Activation of the innate immune response to infection varies signif- *See also p. 695. From the University of British Columbia Critical Care Research Laboratories, James Hogg iCAPTURE Centre/St. Paul’s Hospital, Vancouver, BC, Canada. Supported, in part, by the Canadian Institutes for Health Research, Ottawa, ON, Canada; and the British Columbia Lung Association, Vancouver, BC, Canada. Keith R. Walley is a Michael Smith Foundation for Health Research Distinguished Scholar, Vancouver, BC, Canada. Ainsley M. Sutherland is supported by a Michael Smith Foundation for Health Research Clinical Junior Graduate Studentship. Dr. Russell and Dr. Walley have submitted a patent application regarding TLR2 and clinical phenotypes. Copyright © 2005 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/01.CCM.0000156242.44356.C5 icantly between individuals (1–3), which may have important clinical implications. Genotype has been shown to contribute substantially to outcome of infectious disease (4 – 6). Thus, variation in key innate immunity genes may explain variation in individuals’ responses to infection. If so, it would be useful to evaluate genotypes of critically ill patients to understand individual susceptibility and response to infection, which could lead ultimately to specific patient-tailored therapy based on genotype. Cluster of differentiation (CD)-14 is an innate immunity receptor for lipopolysaccharide, peptidoglycan, and lipoteichoic acid. CD14 is found in association with Toll-like receptors (TLR) on the surface of monocytes, macrophages, neutrophils, and hepatocytes (7, 8) and as a soluble form in serum (9). The serum level of soluble CD14 is increased during septic shock and is associated with greater mortality (10, 11). A C-to-T transition at position 159 in the promoter of the CD14 gene has been associated with increased density of membranebound CD14 on monocytes and increased serum levels of soluble CD14 (12, 13). This may alter host recognition and clearance of pathogens. Mannose-binding lectin (MBL) binds repeating arrays of sugar groups on microbial surfaces with its lectin domains (14). MBL activates the “alternative,” or lectin, complement pathway in an antibody-independent manner. MBL also is an opsonin that enhances phagocytosis (15). MBL binds yeasts, viruses, and Gram-negative bacteria and also Grampositive bacteria with low affinity (16 – 18). Three structural mutations have been found in exon 1 of the MBL gene at codons 52, 54, and 57, and the minor Crit Care Med 2005 Vol. 33, No. 3 638 alleles are referred to as alleles D, B, and C, respectively (19 –21). The A allele indicates the wild-type at each of these positions. G-to-C transversions have also been identified at position 550 (genotype H/L) and 221 (genotype Y/X) in the MBL promoter region 20. MBL polymorphisms occur as six different haplotypes (22–24). These MBL haplotypes are associated with different serum concentrations of MBL (Table 1) (22–24). Individuals having two copies of low MBL level haplotypes have increased incidence of infection and sepsis (25, 26). TLR2 was previously believed to be a receptor for peptidoglycan from the cells walls of Gram-positive bacteria (27–29). It is now recognized that commercial preparations of peptidoglycan used in studies of TLR2 activation contain other cell wall components (30) and that TLR2 was activated by these impurities (31). TLR2 is in fact a pattern recognition receptor for lipoteichoic acid from the cell membrane of Gram-positive bacteria (31, 32). The receptor is a point of direct contact between the host neutrophils, macrophages/monocytes, and T cells with Gram-positive bacteria (33). Binding of lipoteichoic acid to TLR2 causes an intracellular signaling cascade that ultimately leads to activation of the nuclear factor- B pathway and increased transcription of proinflammatory cytokines (32– 34). To date there have been no studies of the association of single nucleotide polymorphisms (SNPs) of CD14, MBL, and TLR2 with clinical outcomes in a single cohort of critically ill patients. We hypothesized that polymorphisms in the genes of these innate immunity receptors are associated with a) increased prevalence of positive bacterial cultures and sepsis; b) type of bacteria (Gram-positive, Gram-negative, other); and c) prevalence of septic shock and 28-day survival of critically ill patients with systemic inflammatory response syndrome (SIRS). We found that specific alleles of CD14, MBL, and TLR2 were each associated with increased risk of infection and sepsis and with specific classes of bacteria but not with altered survival. MATERIALS AND METHODS This study was approved by the Research Ethics Board of Providence Health Care and the University of British Columbia. Study Population All patients admitted to the intensive care unit (ICU) of St. Paul’s Hospital were screened for inclusion. The ICU is a mixed medicalsurgical ICU in a tertiary care, universityaffiliated teaching hospital of the University of British Columbia. Patients were included in the study if they met at least two of four SIRS criteria: a) fever ( 38°C) or hypothermia ( 36°C); b) tachycardia ( 90 beats/min); c) tachypnea ( 20 breaths/min), PaCO2 32 mm Hg, or need for mechanical ventilation; and d) leukocytosis (total leukocyte count 12,000 mm3) or leukopenia ( 4000 mm3) (35). Patients were included in the cohort on the calendar day on which the SIRS criteria were met. To decrease the confounding influence of population admixture secondary to ethnic diversity on associations between genotype and phenotype, only Caucasian patients were studied (36). Of these consecutive critically ill patients admitted to St. Paul’s Hospital ICU, 252 met the inclusion criteria for our study (having at least two of four SIRS criteria and being Caucasian). These patients were genotyped for polymorphisms in CD14, MBL, and TLR2 and were used as our cohort for analysis. ICU (based on the Acute Physiology and Chronic Health Evaluation [APACHE] III diagnostic codes) (37), and the admission APACHE II score (38). Each of the four SIRS criteria was recorded as present or absent on admission to the ICU. Sepsis was defined as the presence of two or more SIRS criteria plus the presence of a known or suspected infection during the 24-hr period. Microbiological cultures were taken as part of routine medical care for any patients who were suspected of having an infection. As this is a cohort of critically ill patients with SIRS, most patients had cultures taken. Cultures that were judged to be positive due to contamination or colonization were excluded. Cultures were categorized as Gram-positive, Gram-negative, fungal, or other. Septic shock was defined by sepsis plus significant hypotension (systolic blood pressure 90 mm Hg or the need for vasopressors). Selection and Genotyping of SNPs CD14. We genotyped the previously described C-159T CD14 promoter polymorphism in our patient cohort (12). The promoter region of the CD14 gene was amplified by polymerase chain reaction (PCR) using a set of primers designed according to the published sequence. PCR was performed and analyzed using the restriction fragment length polymorphism (RFLP) technique. PCR-RFLP fragments were electrophoresed on 2% agarose gel and visualized with ethidium bromide staining and ultraviolet illumination. Clinical Phenotype Our primary outcome variable was the presence of a positive bacterial culture. Secondary outcome variables were sepsis on admission to the ICU, septic shock on admission to the ICU, and 28-day survival. Baseline demographics recorded were age, gender, medical or surgical diagnosis on admission to the Table 1. Mannose-binding lectin (MBL) haplotypes associated with high, moderate, or low levels of serum MBL (Refs. 22–24) MBL Haplotype YA XA O O O O Serum MBL High Moderate Low Low Low Low G-221C (Y/X) Y X Y Y Y Y Codon 52 A/D A A A A D D Codon 54 A/B A A A B A B Codon 57 A/C A A C A A A MBL. We genotyped our patient cohort for the previously described X/Y polymorphism at position 221 in the promoter (22) and structural mutations in exon 1 at codons 52, 54, and 57 (alleles D, B, and C respectively) of MBL (19 –21). The A allele indicates the wild-type structural allele. Haplotypes of these four polymorphisms were inferred for all patients using PHASE software (39). DNA was amplified by site-directed mutagenesis polymerase chain reaction (SDM-PCR) for the exon 1 alleles B, C, and D, and sequencespecific primer PCR (SSP-PCR) for Y/X promoter alleles (22, 23). The B, C, and D alleles were analyzed using the RFLP technique. SDM-PCR products were separated on a 3% agarose gel electrophoresis. Bands were visualized by ethidium bromide staining and ultraviolet illumination. TLR2. We selected an SNP in the TLR2 promoter region using a haplotype-based approach to SNP selection. We used unphased genotypic data from 23 Caucasians from the Coriell Cell Repository 639 Crit Care Med 2005 Vol. 33, No. 3 (from www.innateimmunity.net/IIPGA/ IIPGASNPs) (40) to infer haplotypes of TLR2 using PHASE software (39). We then used MEGA 2 software (41) to infer a phylogenetic tree to identify major haplotype clades. Haplotypes were sorted into clades according to this phylogenetic tree, and this haplotype structure was inspected to choose haplotype clade tag SNPs (htSNP) (42, 43). We chose an htSNP that identified the two major haplotype clades of TLR2 in Caucasians (rs4696480, Fig. 1). This SNP was then genotyped in our patient cohort to define major clades of TLR2. The promoter region of the TLR2 gene was amplified by PCR with SSP-PCR designed according to the published sequence. PCR products were electrophoresed on a 2% agarose gel. Samples that were homozygous for one allele or the other at position 16933 of the TLR2 gene produced only one band, whereas samples that were heterozygous for the 16933T/A SNP produced two bands on separate gels when visualized with ethidium bromide staining and ultraviolet illumination. culture, type of bacterial culture [Grampositive, Gram-negative, other], sepsis and septic shock at admission, and 28-day survival) were compared between genotypes using a chi-square test. Differences in continuous outcome variables (age, APACHE II score) between genotypes were tested using analysis of variance. Twenty-eight-day survival was further compared between genotypes while adjusting for other confounders (age, gender, and medical vs. surgical diagnosis) using a Cox regression analysis in addition to a Kaplan-Meier analysis. Genotype relative risk for survival was calculated for each genotype. Genotype distributions were tested for Hardy-Weinberg equilibrium (44). We report the mean and 95% confidence intervals. Statistical significance was set at p .05. The data were analyzed using SPSS 11.5 for Windows (SPSS, Chicago, IL). The CD14 159TT genotype was associated with significantly increased prevalence of positive bacterial cultures at admission to the ICU (p .02, Fig. 2). Both the CD14 159TT and the 159CT genotypes were associated with significantly increased prevalence of Gram-negative cultures compared with the 159CC genotype (p .03, Fig. 3). CD14 genotypes were not, however, associated with a significantly different prevalence of sepsis or septic shock on admission to the ICU or with 28-day survival (Fig. 4). MBL Two hundred twenty-two patients were successfully genotyped for the chosen alleles within the MBL gene. The MBL X/Y promoter polymorphism and the exon 1 A, B, C, and D structural alleles were each in Hardy-Weinberg equilibrium. Five common and two rare haplotypes of the X/Y polymorphism and the A, B, C, and D structural polymorphisms were inferred in our patient cohort using PHASE software: YAAA (YA), 53.4%; YAAC (O), 2.1%; YABA (O), 15.7%; XAAC (O), 0.2%; YDAA (O), 8.9%; YDBA (O), 0.2%; and XAAA (XA), 19.4%. The frequencies of these haplotypes were similar to frequencies previously reported in Caucasians (22–24). It has been well established that individuals who carry at least one copy of the YAAA haplotype or who have two copies of the XAAA haplotype have higher serum levels of MBL than individuals who carry the XAAA haplotype in combination with an O haplotype or two copies of an O haplotype (Table 1). Therefore, we used this approach to group patients in our cohort into a “high MBL” haplotype group and a “low MBL” haplotype group according to the pair of MBL haplotypes that they carried (Table 2). Patients in the low MBL haplotype group had significantly increased prevalence of positive bacterial cultures at admission to the ICU (p .02, Fig. 5). There was no association of MBL haplotype with a specific class of microorganism (Gram-positive vs. Gram-negative). Patients in the low MBL haplotype group did not have significantly increased rates of sepsis or septic shock at admission to the ICU. Twenty-eight-day survival did not differ significantly between the low MBL haplotype and high MBL haplotype groups (Fig. 6). Crit Care Med 2005 Vol. 33, No. 3 RESULTS Each gene was tested separately for association of genotype with clinical phenotype in the same cohort. There were no significant differences by genotype (for any of the genes evaluated) in age, gender (percent female), medical vs. surgical diagnosis for admission, or severity of illness at time of admission as estimated by APACHE II score (Table 2). Blood Collection and Processing The buffy coat was extracted from whole blood, and samples were transferred into 1.5-mL cryotubes and stored at 80°C. DNA was extracted from the buffy coat using the Qiagen DNA Blood Mini Kit. The genotypic analysis was performed in a blinded fashion, without clinical information. CD14 Two hundred forty-seven patients were successfully genotyped for the C-159T CD14 polymorphism. C-159T CD14 genotypes occurred in our cohort with frequencies similar to those previously reported in Caucasians (12, 45, 46) (Table 2) and are in Hardy-Weinberg equilibrium. Statistical Analysis We used a prospective cohort study design. Rates of dichotomous outcomes (presence or absence of positive bacterial Figure 1. Haplotype structure of the Toll-like receptor (TLR)-2 gene. Haplotypes of the TLR2 gene were inferred from unphased genotype data from 23 Caucasians using PHASE software. Columns are polymorphic sites of TLR2. Rows are haplotypes of TLR2 ordered by phylogenetic relationship. Yellow boxes are minor alleles and blue boxes are major alleles. Cladistic relationships of TLR2 haplotypes were determined using MEGA2 phylogenetic software. There are two major haplotype clades of TLR2, marked clades 1 and 2. T-16933A was chosen as a haplotype clade tag single nucleotide polymorphism (htSNP) as it distinguished between the two major haplotype clades. 640 Table 2. Cluster of differentiation (CD)-14, mannose-binding lectin (MBL), and Toll-like receptor (TLR)-2 genotype/haplotype frequencies and patient baseline characteristics by genotype/haplotype Baseline Characteristics Frequency, % CD14, n CC CT TT p MBL, n High Low p TLR2, n AA TA TT p 247 33 44 23 222 80 20 237 25 50 25 60 59 58 NS 15 17 17 37 33 38 NS 58 76 74 NS 23 21 20 NS 9 8 10 60 56 NS 16 18 37 32 NS 67 73 NS 21 20 NS 8 9 Figure 4. Kaplan-Meier survival analysis by cluster of differentiation (CD)-14 genotype. Patients’ survival was scored for 28 days or until hospital discharge. Kaplan-Meier analysis showed that patients homozygous for CD14 159T did not have different survival rates over the 28-day observation period (p .42) compared with patients homozygous or heterozygous for CD14 159C. 58 61 58 NS 17 15 18 31 39 31 NS 78 66 69 NS 21 21 19 NS 9 8 8 Age, Yrs Mean SD % Medical Diagnosis APACHE II Mean SD % Female APACHE, Acute Physiology and Chronic Health Evaluation; NS, not significant. Figure 2. Prevalence of positive bacterial cultures at intensive care unit admission by cluster of differentiation (CD)-14 C-159T genotype. Bacterial cultures were taken as part of routine medical care for any patients who were suspected of having an infection. Buffy coat was extracted from discarded whole blood. DNA was isolated by Qiagen DNA Blood Mini Kit; the CD14 promoter was amplified by polymerase chain reaction and subjected to restriction fragment length polymorphism. Association of CD14 C 159T genotype with prevalence of positive bacterial cultures was tested by chi-square analysis, and relative risk (RR) was calculated. Figure 3. Prevalence of Gram-negative bacterial cultures at intensive care unit admission by cluster of differentiation (CD)-14 C-159T genotype. Cultures were categorized as Gram-positive, Gram-negative, fungal, or other. Buffy coat was extracted from discarded whole blood. DNA was isolated by Qiagen DNA Blood Mini Kit; the CD14 promoter was amplified by polymerase chain reaction and subjected to restriction fragment length polymorphism. Association of CD14 C 159T genotype with type of bacterial culture was tested by chi-square analysis, and relative risk (RR) was calculated. TLR2 Two hundred thirty-seven patients were successfully genotyped for the TLR2 T-16933A polymorphism. The genotype frequencies of this polymorphism were similar to frequencies reported for other available Caucasian data (from www.innateimmunity.net/IIPGA/IIPGASNPs), and the genotypes were in Hardy-Weinberg equilibrium (Table 2). We found that the TLR2 16933AA genotype was associated with significantly increased prevalence of sepsis on Crit Care Med 2005 Vol. 33, No. 3 admission to the ICU (p .03, Fig. 7) and specifically with increased prevalence of Gram-positive infections (p .04, Fig. 8). TLR2 16933AA was not associated with increased prevalence of positive bacterial cultures or septic shock on admission to the ICU or with a significant difference in 28-day survival (Fig. 9). Figure 5. Prevalence of positive bacterial cultures at intensive care unit admission by mannosebinding lectin (MBL) haplotype pair. Bacterial cultures were taken as part of routine medical care for any patients who were suspected of having an infection. Buffy coat was extracted from discarded whole blood. DNA was isolated by Qiagen DNA Blood Mini Kit. DNA was amplified by site-directed mutagenesis polymerase chain reaction for the exon 1 alleles B, C, and D and subjected to restriction fragment length polymorphism. The Y/X promoter alleles were amplified by sequence-specific primer PCR. Association of low or high MBL haplotype pairs with prevalence of positive bacterial cultures was tested by chisquare analysis, and relative risk (RR) was calculated. DISCUSSION We found that CD14 159TT and low MBL haplotype pairs of MBL were associated with increased rates of positive bacterial cultures on admission to the ICU and that TLR2 16933AA was associated with increased prevalence of sepsis at ICU admission in a cohort of 252 critically ill Caucasians with SIRS. Furthermore, CD14 159CT and 159TT were associated with increased prevalence of Gramnegative infections, whereas TLR2 16933AA was associated with increased prevalence of Gram-positive infections. MBL haplotype pairs were not associated with prevalence of a specific type of infection. None of these polymorphisms of CD14, MBL, or TLR2 were associated with significantly increased prevalence of septic shock at admission or with 28-day survival. 641 Figure 6. Kaplan-Meier survival analysis by mannose-binding lectin (MBL) haplotype pair. Patients’ survival was scored for 28 days or until hospital discharge. Kaplan-Meier analysis showed that patients with low-level haplotype pairs did not have different survival rates over the 28-day observation period (p .34) compared with patients with high-level haplotype pairs. Figure 8. Prevalence of Gram-positive bacterial cultures at intensive care unit admission by Tolllike receptor (TLR)-2 T-16933A genotype. Cultures were categorized as Gram-positive, Gramnegative, fungal, or other. Buffy coat was extracted from discarded whole blood. DNA was isolated by Qiagen DNA Blood Mini Kit. The TLR2 promoter was amplified by polymerase chain reaction with sequence-specific primers so that genomic DNA was only amplified by a sequencespecific primer if the specific allele was contained in the sequence. Association of TLR2 T-16933A genotype with type of bacterial culture was tested by chi-square analysis, and relative risk (RR) was calculated. Figure 7. Prevalence of sepsis at intensive care unit admission by Toll-like receptor (TLR)-2 T-16933A genotype. A patient was considered septic at admission to the intensive care unit if two or more criteria for systemic inflammatory response syndrome were present plus a known or suspected infection. Buffy coat was extracted from discarded whole blood. DNA was isolated by Qiagen DNA Blood Mini Kit. The TLR2 promoter was amplified by polymerase chain reaction with sequence-specific primers so that genomic DNA was only amplified by a sequence-specific primer if the specific allele was contained in the sequence. Association of TLR2 T-16933A genotype with prevalence of sepsis at admission to the intensive care unit was tested by chi-square analysis, and relative risk (RR) was calculated. Figure 9. Kaplan-Meier survival analysis by Tolllike receptor (TLR)-2 T-16933A genotype. Patients’ survival was scored for 28 days or until hospital discharge. Kaplan-Meier analysis showed that patients homozygous for TLR2 16933A and patients homozygous or heterozygous for TLR2 16933T did not have different survival rates over the 28-day observation period (p .77). We have shown that polymorphisms in three key innate immunity receptor genes are associated with increased prevalence of infection in critically adults, marked by either positive microbial cultures or sepsis at admission to the ICU. In contrast, polymorphisms in CD14, MBL, and TLR2 were not associated with increased septic shock or altered survival, suggesting that polymorphisms in innate immunity receptors may increase susceptibility to infection but do not increase risk of shock or death in critically ill patients who have SIRS. 642 To our knowledge, this is the first report of an association of polymorphisms of CD14, MBL, and TLR2 with prevalence and type of infection and outcome in a common prospective cohort of critically ill patients. Several other groups have tested the association of polymorphisms in the CD14 and MBL genes separately with incidence of and outcome from sepsis in separate case-control studies. Two studies reported that the C-159T polymorphism was not associated with risk of severe sepsis in trauma patients or with increased risk of Gram-negative infection in critically ill patients (47, 48), whereas a third reported that the 159T allele was more frequent in patients with septic shock than in healthy controls and that the 159TT genotype was a risk factor for death in septic shock patients (45). Three structural mutations in exon 1 of the MBL gene (Gly54Asp, Gly57Glu, and Cys52Arg) (49, 19) and two promoter variants (G-550C and G-221C) of the MBL gene are associated with altered levels of serum MBL. MBL polymorphisms occur as six different haplotypes or sets of SNPs that are in linkage disequilibrium with one another and are inherited as a unit. These MBL haplotypes are associated with different serum concentrations of MBL (Table 1) (22–24). Low MBL haplotypes have been associated with increased prevalence of sepsis and septic shock and increased mortality rates in ICU patients (26). There have been no studies of the association of the TLR2 T-16933A polymorphism with infection or sepsis. Other polymorphisms in TLR2 have been associated with significantly increased risk of Grampositive infections and hyporesponsiveness to bacterial peptides (50 –52). Our finding of an association of innate immunity receptor genotype and increased prevalence of positive bacterial cultures and sepsis is consistent with recent animal model studies. Anti-CD14 antibodies administered before bacterial challenge caused increased bacterial load in both the lungs and vascular space, but attenuated septic shock, in rabbits (53, 54). CD14 also alters lethality of murine models of Gram-negative and lipopolysaccharide-induced shock (55, 56). TLR2-deficient mice are more susceptible to Gram-positive infection and have reduced clearance of bacteria (57, 58). Taken together, these studies suggest that polymorphisms of innate immunity receptors could be associated with impaired clearance of bacteria. Notably, our finding that polymorphisms in CD14 and TLR2 are associated with increased prevalence of Gramnegative and Gram-positive bacteria, respectively, suggests that polymorphisms of innate immunity receptors could be associated with impaired clearance of bacteria for which they are specific. Interestingly, we did not find that MBL haplotype pairs were associated with prevalence of a particular type of infection. Although MBL’s main role is in the recognition and clearing of Gram-negative bacteria, it also recognizes Gram-positive bacteria, yeasts, and viruses and so it may not be associated with susceptibility to a specific pathogen. Crit Care Med 2005 Vol. 33, No. 3 receptors may alter recognition and clearance of bacteria without changing outcomes of critically ill adults with systemic inflammatory response syndrome. S ingle nucleotide polymorphisms in innate immunity MBL, TLR2) were associated with increased prevalence of infection at admission to the ICU (prevalence of positive bacterial cultures [CD14, MBL]) or sepsis (TLR2) but not with septic shock or survival in critically ill adults. Importantly, we found that polymorphisms in CD14 and TLR2 are associated with increased prevalence of Gram-negative and Grampositive bacteria, respectively. It is possible that specific polymorphisms of innate immunity receptors alter host recognition and clearance of bacteria but surprisingly do not increase risk of septic shock or death in critically ill patients with SIRS. 11. 12. 13. 14. ACKNOWLEDGMENTS There are several strengths of our gene association study that minimized limitations of genetic association studies. First, our prospective study of a large cohort of critically ill patients (n 252) reduced the chance of type I error (finding a spurious association) compared with studies with smaller sample sizes. Second, to avoid spurious associations, we included only Caucasians in our cohort of critically ill adults, thus limiting the chance of population stratification due to ethnic heterogeneity (36). Third, we studied three separate but related innate immunity genes in the same patients, which allows direct comparison of the results of each individual gene association with outcome. There are limitations of our gene association study that are relevant. We did not measure RNA expression or protein levels of CD14, MBL, or TLR2, so we do not know the functional consequences of the polymorphisms of these genes in SIRS. Fortunately, several groups have examined the effects of the CD14 C-159T polymorphism and MBL haplotypes on relevant protein levels. CD14 159T has been associated with increased density of membrane-bound CD14 on monocytes and increased serum levels of soluble CD14 (12, 13, 59). Pairs of low MBL level haplotypes have been associated with decreased serum levels of MBL (22–24) and with an increased incidence of infection, as in chronic necrotizing pulmonary aspergillosis (60). We thank Shelley Dai, Yuliya Shapova, and Ali Alireza for their technical expertise. 15. REFERENCES 1. Burgner D, Levin M: Genetic susceptibility to infectious diseases. Pediatr Infect Dis J 2003; 22:1– 6 2. Bellamy R, Hill AV: Genetic susceptibility to mycobacteria and other infectious pathogens in humans. Curr Opin Immunol 1998; 10: 483– 487 3. Choi EH, Zimmerman PA, Foster CB, et al: Genetic polymorphisms in molecules of innate immunity and susceptibility to infection with Wuchereria bancrofti in South India. Genes Immun 2001; 2:248 –253 4. Majetschak M, Obertacke U, Schade FU, et al: Tumor necrosis factor gene polymorphisms, leukocyte function, and sepsis susceptibility in blunt trauma patients. Clin Diagn Lab Immunol 2002; 9:1205–1211 5. Mira JP, Cariou A, Grall F, et al: Association of TNF2, a TNF-alpha promoter polymorphism, with septic shock susceptibility and mortality: a multicenter study. JAMA 1999; 282:561–568 6. Read RC, Camp NJ, di Giovine FS, et al: An interleukin-1 genotype is associated with fatal outcome of meningococcal disease. J Infect Dis 2000; 182:1557–1560 7. Wright SD, Ramos RA, Tobias PS, et al: CD14, a receptor for complexes of lipopolysaccharide (LPS) and LPS binding protein. Science 1990; 249:1431–1433 8. Chow JC, Young DW, Golenbock DT, et al: Toll-like receptor-4 mediates lipopolysaccharide-induced signal transduction. J Biol Chem 1999; 274:10689 –10692 9. Labeta MO, Durieux JJ, Fernandez N, et al: Release from a human monocyte-like cell line of two different soluble forms of the lipopolysaccharide receptor, CD14. Eur J Immunol 1993; 23:2144 –2151 10. Landmann R, Zimmerli W, Sansano S, et al: Increased circulating soluble CD14 is associ- 16. 17. 18. 19. 20. 21. 22. 23. 24. CONCLUSIONS We found that polymorphisms of three important innate immunity genes (CD14, Crit Care Med 2005 Vol. 33, No. 3 ated with high mortality in Gram-negative septic shock. J Infect Dis 1995; 171:639 – 644 Burgmann H, Winkler S, Locker GJ, et al: Increased serum concentration of soluble CD14 is a prognostic marker in Grampositive sepsis. Clin Immunol Immunopathol 1996; 80:307–310 Hubacek JA, Rothe G, Pit’ha J, et al: C(260)– T polymorphism in the promoter of the CD14 monocyte receptor gene as a risk factor for myocardial infarction. Circulation 1999; 99:3218 –3220 Baldini M, Lohman IC, Halonen M, et al: A polymorphism* in the 5' flanking region of the CD14 gene is associated with circulating soluble CD14 levels and with total serum immunoglobulin E. Am J Respir Cell Mol Biol 1999; 20:976 –983 Weis WI, Drickamer K: Trimeric structure of a C-type mannose-binding protein. Structure 1994; 2:1227–1240 Kuhlman M, Joiner K, Ezekowitz RA: The human mannose-binding protein functions as an opsonin. J Exp Med 1989; 169: 1733–1745 Tabona P, Mellor A, Summerfield JA: Mannose binding protein is involved in first-line host defence: Evidence from transgenic mice. Immunology 1995; 85:153–159 Schelenz S, Malhotra R, Sim RB, et al: Binding of host collectins to the pathogenic yeast Cryptococcus neoformans: Human surfactant protein D acts as an agglutinin for acapsular yeast cells. Infect Immun 1995; 63: 3360 –3366 van Emmerik LC, Kuijper EJ, Fijen CA, et al: Binding of mannan-binding protein to various bacterial pathogens of meningitis. Clin Exp Immunol 1994; 97:411– 416 Madsen HO, Garred P, Kurtzhals JA, et al: A new frequent allele is the missing link in the structural polymorphism of the human mannan-binding protein. Immunogenetics 1994; 40:37– 44 Lipscombe RJ, Sumiya M, Hill AV, et al: High frequencies in African and non-African populations of independent mutations in the mannose binding protein gene. Hum Mol Genet 1992; 1:709 –715 Sumiya M, Super M, Tabona P, et al: Molecular basis of opsonic defect in immunodeficient children. Lancet 1991; 337:1569 –1570 Madsen HO, Garred P, Thiel S, et al: Interplay between promoter and structural gene variants control basal serum level of mannan-binding protein. J Immunol 1995; 155: 3013–3020 Steffensen R, Thiel S, Varming K, et al: Detection of structural gene mutations and promoter polymorphisms in the mannanbinding lectin (MBL) gene by polymerase chain reaction with sequence-specific primers. J Immunol Methods 2000; 241:33– 42 Crosdale DJ, Ollier WE, Thomson W, et al: Mannose binding lectin (MBL) genotype distributions with relation to serum levels in UK Caucasoids. Eur J Immunogenet 2000; 27:111–117 643 25. Fidler KJ, Wilson P, Davies JC, et al: Increased incidence and severity of the systemic inflammatory response syndrome in patients deficient in mannose-binding lectin. Intensive Care Med 2004; 30:1438 –1445 26. Garred P, Strom J, Quist L, et al: Association of mannose-binding lectin polymorphisms with sepsis and fatal outcome, in patients with systemic inflammatory response syndrome. J Infect Dis 2003; 188:1394 –1403 27. Lien E, Sellati TJ, Yoshimura A, et al: Tolllike receptor 2 functions as a pattern recognition receptor for diverse bacterial products. J Biol Chem 1999; 274:33419 –33425 28. Aliprantis AO, Yang RB, Mark MR, et al: Cell activation and apoptosis by bacterial lipoproteins through toll-like receptor-2. Science 1999; 285:736 –739 29. Brightbill HD, Libraty DH, Krutzik SR, et al: Host defense mechanisms triggered by microbial lipoproteins through toll-like receptors. Science 1999; 285:732–736 30. Takeda K, Akira S: Toll receptors and pathogen resistance. Cell Microbiol 2003; 5:143–153 31. Travassos LH, Girardin SE, Philpott DJ, et al: Toll-like receptor 2-dependent bacterial sensing does not occur via peptidoglycan recognition. EMBO Rep 2004; 5:1000 –1006 32. Lotz S, Eresso A, Wilde I, et al: Highly purified lipoteichoic acid activates neutrophil granulocytes and delays their spontaneous apoptosis via CD14 and TLR2. J Leukoc Biol 2004; 75:467– 477 33. Ellingsen E, Morath S, Flo T, et al: Induction of cytokine production in human T cells and monocytes by highly purified lipoteichoic acid: Involvement of Toll-like receptors and CD14. Med Sci Monit 2002; 8:BR149 –BR156 34. Morath S, Stadelmaier A, Geyer A, et al: Synthetic lipoteichoic acid from Staphylococcus aureus is a potent stimulus of cytokine release. J Exp Med 2002; 195:1635–1640 35. Bone RC: The sepsis syndrome. Definition and general approach to management. Clin Chest Med 1996; 17:175–181 36. Cardon LR, Bell JI: Association study designs for complex diseases. Nat Rev Genet 2001; 2:91–99 37. Knaus WA, Wagner DP, Draper EA, et al: The APACHE III prognostic system. Risk predic- 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. tion of hospital mortality for critically ill hospitalized adults. Chest 1991; 100: 1619 –1636 Knaus WA, Draper EA, Wagner DP, et al: APACHE II: A severity of disease classification system. Crit Care Med 1985; 13: 818 – 829 Stephens M, Smith NJ, Donnelly P: A new statistical method for haplotype reconstruction from population data. Am J Hum Genet 2001; 68:978 –989 Rats RU: TLR2. Innate immunity PGA. Bethesda, MD, NHLBI Program in Genomic Applications, 2002 Kumar S, Tamura K, Jakobsen IB, et al: MEGA2: Molecular evolutionary genetics analysis software. Bioinformatics 2001; 17: 1244 –1245 Johnson GC, Esposito L, Barratt BJ, et al: Haplotype tagging for the identification of common disease genes. Nat Genet 2001; 29: 233–237 Gabriel SB, Schaffner SF, Nguyen H, et al: The structure of haplotype blocks in the human genome. Science 2002; 296:2225–2229 Guo SW, Thompson EA: Performing the exact test of Hardy-Weinberg proportion for multiple alleles. Biometrics 1992; 48: 361–3672 Gibot S, Cariou A, Drouet L, et al: Association between a genomic polymorphism within the CD14 locus and septic shock susceptibility and mortality rate. Crit Care Med 2002; 30:969 –973 Lichy C, Meiser H, Grond-Ginsbach C, et al: Lipopolysaccharide receptor CD14 polymorphism and risk of stroke in a South-German population. J Neurol 2002; 249:821– 823 Heesen M, Bloemeke B, Schade U, et al: The -260 C– T promoter polymorphism of the lipopolysaccharide receptor CD14 and severe sepsis in trauma patients. Intensive Care Med 2002; 28:1161–1163 Agnese DM, Calvano JE, Hahm SJ, et al: Human toll-like receptor 4 mutations but not CD14 polymorphisms are associated with an increased risk of Gram-negative infections. J Infect Dis 2002; 186:1522–1525 Wallis R, Cheng JY: Molecular defects in variant forms of mannose-binding protein asso- 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. ciated with immunodeficiency. J Immunol 1999; 163:4953– 4959 Bochud PY, Hawn TR, Aderem A: Cutting edge: a Toll-like receptor 2 polymorphism that is associated with lepromatous leprosy is unable to mediate mycobacterial signaling. J Immunol 2003; 170:3451–3454 Kang TJ, Lee SB, Chae GT: A polymorphism in the toll-like receptor 2 is associated with IL-12 production from monocyte in lepromatous leprosy. Cytokine 2002; 20:56 – 62 Lorenz E, Mira JP, Cornish KL, et al: A novel polymorphism in the toll-like receptor 2 gene and its potential association with staphylococcal infection. Infect Immun 2000; 68: 6398 – 6401 Opal SM, Palardy JE, Parejo N, et al: Effect of anti-CD14 monoclonal antibody on clearance of Escherichia coli bacteremia and endotoxemia. Crit Care Med 2003; 31:929 –932 Frevert CW, Matute-Bello G, Skerrett SJ, et al: Effect of CD14 blockade in rabbits with Escherichia coli pneumonia and sepsis. J Immunol 2000; 164:5439 –5445 Tasaka S, Ishizaka A, Yamada W, et al: Effect of CD14 blockade on endotoxin-induced acute lung injury in mice. Am J Respir Cell Mol Biol 2003; 29:252–258 Ferrero E, Jiao D, Tsuberi BZ, et al: Transgenic mice expressing human CD14 are hypersensitive to lipopolysaccharide. Proc Natl Acad Sci U S A 1993; 90:2380 –2384 Echchannaoui H, Frei K, Schnell C, et al: Toll-like receptor 2-deficient mice are highly susceptible to Streptococcus pneumoniae meningitis because of reduced bacterial clearing and enhanced inflammation. J Infect Dis 2002; 186:798 – 806 Drennan MB, Nicolle D, Quesniaux VJ, et al: Toll-like receptor 2-deficient mice succumb to Mycobacterium tuberculosis infection. Am J Pathol 2004; 164:49 –57 Karhukorpi J, Yan Y, Niemela S, et al: Effect of CD14 promoter polymorphism and H. pylori infection and its clinical outcomes on circulating CD14. Clin Exp Immunol 2002; 128:326 –332 Yoon HJ, Shin JH, Yang SH, et al: Association of the CD14 gene -159C polymorphism with progression of IgA nephropathy. J Med Genet 2003; 40:104 –108 644 Crit Care Med 2005 Vol. 33, No. 3