|Home | About | Journals | Submit | Contact Us | Français|
Gene polymorphisms, giving rise to low serum levels of mannose-binding lectin (MBL) or MBL-associated protease 2 (MASP2), have been associated with an increased risk of infections. The objective of this study was to assess the outcome of intensive care unit (ICU) patients with systemic inflammatory response syndrome (SIRS) regarding the existence of functionally relevant MBL2 and MASP2 gene polymorphisms. The study included 243 ICU patients with SIRS admitted to our hospital, as well as 104 healthy control subjects. MBL2 and MASP2 single nucleotide polymorphisms were genotyped using a sequence-based typing technique. No differences were observed regarding the frequencies of low-MBL genotypes (O/O and XA/O) and MASP2 polymorphisms between patients with SIRS and healthy controls. Interestingly, ICU patients with a noninfectious SIRS had a lower frequency for low-MBL genotypes and a higher frequency for high-MBL genotypes (A/A and A/XA) than either ICU patients with an infectious SIRS or healthy controls. The existence of low- or /high-MBL genotypes or a MASP2 polymorphism had no impact on the mortality rates of the included patients. The presence of high-MBL-producing genotypes in patients with a noninfectious insult is a risk factor for SIRS and ICU admission.
Sepsis is the main cause of death in intensive care units (ICU), with mortality rates above 50% in patients with septic shock (54). Increasing evidence suggests that variations in genes encoding different components of the immune system influence an individual's capacity to respond adequately to infections. Genetic polymorphisms of several molecules of the innate immune system, such as tumor necrosis factor alpha (TNF-α) (35), interleukin-1 receptor antagonist (IL-1RA) (2), and more recently plasminogen activator inhibitor 1 (PAI-1) (12), have been associated with increased mortality in patients with severe sepsis and septic shock.
The mannose-binding lectin (MBL) is an important element of the innate immune defense system. The MBL is a circulating C-type plasma lectin, mainly produced by the liver, which binds to the specific carbohydrates present on the surface of different microorganisms (21, 38). In serum, MBL is present as oligomers (mainly trimers and tetramers) bound to the MBL-associated serine proteases (MASPs, which are also produced in the liver), mainly MASP2. Once bound to the carbohydrate residues, the MBL/MASP2 complex acts as an opsonin for phagocytosis for numerous pathogens and activates complement (21, 34, 38, 56). MBL is not only involved in complement activation but also is a potent modulator of proinflammatory cytokine production (27). Additionally, MBL is capable of increasing the clearance of endotoxin via Kupffer cells (40) and increasing the turnover of fibrinogen by cleavage of prothrombin, generating thrombin (29).
Three missense single nucleotide polymorphisms (SNP) have been reported within exon 1 of the MBL2 gene, introducing amino acid replacements at codon 52 (allele D), 54 (allele B), or 57 (allele C), which cause a reduction of the MBL levels due to impaired assembly of MBL monomers into functional oligomers (14). In addition to these structural variant alleles, three SNP in the promoter region of the MBL2 gene at positions −550 (H/L), +4 (P/Q) and, particularly, −221 (Y/X), influence the rate of transcription and are also associated with low concentrations of serum MBL (32, 47). Genetically defined MBL deficiency is common and appears to predispose to serious infections (9), particularly during early childhood (28), in patients undergoing chemotherapy (41) as well as in adults with concomitant diseases (15, 16). In the ICU setting, although several studies have suggested the existence of a relationship between low MBL serum levels and an increased risk of infections, the association with death has yielded conflicting results (17, 19, 24, 25, 50).
In addition to the MBL2 polymorphisms, an inherited deficiency of MASP2 has also been reported. This deficiency is due to a homozygous mutation in exon 3 of the MASP2 gene, resulting in a change of aspartic acid to glycine at position 105 (Asp105Gly) of the CUB1 protein domain, which is an essential region for the formation of functional MBL/MASP2 complexes. This mutation renders MASP2 incapable of binding to MBL and therefore interrupts the MBL pathway of complement activation and also reduces the plasma concentration of MASP2 (48). Patients heterozygous for the MASP2 Asp 105Gly SNP have no impairment in the lectin complement pathway (10, 11, 51). Several additional variants have been identified in exon 3 of the MASP2 gene that do not cause a reduction of the levels or activity of the protein (31, 52).
The possible association between MASP2 deficiency and susceptibility to infections remains largely unknown. Schlapbach et al. recently demonstrated that MASP2 deficiency increases the risk of neutropenic fever in pediatric cancer patients (43). Granell et al. found an increased risk of invasive aspergillosis following allogeneic stem cell transplantation in adult patients heterozygous for the Asp105Gly SNP (20). On the other hand, three recent studies were unable to demonstrate an increased frequency for the Asp105Gly SNP in adult patients with community-acquired pneumonia, pneumococcal bacteremia in HIV-infected patients, and among renal transplant recipients with infectious complications (5, 10, 26).
The aim of the present study was to investigate the implications of MBL2 and MASP2 polymorphisms in the outcome of ICU patients with systemic inflammatory response syndrome (SIRS).
We prospectively collected blood samples from 243 Caucasian patients admitted to the medical ICU of the Hospital Clinic of Barcelona between January 2003 and January 2004. Inclusion criteria for patients were age of >18 with a minimum ICU stay of 24 h and meeting the criteria for SIRS (see below). The patients were included and followed until hospital discharge. For further comparison, 104 healthy Caucasian blood donors from the geographic area of Barcelona were also included in the study. The present study was conducted with the approval of the hospital Ethics Committee and informed consent from the patients or their relatives within 24 h after admission.
Definitions for community- or nosocomial-acquired infection in ICU patients were established according to the Sepsis Forum Consensus Conference and the CDC guidelines (4, 13). The criteria for SIRS, sepsis, severe sepsis, and septic shock were defined according to the SCCM/ESICM/ACCP/ATS/SIS consensus conference (30). Clinical data, including demographic details and the severity indexes (Acute Physiology and Chronic Health Evaluation II [APACHE II], Simplified Acute Physiology Score II [SAPS II], and Sequential Organ Failure Assessment [SOFA]) were recorded for each patient at ICU admission and thereafter on a daily basis. Multiorgan failure was considered in cases of acute progressive dysfunction of two or more organ systems, with a minimum failure score of 3 points for each organ.
Genomic DNA was extracted from 1.5 ml of EDTA-treated whole blood samples by using the QIAamp DNA blood minikit according to the manufacturer's instructions (Qiagen GmbH, Hilden, Germany) and then stored at −80°C until used. Genotyping of the MBL2 and the MASP2 genes was performed by using a sequencing-based typing method according to the published sequences (GenBank accession numbers AF360991 for the MBL2 gene and NG_007289 for the MASP2 gene). Six SNP in the MBL2 gene (−550 G/C [rs11003125], −221 C/G [rs7096206], +4 C/T [rs7095891], codon 52 CGT/TGT [rs5030737], codon 54 GGC/GAC [rs1800450], and codon 57 GGA/GAA [rs1800451]) within the promoter and exon 1 of the MBL2 gene were analyzed as previously reported (31). Briefly, a 969-bp fragment encompassing from the promoter to the end of exon 1 of the MBL2 gene was obtained by PCR amplification using the sense primer 5′-GGGGAATTCCTGCCAGAAAGT-3′ and antisense primer 5′-CATATCCCCAGGCAGTTTCCTC-3′ and the Expand 20 kbPLUS PCR system (Roche Diagnostics GmbH, Mannheim, Germany). Five SNP in the MASP2 gene (Pro111Leu [rs56392418], Asp105Gly [rs72550870], Arg84Gln [rs61735600], Thr73Met [rs61735601], and Arg103Cys [rs id not submitted]) within exon 3 of the MASP2 gene were analyzed as previously reported (31). Similarly, a 354-bp fragment from exon 3 of the MASP2 gene was PCR amplified by using the sense primer 5′-GCGAGTACGACTTCGTCAAGG-3′ and antisense primer 5′-CTCGGCTGCATAGAAGGCCTC-3′ and the Expand high-fidelity PCR system (Roche Diagnostics GmbH, Mannheim, Germany). The cycling conditions for amplification were 94°C for 8 min, 35 cycles at 94°C for 45 s, 58°C for 30 s, and 72°C for 90 s, and finally 72°C for 10 min. Five microliters of the resulting PCR product was treated with ExoSAP-IT (USB Corporation, Cleveland, OH) and then subjected to direct sequencing with the BigDye Terminator v1.1 cycle sequencing kit (Applied Biosystems, Warrington, United Kingdom) following the manufacturer's instructions with the sense and antisense gene-specific primers mentioned above.
The SNP at exon 1 of the MBL2 gene are in strong linkage disequilibrium with those at the promoter and give rise to seven common haplotypes (HYPA, LYQA, LYPA, LXPA, LYPB, LYQC, and HYPD). The HY haplotype induces high MBL concentrations, whereas exon 1 mutations (O variants) and the LX haplotypes cause reduced MBL concentrations (47). Therefore, the analysis of exon 1 and the promoter region of the MBL2 gene allowed the categorization of individuals into three groups according to their MBL2 genotype. Group I included genotypes responsible for high MBL levels (A/A and A/XA), group II genotypes were responsible for intermediate MBL levels (A/O and XA/XA) and, finally, group III genotypes were responsible for low MBL levels (O/O and XA/O), according to previously published results (32). The MASP2 genotype was divided into wild type or mutant (Asp105Gly).
Continuous variables were compared with Student's t test or the Mann-Whitney U test when the distribution departed from normality and are described as means (±standard deviations) or medians (and ranges of values), respectively. Categorical data were compared by the chi-square or the Fisher's exact test as appropriate. Deviations from Hardy-Weinberg expectations were tested using the chi-square test for comparing observed and expected values. Statistical significance was defined as a two-tailed P value of <0.05. Statistical analysis was carried out with the program SPSS (version 15.0; SPSS, Inc., Chicago, IL).
The study population consisted of 243 consecutive adult patients (mean age, 62.8 ± 15.7 years; range, 18 to 85 years; 63% men). The mean APACHE II, SAPS II, and SOFA scores at ICU entry were 17.4 ± 6.4, 37 ± 11.9, and 8.7 ± 3.1, respectively. Within 24 h of ICU admission 101 patients (41.5%) met the criteria for septic shock and 28 (11.5%) for severe sepsis. For the remaining patients, 54 (22.2%) met the criteria for sepsis and 60 (24.7%) for a noninfectious SIRS. The main clinical characteristics of the patients included in the study are shown in Table Table1.1. Most of the patients had a medical indication for ICU admission, mainly a community-acquired pneumonia. In the group of septic shock patients, nosocomial pneumonia was the focus of infection in nearly 10% of ICU admissions. Additionally, 13% of the patients developed nosocomial pneumonia during ICU hospitalization.
MBL2 was genotyped in 216 patients (88.8% of the included patients) and MASP2 in 240 patients (98.7% of the included patients). MBL2 and MASP2 were also genotyped in the 104 healthy blood donor volunteers. Genotype frequencies found for patients and control groups did not differ from those predicted by the Hardy-Weinberg expectations (P = 0.43 and P = 0.072 for healthy controls and ICU patients with SIRS, respectively, regarding MBL2 genotypes; P = 0.88 and P = 0.66 for healthy controls and ICU patients with SIRS, respectively, regarding MASP2 genotypes). Table Table22 shows the frequencies for the MBL2 and MASP2 genotypes found among the ICU patients and the healthy control group. No significant differences in the frequencies for the different MBL2 haplotypes were found between the ICU patients with SIRS and the healthy controls (data not shown). LYPB was the predominant MBL2 variant haplotype in both the ICU patients with SIRS and in the healthy controls.
No overall statistically significant differences were observed among the patients with SIRS and the healthy controls, respectively, for the frequencies of the genotypes high MBL2 (48.6% versus 54.8%; P = 0.29), intermediate MBL2 (38.4% versus 29.8%; P = 0.13), and low MBL2 (12.9% versus 15.3%; P = 0.55) (Table (Table2).2). The group of patients with an infectious SIRS, the analysis of which always included patients with sepsis, severe sepsis, and septic shock, had an increased frequency for low-MBL2 genotypes (15.9%) when compared with the frequency found among patients with a noninfectious SIRS (3.8%; P = 0.025). When comparisons were established with the healthy control group, no differences could be found (15.9% versus 15.3%; P = 0.91). In view of the results we decided to evaluate the frequencies for high-MBL2 genotypes and observed a higher prevalence of these genotypes among ICU patients with a noninfectious SIRS compared to the frequencies found in ICU patients with an infectious SIRS (63.4% versus 43.9%; P = 0.014). A higher frequency for high-MBL2 genotypes was also found when comparisons were established between ICU patients with a noninfectious SIRS and the healthy control group, although these differences did not reach statistical significance (63.4% versus 54.8%; P = 0.3).
We also analyzed the severity indexes at ICU entry among patients with a noninfectious SIRS and observed higher APACHE II (19.7 ± 5.5 versus 10 ± 1.4; P = 0.02), SAPS II (34 ± 8.7 versus 32 ± 8.5; P = 0.75), and SOFA scores (8.8 ± 2.8 versus 5 ± 0; P = 0.04) in patients with high-MBL2 genotypes than in patients with low-MBL2 genotypes. Although infectious SIRS patients with high-MBL2 genotypes also had higher APACHE II (16 ± 6.8 versus 16 ± 7.3; P = 0.9), SAPS II (37.7 ± 12.2 versus 35.6 ± 13.7; P = 0.54), and SOFA (9.1 ± 3.4 versus 8.9 ± 3.4; P = 0.84) scores than patients with low-MBL2 genotypes, differences were not statistically significant.
The analysis of exon 3 of MASP2 revealed no homozygous Asp105Gly carriers among patients or controls. As shown in Table Table2,2, no significant differences were found for the frequency of the heterozygous MASP2 Asp105Gly SNP between the ICU patients with SIRS (5%) and the healthy controls (2.8%; P = 0.56). One ICU patient included in the sepsis group was heterozygous for the Pro111Leu polymorphism, while none of the healthy controls had it. No differences were found in the frequency for the MASP2 Asp105Gly SNP between the group of ICU patients with an infection (5%) and those with a noninfectious SIRS (5%).
Table Table33 shows the outcomes for the patients during ICU stay according to the existence of normal- or low-MBL2-producing genotypes and the presence of the wild-type MASP2 or the heterozygous Asp105Gly SNP. No differences in the length of hospital or ICU stay were seen between the different groups. The frequencies and the total days of invasive mechanical ventilation or for renal replacement techniques were also similar among the groups. Fifty-three (21.8%) patients acquired a nosocomial infection while hospitalized in the ICU, 32 cases (13%) of which were a pneumonia. No differences in the global incidence of nosocomial infection or in the frequency of hospital-acquired pneumonia were seen between the group of ICU patients with low-MBL2 genotypes or with the heterozygous Asp105Gly SNP and the rest of the patients. We also compared the clinical outcomes of ICU patients with high-MBL2 genotypes and those with low-MBL2 genotypes, excluding from the analysis those with intermediate-MBL2 genotypes with regard to the clinical variables exposed in Table Table3,3, and we found nonsignificant differences between them (data not shown).
Regarding the mortality rate, 73 (30%) of the patients died during their hospital stay, 54 (22.2%) of whom died during their ICU stay. Multiorgan failure was the main cause of death (68.5%), although its frequency was different among ICU patients with a noninfectious SIRS and those with an infectious cause of SIRS (38.8% versus 76.3%; P = 0.003). Again, the existence of low-producing MBL2 genotypes or the heterozygous MASP2 Asp105Gly SNP had no influence on the mortality rate. When mortality was analyzed with consideration taken for the existence of high, intermediate, or low MBL2-producing genotypes, no significant differences were found (21.4%, 33.7%, and 29.5% respectively).
Microbial samples obtained at ICU entry were positive in 93 (50.8%) of the patients with an infectious SIRS. No differences were found regarding the frequencies for Gram-positive, Gram-negative, or fungi among patients with low-MBL2 genotypes or heterozygous for the Asp105Gly SNP (Table (Table33).
Despite the intense efforts that have been made to reduce the mortality related to sepsis, mainly driven through the implementation of the “Surviving Sepsis Campaign” guidelines (6), mortality rates continue to be high, around 40 to 70% in patients with septic shock (42). It has been estimated that each year in the United States 750,000 new cases of sepsis are diagnosed, and its frequency is rising due to an increasing aging population, the use of invasive diagnostic procedures, and aggressive therapies, together with the high prevalence of chronic diseases (33).
Although both the innate and adaptive immune systems are involved in the pathogenesis of sepsis, the innate immune system plays a pivotal role. Moreover, it has been suggested that susceptibility and response to infectious disease might be inheritable (3, 45). Numerous studies have evaluated different components of the innate immune system, including the MBL, searching for SNP associated with an increased risk for developing more severe forms of sepsis, organ dysfunction, or death in patients with sepsis (2, 17, 19, 24, 25, 35, 50).
The serum concentration and functional activity of MBL is determined by SNP at the promoter and exon 1 of the MBL2 gene, while SNP at exon 3 of the MASP2 gene, mainly the Asp105Gly SNP, cause a reduction in the serum levels of MASP2. In the present study nonsignificant differences for the frequencies of MBL2 haplotypes and genotypes were observed between ICU patients with SIRS and healthy controls. The prevalence of low-MBL2 genotypes was similar to the frequency observed by Garcia-Laorden et al. in Spain (10) and others in previous studies (17, 18). LYPB, as previously reported in other Caucasian populations (1, 46), was the predominant variant type of haplotype both in the group of patients with SIRS and in the healthy blood donors. Regarding the MASP2 gene SNP, neither patients nor healthy controls were homozygous for the Asp105Gly SNP. No significant differences in the frequency for the heterozygous MASP2 Asp105Gly were seen between ICU patients with SIRS and the healthy controls, the prevalence of which was in turn similar to that found in previous studies (10, 48). One Caucasian patient carried the MASP2 Pro111Leu SNP, as primarily described by Lozano et al. in North African individuals, which is not capable of causing reductions in the serum levels or activity of the MASP-2 (31, 52).
Previous studies have demonstrated a high prevalence of MBL2-deficient genotypes among patients with sepsis admitted to ICU (17, 19, 50). Accordingly, our data showed that patients with an infectious SIRS had a higher frequency for low-MBL2 genotypes than patients with a noninfectious SIRS. This was due to the higher prevalence of high-MBL2 genotypes observed in ICU patients with a noninfectious SIRS. The observation of a higher frequency for high-MBL2 genotypes among ICU patients with a noninfectious SIRS is coincidental with that previously reported by Garred et al. (17).
SIRS describes physiological and laboratory abnormalities that accompany inflammation independently from the original cause (30). In a recent study, Dulhunty et al. demonstrated that patients with a noninfectious SIRS present clinical differences from patients with sepsis (7). According to this study patients with sepsis died more frequently from multiorgan failure than patients with nonseptic SIRS, which was also observed in our study. We hypothesize that some of the clinical differences observed between noninfectious and infectious SIRS patients might reflect, at least in part, the existence of different subjacent physiological mechanisms.
The results of our study suggest that patients with a noninfectious insult and high-MBL2 genotype are at risk of developing SIRS and require ICU admission without having significant infectious complications. High levels of functional MBL could directly be associated with the proinflammatory adverse effects following uncontrolled complement activation, which has been previously demonstrated in ischemic-reperfusion experimental mice models (22, 36, 55). According to these models, excess production of MBL or the administration of exogenous MBL following induced ischemia causes organ (myocardial, kidney, intestine) reperfusion injury due to complement activation, while low MBL levels are protective. One of the hallmarks of reperfusion to ischemic tissues is the severe oxidative stress that occurs at the level of the endothelium, which causes vascular injury. It has been suggested that the lectin pathway initiates complement activation following oxidative stress, particularly after myocardial, intestinal, and skeletal muscle ischemic-reperfusion-induced injury, at least in experimental models (23). Additionally, MBL has been reported to recognize apoptotic and necrotic cells (37, 39), which could also trigger activation of complement through the lectin pathway in patients with high-MBL2 genotypes and a noninfectious SIRS. In fact a defective apoptotic clearance has been found in MBL-defective mice (49). Unfortunately we did not measure the level of inflammation or complement activation in our patients, and therefore we could not demonstrate a higher degree of activation, as we would have expected from our results, in patients with a noninfectious SIRS and high-MBL2 genotypes. However, this group of patients had higher severity index scores, particularly a higher SOFA score, which measures organ failure due to endothelium injury.
The analysis of the outcomes of the ICU patients with SIRS, particularly for mortality, revealed no differences regarding the existence of low-MBL2 genotypes. Also, no differences were seen between low- and high-MBL2 genotypes (data not shown). The relationship between mortality and the presence of low-MBL2 genotype continues to be a controversial issue, with studies supporting this association (17, 19, 24) and others like the present one finding no relationship (25, 50). In a recent meta-analysis by Eisen et al. (8), only a trend toward an increased mortality among patients with low-MBL2 genotypes and bacterial infections was found. We could not observe a better outcome in patients with intermediate-MBL2 genotypes, as demonstrated by Helleman et al. (24). It should be noted that one of the limitations of our study is the short follow-up period of the patients. Nowadays, with the application of increasingly effective organ support treatments, withdrawal of life support therapy is the most common terminal event in ICU patients, and thus longer-term mortality analyses are recommended (53).
Regarding the genotyping of exon 3 of MASP2, no significant differences were found in the frequency of the heterozygous MASP2 Asp105Gly SNP between patients with an infectious versus a noninfectious SIRS. The MASP2 Asp105Gly SNP was not associated with any of the clinical variables associated with an increased severity of disease at ICU admission, including mortality. Our results were similar to those reported by Garcia-Laorden et al. in patients with community-acquired pneumonia (10), but not to the ones from the study of Henckaets et al., who observed an association between the existence of the heterozygous MASP2 Asp105Gly SNP and higher mortality rates in ICU patients (25).
Our study was not capable of finding an increased prevalence of any microorganism or between Gram-positive and -negative bacteria among patients with low-MBL2 genotypes. While some of the published studies have found an association between Gram-positive infections and MBL deficiency (8, 24), others have observed a link between Gram-negative infections and MBL deficiency (44).
The present study has important design limitations that have to be taken into consideration. First, our control group was composed of healthy unmatched blood donors. Although the use of healthy blood donors as a control group has commonly been used in genetic studies, it has the potential risk of misclassification of individuals, particularly when evaluating a condition with a high penetrance, such as MBL deficiency. Second, due to the retrospective design of the study, no power calculation was performed, which could seriously affect its capacity to detect small differences, particularly when analyzing the different subgroups of SIRS patients. Therefore, additional independent studies are required to replicate our observations.
In conclusion, this study shows that low-MBL2 genotypes were significantly underrepresented among patients with a noninfectious SIRS and were not associated with more severe forms of sepsis or with death. The presence of MASP2 polymorphisms was not related to infectious SIRS and had no impact on the prognosis of ICU patients with SIRS. Thus, high-MBL producers seem to be at risk of developing a noninfectious SIRS, while low-MBL producers are not likely to be at higher risk of developing severe forms of infectious SIRS.
This research was supported by a grant from the Fondo de Investigación Sanitaria (050164).
Published ahead of print on 30 December 2009.