From the Department of Medicine (Drs Lorenz and Schwartz and Ms Frees) and the Department of Veterans Affairs Medical Center (Drs Lorenz and Schwartz), The University of Iowa, Iowa City; and Medical Intensive Care Unit, Cochin University Hospital, Paris, France (Dr Mira). Dr Schwartz is now with Pulmonary and Critical Care Medicine, Duke University Medical Center, Durham, NC.
Septic shock remains a significant health concern worldwide, and despite progress in understanding the physiological and molecular basis of septic shock, the high mortality rate of patients with septic shock remains unchanged. We recently identified a common polymorphism in toll-like receptor 4 (TLR4) that is associated with hyporesponsiveness to inhaled endotoxin or lipopolysaccharide in humans.
Since TLR4 is a major receptor for lipopolysaccharide in mammals and gram-negative bacteria are the prevalent pathogen associated with septic shock, we investigated whether these specific TLR4 alleles are associated with a predisposition to a more severe disease outcome for patients with septic shock. We genotyped 91 patients with septic shock as well as 73 healthy blood donor controls for the presence of the TLR4 Asp299Gly and TLR4 Thr399Ile mutations.
We found the TLR4 Asp299Gly allele exclusively in patients with septic shock (P = .05). Furthermore, patients with septic shock with the TLR4 Asp299Gly/Thr399Ile alleles had a higher prevalence of gram-negative infections.
Mutations in the TLR4 receptor may predispose people to develop septic shock with gram-negative microorganisms.
DESPITE SIGNIFICANT advances in understanding the molecular basis of sepsis and its associated immunologic response, sepsis remains a problem worldwide and is associated with a high mortality. Annually, septic shock, the most severe form of sepsis, causes the death of more than 100 000 people in the United States.1,2 Gram-negative bacteria are the most common pathogens associated with bacterial infections in diseases such as meningitis. Endotoxin or lipopolysaccharide (LPS), the main component of the cell wall of gram-negative bacteria, has been shown to elicit an inflammatory response that mimics all of the features described in septic shock.3 In vivo and in vitro, LPS activates macrophages and monocytes, as shown by increased secretion of cytokines such as tumor necrosis factor (TNF) α, interleukin (IL) 1, IL-6, and IL-8.4,5 Activation of an innate immune mechanism causes proinflammatory cytokine release to be magnified to dangerous levels, resulting in hypotension, organ failures, and even death in a significant number of patients.1,2 A likely target for novel therapeutic intervention is the initial recognition of pathogens and the initiation of an early innate immune response.
The molecular events leading to cell recognition and response to LPS are becoming more clearly defined. Recently, the toll family of proteins, which consists of at least 10 type 1 transmembrane receptor proteins, has been identified.6 The toll proteins share a highly homologous cytoplasmic domain, similar to the IL-1 receptor, a very short transmembrane domain, and an extracellular portion consisting of a various number of leucine-rich repeats.7 Toll-like receptor 4 (TLR4), in particular, has been shown to be a major LPS recognition receptor.8 In vivo, mouse strain C3H/HeJ (which has a proline-to-histidine substitution at amino acid 712 of TLR4) and C57BL/10ScCR (which has a deletion of the TLR4 locus and does not express TLR4 RNA) are both hyporesponsive to LPS.9 In addition, macrophages and B cells from TLR4−/− mice are unable to mount a cellular response to LPS.10,11 In vitro data from transfection of mammalian cell lines further proved that TLR4 is able to enhance the LPS responsiveness of cells, a process augmented by CD14 and leading to activation of transcription factors nuclear factor–κB and activator protein 1.12 These findings indicate that TLR4 is a recognition molecule for gram-negative pathogens and specifically LPS.
The cloning of TLR4 and its identification as a primary LPS receptor in mammals prompted us recently to investigate whether mutations in TLR4 would modulate the response to LPS in humans. Using a reliable model of LPS-induced airflow obstruction,13 we demonstrated that specific mutations in TLR4 (Asp299Gly and Thr399Ile) are associated with a diminished airway response to inhaled LPS in normal human volunteers.14
Since mice that contain mutations in TLR4 are hyporesponsive to LPS and are also more susceptible to gram-negative bacterial challenge, we hypothesized that the mutations in human TLR4 associated with hyporesponsiveness to inhaled endotoxin would also enhance the susceptibility of humans to gram-negative sepsis. To investigate this hypothesis, we studied the distribution of mutations in TLR4 in healthy blood donors and in patients with septic shock. Our findings indicate that, although the overall frequency of TLR4 mutations was similar in both populations, within the septic shock group patients with TLR4 mutations had more severe disease and an increased frequency of gram-negative infections. In addition, mutations of TLR4 involving only the 299 residue occurred exclusively in those with septic shock, and all of the patients with this isolated mutation had gram-negative septic shock. These findings indicate that mutations of the TLR4 receptor may predispose individuals to gram-negative septic shock.
The patient samples were collected as part of a multicenter study conducted in 7 academic adult intensive care units in France between March 1, 1996, and November 30, 1997. The protocol was approved by the institutional review board of Cochin University Hospital, Paris, France. Informed consents were obtained from control subjects and patients or their relatives.
The control group comprised 73 healthy unrelated white blood donors to the Cochin University Hospital blood bank; 42 of the controls had been previously genotyped for a study on TNF-α polymorphisms.15 An additional 31 healthy blood donors were collected as control subjects for this study. The septic shock group, defined by the criteria of the consensus conference,16 included 91 intensive care unit (ICU) patients. Of these patients, 88 had been included in a previously published study on the effect of TNF-α polymorphism on septic shock,15 and 3 patients were entered into the study after June 1997.
To be eligible for enrollment, the patients with septic shock had to be white and to have 6 of the following inclusion criteria of septic shock within a 12-hour period: (1) clinical evidence of infection; (2) hyperthermia (temperature >38°C) or hypothermia (temperature <35.6°C); (3) tachycardia (heart rate >90 beats/min); (4) tachypnea (respirations 120/min) or need for mechanical ventilation; (5) use of vasopressor to maintain systolic blood pressure higher than 90 mm Hg, or hypotension, defined as systolic blood pressure less than 90 mm Hg for more than 30 minutes or a decrease in systolic blood pressure of more than 40 mm Hg from previously established values for more than 30 minutes (hypotension had to be present at enrollment and refractory to an intravascular volume challenge of at least 500 mL); and (6) evidence of inadequate organ function or perfusion within 12 hours of enrollment, as manifested by at least 1 of the following syndromes (previously described): acute deterioration of the patient's mental status; arterial hypoxemia (PaO2/fraction of inspired oxygen <280); plasma lactate concentration above the normal range or metabolic acidosis; oliguria; and disseminated intravascular coagulation. The exclusion criteria were the following: (1) age greater than 80 years; (2) cardiac failure (class III or IV); (3) liver insufficiency (Child class C); (4) bone marrow aplasia (white blood cell count <500/µL); or (5) immunosuppression (positive human immunodeficiency virus serologic result, current immunosuppressive therapy including corticosteroids [equivalent prednisone >0.5 mg/kg per day], or cancer).
Patients were followed up throughout their stays in the ICU. Age, sex, primary site of infection, infection-related organisms, and severity indexes including Simplified Acute Physiology Score (SAPS II),17 which uses a total of 15 variables to assess the severity of ICU patients, and Organ System Failure score,18 were collected at each patient's entry into the ICU. All the samples and information used in the study were coded and patient confidentiality was preserved according to the guidelines for studies of human subjects.
To determine the TLR4 genotype of the samples with respect to residues 299 and 399, the genomic DNA was amplified with primers (forward primer: 5′-tctagagggcctgtgcaatt-3′; reverse primer, 5′-tgaaaactcactcatttgtttcaa-3′) that span the region containing the 2 polymorphisms. Polymerase chain reaction conditions were as follows: 5-minute initial denaturation at 95°C, followed by 30 cycles (95°C for 30 seconds, 55°C for 30 seconds, and 72°C for 1 minute). After polymerase chain reaction amplification, fragments were purified (Gel Purification Kit; Qiagen, Valencia, Calif) and sequenced. The sequence of the genomic DNA from each subject was determined with an automated DNA sequencer (Model 377; Perkin Elmer, Norwalk, Conn). The sequence analysis was performed blinded to the phenotype of the study subjects.
The statistical analysis was designed to determine whether specific mutations in the TLR4 gene were associated with several factors related to septic shock. We used a Fisher 1-sided exact test (consistent with our hypothesis) to determine whether specific mutations of the TLR4 gene occurred more frequently in cases than controls.19 Similar comparisons were made between TLR4 allele type and disease characteristics among those with septic shock. A Mann-Whitney test was used to determine the significance of continuous variables.19
We studied 91 patients with septic shock and 73 healthy control subjects. The control subjects were on average younger (mean, 36.9 years; SD, 11.6 years) than the patients (average age, 57.7 years; SD, 15.0 years). The sex was evenly distributed between both study populations. Eleven patients (12%) and 8 control subjects (11%) had either the 299 or the 399 mutation in TLR4 (Table 1), a frequency similar to what we reported previously.14 When the populations were tested for occurrence of the TLR4 mutation (Asp299Gly/Thr399Ile), we observed a similar prevalence of the double mutation Asp299Gly Thr399Ile in both the patients with septic shock and the control group (Table 1). However, the TLR4 Asp299Gly mutation in the absence of a cosegregating mutation at residue 399 was found to occur only in the septic shock group. Importantly, 5 patients with septic shock (5.5%) had the Asp299Gly TLR4 mutation, while this occurred in none of the control subjects (P = .05) (Table 1).
Among the patients with septic shock, we investigated the relationship between specific TLR4 alleles and clinical characteristics of the disease process. While no significant differences were observed for age, sex, or treatment site (medical or surgical ICU) (data not reported), infections due to gram-negative organisms ranged from about 35% for patients carrying the wild-type TLR4 locus or the TLR4 Asp299Gly/Thr399Ile allele to 80% in patients with only the TLR4 Asp299Gly allele (P = .06; Table 2). When we looked at gram-negative organisms occurring with/without polymicrobial infections, the numbers increased to 50% in the TLR4 wild-type group and to 100% in patients with the TLR4 Asp299Gly allele (P = .04; Table 2), whereas the percentage remained unchanged for patients carrying the TLR4 Asp299Gly Thr399Ile mutation.
We next investigated the relationship between the different alleles of TLR4 and the primary sites of infection (Table 3). Although we found that meningitis occurred more frequently in those with both the TLR4 Asp299Gly and the Thr399Ile mutations, the number of study subjects was small, and this association was no longer evident when the analysis was limited to patients with gram-negative sepsis. Since our in vitro data had suggested that the mutation at residue 299 had a more severe effect on TLR4 function,14 we tested whether the different TLR4 mutations had an effect on disease severity as measured by SAPS II, Organ System Failure score, and mortality (Table 4). No significant differences were observed between the TLR4 genotype and SAPS II values or mortality. While patients with the TLR4 Asp299Gly allele had a higher Organ System Failure value than the other groups (Table 4), the difference was not significant (P = .24).
Of note, the only patient who was homozygous for the TLR4 Asp299Gly allele in our study population was a 15-year-old girl who died of Escherichia coli pyelonephritis. This patient was admitted to the ICU with altered mental status, fever (temperature, 41°C), and abdominal pain. She was found to have pyelonephritis and was treated with fluids and appropriate antibiotics. Despite these treatments, she developed a dramatic septic shock and multiorgan failure (adult respiratory distress syndrome, anuria, refractory shock) and died at day 3 of hospitalization. Subsequent bacteriologic analysis showed that the antibiotic treatment was appropriate for a multisensitive E coli that was cultured from her blood and urine.
Previously, the −308 TNF-α polymorphism, TNF2, had been shown to be associated with sepsis in this population.15 We therefore investigated whether the presence of the −308 TNF-α polymorphism altered the relationship between TLR4 polymorphisms and septic shock. In the patient population, the values for the TNF2 polymorphism ranged from 0% in the patients with the TLR4 Asp299Gly mutation to 42.5% in the patients with wild-type TLR4 (P = .07). Stratifying the septic subjects by those with (n = 32) and without (n = 59) the TNF2 (−308 TNF-α polymorphism), we did not observe any novel associations between the TLR4 alleles and the type or severity of septic shock.
Our findings demonstrate that mutations in the human TLR4 gene appear to be associated with an increased risk for septic shock and a higher prevalence of gram-negative bacterial infection. These findings suggest that mutations in the TLR4 gene, which impair its responsiveness to LPS, could lead to an increased risk of subsequent bacterial infections, analogous to the in vivo findings in the mouse strain C3H/HeJ.20 This hypothesis could explain our finding that the higher percentage of carriers for the TLR4 polymorphisms may be associated with a higher risk of developing gram-negative sepsis in the patient population. Even though the patient and control populations differed significantly in age (see Table 1), the difference in polymorphism frequency is not age related, and the occurrence of the TLR4 Asp299Gly polymorphism exclusively in the patient population indicates an association of specific TLR4 mutations with septic shock. It is important to note that in previous studies the TLR4 Asp299Gly polymorphism occurred solely in combination with the mutation at residue 399.14 The septic shock population used in this study constitutes the first example, to our knowledge, of a white population in which the TLR4 Asp299Gly mutation occurs regularly in the absence of a second mutation at residue 399. Moreover, 1 case in our study, that of a 15-year-old girl homozygous for the TLR4 Asp299Gly allele, who died of E coli pyelonephritis, illustrates the potentially detrimental effects of the TLR4 Asp299Gly mutation in a homozygous state. It is possible that the increased severity of the 299 mutation in terms of impairing protein function, as shown by our previous in vitro analysis,14 causes a more severe effect in carriers for this mutation alone, making these patients more susceptible to bacterial infection.
The increased occurrence of the TLR4 Asp299Gly genotype in the septic patients suggests that mutations of the TLR4 receptor enhance the susceptibility to gram-negative sepsis because of a decrease in endotoxin responsiveness. These in vivo findings confirm our in vitro studies that suggest that the mutations at residues 299 and 399 interrupt LPS signaling.14 In this previous study, we demonstrated that the TLR4 Asp299Gly allele and, to a much lesser degree, the TLR4 Thr399Ile allele caused THP-1 cells to be hyporesponsive to LPS in culture. The difference in mutation severity shown in vitro could potentially explain why the TLR4 Asp299Gly allele is more frequent in this septic shock patient population than the TLR4 Thr399Ile allele alone.
Since TLR4 is a transmembrane protein, it is not surprising to find a putative signal peptide at the N-terminus, presumably responsible for proper trafficking to the cell membrane.8 This raises the possibility that sequence variants in the extracellular domain of TLR4 can disrupt trafficking of this receptor to the cell membrane and enhance proteolysis.21 Mutations of this kind could give rise to a range of phenotypes from almost normal to completely unresponsive, depending on the severity of the trafficking defect, which would indirectly affect receptor density on the cell surface. Potential other effects of the mutant TLR4 receptor include decreased ligand binding or changes in the receptor conformation that could impair signal transduction, once the ligand is bound. In the absence of any structural information on the extracellular domain of the TLR4 receptor, the potential structural effect of the TLR4 mutation remains unclear, although the immunostaining of the airway epithelial cells suggests an effect on receptor density.14
Another finding in this study is the high prevalence of gram-negative infections in patients with TLR4 receptor mutations. In addition to all patients with TLR4 receptor mutations having a higher prevalence of gram-negative infections, the 5 patients carrying the TLR4 Asp299Gly variant had solely gram-negative infections. Although this finding may not be statistically significant in a larger patient population, it could suggest that the increased severity of the TLR4 Asp299Gly could more severely attenuate the immune response to gram-negative infections among these individuals, thereby increasing the susceptibility to infections. Since TLR4 and LPS binding is an important first step in the defense against some gram-negative bacteria, TLR4 plays a crucial role in mediating the cellular response to recognize and eliminate these pathogens. The correlation between a potential lack of receptor function and an increased susceptibility to infection is analogous to what is seen in the TLR4−/− mice11 and the C3H/HeJ strain, which are both hyporesponsive to LPS but more susceptible to gram-negative infections.20
The −308 TNF-α polymorphism and the TLR4 Asp299Gly polymorphism appear to be independent predictors of septic shock. While the −308 TNF-α polymorphism was found to occur more frequently among those with septic shock and found to be associated with enhanced mortality,15 in the current study we found that the TLR4 polymorphisms were associated with gram-negative septic shock. The absence of an interaction between these genotypes suggests that the effect of these polymorphisms on sepsis may be additive. Future investigations examining the interaction between several biologically relevant polymorphisms may provide important information to risk stratify patients with life-threatening infections.
In summary, we show a potential link between mutations in the TLR4 receptor and gram-negative septic shock. This preliminary finding warrants further studies with a larger patient population to determine whether some of the findings, such as a more severe disease phenotype, increased presence of gram-negative bacteria in particular, will reach statistical significance.
Accepted for publication September 6, 2001.
This study was supported by grants from the Department of Veterans Affairs (Merit Review), Washington, DC; grants ES11375, ES07498, and ES09607 from the National Institute of Environmental Health Sciences, Bethesda, Md; and grant HL62628 from the National Heart, Lung, and Blood Institute, Bethesda.
Corresponding author and reprints: David A. Schwartz, MD, Pulmonary and Critical Care Medicine, Duke University Medical Center, Research Drive, Room 275 MSRB, DUMC Box 2629, Durham, NC 27710 (e-mail: email@example.com).
Thank you for submitting a comment on this article. It will be reviewed by JAMA Internal Medicine editors. You will be notified when your comment has been published. Comments should not exceed 500 words of text and 10 references.
Do not submit personal medical questions or information that could identify a specific patient, questions about a particular case, or general inquiries to an author. Only content that has not been published, posted, or submitted elsewhere should be submitted. By submitting this Comment, you and any coauthors transfer copyright to the journal if your Comment is posted.
* = Required Field
Disclosure of Any Conflicts of Interest*
Indicate all relevant conflicts of interest of each author below, including all relevant financial interests, activities, and relationships within the past 3 years including, but not limited to, employment, affiliation, grants or funding, consultancies, honoraria or payment, speakers’ bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued. If all authors have none, check "No potential conflicts or relevant financial interests" in the box below. Please also indicate any funding received in support of this work. The information will be posted with your response.
Some tools below are only available to our subscribers or users with an online account.
Download citation file:
Web of Science® Times Cited: 442
Customize your page view by dragging & repositioning the boxes below.
Users' Guides to the Medical Literature
Sequencing the human genomeidentifying the entire sequence of base pairs in the 25 ...
All results at
Enter your username and email address. We'll send you a link to reset your password.
Enter your username and email address. We'll send instructions on how to reset your password to the email address we have on record.
Athens and Shibboleth are access management services that provide single sign-on to protected resources. They replace the multiple user names and passwords necessary to access subscription-based content with a single user name and password that can be entered once per session. It operates independently of a user's location or IP address. If your institution uses Athens or Shibboleth authentication, please contact your site administrator to receive your user name and password.