Volume 15, Issue 9 , Pages 1130-1133, September 2009
Toll-Like Receptor 4 Polymorphisms and Risk of Gram-Negative Bacteremia after Allogeneic Stem Cell Transplantation. A Prospective Pilot Study
Article Outline
The Toll-like receptor 4 (TLR4) gene is a major recognition receptor for lipopolysaccharide (LPS). In a pilot prospective study, we examined the association of 2 TLR4 polymorphisms (Asp299Gly and Thr399Ile) in the donor or the recipient with Gram-negative bloodstream infection (BSI) in 77 allogeneic hematopoietic stem cell transplant (HSCT) patients. Heterozygosity at both loci was defined as “risk genotype.” The cumulative incidence of infection was estimated by treating death prior to infection as a competing risk event and compared between relevant groups using a modified chi-square test. Nine patients had the risk genotype based on donor and 5 based on recipient genotype data. Donor risk genotype showed marginal statistical significance (0.06) in univariate analysis, but not in multivariate analysis. A larger study is required to validate our findings and define genetic susceptibility to this serious infection in HSTC patients.
Key Words: Innate immunity, Toll-like receptor, Mutation, Polymorphism, Gram-negative infection, Bone marrow transplantation
Introduction
Bloodstream infection (BSI) by Gram-negative organisms occurs in approximately 20% of allogenic hemopoietic stem cell transplant (HSCT) recipients [1] .
Toll-like receptor 4 (TLR4) has been shown to be a major recognition receptor for lipopolysaccharide (LPS), a highly conserved component of the cell wall of Gram-negative pathogens. The mouse strains C3H/HeJ (TLR4 Pro712His) and C57BL/10ScCR(deletion of TLR4 locus) are hyporesponsive to LPS [2]. Interestingly, these strains are very susceptible to fatal infections by Gram-negative bacteria.
Two polymorphic point mutations at nucleotides A896G and C1196T in the cDNA of human TLR4 encode for amino-acid substitutions Asp299Gly and Thr399Ile, respectively, within the extra-cellular domain of TLR4. Both mutations are predicted to affect a ligand binding region and a co-receptor biding region, respectively [3]. Primary human epithelial cells and alveolar macrophages with these mutations have blunted responses to LPS. Individuals with these mutations have hyporesponsiveness to inhaled LPS [4].
TLR4Asp299Gly and the co-segregating mutation TLR4Thr399Ile are found in approximately 6% to 10% of Caucasians 5, 6. These polymorphic forms of TLR4 have been associated with an increased incidence of Gram-negative infections and sepsis, and more recently with an increased susceptibility to Crohn's disease 5, 7. Conversely, individuals with these mutations appear to be at decreased risk for certain inflammatory diseases including atherosclerosis and rheumatoid arthritis 8, 9. Thus TLR4 signaling appears to play an important role in the balance between the ability to control infections and the pathogenesis of inflammatory diseases. Functional polymorphisms with no apparent effect on immunocompetent individuals may become clinically relevant in immunocompromised patients, such as allogeneic HSCT recipients.
We studied prospectively a cohort of 105 allogeneic HSCT to determine any association between the TLR4Asp299Gly and Thr399Ile mutations in the donor or the recipient with development of Gram-negative bloodstream infection post HSCT.
Materials and Methods
Study Patients
The study was approved by the Memorial Sloan-Kettering Cancer Center (MSKCC) Institutional Review Board. One hundred and five non-consecutive, non-selected patients who received allogenic HSCT at MSKCC from October 2003 through December 2004 were enrolled in a prospective study. The objective of the study was to assess the role of genetic polymorphisms in innate immune genes in susceptibility to specific infections post HSCT.
Case Definition
Gram-negative BSI was defined as any positive blood culture for a Gram-negative organism treated with antibiotics. Follow-up of the patients ranged from 60 days to 2 years post HSCT.
Genotyping
DNA sourceAfter signing informed consent, 5-10
mL of peripheral blood was obtained from the patients once prior to stem cell infusion (for mutational analysis of the recipient) and once after full chimerism was documented (for mutational analysis of the donor). Genomic DNA was extracted from the peripheral blood using the PureGene™ (Gentra Systems, Minneapolis, MN) and stored at −20°C.
A 1420-bp fragment of TLR4 containing TLR4 896A>G (Asp299Gly) and 1196C>T (Thr399Ile) was amplified from genomic DNA by the polymerase chain reaction (PCR) using the primers forward: 5′-TTTCTAATCTGACCAATCTAG-3′ and reverse: 5′-ACTACAAGCACACTGAGGA-3′.
The PCR reaction was performed in a final volume of 50 μL containing 100 ng of DNA, dNTPs at a final concentration of 3mM, 15 picomoles of each primer, and 1.5 units of Taq DNA polymerase (Mercury Products, Continental Lab Products, San Diego, CA). Polymerase chain reaction (PCR) conditions were: denaturation at 94°C for 20 seconds, annealing at 60°C for 20 seconds, and extension at 72°C for 1minute for 35 cycles. Final extension was conducted at 72°C for 5minutes. The PCR product was purified for sequencing reaction using the Montage PCR96 plates (Millipore, Bedford, MA). Both strands of the PCR product were sequenced using primers forward: 5′-TTGGTTCTGGGAGAATTTAG-3′ and reverse: 5′-TGAAAACTCACTCATTTGTTT-3′. Sequencing was performed on an ABI 377 sequencer using Big Dye terminator chemistry (Applied Biosystems, Bedford, MA) according to the manufacturer's recommendations. Electropherograms were analyzed using the Phred and Phrap softwares (www.phrap.org). The reference sequence NM_138554 was used for TLR4 (http://www.ncbi.nlm.nih.gov/). Screening was performed blinded to the identity of the samples.
Statistical Analysis
Genotypes of the donor and the recipient were recorded.
For the analysis the data from the 2 single nucleotide polymorphisms (SNPs) were summarized into 1 SNP score as follows: Genotypes heterozygous for the minor allele at both SNP were defined as “risk genotype. All other genotypes were defined as “no risk genotype.”
The association between Gram-negative BSI and risk genotype was investigated separately for the genotypes of donors and recipients. The outcome of interest was Gram-negative BSI, measured as the time elapsed (in days) between the date of stem cell infusion and date of Gram-negative BSI or date of death (if the patient died prior to the onset of infection) or the last follow-up date (if the patient was alive and infection free). The cumulative incidence of infection was estimated by treating death prior to the onset of infection as a competing risk event, and compared between relevant groups using a modified chi-squared test. P-value <.05 was considered statistically significant [10]. Univariate and multivariate analysis were used to examine the relationship between BSI and putative risk factors.
Results
Study Patients
Genotypes were available for 77 donor-recipient pairs. Thus the sample size is 77 patients. The characteristics of the patients are shown on Table 1. The risk genotype was present in 9 (11.6%) patients based on donor genotype data and 5 (6.4%) patients based on recipient genotype data. Fifteen patients developed BSI during the study period. Five BSI occurred before neutrophil engraftment and 10 after neutrophil engraftment a median 79 days (range: 37-381 days) after HSCT. The causative organisms were enteric Gram-negatives (11) and non-fermenting Gram-negative rods (4).
Table 1. Characteristics of the Patients (n = 77)
| Variable | Total Number of Patients | |
|---|---|---|
| Race | White non-Hispanic | 57 |
| Other | 20 | |
| Conditioning type | Myelo ablative | 64 |
| Reduced intensity | 13 | |
| Donor type | Matched related | 33 |
| Matched unrelated | 27 | |
| Mismatched | 17 | |
| Acute GVHD grade | II,III,IV | 13 |
| 0,1 | 64 | |
| missing | 2 | |
| T cell depletion | Yes | 46 |
| No | 31 | |
| “Risk genotype”∗ donor | Yes | 9 |
| No | 68 | |
| “Risk genotype” recipient | Yes | 5 |
| No | 72 | |
| Bloodstream infection | Yes | 15 |
| No and alive | 48 | |
| Dead without infection | 14 | |
Four groups of patients were defined based on the presence of risk genotype in donor and recipient (Table 2). Fifteen patients had BSI, 14 died prior to infection, and 48 patients were alive at the end of the follow-up period. Among the 15 patients with BSI, 4 (26.6%) had risk genotypes based on the donor, and none (0%) on the recipient. Among the 14 patients who died, none had risk genotypes based on the donor, and 1 based on the recipient. Among the 48 patients alive without infection, 5 (10.6%) had a risk genotype based on the donor, and 4 (8.5%) based on the recipient.
Table 2. Infection by Risk Genotype in Donor/Recipient
| Donor/Recipient Risk Genotype | ||||
|---|---|---|---|---|
| Status | No/No | Yes/No | No/Yes | Yes/Yes |
| N=66 | N=6 | N=2 | N=3 | |
| Alive/no BSI | 42 | 2 | 1 | 3 |
| BSI | 11 | 4 | 0 | 0 |
| Dead/no BSI | 13 | 0 | 1 | 0 |
Univariate analysis results of each putative risk factor for BSI are given in Table 3. The only factor that showed marginal statistical significance was presence of risk genotype in the donor.
Table 3. Univariate Analysis of Risk Factors for Bloodstream Infection.
| Variable | P-Value from Modified Chi-Squared Test | |
|---|---|---|
| Race | Other | |
| White non-Hispanic | .89 | |
| T cell depletion | No | |
| Yes | .51 | |
| Stem cell source | Nonblative | |
| Ablative | .69 | |
| Donor type | Other | |
| Matched relative | .28 | |
| Matched nonrelated | .86 | |
| Acute GVHD | 2,3,4 | |
| (2 missing subjects excluded) | 0,1 | .71 |
| Donor risk genotype | No | |
| Yes | .06 | |
| Recipient risk genotype | No | |
| Yes | .25 | |
Since risk genotype in the donor (but not in the recipient) showed marginal statistical significance in the univariate analysis, we included donor risk genotype in the multivariate model. All other risk factors were also included in the multivariate model because they are deemed putative risk factors, even though they did not exhibit statistical significance in the univariate analyses (Table 4). Donor risk genotype did not show statistical significance in multivariate analysis.
Table 4. Multivariate Analysis of Risk Factors for Bloodstream Infection
| Variable | Hazard Ratio | 95% CI | P-Value | |
|---|---|---|---|---|
| Race | Other | 1.00 | ||
| White non-Hispanic | 0.57 | 0.13-2.71 | .49 | |
| T cell depletion | No | 1.00 | ||
| Yes | 1.04 | 0.33-3.28 | .95 | |
| Stem cell source | Other | 1.00 | ||
| Myeloablative | 0.85 | 0.18-3.92 | .84 | |
| Donor type | Mismatched | 1.00 | ||
| Matched related | 0.40 | 0.03-5.45 | .49 | |
| Matched unrelated | 1.07 | 0.56-2.03 | .84 | |
| Acute GVHD | 2,3,4 | 1.00 | ||
| 0,1 | 0.94 | 0.24-3.66 | .93 | |
| Donor risk genotype | No | 1.00 | ||
| Yes | 3.38 | 0.84-13.63 | .09 | |
Discussion
Genetic variants in immune response genes can influence the outcome of HSCT independent of HLA match status [11]. BSIs by Gram-negative organisms are associated with substantial morbidity and mortality. The inflammatory response to Gram-negative bacteria has been mainly attributed to lipopolysaccharide (LPS), a highly conserved glycolipid component of the cell wall [12]. Key components of the innate immune response to LPS are TLR4, the lipopolysaccharide binding protein (LBP) the secreted protein MD-2, and the TLR4 co-receptor CD14 [13]. It is likely that genetic variants in 1 or more genes affecting the interaction of these molecules may contribute to susceptibility or outcome of infections by Gram-negative organisms.
A retrospective study in HSCT, reported a trend towards reduced graft-versus-host disease (GVHD) in HSCT recipients heterozygous for the TLR4Asp299Gly mutation. The same study showed a trend for higher incidence of Gram-negative infections if the donor or the recipient were heterozygous for the same mutation [14].
In our prospective study study we analyzed 77 HSCT patients to determine any association between 2 functional TLR4 mutations (Asp299Gly and Thr399Ile) with Gram-negative BSI. In our patients the frequency of minor alleles at both loci were within the range reported in the literature for Caucasians (6%-10%) [6]. Because these mutations are in linkage disequilibrium and are predicted to affect TLR4 signaling, we defined as “risk genotype,” the genotype heterozygous for the minor allele in both loci.
We did not find any association between risk genotype in the donor or recipient and development of BSI posttransplant. The presence of the risk genpotype in the donor showed marginal statistical significance in univariate analysis. Our negative findings may be due to our limited sample size. To design a larger prospective study to address the association of TLR4 SNPs with Gram-negative BSI post-HSCT we used the power calculation method of Schoenfeld [15] to calculate the sample size required to estimate a desired hazard ratio with 80% power and 5% significance level. Our observed data shows that 9 (12%) of 77 donor specimens have a risk genotype. The remaining 68 specimens (88%) have no risk genotype in the donor. In the absence of any competing cause of infection, the total number of infections required to detect a hazard ratio of 3 with 80% power, and 5% significance level is 62. If 20% of the patients develop an infection, then the required total sample size is 62/0.2=310 patients. Ideally, the analysis should take into account the amount of LPS produced by the pathogens of BSI and host chimerism at the time of infection.
We report the first prospective study examining the association of TLR4 polymorphisms with Gram-negative bloodstream infection after allogenic HSCT. The ability to predict genetic susceptibility to a common infection with substantial morbidity is of direct relevance to the design of strategies to protect the most vulnerable patients.
Acknowledgments
Financial disclosure: The study was supported by a grant from Dana Foundation to GAP and EGP. The authors thank Ellen King for her assistance with the Phred and Phrap software.
References
- Pre- and post-engraftment bloodstream infection rates and associated mortality in allogeneic hematopoietic stem cell transplant recipients. Transpl Infect Dis. 2005;7:11–17
- Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene. Science. 1998;282:2085–2088
- TLR4 mutations are associated with endotoxin hyporesponsiveness in humans. Nat Genet. 2000;25:187–191
- . Haplotype variation in bovine Toll-like receptor 4 and computational prediction of a positively selected ligand-binding domain. Proc Natl Acad Sci USA. 2003;100:10364–10369
- . Relevance of mutations in the TLR4 receptor in patients with Gram-negative septic shock. Arch Intern Med. 2002;162:1028–1032
- . Determination of the TLR4 genotype using allele-specific PCR. BioTechniques. 2001;31:22–24
- . Polymorphisms of the lipopolysaccharide-signaling complex in inflammatory bowel disease: association of a mutation in the Toll-like receptor 4 gene with ulcerative colitis. Clin Immunol. 2004;112:85–91
- The Toll-like receptor 4 Asp299Gly functional variant is associated with decreased rheumatoid arthritis disease susceptibility but does not influence disease severity and/or outcome. Arthritis Rheum. 2004;50:999–1001
- Toll-like receptor 4 polymorphisms and atherogenesis. N Engl J Med. 2002;347:185–192
- . A proportional hazards model for the sub-distribution of a competing risk. J Am Stat Assoc. 1999;94:496–509
- . Beyond HLA: the significance of genomic variation for allogeneic hematopoietic stem cell transplantation. Blood. 2007;109:1355–1362
- . Metropolitan home living conditions associated with indoor endotoxin levels. J Allergy Clin Immunol. 2001;107:790–796
- . Bacterial lipopolysaccharides and innate immunity. J Endotoxin Res. 2001;7:167–202
- Association of TLR4 mutations and the risk for acute GVHD after HLA-matched-sibling hematopoietic stem cell transplantation. Biol Blood Marrow Transplant. 2001;7:384–387
- . Sample-size formula for the proportional-hazards regression model. Biometrics. 1983;39:499–503
Present address: FIRC Institute of Molecular Oncology, Milan, Italy.
Financial disclosure: See Acknowledgments on page 1133.
PII: S1083-8791(09)00212-2
doi:10.1016/j.bbmt.2009.04.012
© 2009 American Society for Blood and Marrow Transplantation. Published by Elsevier Inc. All rights reserved.
Refers to erratum:
- Erratum , 25 September 2009
Volume 15, Issue 9 , Pages 1130-1133, September 2009
