CD14 C(-159)T Polymorphism Is a Risk Factor for Development of Pulmonary Tuberculosis
- Adrian G. Rosas-Taraco1,
- Agnès Revol1,
- Mario C. Salinas-Carmona1,
- Adrian Rendon2,
- Guillermo Caballero-Olin3 and
- Dr. Alma Y. Arce-Mendoza1
- 1 Departamento Inmunologia, Facultad de Medicina, Monterrey, Nuevo León, México
- 2 CIPTIR (Centro de Investigacion, Prevencion y Tratamiento de Infecciones Respiratorias), Hospital Universitario, UANL, Monterrey, Nuevo León, México
- 3 Instituto Mexicano del Seguro Social, Monterrey, Nuevo León, México
- Reprints or correspondence: Dr. Alma Yolanda Arce-Mandoza Departamento Inmunologfa Facultad Medicina Universidad Autonoma de Nuevo León Gonzalitos 235 Norte Mitras Centro Monterrey Nuevo León México 64460 (aya_mayola{at}yahoo.com).
Abstract
Background. Neither the expression of CD14 and Toll-like receptor 4 (TLR4) on monocytes' surface nor the mutations CD14 — 159TT and TLR4 Asp299Gly have yet been evaluated as risk factors for development of pulmonary tuberculosis (TB) in the Mexican population.
Methods. Level of membrane CD14 (mCD14) and membrane TLR4 (mTLR4) were determined by flow cytometry, in 104 patients with pulmonary TB (before and after treatment), 67 household contacts, and 114 healthy control subjects. Genotype/allele frequencies in CD14 -159 and TLR4 Asp299Gly were obtained by polymerase chain reaction—restriction-fragmentlength polymorphism. Levels of soluble CD14 (sCD14) in sera were quantified by ELISA.
Results. Higher levels of mCD14/sCD14 and mTLR4 were observed in the patients and the household contacts than in the control subjects (P < .05) and decreased in the patients after the infection was resolved. The frequency of the CD14 — 159TT genotype was higher in the patients than in the control subjects (35.6% vs. 12.3%, respectively). Patients who were homozygous for allele T of the CD14 promoter gene had a significantly higher risk for development of pulmonary TB, with an odds ratio of 2.267 (95% confidence interval, 1.5%–3.3%). Levels of sCD14 or mCD14 were not associated with the CD14 — 159TT genotype (P > .05).
Conclusions. No association between TLR4 Asp299Gly and pulmonary TB was found. CD14 — 159TT is a risk factor for development of pulmonary TB, whereas mCD14/sCD14 and mTLR4 are possible biomarkers for the prognosis for TB disease.
Clinical Trial Protocol ID: SA1168-05.
Approximately one-third of the world's population is infected with Mycobacterium tuberculosis (Mtb), a causative agent of tuberculosis (TB), making it a significant public-health problem [1]. It has been reported that pattern-recognition receptors (PRRs) interact with pathogen-associated molecular patterns involved in macrophage uptake of different microorganisms [2]. In the case of Mtb, there are several groups of PRRs that bind to different conserved structures of the bacilli, including complement receptors (CR1, CR3, and CR4), mannose receptor, CD14, DC-SIGN, Toll-like receptor (TLR)-4, surfactant-protein receptors, CD44, and scavenger receptors [3–7].
CD14 is a contributor to receptor-mediated ingestion of nonopsonized Mtb by macrophages, via recognition of components of the cell wall, such as lipoarabinomannan [8]. It is a 55-kDa glycosyl phosphatidylinositolanchored glycoprotein expressed on the surface of monocytes, macrophages, and polymorphonuclear leukocytes [9]. CD14 has been reported to bind to the lipopolysaccharide (LPS) receptor, leading to NF-κB activation and cytokine expression mediated by the TLR4/ MD2 complex [10]. Soluble CD14 (sCD14) is produced by enzymatically cleaved membrane CD14 (mCD14), mediated mainly by phospholipase C, and via secretion of CD14 [11]. sCD14 is considered to be an acute-phase protein, and its concentration in serum has been found to increase in several clinical pathologies, such as septic shock [12], Kawasaki disease [13], atopic dermatitis [14], liver disease [15], rheumatoid arthritis [16], systemic lupus erythematosus [17], brucellosis [18], and TB [19].
Toll-like receptors too are PRRs that trigger innate immune response. It is well known that TLR4 recognizes LPS and several culture-filtrate proteins of Mtb [20]. The results of previous reports suggest that CD14 and TLR4 play important roles in the activation of the immune response to Mtb, especially in the case of chronic infection [21].
A common single-nucleotide polymorphism (SNP) is found at position -159 in the CD14 promoter, where a C→T change occurs. Baldini et al. found that this SNP is associated with elevated levels of sCD14 and is correlated with low levels of IgE and IL-4 in allergic patients [22]. Other studies have shown an association between the CD14 — 159TT genotype and chronic periodontitis [23], myocardial infarction [24], Crohn disease [25], and, more recently, susceptibility to brucellosis [26].
Asp299Gly, found in the TLR4 gene, is an important SNP that is associated with acute coronary syndromes [27]. Rudofsky et al. found that both this SNP and another, TLR4 Thr399Ile, are associated with reduced prevalence of diabetic neuropathy in type 2 diabetes [28], whereas Franchimont et al. have found that ulcerative colitis and Crohn disease are associated with TLR4 Asp299Gly [29]. Susceptibility to infectious diseases such as malaria [30] and brucellosis [31] is also associated with TLR4 Asp299Gly.
Although CD14 and TLR4 have been reported to play a crucial role in TB, little is known about their SNPs and their possible relationship to the development of TB. The aim of the present study was to determine whether CD14 and TLR4 are biomarkers for susceptibility to development of TB.
Subjects and Methods
Study population. During a 29-month period (2004–2006), 104 patients with active pulmonary TB who presented at the Instituto Mexicano del Seguro Social and Centro de Investigatión, Preventión y Tratamiento de Infecciones Respiratorias (CIPTIR), Hospital Universitario, in Monterrey, Nuevo León, México, were enrolled prospectively in the present study. Also enrolled were 67 household contacts—that is, persons who were or were not genetically related to the patients but who, during the study, cohabited with the patients; these household contacts did not, at the time of enrollment, present with any signals or symptoms of pulmonary TB, and tests of sputum samples from them were negative for the disease. The study also included 114 healthy individuals as control subjects. All participants were >18 years old and Mexican, had been residents of Nuevo León State for 2–3 generations, and were from white and Mestizo ethnic groups. Pulmonary TB was diagnosed on the basis of clinical findings and a smear or culture positive for pulmonary TB. All participants were negative for HIV and did not have diabetes, and none had been treated with steroids or immunosuppressive agents. The Bioethics Committee from the Facultad de Medicina at the Universidad Autonoma de Nuevo León approved the project, ensuring that it was in accordance with Institutional Review Board criteria. Blood samples were obtained from all participants; in the case of the patients, they were collected both before and 6–7 months after anti-TB treatment was initiated. Initial treatment for all the cases of pulmonary TB included isoniazid, rifampin, pyrazinamide, and ethambutol for 2 months and then isoniazid plus rifampin for 4 additional months (6 months in total).
Flow-cytometry analysis of mCD14 and membrane TLR4 (mTLR4) expression. A sample of heparinized whole blood from each patient was used for flow-cytometry analysis. The samples were incubated with the following monoclonal antibodies: isotype control IgG1-conjugated fluorescein isothiocyanate (FITC) and IgG2a-conjugated R-phycoerythrin (PE) (BD Biosciences Pharmingen), human anti-CD14-PE (BD Biosciences Pharmingen), human anti-TLR4-PE (Santa Cruz, California), and human anti-CD45-FITC (Santa Cruz Biotechnology, California). Erythrocytes were lysed, and leukocytes were recovered, after centrifugation at 220–240 g for 8 min. Finally, the cells were resuspended in FACS flow solution (BD Biosciences Pharmingen), for cytofluorometric analysis (FACS Sort Calibur; BD Biosciences Pharmingen). A total of ∼10,000 cells were analyzed, the percentage of cells and the fluorescence intensity were obtained, and the nonspecific fluorescence was subtracted.
ELISA for sCD14. The concentration of sCD14 in blood samples was measured by commercial kits using sandwich ELISA, according to the manufacturer's instructions (Biosource).
Genetic analysis. Genomic DNA was extracted from the blood samples by use of phenol/chloroform followed by ethanol precipitation [32]. The CD14 C(-159)T and TLR4 Asp299Gly polymorphisms were detected by a polymerase chain reaction (PCR)-restriction-fragment length-polymorphism method that has been described elsewhere [22, 33]. A 497-bp PCR product was obtained for the CD14 promoter and was digested by commercially available AvaII endonuclease (Promega), for detection of SNPs in CD14 -159; as well, a 249-bp PCR product was obtained for TLR4 and was cut with NcoI (Promega), for detection of SNP Asp299Gly. The digested fragments were separated by electrophoresis on a 2%–3.5% agarose gel and were visualized after ethidium-bromide staining. To avoid the possibility of incomplete digestion, 10% of the samples were analyzed again; all of them gave the same results.
Statistical analysis. The Kolmogorov-Smirnov test for normality was used for the continuous-response variables. Analysis of variance and the nonparametric test were used to evaluate the results. The Kruskal-Wallis test follow by the post-hoc Dunn's multiple-comparison test and the Mann-Whitney test were used for comparisons of the levels of mCD14/sCD14 and TLR4 in each study group—for example, for analysis of the interaction between CD14 genotypes and the levels of mCD14/ sCD14 in the patients with pulmonary TB. Student's t test was used to evaluate the values of mCD14/sCD14 and TLR4 in 24 of the patients, both before and 6–7 months after treatment. Allele and genotype frequencies were analyzed by Fisher's exact test, and odd ratios and their 95% confidential intervals were obtained. Associations and differences for which P <.05 were considered to be significant. Data were analyzed by SPSS software (version 10.1; SPSS) and GraphPad Prism 4 (San Diego, California).
Results
Characteristics of participants. Similar numbers of males and females were included in each of the 3 groups: females constituted 50.9% of the patients with TB, 58.2% of the household contacts of the patients, and 53.5% of the control subjects. The average age was 39.5 ± 17.9 years (range, 18–70 years) in the patients, 34.3 ± 11.3 years (range, 18–62 years) in the household contacts, and 26.3 ± 8.5 years (range, 21–56 years) in the control subjects. Initially, 81% of the patients were positive by tuberculin skin test (TST); however, during treatment, the patients who initially had been negative became positive. All of the household contacts and 48.2% of the control subjects were positive by TST; 7 (10.4%) of the 67 household contacts developed pulmonary TB 10–14 months after they were enrolled in the study.
Expression ofCD14 and TLR4 on monocytes. The levels of expression of mCD14 and mTLR4 were similar in both the TST-positive control subjects and the TST-negative control subjects (P > .05) (data not shown). Before the patientswith pulmonary TB started anti-TB treatment, their levels of mCD14 on monocytes were higher than those in the control subjects (mean fluorescence intensity [MFI], 60 and 29.1, respectively [P < .001]). The levels of mCD14 in the household contacts who did or did not develop TB were higher than those in the control subjects (MFI, 67.5 and 53.7, respectively [P < .001], in both cases); there was no difference between the 2 household-contacts groups themselves (P > .05). In the patients, the levels of expression of mCD14 6–7 months after initiation of anti-TB treatment were lower than those before treatment (MFI, 32 [P < .001]) (see figure 1) and did not show significant differences from those in the control subjects (P > .05).
Figure 1.
A, Expression of membrane CD14 (mCD14) on monocytes from control subjects (n = 114); from patients with pulmonary tuberculosis (PTB), before (n = 104) and after (n = 24) treatment; and from household contacts without (wo/TB; n = 60) or with (w/TB; n = 7) PTB. High levels of mCD14 were found in the patients and in their household contacts. Values were analyzed by analysis of variance and by Kruskal-Wallis, Dunn's, and Mann-Whitney tests; the horizontal lines represent mean values. B, Expression of mCD14 on monocytes from 24 patients with PTB, before and after 6–7 months of treatment. Values were analyzed by Student's t test.
Before the patients with TB started anti-TB treatment, their levels of expression of mTLR4 were 1.6-fold higher than those in the control subjects (MFI, 29.4 and 19.5, respectively [P < .001]); 6–7 months after anti-TB treatment had been initiated in them, the patients had levels of expression of mTLR4 that were similar to those in the control subjects (MFI, 20.7 [P > .05]). In contrast, the household contacts who did or did not develop pulmonary TB had levels of expression of mTLR4 that were higher than those in the control subjects (MFI, 33.8 and 35, respectively [P < .01] and P <.001, respectively]); there was no difference between the 2 household-contacts groups themselves (P > .05) (see figure 2).
Figure 2.
A, Expression of membrane Toll-like receptor 4 (mTLR4) on monocytes from control subjects (n = 114); from patients with pulmonary tuberculosis (PTB), before (n = 104) and after (n = 24) treatment; and from household contacts without (wo/TB; n = 60) or with (w/TB; n = 7) PTB. High levels of mTLR4 were found in the patients and in their household contacts. Values were analyzed by analysis of variance and by Kruskal-Wallis, Dunn's, and Mann-Whitney tests; the horizontal lines represent mean values. B, Expression of mTLR4 on monocytes from 24 patients with PTB, before and after 6–7 months of treatment. Values were analyzed by Student's t test.
Levels of sCD14 in serum. Before the patients with TB started anti-TB treatment, their levels of sCD14 were higher than those in the control subjects (9406 ng/mL and 5663 ng/mL, respectively [P < .001]); 6–7 months after anti-TB treatment had been initiated in them, the patients had levels of sCD14 that were lower than those when they started anti-TB treatment (6198 ng/mL [P < .001]) and that were very similar to those in the control subjects. The levels of sCD14 in the household contacts who did or did not develop TB were higher than those in the control subjects (8376 ng/mL and 7831 ng/mL, respectively [P < .001, in both cases]) (figure 3).
Figure 3.
A, Levels of soluble CD14 (sCD14) in sera from control subjects (n = 114); from patients with pulmonary tuberculosis (PTB), before (n = 104) and after (n = 24) treatment; and from household contacts without (wo/TB; n = 60) or with (w/TB; n = 7) PTB. High levels of sCD14 were found in patients and in their household contacts. Values were analyzed by analysis of variance and by Kruskal-Wallis, Dunn's, and Mann-Whitney tests; the horizontal lines represent mean values. B, Levels of sCD14 in sera from 24 patients with PTB, before and after 6–7 months of treatment. Values were analyzed by Student t test.
Allele and genotype frequencies in pulmonary TB. Genotypes and alleles in the 114 control subjects (TST positive and TST negative), the 104 patients with pulmonary TB, and the 67 household contacts were analyzed for the presence of the CD14 — 159 and TLR4 Asp299Gly polymorphisms. The frequency of the homozygous CD14 — 159TT genotype was highest in the patients with pulmonary TB (35.6%, compared with 12.3% in the control subjects [P < .0001]), whereas the frequency of the heterozygous (i.e., CT) genotype in these 2 groups was similar (49% and 55.3%, respectively). The frequency of the CD14 -159TT genotype in the household contacts with TB (n = 7) was 57%, whereas it was only 18.4% in the household contacts without TB. The highest frequencies of the CD14 — 159T allele were 71%, in the household contacts who developed TB, and 60%, in the patients with pulmonary TB (P < .0001); in contrast, the frequencies were only40% and 39.2%, in the case ofthe control subjects and the household contacts without TB, respectively (table 1). In neither the patients with pulmonary TB nor the control subjects was there any association between the CD14–159 genotype and the levels of mCD14 and sCD14 (P > .05) (see figure 4).
Figure 4.
Effect of CD14 — 159 genotype on levels of membrane CD14 (mCD14) (upper graph) and soluble CD14 (sCD14) (lower graph), in patients with pulmonary tuberculosis (PTB). CD14 — 159 genotypes identified by polymerase chain reaction-restriction-fragment length polymorphism and either mCD14, identified by flow cytometry, or sCD14, identified by ELISA, were compared. No association between genotype CD14 — 159TT and the levels of mCD14 and sCD14 was found (P > .05).
In neither the patients with pulmonary TB nor the control subjects was there an association between genotypes and allele frequencies of TLR4 Asp299Gly (see table 2).
Table 1.
Genotype and allele frequencies of CD14 -159C→T in patients with pulmonary tuberculosis (TB), household contacts, and control subjects.
Table 2.
Genotypes and allele frequencies of Toll-like receptor 4 (TLR4) Asp299Gly in patients with pulmonary tuberculosis (TB), household contacts, and control subjects.
Discussion
The present study has provided data showing high levels of sCD14/mCD14 and mTLR4 in sera from the patients with pulmonary TB and their household contacts, compared with the control subjects. The levels of mCD14 and sCD14 were diminished after the patients had received 6–7 months of anti-TB treatment. In 1995, Hoheisel et al. reported similar results for bronchoalveolar-lavage fluids from patients with pulmonary TB [19]. The results of the present study are supported by Juffermans et al.'s [34] and Pacheco et al.'s [35] reports of high levels of sCD14 during active TB that decrease after initiation of anti-TB treatment. Also consistent with the results of the present study are those of our previous report of high levels of mCD14 on monocytes from patients with either pulmonary or extrapulmonary TB [36]. In contrast to the results reported by Juffermans et al., the present study found differences that the levels of sCD14 in the control subjects were different from those in the household contacts. This difference may be explained by how the household contacts were recruited: Juffermans et al. included close contacts who did not cohabit with the patients, whereas the present study included only household contacts, who, by definition, did cohabit with the patients. Also, the present study considered used a larger contact group—67 individuals, versus the 16 considered by Juffermans et al. These 2 important factors could influence the results obtained by the studies. On the other hand, an increase in the level of sCD14 is not specific to TB disease; similar results have been reported for patients with brucellosis [18], septic shock [37], or sarcoidiosis [38] and for patients infected with HIV [39] or autoimmune diseases [40].
The results of the present study show that the level of mTLR4 is diminished after 6–7 months of anti-TB treatment. In 2004, Yan et al. reported that LPS induces higher levels of expression of mTLR4 on the surface of mononuclear cells [41]. Like the level of sCD14, the level of TLR4 is high in several pathologies, such as sepsis, severe sepsis, septic shock, and infection with the hepatitis C virus [37, 42].
In the present study, the household contacts without TB had elevated levels of CD14 and TLR4. It is well known that TB develops in only 5%–15% of people in close contact with a patient who presents as TB positive by smear [43]. Because both CD14 and TLRs are involved in the innate immune response, they may have been up-regulated in the household contacts without TB, as a way to avoid development of TB disease.
In the present study, the frequency of the T allele in the control subjects (40%) was similar to that reported by other studies, which were performed with American, Czech, Iranian, or German healthy subjects [22, 23, 26, 44]. In 2001, LeVan et al. reported that the CD14 — 159TT genotype diminishes the affinity of specificity proteins (nuclear factor) binding to the CD14 promoter, thereby enhancing the transcriptional activity of CD14 [45]. However, the results of the present study do not reveal any differences between any genotype of CD14 -159 and the level of either mCD14 or sCD14. Corroborating the results of the present study, Pacheco et al. reported no association between CD14 polymorphism and the level of sCD14 in the Colombian population [35]; in addition, von Aulock et al. found no association between CD14 -159 genotype and either release of tumornecrosis factor a or mCD14 [46]. Although the biological effect of the C→T SNP remains unclear, the CD14 — 159-genotype promoter polymorphism has been associated with several diseases, such as brucellosis in the Iranian population [26], chronic periodontitis in the Czech population [23], chronic chlamydial infection [47], and Crohn disease [44]. In the present study, the frequency of the CD14 — 159TT genotype was higher (35.6%) in the patients with pulmonary TB than in the control subjects, a result similar to that found for the aforementioned pathologies, particularly with brucellosis (which, like TB, is caused by an intracellular pathogen). So the CD14 — 159TT genotype maybe considered to be a risk factor for development of TB. The household contacts, who were homozygous for allele T of the CD14 promoter, had a significantly higher risk for development of pulmonary TB, a result supported by the fact that this allele was found in 57.1% of the household contacts who developed TB (although their number was low). In contrast to the results of the present study, Pacheco et al. did not find any association between CD14 polymorphism and different clinical forms of TB disease in white and Mestizos ethnic groups from Medellin, Colombia. The present study is the first to suggest the possible association between a CD14 — 159 SNP and TB, and it produced some interesting findings, although some considerations have to be mentioned. First, the control group included both TST-negative and TST-positive individuals, because a large part ofthe Mexican healthy population is positive for TST. It should be noted that because these 2 subgroups did not differ with respect to the data on receptor expression and sCD14, both of them were used in the analysis. Also, many of the patients with TB left the study after the first serum sample was obtained, either because they did not want to be included in the follow-up or because they changed their address without telling us, resulting in fewer samples being obtained after treatment; fortunately, however, it was still possible to analyze 24 samples. Although the number of household contacts in the present study was limited, their inclusion provides very interesting findings. Future studies should include a larger number of household contacts.
Several mutations in TLR4, such as Asp299Gly and Thr399Ile, have been demonstrated to be involved in hyporesponsiveness to endotoxin [48]: In 2006, Mockenhaupt et al. reported that these 2 SNPs are frequently found in African children with severe malaria (caused by Plasmodium falciparum, an intracellular pathogen) [30]. Also in 2006, Rezazadeh et al. reported that patients who were heterozygous for allele G of TLR4 Asp299Gly had a higher risk for brucellosis [31]. In 2002, Raby et al. did not find a link between TLR4 Asp299Gly and asthma [49]. In agreement with the results of the present study, Newport et al. did not find any association between TLR4 Asp299Gly and pulmonary TB in a Gambian population [50]. Because the frequency of the G allele was found to be very low in the population of northeastern México, it should be interesting to increase the number of subjects to improve the power of analysis of TLR4.
The results of both the aforementioned reports and the present study suggest that increases in the levels of CD14 and TLR4 are part of an inflammatory process that is common in several diseases. After treatment, the levels of both molecules decrease to normal. Consequently, mCD14/sCD14 and mTLR4 are possible biomarkers for the prognosis for TB disease. In addition, although CD14 — 159TT appear to be a risk factor for the development of pulmonary TB, no link between TLR4 Asp299Gly and the illness was found. In the present study, the susceptibility to develop TB was found to be associated with CD14 — 159TT but not with an increase in the expression of CD14; therefore, it will be important to analyze a possible association between the CD14 — 159TT genotype and some cytokines such as interferon-γ, which is important in the immune response against infection by M. tuberculosis.
Acknowledgments
We thank Susana Covarrubias for technical support, Dr. Marco Gomez for statistic support, and Dr. Roisin Owens for critical reading of the manuscript.
Footnotes
-
Potential conflicts of interest: none reported.
-
Financial support: Programa de Apoyo a la Investigacion Cientffica y Tecnologica (PAICYT), Universidad Autonoma de Nuevo León (grant SA1168–05 to A.Y.A.-M.).
- Received January 30, 2007.
- Accepted June 22, 2007.
- © 2007 by the Infectious Diseases Society of America
References
This Article
- AbstractFree
- » Full Text (HTML)Free
- Full Text (PDF)Free
Classifications
-
- Major Articles and Brief Reports
- Bacteria
- Major Article
- Bacteria
- Major Articles and Brief Reports
Services
Citing Articles
Google Scholar
PubMed
Related Content
Share
- Email this article
-
CiteULike
Connotea
Delicious
Facebook
Google+
Mendeley
Twitter
Published on behalf of
IDSA Members: For your free access to this journal, log in via the IDSA members area.
The Journal
Impact Factor: 6.288
5-Yr impact factor: 6.103
For The Media
For Authors
- Services for authors
- Self-archiving policy
- IDSA Conflict of Interest policy
- Instructions for Authors
- ICMJE Form
- Submit Now!
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy
Corporate Services
Alerting Services
Related IDSA Publications
