Cardiovascular disease (CVD) is an important contributor to morbidity and mortality in American Indian communities. The Strong Heart Study (SHS) was initiated in response to the need for population based estimates of cardiovascular disease in American Indians. Previous studies within SHS have identified correlations between heart disease and deficiencies in mannose binding lectin (MBL), a motif recognition molecule of the innate immune system. MBL mediates the immune response to invading pathogens including Chlamydia pneumoniae (Cp), which has also been associated with the development and progression of CVD. However, a link between MBL2 genotype and Cp in contributing to heart disease has not been established. To address this, SHS collected baseline Cp antibody titers (IgA and IgG) and MBL2 genotypes for common functional variants from 553 individuals among twelve participating tribes. A single nucleotide polymorphism (SNP) in the promoter, designated X/Y, correlated significantly with increased Cp IgG titer levels, whereas another promoter SNP (H/L) did not significantly influence antibody levels to Cp. Two variants within exon 1 of MBL2, the A and B alleles, also displayed significant association with Cp antibody titers. Some MBL2 genotypes were absent from the population, suggesting linkage disequilibrium may be operating within the SHS cohort. Additional factors, such as increasing age and socioeconomic status, were also associated with increased Cp IgG antibody titers. This study demonstrates that MBL2 genotype associates with immune reactivity to C. pneumoniae in the SHS cohort. Thus, MBL2 may contribute to the progression of cardiovascular disease (CVD) among American Indians indirectly through pathogen interactions in addition to its previously defined roles.
Citation: Monsey L, Best LG, Zhu J, DeCroo S, Anderson MZ (2019) The association of mannose binding lectin genotype and immune response to Chlamydia pneumoniae: The Strong Heart Study. PLoS ONE 14(1): e0210640. https://doi.org/10.1371/journal.pone.0210640
Editor: Ray Borrow, Public Health England, UNITED KINGDOM
Received: May 10, 2018; Accepted: December 28, 2018; Published: January 10, 2019
Copyright: © 2019 Monsey et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: This work was supported by cooperative agreement grants U01-HL65520, U01-HL41642, U01-HL41652, U01-HL41654, U01-HL65521, and R01HL62233 from the National Heart, Lung, and Blood Institute, Bethesda, MD (https://www.nhlbi.nih.gov/). LM was supported by an ASC Undergraduate Research Scholarship from The Ohio State University (https://aschonors.osu.edu/opportunities/scholarships/undergrad). Missouri Breaks Industries Research, Inc, provided support in the form of salaries for author LGB, but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.
Competing interests: I have read the journal's policy and the authors of this manuscript have the following competing interests: LGB receives compensation in the form of salary support from Missouri Breaks Industries Research, Inc. This does not alter our adherence to PLOS ONE policies on sharing data and materials.
Cardiovascular disease contributes substantially to poor global health. In the United States, cardiovascular disease (CVD) is the leading cause of death, responsible for 1 in 4 deaths . Although decreasing considerably over three decades starting in the 1980's, total mortality rate due to CVD increased by 3% between 2011 and 2014 and has been consistently higher than cancer-related deaths . Clinical manifestations include damage to the epithelial lining of blood vessels, inflammation of the heart or arterial lining; and eventually restricted blood flow, which results in myocardial infarction, stroke, and numerous other manifestations
Cardiovascular disease risk is largely composed of two primary sources, environmental contributions and genetic predisposition. Environmental contributors to CVD include diet, lifestyle, air quality, and infectious agents . The relative contributions of these factors vary across time, geography, and socioeconomic status. Consequently, environmental factors shape the impact of CVD on specific populations differently. Some factors such as infectious disease outbreaks and air quality are communal and require large-scale concerted efforts to combat while diet and lifestyle include individualized components. Preventative therapies exist that alleviate the impact of personalized risks such as smoking cessation programs and lifestyle/dietary intervention  but do not address factors that impact whole populations. For example, previous work has linked locally endemic infectious agents to CVD risk including HIV, Trypanasoma cruzi, and Chlamydia pneumoniae, which are highly prevalent within particular global regions [5, 6].
Cardiovascular disease is a complex genetic disease influenced by the contribution of thousands of variant positions across hundreds of loci . Immunomodulatory molecules comprise a significant portion of genes implicated in CVD and include genes such as C-reactive protein (CRP), apolipoprotein E (apoE), and mannose binding lectin (MBL2) [8–10]. MBL is a major circulating C-type lectin of the innate immune system involved in self versus non-self recognition. As such, MBL plays a significant role in identification of dead, dying, or cancerous cells for clearance by professional phagocytes . Similarly, MBL binds to a variety of pathogen-associated molecular patterns (PAMPs) leading to pathogen recognition and destruction [12, 13]. MBL2 dysregulation is associated with coronary heart disease (CHD) and increased susceptibility to microbial infection [14, 15].
One of the pathogens most consistently associated with CVD onset and progression is Chlamydia pneumoniae (Cp), an obligate intracellular bacterium that infects humans, and specifically targets endothelial cells [6, 16–19]. C. pneumoniae is capable of producing both active, acute disease as well as latent infections that persist throughout the life of the host with the potential to re-emerge. During active infection, accumulation of cellular debris during rupture of infected cells as well as attraction of circulating phagocytes can lead to plaque formation on blood vessel walls, inducing vascular complications such as coronary heart disease (CHD) . MBL2 operates as a recognition receptor for mannose and N-acetylglocusamine, which are major components of the Cp cell membrane. Importantly, reduced MBL function has been linked to CVD severity and progression in patients infected with Chlamydia pneumoniae [17, 20, 21].
Variants within the MBL2 coding sequence and promoter region, (see S1 Table), have significant consequences on its molecular function. Several well characterized MBL2 promoter and coding missense SNPs significantly alter circulating levels of MBL and are associated with clinical disease [22, 23]. The frequency of these coding alleles within groups varies depending on population origin. For example, the B allele frequency is approximately 45% in South American Mapuche and Chiriguano communities, whereas it is present in less than 1% of a Mozambican population [24, 25]. Such population-based variation in allele frequency may contribute to differences in disease risk among specific demographics.
The Strong Heart Study (SHS) is the largest epidemiological study within American Indian populations and is focused on cardiovascular health . Twelve tribes have participated in the Strong Heart Study to develop reliable, population-based data related to cardiovascular health among American Indians. Previous studies within the SHS cohort identified an association between MBL2 genotypes and coronary artery disease . However, the relationship between the immune response to Cp and MBL2 genotypes associated with CVD has not yet been established.
Here we investigated the relationship between Cp and MBL2 within the SHS cohort. MBL2 promoter and coding sequence genotype frequencies do not reflect majority Caucasian populations. Cp antibody titers are significantly affected by the X/Y promoter variant and the copy number of MBL2 ‘A’ and ‘B’ alleles. Several additional descriptive factors correlate with Cp antibody levels including age and education. Thus, development of CVD may be due, in part, to the interaction of the host immune system with infectious agents such as Cp.
This ancillary study was conducted between August 1, 2000, and November 30, 2000, among individuals enrolled in the Strong Heart Study (SHS) at ages 45 to 74 years between July 1, 1989, and January 31, 1992. It was approved by the Indian Health Service, MedStar Research Institute Institutional Review Boards, and by the participating tribes. All investigative methods conformed to the principles set out by the Declaration of Helsinki and involved informed consent of the participants. Since these data were collected, one tribe has declined further participation and those tribal members have been excluded from the analysis. All studies are reviewed and approved by the SHS Review Board composed of tribal representatives prior to study initiation and publication.
The original study’s design and methods have been reported . An ancillary study of the relationship between infectious disease and CVD measured baseline antibodies to C pneumoniae and other pathogens on a subset of 421 CVD cases and an equal number of controls, matched by study center, sex and age (+/- 5 years). From this substudy, MBL2 genotypes were available from 553 individuals. Medical record review by trained medical record abstractors and a physician reviewer committee determined that 211 had experienced definite CVD, 161 possible CVD and 181 no CVD in subsequent follow-up. Risk factors for CVD were determined at time of entry into the SHS and at two subsequent follow-up examinations.
Blood specimen collections
Blood specimens collected during the study were separated into serum samples, frozen at -80°C, and aliquots were thawed only when performing specific tests.
MBL2 locus genotyping
Genotyping was done at the University of Pittsburgh. A total of 553 individuals were assessed for the presence of the MBL2 B allele (G54D, rs1800450), C allele (G57E, rs1800451) or D allele (R52C, rs5030737) structural variations, and two promoter polymorphisms, one a G/C transition at—550 bp (the H/L alleles, rs11003125) and the other, a G/C transition at -221 (X/Y alleles, rs7096206). The most common coding allele has been conventionally labelled "A" and the three structural variants collectively labelled "O". MBL2 genotypes were determined by the oligonucleotide ligation assay as described by Nickerson and colleagues . Quality control, duplicate, genotyping was performed by direct DNA sequencing. The structural variations were assumed to occur on opposing chromosomes. A number of promoter variants and structural alleles have been found to be in complete linkage disequilibrium and genotypes were checked against these established relationships .
Haplotypes were built using previously described methods . Briefly, homozygous positions could be inherently linked to alleles at other variant positions. Subsequently, unresolved haplotypes were built using previously described association including the B and C coding alleles and X promoter variant being found with the L promoter variant, the D MBL2 allele linked to the H promoter variant, etc.,
Cp antibody titer
Serum IgG and IgA antibody titers for C. pneumoniae were determined by microimmunofluorescence (MIF) at the International Chlamydia Laboratory at Johns Hopkins School of Medicine. This test uses the application of purified elementary bodies (EBs) from high titer chlamydia EB antigen preparations (Washington Research Foundation, Seattle, WA). There is only one serovar of C. pneumoniae and the purified antigen is made from strain AR39. Antigen dots for C. trachomatis were also be included in the series of antigen dots, so that the specificity of the anti-C. pneumoniae antibody could be confirmed. The highest dilution of serum demonstrating good even fluorescence of the EBs was recorded as the titer for each group. Separate assays were performed for the determination of IgG, and IgA. The laboratory has participated with others in quality assurance studies for MIF and demonstrated consistency and comparability for this assay. The Cp antibody titers could be determined for 514 individuals.
The MBL2 gene is a key component of innate immunity whose product recognizes pathogen-associated molecular patterns of invading organisms and marks them for destruction. The functional capacity of MBL2 is associated with five polymorphic sites, two in the promoter and three within the first exon . The two promoter variant positions, denoted as H/L and X/Y, are located 550 and 221 nucleotides upstream of the start codon, respectively (Fig 1A). Polymorphisms at these positions impact the level of MBL2 transcription with the H and Y variants producing more robust expression . Additionally, variants of exon 1 referred to as alleles B, C, and D encode changes at amino acid 52, 54, and 57 compared to the reference A allele. The B and C alleles produce proteins with reduced function due to a decrease in oligomerization capability required for function and the D allele is especially weakened in its immunological role .
A. The MBL2 gene is composed of 5 exons comprised of both noncoding (grey) and coding (black) sequences. Two prominent polymorphic sites associated with MBL2 function exist in the promoter at positions -550 (H/L variant) and -221 (X/Y variant). Additionally, the genotype of exon 1 impacts gene function based on three variant alleles B, C, and D corresponding to amino acid changes at codon 52, 54, and 57, respectively, as well as the canonical allele A. The frequency of exon 1 genotypes for allele A (B), non-A alleles (O) (C), or promoter variants (D) was plotted for five previously reported populations and the SHS cohort. The number of individuals plotted are: SHS (553), Dutch (212), Inuit (148), Mozambiquan (185), and Korean (128).
MBL2 genotypes and distributions
Previous investigations within the Strong Heart Study cohort identified links between heart disease and low functioning MBL2 genotypes [17, 27]. To determine the distribution of MBL2 alleles within the SHS cohort using an expanded panel of polymorphic positions, individuals were genotyped at both promoter polymorphic sites and exon 1. The majority of individuals encoded the functional A/A MBL2 genotype (69.6%) with a smaller subset either heterozygous or homozygous for a "non-A" (O) allele (27.5% and 2.9%, respectively) (Table 1). The B variant was the most common O allele (14.3%) followed by the D and C alelles (1.6% and 0.7%, respectively).
Similar exon 1 allele frequencies are observed among distinct populations from different geographic regions for the A allele (Fig 1B). In contrast, the distribution of non-A alleles within exon 1 differed widely among distinct populations. The SHS cohort, Inuit, and South Korean populations harboured high frequencies of allele B, the Mozambican population primarily contained C allele variants, and the United States population had an even distribution among all three non-A alleles (Fig 1C). Furthermore, promoter alleles frequencies varied considerably among representative populations (Fig 1D). In the SHS cohort population the H and Y variants comprised 66.2% and 94.3% of the population, respectively (Table 1). Alleles within all genotyped variant positions conformed to Hardy-Weinberg equilibrium (S2 and S3 Tables). The SHS cohort was most similar to the Inuit and South Korean populations, suggesting that the frequency of MBL2 genotypes among American Indians significantly differs from Americans of European descent.
MBL2 haplotypes were constructed from allelic genotypes to assess association of promoter and coding variants. Haplotype construction revealed a majority of HYA haplotypes among SHS participants associated with robust Mbl2 function (Table 2). Thus, relatively few haplotypes containing L variants emerged from construction and X variant haplotypes were very rare. Consequently, all assembled haplotypes included at least one high function allele from either the promoter or exon 1.
Comparison to MBL2 haplotypes from other global populations most resembled the Inuit and South Korean populations, similar to analysis of individual variant positions.
Association of MBL2 genotype with C. pneumoniae
To test for an association between MBL2 genotype and immune reactivity to C. pneumoniae, antibody titers were measured for systemic IgG and mucosa-enriched IgA. C. pneumoniae IgG titers followed a Gaussian distribution centered on 1:128, indicating relatively strong systemic reactivity (Fig 2A). Participants expressed lower levels of IgA to C. pneumoniae, however, with more than half exhibiting no detectable antibody (Fig 2B). A positive linear correlation exists between Cp IgG and IgA within the SHS cohort (Spearman, rs = 0.47, p<0.01, Fig 2C). Consequently, the majority of our subsequent analysis relied upon IgG antibody titers as an indicator of immune reactivity.
The C. pneumoniae antibody titer of 514 individuals within the SHS cohort was plotted for IgG (A) or IgA (B). C. Antibody titers for IgG and IgA are significantly correlated within individuals (Spearman, rs = 0.47, p<0.01).
To determine the influence of MBL2 polymorphism on Cp antibody titers, we assessed the distribution of antibody levels for individual SNP positions. Analysis of covariance revealed additional copies of the X allele are significantly correlated with elevated Cp IgG titers (F(1,544) = 3.895, p = 0.049, Fig 3A, Table 3).
A. The mean C. pneumoniae antibody titer (red) is significantly correlated with MBL2 promoter genotypes (F(1,644) = 3.895, p = 0.049). Increased copies of the X (X/Y) variant resulted in elevated antibody titers among these 514 genotyped individuals. B. Genotyping of 514 participants indicated the presence of the canonical A allele of exon 1 was associated with increased C. pneumoniae antibody titer (F(1,644) = 4.39, p = 0.037). The mean is denoted in red. C. The mean antibody titer for all 553 MBL2 genotypes within the SHS cohort increases with elevated copy number for the H, X, and A alleles.
The H/L polymorphic site trended towards increased Cp antibody titers with increased dosage of the L allele but did not reach significance (F(1, 544) = 1.934, p = 0.165, Fig 3A).
The coding variants of exon 1 also impacted antibody levels to C. pneumoniae, with Cp antibody titer increasing with each additional A allele (F(1,544) = 4.39, p = 0.037, Fig 3B, Table 4). The B allele was also predictive of Cp IgG antibody titers (F(1,544) = 4.48, p = 0.035, Table 4). Neither the C nor D allele displayed any association with Cp titers although this may be due, in part, to their low prevalence and hence, lack of power. Taken together, both promoter and coding sequence alleles correlate with Cp antibody titers in the SHS cohort.
The aggregate contributions of individual polymorphic sites to Cp antibody titers can either be additive or show more complex interactions. To investigate this, the average Cp IgG titers were plotted for each represented genotype in the SHS cohort (Fig 3C, S1 Fig). No specific position determined the relative antibody levels within the cohort, suggesting that the combined effects of each variant contribute to the overall immune reactivity. For example, the MBL2 A and L alleles were enriched among higher Cp antibody titer groups and co-occurred in individuals with the highest average antibody titers. Additionally, the X promoter variant clustered with genotypes displaying higher Ab titers. Interestingly, the reported strength of MBL2 function for variant positions did not seem consistent with Cp antibody titers; the high-functioning MBL2 A allele corresponded to increased antibody levels, while promoter variants with reduced function (X and L) also correlated with higher Cp antibody titers. Across variant positions, the MBL2 A allele displayed the strongest association with Cp antibody titers (S2 Fig). Furthermore, analysis of constructed haplotypes including promoter and coding variants also failed to identify a significant association with Cp antibody titer (F(6, 1094) = 1.575, p = 0.151). Thus, individual promoter and coding sequence variants but not complete genotypes correlate with antibody production to Cp.
Associations of covariates
Changes in Cp antibody titers can be influenced by multiple non-genetic factors that are produced by lifestyle and the environment . We tested the association of Cp IgG titers against 14 variables including age, sex, education, smoking, alcohol use, ACR, and prior diagnoses of cardio vascular disease and diabetes (S4 Table). Three features showed significant associations with Cp IgG antibody levels, age (Spearman’s correlation; rs = 0.09 p < 0.041), smoking (Kruskal Willis; H(2) = 7.185, p = 0.028), and sex (Mann Whitney; U = 3.27, p = 0.001) (Fig 4). Additionally, the presence of the Y or H variants in the MBL2 promoter strongly correlated with a prior diagnoses of diabetes (rs = -0.106, .111; p = <0.01 for both) (S4 Table).
A. The mean C. pneumoniae antibody titer increased for each represented age within the SHS cohort (Spearman’s correlation; rs = 0.09 p < 0.041). B. The mean C. pneumoniae antibody titer (red) was elevated for men compared to women (Mann Whitney; U = 3.27, p = 0.001). N = 514 individuals. C. The mean C. pneumoniae antibody titer (red) increased among current smokers compared to non-smokers or previous smokers (Kruskal Willis; H(2) = 7.185, p = 0.028). N = 514 individuals.
Association of Cp with CVD
Cp infection has been previously associated with autoimmune disorders including CVD [17, 20]. To test for this association, participant genotypes were separated based on CVD diagnosis as either present, possible, or absent. No relationship between CVD and Cp antibody titers existed among SHS participants with diagnosed CVD versus healthy individuals (Mann-Whitney U-test (MW); F(1,369) = 1.135, p = 0.256, S4 Table). Furthermore, a test for association between Cp antibody titers and mortality due to CVD failed to identify a relationship (MW; F(1,369) = 0.879, p = 0.379). However, the H/L MBL2 promoter variant associated with CVD in which increased copies of low expression L alleles associated with increased CVD diagnosis as would be expected (S4 Table). Thus, no clear relationship between immune reactivity to Cp and CVD within this SHS population can be established.
Here, we have shown that MBL2 genotypes within an American Indian cohort correlated with the immunological response to the intracellular bacteria C. pneumoniae. MBL2 promoter variants within the SHS cohort were not found at similar frequencies compared to majority Caucasian populations. Both MBL2 promoter and coding variants were associated with the prevalence of antibody following exposure to C. pneumoniae, suggesting differences in immunological response although constructed haplotypes diminished the associations of individual variants with Cp antibody titers. The degree of immunogenicity was also correlated with more complex social and biological processes such as aging and education. These factors may contribute to the elevated risk of CVD within Native American communities.
Participants reported here had unique MBL2 genotype frequencies consistent with another Indigenous population in North America and similar to East Asian populations [25, 30, 31, 35, 36]. While the SHS prevalence of the majority A allele of MBL2 is similar to but slightly elevated compared most global populations, the B allele was much more common compared to populations of majority European or African ancestry [25, 32, 37]. Furthermore, the distribution of promoter variants and haplotypes in the SHS cohort differed markedly from the majority Caucasian population in the US [32, 38] and most closely resembled an Inuit and Korean population [30, 31]. Increased prevalence of high expression H and Y promoter variants in the SHS cohort could significantly influence the response to pathogens through MBL2, as well as impact other, intrinsic immunological functions that may contribute to chronic inflammation and hence influence CVD among American Indian populations.
Within the SHS cohort, all allelic variants were present but some specific genotypes were missing. Combinations of MBL2 “C” and “D” alleles were present only with “A” or “B” despite being found together in similarly sized cohorts within the US . Additionally, certain promoter genotypes were absent from the population despite being present in combination with other alleles (i.e., HHXX). Initial studies of MBL2 haplotypes also reported an absence of the HX haplotype ; however, recent studies have described its presence among a number of distinct global populations [40, 41]. Interestingly, none of these reports include Indigenous peoples of the Americas, suggesting that these genotypes may be absent because of maintained haplotype blocks that do not recombine or are selected against when present. It is tempting to speculate that selective pressures have modulated MBL activity by altering genotypic frequencies, reflecting the dual roles of controlling infectious agents and promoting cardiovascular function.
MBL2 functions as a “first responder” within the innate immune system in recognizing pathogens such as C. pneumoniae, which have been implicated in promoting CVD. Two particular MBL2 variant positions correlated with antibody production to Cp, the X/Y promoter variant and exon 1. Surprisingly, the high functioning A allele of MBL2 correlated with elevated Cp antibody titers; in a consistent manner, the lower functioning B allele correlated with reduced Cp antibody titers. The high prevalence of the MBL2 B allele and its corresponding reduced function may play a role in the development of CVD within American Indian populations specifically . Within the promoter, both low expression X and L alleles were associated with increased antibody production. Similarly, immunization of mice lacking MBL2 led to increased IgG production compared to wildtype mice . It is unclear how MBL2 variants that promote opposing levels of expression and/or function both increase Cp antibody titers but suggests that production of antibody through MBL regulation is complex, which is further supported by the non-additive nature of individual alleles in haplotype construction with immune reactivity. Additionally, a large prospective study demonstrated the unexpected relationship between high prevalence of MBL2 genotypes associated with high expression and increased risk of CVD . Thus, additional factors such as Interleukin-6 (IL-6) and C-reactive protein (CRP), which contribute to Cp antibody production, and other host-specific factors may contribute to these discrepancies .
Other environmental factors contributed significantly to Cp antibody titers among the SHS cohort. Our data indicated a linear relationship between age and higher amounts of circulating Cp antibody. This trend is consistent with previous studies that found an increase in seroprevalence transitioning from childhood into adulthood [34, 45, 46] and extends those studies to suggest that immune response to Cp continues to accumulate later in life as well. Association between Cp antibody titers and smoking have also been noted previously although the mechanism is unclear . The impact of gender on Cp immune reactivity is less well defined as studies have found both higher antibody titers in males as we observed  or no difference between sexes , suggesting this may vary depending on the sampled population.
The previously observed association between MBL2 genotype and CVD may include both the direct action of MBL in regulation of host cellular maintenance and its additional role in immune surveillance of pathogens . For example, some MBL2 variants correlated with increased levels of Cp antibodies were also shown to associate with CVD risk. Thus, MBL2 genotyping could be used in the clinic to screen individuals for host- and microbe-associated risk factors in CVD development. As a result of screening, physicians could introduce preventative strategies earlier to reduce the likelihood of disease onset and progression within American Indians populations.
S1 Fig. C. pneumoniae antibody titers for SHS cohort genotypes.
The C. pneumoniae antibody titer for IgG (A) or IgA (B) are plotted for all genotypes among the 553 individuals found within the SHS cohort.
S2 Fig. Association of promoter and coding allelic variants are associated in MBL2.
The relationship between MBL2 genetic variants and Cp antibody titers was constructed using hierarchical clustering and Euclidean distances.
S1 Table. Nomenclature of studied MBL2 variant positions.
S2 Table. MBL2 allele frequencies in the SHS cohort conformed to the Hardy-Weinberg equilibrium.
S3 Table. MBL2 Exon 1 frequencies in the SHS cohort conformed to the Hardy-Weinberg equilibrium.
S4 Table. Correlation analysis of genotypic and phenotypic traits.
Pairwise associations of traits were performed using a Spearman’s correlation analysis for all genetic polymorphisms assayed and a number of important covariates. Pink indicates p<0.05 and red indicates p<0.01.
We would like to thank Michael Davidson at MedStar Health and Robert Ferrell at the University of Pittsburgh for important intellectual contribution to the design and instigation of this study. We would also like to acknowledge the indigenous people that consented to participating this study for their fundamental role in its progression. Finally, we must thank the Strong Heart Study IRB and tribal review for their roles in overseeing the work.
We thank the SHS participants, Indian Health Services facilities, and participating tribal communities for their extraordinary cooperation and involvement, which has contributed to the success of the Strong Heart Study.
- 1. CDC N. Underlying Cause of Death 1999–2013 on CDC WONDER Online Database, released 2015. Data are from the Multiple Cause of Death Files, 1999–2013, as compiled from data provided by the 57 vital statistics jurisdictions through the Vital Statistics Cooperative Program. 2015. Epub Feb. 3, 2015.
- 2. Heron MaA, R.H. Changes in the Leading Cause of Death: Recent Patterns in Heart Disease and Cancer Mortality. NCHS Data Brief. 2016;(254):1–8. Epub August 2016. pmid:27598767
- 3. Lim SS, Vos T, Flaxman AD, Danaei G, Shibuya K, Adair-Rohani H, et al. A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet. 2012;380(9859):2224–60. pmid:23245609; PubMed Central PMCID: PMCPMC4156511.
- 4. de Waure C, Lauret GJ, Ricciardi W, Ferket B, Teijink J, Spronk S, et al. Lifestyle interventions in patients with coronary heart disease: a systematic review. Am J Prev Med. 2013;45(2):207–16. pmid:23867029.
- 5. Triant VA. Cardiovascular disease and HIV infection. Curr HIV/AIDS Rep. 2013;10(3):199–206. Epub 2013/06/25. pmid:23793823; PubMed Central PMCID: PMCPMC3964878.
- 6. Fong IW. New perspectives of infections in cardiovascular disease. Curr Cardiol Rev. 2009;5(2):87–104. pmid:20436849; PubMed Central PMCID: PMCPMC2805819.
- 7. Nikpay M, Goel A, Won HH, Hall LM, Willenborg C, Kanoni S, et al. A comprehensive 1,000 Genomes-based genome-wide association meta-analysis of coronary artery disease. Nat Genet. 2015;47(10):1121–30. pmid:26343387; PubMed Central PMCID: PMCPMC4589895.
- 8. Madsen HO, Videm V, Svejgaard A, Svennevig JL, Garred P. Association of mannose-binding-lectin deficiency with severe atherosclerosis. Lancet. 1998;352(9132):959–60. pmid:9752823.
- 9. Moazed TC, Campbell LA, Rosenfeld ME, Grayston JT, Kuo CC. Chlamydia pneumoniae infection accelerates the progression of atherosclerosis in apolipoprotein E-deficient mice. J Infect Dis. 1999;180(1):238–41. pmid:10353889.
- 10. O'Donnell CJ, Nabel EG. Genomics of cardiovascular disease. N Engl J Med. 2011;365(22):2098–109. pmid:22129254.
- 11. Ogden CA, deCathelineau A, Hoffmann PR, Bratton D, Ghebrehiwet B, Fadok VA, et al. C1q and mannose binding lectin engagement of cell surface calreticulin and CD91 initiates macropinocytosis and uptake of apoptotic cells. J Exp Med. 2001;194(6):781–95. pmid:11560994; PubMed Central PMCID: PMCPMC2195958.
- 12. Ali YM, Lynch NJ, Haleem KS, Fujita T, Endo Y, Hansen S, et al. The lectin pathway of complement activation is a critical component of the innate immune response to pneumococcal infection. PLoS Pathog. 2012;8(7):e1002793. pmid:22792067; PubMed Central PMCID: PMCPMC3390405.
- 13. Stuart LM, Ezekowitz RA. Phagocytosis: elegant complexity. Immunity. 2005;22(5):539–50. pmid:15894272.
- 14. Rantala A, Lajunen T, Juvonen R, Bloigu A, Paldanius M, Silvennoinen-Kassinen S, et al. Low mannose-binding lectin levels and MBL2 gene polymorphisms associate with Chlamydia pneumoniae antibodies. Innate Immun. 2011;17(1):35–40. pmid:19969625.
- 15. Pesonen E, Hallman M, Sarna S, Andsberg E, Haataja R, Meri S, et al. Mannose-binding lectin as a risk factor for acute coronary syndromes. Ann Med. 2009;41(8):591–8. pmid:19711212.
- 16. Assar O, Nejatizadeh A, Dehghan F, Kargar M, Zolghadri N. Association of Chlamydia pneumoniae Infection With Atherosclerotic Plaque Formation. Glob J Health Sci. 2015;8(4):260–7. pmid:26573036; PubMed Central PMCID: PMCPMC4873590.
- 17. Rugonfalvi-Kiss S, Endresz V, Madsen HO, Burian K, Duba J, Prohaszka Z, et al. Association of Chlamydia pneumoniae with coronary artery disease and its progression is dependent on the modifying effect of mannose-binding lectin. Circulation. 2002;106(9):1071–6. pmid:12196331.
- 18. Fong IW, Chiu B, Viira E, Tucker W, Wood H, Peeling RW. Chlamydial heat-shock protein-60 antibody and correlation with Chlamydia pneumoniae in atherosclerotic plaques. J Infect Dis. 2002;186(10):1469–73. Epub 2002/10/31. pmid:12404163.
- 19. Fong IW, Chiu B, Viira E, Jang D, Mahony JB. De Novo induction of atherosclerosis by Chlamydia pneumoniae in a rabbit model. Infect Immun. 1999;67(11):6048–55. Epub 1999/10/26. pmid:10531266; PubMed Central PMCID: PMCPMC96992.
- 20. Nagy A, Kozma GT, Keszei M, Treszl A, Falus A, Szalai C. The development of asthma in children infected with Chlamydia pneumoniae is dependent on the modifying effect of mannose-binding lectin. J Allergy Clin Immunol. 2003;112(4):729–34. Epub 2003/10/18. pmid:14564351.
- 21. Asner SA, Morre SA, Bochud PY, Greub G. Host factors and genetic susceptibility to infections due to intracellular bacteria and fastidious organisms. Clin Microbiol Infect. 2014;20(12):1246–53. Epub 2014/11/05. pmid:25366416.
- 22. Babovic-Vuksanovic D, Snow K, Ten RM. Mannose-binding lectin (MBL) deficiency. Variant alleles in a midwestern population of the United States. Ann Allergy Asthma Immunol. 1999;82(2):134–8, 41; quiz 42–3. pmid:10071515.
- 23. Jakab L, Laki J, Sallai K, Temesszentandrasi G, Pozsonyi T, Kalabay L, et al. Association between early onset and organ manifestations of systemic lupus erythematosus (SLE) and a down-regulating promoter polymorphism in the MBL2 gene. Clin Immunol. 2007;125(3):230–6. pmid:17942372.
- 24. Madsen HO, Satz ML, Hogh B, Svejgaard A, Garred P. Different molecular events result in low protein levels of mannan-binding lectin in populations from southeast Africa and South America. J Immunol. 1998;161(6):3169–75. pmid:9743385.
- 25. Valles X, Sarrias MR, Casals F, Farnos M, Piner R, Suarez B, et al. Genetic and structural analysis of MBL2 and MASP2 polymorphisms in south-eastern African children. Tissue Antigens. 2009;74(4):298–307. Epub 2009/09/25. pmid:19775369.
- 26. Lee ET, Welty TK, Fabsitz R, Cowan LD, Le NA, Oopik AJ, et al. The Strong Heart Study. A study of cardiovascular disease in American Indians: design and methods. Am J Epidemiol. 1990;132(6):1141–55. pmid:2260546.
- 27. Best LG, Davidson M, North KE, MacCluer JW, Zhang Y, Lee ET, et al. Prospective analysis of mannose-binding lectin genotypes and coronary artery disease in American Indians: the Strong Heart Study. Circulation. 2004;109(4):471–5. pmid:14732744.
- 28. Madsen HO, Garred P, Thiel S, Kurtzhals JA, Lamm LU, Ryder LP, et al. Interplay between promoter and structural gene variants control basal serum level of mannan-binding protein. J Immunol. 1995;155(6):3013–20. pmid:7673719.
- 29. Bernig T, Breunis W, Brouwer N, Hutchinson A, Welch R, Roos D, et al. An analysis of genetic variation across the MBL2 locus in Dutch Caucasians indicates that 3' haplotypes could modify circulating levels of mannose-binding lectin. Hum Genet. 2005;118(3–4):404–15. Epub 2005/10/07. pmid:16208516.
- 30. Hegele RA, Busch CP, Young TK, Connelly PW, Cao H. Mannose-binding lectin gene variation and cardiovascular disease in Canadian Inuit. Clin Chem. 1999;45(8 Pt 1):1283–5. Epub 1999/08/03. pmid:10430797.
- 31. Lee SG, Yum JS, Moon HM, Kim HJ, Yang YJ, Kim HL, et al. Analysis of mannose-binding lectin 2 (MBL2) genotype and the serum protein levels in the Korean population. Mol Immunol. 2005;42(8):969–77. pmid:15829288.
- 32. Stevenson HL, Amador A, McCue J, Weppler D, Tryphonopoulos P, Roth D, et al. Mannose binding lectin (mbl2) haplotype frequencies in solid organ transplant patients and correlation with MBL protein levels—evaluation of complement-mediated effector pathway deficiency. Transpl Immunol. 2013;28(2–3):73–80. Epub 2013/02/27. pmid:23439277.
- 33. Garred P, Larsen F, Seyfarth J, Fujita R, Madsen HO. Mannose-binding lectin and its genetic variants. Genes Immun. 2006;7(2):85–94. pmid:16395391.
- 34. Paldanius M, Bloigu A, Alho M, Leinonen M, Saikku P. Prevalence and persistence of Chlamydia pneumoniae antibodies in healthy laboratory personnel in Finland. Clin Diagn Lab Immunol. 2005;12(5):654–9. pmid:15879028; PubMed Central PMCID: PMCPMC1112086.
- 35. Sasaki K, Tsutsumi A, Wakamiya N, Ohtani K, Suzuki Y, Watanabe Y, et al. Mannose-binding lectin polymorphisms in patients with hepatitis C virus infection. Scand J Gastroenterol. 2000;35(9):960–5. Epub 2000/11/04. pmid:11063157.
- 36. Tsutsumi A, Sasaki K, Wakamiya N, Ichikawa K, Atsumi T, Ohtani K, et al. Mannose-binding lectin gene: polymorphisms in Japanese patients with systemic lupus erythematosus, rheumatoid arthritis and Sjogren's syndrome. Genes Immun. 2001;2(2):99–104. Epub 2001/06/08. pmid:11393663.
- 37. Garred P, Thiel S, Madsen HO, Ryder LP, Jensenius JC, Svejgaard A. Gene frequency and partial protein characterization of an allelic variant of mannan binding protein associated with low serum concentrations. Clin Exp Immunol. 1992;90(3):517–21. pmid:1458688; PubMed Central PMCID: PMCPMC1554582.
- 38. Bernig T, Taylor JG, Foster CB, Staats B, Yeager M, Chanock SJ. Sequence analysis of the mannose-binding lectin (MBL2) gene reveals a high degree of heterozygosity with evidence of selection. Genes Immun. 2004;5(6):461–76. pmid:15306844.
- 39. Crosdale DJ, Ollier WE, Thomson W, Dyer PA, Jensenious J, Johnson RW, et al. Mannose binding lectin (MBL) genotype distributions with relation to serum levels in UK Caucasoids. Eur J Immunogenet. 2000;27(3):111–7. Epub 2000/08/12. pmid:10940077.
- 40. Su C, Lin Y, Cai L, Mao Q, Niu J. Association between mannose-binding lectin variants, haplotypes and risk of hepatocellular carcinoma: A case-control study. Sci Rep. 2016;6:32147. Epub 2016/08/26. pmid:27557564; PubMed Central PMCID: PMCPMC4997250.
- 41. Speletas M, Gounaris A, Sevdali E, Kompoti M, Konstantinidi K, Sokou R, et al. MBL2 genotypes and their associations with MBL levels and NICU morbidity in a cohort of Greek neonates. J Immunol Res. 2015;2015:478412. Epub 2015/04/17. pmid:25879044; PubMed Central PMCID: PMCPMC4387979.
- 42. Guttormsen HK, Stuart LM, Shi L, Carroll MC, Chen J, Kasper DL, et al. Deficiency of mannose-binding lectin greatly increases antibody response in a mouse model of vaccination. Clin Immunol. 2009;130(3):264–71. Epub 2008/11/11. pmid:18996748; PubMed Central PMCID: PMCPMC2660855.
- 43. Keller TT, van Leuven SI, Meuwese MC, Wareham NJ, Luben R, Stroes ES, et al. Serum levels of mannose-binding lectin and the risk of future coronary artery disease in apparently healthy men and women. Arterioscler Thromb Vasc Biol. 2006;26(10):2345–50. Epub 2006/08/12. pmid:16902159.
- 44. Cortes C, Rzomp KA, Tvinnereim A, Scidmore MA, Wizel B. Chlamydia pneumoniae inclusion membrane protein Cpn0585 interacts with multiple Rab GTPases. Infect Immun. 2007;75(12):5586–96. pmid:17908815; PubMed Central PMCID: PMCPMC2168330.
- 45. Blasi F, Tarsia P, Arosio C, Fagetti L, Allegra L. Epidemiology of Chlamydia pneumoniae. Clin Microbiol Infect. 1998;4 Suppl 4:S1–S6. pmid:11869264.
- 46. Branden E, Koyi H, Gnarpe J, Gnarpe H, Tornling G. Chronic Chlamydia pneumoniae infection is a risk factor for the development of COPD. Respir Med. 2005;99(1):20–6. Epub 2005/01/28. pmid:15672844.
- 47. Phoon MC, Desbordes C, Howe J, Chow VT. Linoleic and linolelaidic acids differentially influence proliferation and apoptosis of MOLT-4 leukaemia cells. Cell Biol Int. 2001;25(8):777–84. Epub 2001/08/03. pmid:11482901.