| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Lipids, Lipoproteins, and Cardiovascular Risk Factors |
Unit of Diabetes, Endocrinology and Nutrition, Institut d’Investigació Biomédica de Girona, and CIBER Fisiopatologia Obesidad y Nutricion (CB06/03/010), Instituto de Salud Carlos III, Girona, Spain.
aAddress correspondence to these authors at: Unit of Diabetes, Endocrinology and Nutrition, Hospital de Girona "Dr Josep Trueta," ctra. França s/n, 17007 Girona, Spain. Fax 34-972-227 443; e-mail uden.jmfernandezreal{at}htrueta.scs.es. (J.M.F.-R.), h416ummn{at}htrueta.scs.es (J.M.M.-N.).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: We evaluated 2 nonsynonymous LTF polymorphisms (rs1126477 and rs1126478) and measured circulating lactoferrin concentrations by ELISA under nonstressed conditions in healthy Caucasian men (n = 188) and male patients with an altered glucose tolerance (n = 202). We also studied the association of lactoferrin concentration with vascular reactivity via high-resolution ultrasound analysis of the brachial artery in a subsample of study participants.
Results: Circulating lactoferrin concentration was inversely associated with fasting triglyceride concentration (r = –0.24; P = 0.001), body mass index (BMI) (r = –0.20; P = 0.007), waist-to-hip ratio (r = –0.35; P <0.001), and fasting glucose concentration (r = –0.18; P = 0.01), and directly correlated with HDL cholesterol concentration (r = 0.21; P = 0.004). Control AG heterozygotes for rs1126477 had significantly decreased fasting triglyceride concentrations (P = 0.001). Similarly, control individuals who were G carriers for rs1126478 had significantly lower fasting triglyceride concentrations (P = 0.044) and significantly higher HDL cholesterol concentrations (P = 0.028) than AA homozygotes. These associations remained significant after controlling for age, BMI, waist-to-hip ratio, fasting glucose concentration, smoking status, and alcohol intake. Circulating lactoferrin concentration was not significantly associated with endothelium-dependent vasodilatation (EDVD) in the individuals studied (n = 95); however, lactoferrin was positively associated with EDVD in obese participants with an altered glucose tolerance (r = 0.54; P = 0.04).
Conclusions: We have identified associations among LTF polymorphisms, circulating lactoferrin concentration, fasting triglyceride concentration, and vascular reactivity in humans.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-defensins, major constituents of neutrophils with antimicrobial functions, also possess antiinflammatory properties, and others have associated the serum concentrations of these proteins with several metabolic markers, including plasma triglycerides, HDL cholesterol, and LDL cholesterol (7)(8). Lactoferrin, an 80-kDa monomeric multifunctional glycoprotein that binds nonheme iron, consists of 2 lobes, each of which binds a ferric ion. Lactoferrin is produced by neutrophils and by epithelial glands. It is present in all body fluids, being abundant in milk (particularly the colostrum) and other secretions, such as tears and saliva. The physiological roles that have been proposed for lactoferrin include antiinflammatory, immunomodulatory, antimicrobial, antiviral, and antitumoral functions. For this reason, lactoferrin is regarded as a host-defense mediator. Specific lactoferrin receptors exist in a variety of cells, including monocytes, lymphocytes, adipocytes, hepatocytes, and endothelial cells (9).
Bovine lactoferrin has been found to display beneficial effects on plasma lipid concentrations. Its administration in rodents has led to increased plasma HDL cholesterol concentrations, decreased plasma concentrations of triacylglycerol and nonesterified fatty acids, and decreased hepatic cholesterol and triacylglycerol concentrations (10)(11)(12)(13)(14)(15).
Lactoferrin also directly interacts with modified LDL to prevent its interaction with scavenger receptors (16)(17). A region rich in basic amino acid residues near the lactoferrin N terminus is responsible for the interaction with acetylated or oxidized LDL. This cationic part of lactoferrin strongly binds modified LDLs via electrostatic interaction (16) with positively charged Arg residues (isoelectric point, 9) at physiological pHs (18).
Two nonsynonymous single-nucleotide polymorphisms (SNPs)1
in the lactoferrin gene (LTF),2
lactotransferrin) have been associated with amino acid changes in the N-terminal region of lactoferrin. These SNPs lead to a change of Lys to Arg at amino acid residue 47 (rs1126478) and Ala to Thr at residue 29 (rs1126477). We hypothesized that these amino acid modifications could affect lactoferrin interactions with acetylated or oxidized LDL or with LRP1 [LDL- related protein 1 (2-macroglobulin receptor)] to produce differential lipid clearance and altered circulating lipid concentrations.
Our aim in this study was to investigate whether these LTF polymorphisms and the circulating lactoferrin concentration were associated with plasma lipid concentrations and obesity in men with a nonpathologic or an altered glucose tolerance. We also investigated the association of circulating lactoferrin concentration with vascular reactivity in a subsample of the study participants.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Inclusion criteria were (a) a body mass index (BMI) <40 kg/m2, (b) absence of systemic disease, (c) absence of infection within the previous month, and (d) a serum ferritin concentration >10 µg/L and a nonpathologic blood hemoglobin concentration (i.e., >120 g/L) to exclude iron deficiency. None of the control individuals were taking medication or had evidence of metabolic disease other than obesity. Alcohol and caffeine were withheld within 12 h of the oral glucose-tolerance test. Smokers were defined as any person consuming at least 1 cigarette per day in the previous 6 months. Resting blood pressure was measured as previously reported (7)(8). Liver disease and thyroid dysfunction were specifically excluded in a biochemical workup.
To increase the statistical power of the group of patients with type 2 diabetes, we prospectively recruited 90 patients from diabetes outpatient clinics who demonstrated stable metabolic control in the previous 6 months as defined by stable values for glycosylated hemoglobin (HbA1c). Data from these patients were merged with those of patients with recently diagnosed type 2 diabetes. Exclusion criteria for these patients included the following: (a) clinically significant hepatic, neurologic, endocrinologic, or other major systemic disease, including malignancy; (b) a history of or current clinical evidence for hemochromatosis; (c) a history of drug or alcohol abuse, as defined as >80 g of alcohol per day in men and >40 g/day in women, or a serum transaminase activity greater than twice the upper limit of the reference range; (d) an increased serum creatinine concentration; (e) an acute major cardiovascular event in the previous 6 months; (f) acute illnesses and current evidence of acute or chronic inflammatory or infectious diseases; and (g) mental illness rendering the individual unable to understand the nature, scope, and possible consequences of the study. The pharmacologic treatments for the patients were as follows: insulin, 31 patients; metformin, 37 patients; sulfonylureas, 16 patients; statins, 34 patients; fibrates, 9 patients; blood pressure–lowering agents, 38 patients; aspirin, 42 patients; and allopurinol, 3 patients. All participants provided written informed consent after the purpose of the study was explained to them. The institution’s review board approved the protocol. Anthropometric measurements and biochemical assays were performed at the same time.
anthropometric measurements
We studied participants in the postabsorptive state. BMI was calculated as the weight in kilograms divided by height in meters squared. The waists of participants were measured with a soft tape midway between the lowest rib and the iliac crest, and hip circumference was measured at the widest part of the gluteal region. We calculated the waist-to-hip ratio accordingly. We measured blood pressure on the right arm after the participant had rested for 10 min in the supine position; we used a standard sphygmomanometer of appropriate cuff size and recorded the first and fifth phases. Values used in analyses are the means of 3 readings taken at 5-min intervals.
brachial artery vascular reactivity
We assessed vascular reactivity in a subset of 95 consecutive, apparently healthy individuals who agreed to participate further in the study (their clinical and biochemical characteristics did not differ significantly from the entire population).
A high-resolution external ultrasound instrument (Acuson 128XP/10) with a 7.5-MHz linear array transducer (Toshiba SSH-140A) was used to measure changes in brachial artery diameter in response to reactive hyperemia [endothelium-dependent vasodilatation (EDVD)], as described by Celermajer et al. (19). The diameter of the artery lumen was defined as the distance between the leading edge of the echo of the near wall–lumen interface to the leading edge of the echo of the far wall–lumen interface. All scans were taken as electrocardiogram-triggered coincident with the R wave, which corresponds to end-diastole at the brachial artery. All images were recorded with a SuperVHS videotape (Panasonic MD-830AG). EDVD was secondary to hyperemia that was induced by the inflation of a pneumatic tourniquet placed around the forearm distal to the scanned part of the artery and at a pressure of 300 mmHg for 5 min, followed by sudden deflation. EDVD is expressed as the percentage change in arterial diameter 1 minute after hyperemia. Reactive hyperemia is calculated as the percent difference between the maximum flow recorded in the first 15 s after cuff deflation and the flow during the resting scan.
The first scan was recorded in a quiet room after the participant had rested for 10 min in the supine position. The tourniquet was then inflated for 5 min. A second scan was recorded for 90 s, beginning 10 s before cuff deflation. Arterial diameter was assessed in 4 different cardiac cycles for each condition, and the measurements were averaged. The reproducibility of this technique at our center has previously been reported (8). Because the agreement of EDVD measurements among our trained personnel was high, a single observer who was blinded to the participants’ clinical and biochemical characteristics analyzed the scans recorded in the present study.
biochemical assays
Glucose concentrations in serum samples were measured in duplicate with a Beckman Glucose Analyzer II (Beckman Instruments) by the glucose oxidase method. We used a Roche Hitachi 747 instrument to measure total serum cholesterol by the cholesterol esterase/cholesterol oxidase/peroxidase reaction. We quantified HDL cholesterol following precipitation with polyethylene glycol at room temperature, used the Friedewald formula to calculate the concentration of LDL cholesterol, and measured total serum triglycerides by monitoring the reaction of glycerol/phosphate/oxidase and peroxidase (20)(21).
Plasma lactoferrin concentrations were measured with the BIOXYTECH Lactof EIA reagent set (OxisResearch). Plasma samples were diluted and assayed according to the manufacturer’s instructions. Intra- and interassay CVs were between 5% and 10%. The lower detection limit of the assay is 1 µg/L. The degree of cross-reactivity with transferrin was <1%. Intraassay imprecision (CV) was 2.1% for a pool of plasma samples (n = 20) with a median lactoferrin concentration of 404 µg/L and was 3.8% at 774 µg/L. The interassay CVs were 9.6% at 425 µg/L and 6.4% at 825 µg/L. All samples were stored at –80 °C and analyzed on 3 consecutive days.
snp analyses
Genomic DNA was extracted from peripheral blood leukocytes with standard procedures (QIAamp DNA Blood Mini Kit; Qiagen). (The DNA purification protocol can be found in the Data Supplement that accompanies the online version of this article at http://www.clinchem.org/content/vol54/issue2 .) To detect LTF polymorphisms rs1126477 and rs1126478 (National Center for Biotechnology Information), we used a TaqMan-based technology suitable for distinguishing alleles (ABI Prism 7000 Sequence Detection System; Applied Biosystems). We used an Applied Biosystems TaqMan assay with minor groove–binding reporter probes and an end-read protocol. We used PCR conditions recommended by the manufacturer and used a sample containing water (instead of DNA) as a negative control for each PCR run. (Further details regarding the PCR procedure can be found in the online Data Supplement.)
statistical analyses
Statistical analyses were performed with SPSS 12.0 software. Unless otherwise stated, descriptive results for continuous variables are expressed as the mean (SD) for gaussian variables and as the median (interquartile range) for nongaussian variables. Variables that were not normally distributed were logarithmically transformed for subsequent analyses. We used simple correlation (Pearson test) and general linear models to analyze relationships between variables and used unpaired Student t-tests to compare individuals with nonpathologic and altered glucose tolerances. Statistical significance was set at a P value of <0.05. We used the 
2 test to compare genotype and allele frequencies according to glucose-tolerance status.
For a P value of 0.05, the study had a 98% power to detect significant correlations between plasma lactoferrin and metabolic variables (Pearson coefficient of at least 0.30) in bilateral tests in all participants studied and had a 91% power for individuals with an altered glucose tolerance. The study also had a 99% power to detect significant differences in plasma triglyceride concentrations according to LTF polymorphism in individuals with a nonpathologic glucose tolerance.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and 2
summarize the main characteristics of the individuals included in this study. Both the G5385A and A5440G LTF polymorphisms were in Hardy–Weinberg equilibrium, and the distributions of the genotypes were similar for participants with and without an altered glucose tolerance or an obesity status. Individuals with a nonpathologic glucose tolerance who were AG heterozygotes (we detected no homozygous individuals) for the G5385A LTF polymorphism (rs1126477) had significantly decreased fasting triglyceride concentrations (P = 0.001; Table 1
). These associations remained significant after we controlled for age, BMI, waist-to-hip ratio, fasting glucose concentration, alcohol intake, and smoking status. In fact, an analysis of a general linear model for predicting the fasting triglyceride concentration revealed that waist-to-hip ratio (P = 0.004), smoking status (P = 0.004), and the G5385A LTF polymorphism (P = 0.02) contributed independently to 10.2% of the variance in fasting triglyceride concentration (Table 3
).
|
|
|
Similarly, individuals with a nonpathologic glucose tolerance who were G carriers for the A5440G LTF polymorphism (rs1126478) had significantly lower fasting triglyceride concentrations and significantly higher HDL cholesterol concentrations than AA homozygotes (Table 2
). This association remained significant (P = 0.04) after we controlled for age, BMI, waist-to-hip ratio, fasting glucose concentration, smoking status, and alcohol intake; this polymorphism independently contributed to 9% of the variance in fasting triglyceride concentration (Table 3
).
We observed no significant associations with LTF polymorphisms, including in analyses with the general linear model, in study participants with an altered glucose tolerance. Associations remained nonsignificant after including or excluding individuals taking statins or fibrates and after controlling for different hypoglycemic agents.
We next evaluated circulating lactoferrin concentration with respect to the clinical and biochemical variables of 186 consecutive individuals (71 with a nonpathologic glucose tolerance and 115 individuals with an altered glucose tolerance, Table 4
) for whom we had access to plasma samples. Lactoferrin spuriously increases in serum because of neutrophil degranulation during clot retraction. Interestingly, plasma lactoferrin concentration was inversely correlated with BMI, waist-to-hip ratio, and fasting glucose and triglyceride concentrations (Fig. 1
, upper panel) and was directly correlated with HDL and LDL cholesterol concentrations (Table 4
). These associations were strengthened in the individuals with an altered glucose tolerance (Table 4
), in whom the mean circulating lactoferrin concentration was significantly decreased (Table 1
).
|
|
LTF polymorphism was not significantly associated with circulating lactoferrin concentration, even after we controlled for different confounding factors (Table 2
); however, the associations between circulating lactoferrin concentration and plasma lipids were different, depending on the rs1126478 genotype, in individuals with an altered glucose tolerance. Although the association between lactoferrin and fasting triglyceride concentrations was significant in G carriers [r = –0.34; P = 0.005 (n = 69)], it was nonsignificant in AA homozygotes [r = 0.13; P = 0.3 (n = 55)]. Similarly, although the association between lactoferrin and HDL cholesterol concentrations was significant in G carriers (r = 0.25; P = 0.04), it was nonsignificant in AA homozygotes (r = 0.07; P = 0.6). In fact, G carriers with a lactoferrin concentration greater than the median of 408 µg/L had a significantly lower mean (SD) fasting triglyceride concentration {1.539 (0.817) mmol/L [1363 (724) mg/L] vs 2.113 (1.191) mmol/L [1872 (1055) mg/L]; P = 0.02} and a significantly higher HDL cholesterol concentration {1.37 (0.325) mmol/L [529 (126) mg/L] vs 1.17 (0.331) mmol/L [453 (128) mg/L]; P = 0.02} than AA homozygotes.
lactoferrin and endothelial function
Ninety-five consecutive participants for whom plasma samples were available agreed to a study of vascular reactivity. The clinical characteristics of these individuals did not differ significantly from those shown in Table 2
. We observed no association between circulating lactoferrin concentration and EDVD (r = 0.03; P = 0.7); however, in the subset of obese participants with an altered glucose tolerance (n = 15), we found circulating lactoferrin concentration to be positively associated with EDVD (r = 0.54; P = 0.04).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Virtually all body fluids contain lactoferrin, but it is especially abundant in milk. Ingestion of protein from milk especially affects plasma lipid concentrations, and whey protein has also been shown to lower plasma lipids (10). In 2004, Takeuchi et al. were the first to demonstrate that milk-derived bovine lactoferrin mixed with a standard commercial diet reduced the plasma and hepatic concentrations of cholesterol and triglycerides in mice and increased plasma concentrations of HDL cholesterol; however, lactoferrin had no significant effects on lipids in mice fed a high-fat diet (11).
Neutrophils also produce lactoferrin and store it in lactoferrin granules as they mature, and neutrophils are the main source of the lactoferrin measured in this study. Lactoferrin has previously been described to inhibit the selective uptake of HDL cholesteryl esters by 35%–50% in human primary adipocytes and SW872 liposarcoma cells (22). This action is mediated via interaction with LRP, which contributes physiologically to the selective uptake of HDL cholesteryl ester in adipocytes. Lactoferrin also inhibits the interaction of lipoprotein lipase with LRP.
A striking feature of lactoferrin is its ability to inhibit, both in vivo and in vitro, the binding and uptake of apolipoprotein E–bearing lipoproteins by parenchymal liver cells. Lactoferrin Arg residues are crucial for its recognition by parenchymal liver cells and its capacity to inhibit the hepatic uptake of apolipoprotein E–bearing lipoproteins. This conclusion is based on experiments in which lactoferrin Arg residues were selectively modified by 1,2-cyclohexanedione. This modification produced a marked reduction in the liver uptake of lactoferrin. We have described how healthy individuals carrying an SNP associated with substitution of an Arg residue at position 47 of lactoferrin (an A5440G nonsynonymous gene polymorphism) had significantly higher HDL cholesterol concentrations and significantly lower fasting triglyceride concentrations than individuals with a Lys residue at this position in the lactoferrin molecule. We hypothesize that this change may lead to differential interactions with apolipoprotein E and LRP1, leading to changes in plasma lipid concentrations. In the G5385A nonsynonymous LTF polymorphism, healthy individuals with the AG genotype (associated with an Ala-to-Thr change at position 29) had significantly and markedly lower fasting triglyceride concentrations than individuals with the GG genotype (the amino acid change is found within the sequence that interacts with LRP1; unfortunately, we found no AA homozygotes for this polymorphism) (14)(23). It is possible, therefore, that Thr and Arg residues at these 2 positions favor a stronger interaction with LRP1.
The absence of differences in the plasma lipid profile according to LTF polymorphisms in patients with an altered glucose tolerance could be attributed to the concomitant treatments of these individuals (with statins, fibrates, insulin, and oral antidiabetic drugs). Even after controlling for these factors, we found no significant differences. One could argue that the heterogeneity of the individuals with an altered glucose tolerance and the heterogeneity in their treatments could be behind this absence of such associations in this group; however, it is interesting that the decreased lactoferrin concentrations in these individuals were significantly associated with HDL cholesterol and fasting triglyceride concentrations among carriers of the polymorphisms. We speculate that the preservation of lactoferrin production in individuals with an altered glucose tolerance is critical for maintaining an adequate lipid profile among carriers of these polymorphisms.
Other mechanisms through which lactoferrin concentration is associated with the lipid profile cannot be excluded. In fact, bovine lactoferrin was found to reduce the accumulation of cholesteryl esters in macrophages incubated with acetylated LDL by more than 80% compared with the control value. Treatment with bovine lactoferrin reportedly also leads to decreased intestinal absorption of triacylglycerols via lymphatic pathways (11)(24).
The circulating concentration of lactoferrin was also positively associated with EDVD among the subsample of obese men with an altered glucose tolerance. There is evidence that lactoferrin increases nitric oxide production (25). In fact, lactoferrin has been shown to affect peripheral opioid-mediated antinociception via nitric oxide (26)(27).
In summary, the plasma concentration of lactoferrin and LTF polymorphisms were associated with the plasma lipid profile and EDVD. The decreased values for lactoferrin in individuals with obesity, a high waist-to-hip ratio, and an altered glucose tolerance may contribute to dyslipidemia in these individuals. Given the relatively small sample size in this study, however, our findings require replication in other large, prospective studies. The mechanisms responsible for the different associations should be investigated further.
|
Acknowledgments |
|---|
Financial Disclosures: None declared.
Acknowledgments: We thank Dr. Maria Garcia for reviewing the statistics of this report.
|
Footnotes |
|---|
2 Human genes: LTF, lactotransferrin.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  J. M. Moreno-Navarrete, F. J. Ortega, J. Bassols, W. Ricart, and J. M. Fernandez-Real Decreased Circulating Lactoferrin in Insulin Resistance and Altered Glucose Tolerance as a Possible Marker of Neutrophil Dysfunction in Type 2 Diabetes J. Clin. Endocrinol. Metab., October 1, 2009; 94(10): 4036 - 4044. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
