Abstract
Context:
Corticosteroid-binding globulin (CBG) is the binding protein for cortisol. Rare kindreds with CBG mutations reducing CBG levels or altering binding affinity have been described, along with clinical manifestations encompassing fatigue, chronic pain, and hypotension. The largest kindred, exhibiting two mutations (null and Lyon) were Australian immigrants from Italy.
Objective:
Our objective was to determine the prevalence of the null/Lyon mutations in the village where the original null/Lyon family emigrated and compare subjects with and without CBG mutations, without previous knowledge of their mutation status.
Design, Setting, and Participants:
We conducted a survey field study that included 495 adult residents.
Main Outcomes:
We assessed clinical history, CBG mutation analysis, plasma CBG, salivary cortisol, body mass index, waist circumference, blood pressure, and the Krupp fatigue scale.
Results:
Eighteen of 495 participants (3.6%, seven males and 11 females) had one of two function-altering CBG mutations. All were heterozygous for the null (n = 6) or Lyon mutations (n = 12). Of 12 Lyon participants (four males and eight females), eight (two males and six females) had chronic widespread pain and five osteoarthritis with associated pain (one male and four females). Of six null participants (three males and three females), three (one male and two females) had chronic pain and four osteoarthritis with associated pain (two males and two females).
Conclusions:
A high combined prevalence (3.6%) of these two CBG mutations was detected. The presence of either mutation conferred a propensity to chronic pain. In other communities, individuals with the same genetic background complain more of fatigue than pain, suggesting an environmental effect on the phenotype. These findings, combined with animal CBG gene knockout and human CBG single-nucleotide polymorphism haplotype studies, suggest that CBG influences the endocrine and neurobehavioral response to stress, including the development of pain/fatigue syndromes.
Corticosteroid-binding globulin (CBG), or transcortin, is the 383-amino-acid circulating glycoprotein that binds cortisol with high affinity (1). Under basal unstressed conditions, 80% of plasma cortisol is CBG bound, 10–20% is albumin bound, and 5–10% is free (2). Human CBG protein is encoded by a 19-kb gene located at chromosome 14q31-q32.1, among a group of related serine protease inhibitor genes, with the complete coding sequence spanning exons 2–5 (3, 4).
The roles of CBG include cortisol transport and a reservoir for cortisol in plasma. Other putative roles involve specific tissue delivery of cortisol, participation in the acute stress response, and a role in metabolism. CBG is cleaved by neutrophil elastase, allowing selective cortisol delivery at inflammatory sites (5). CBG concentrations fall in acute inflammation, partly through inhibition of CBG synthesis by inflammatory cytokines such as IL-6 (6, 7). Consequent enhanced elevation of free cortisol levels through suppression of CBG may amplify the hypothalamic-pituitary-adrenal (HPA) axis response. CBG may influence metabolism; levels are reduced in association with elevated IL-6 levels in obesity (8) and fall precipitously in synchrony with HPA activation in severe weight loss/starvation but not in modest weight loss/caloric restriction (9, 10).
Four CBG gene mutations have been described in five kindreds; three of these (Leuven, Lyon, and c776,G→T) reduce cortisol binding affinity, and one (null) reduces CBG levels by 50% (heterozygous) or 100% (homozygous) (11–15). Observations of the clinical manifestations of these mutations have repeatedly suggested fatigue, asthenia, pain, and, in some cases, low blood pressure relative to population norms (16). Typically, probands present with fatigue, leading to cortisol measurements (16). Often the lack of a fully developed clinical picture of hypocortisolism or a normal urine free cortisol in the face of low plasma cortisol then led to suspicion of CBG deficiency (16). Ascertainment bias is therefore possible.
The largest kindred reported was a 39-member, four-generation, Italian-Australian family, which independently segregated for the Lyon mutation as well as a new mutation, the null mutation (12). The null mutation (G121A) leads to a premature stop codon (Trp-12X) in the first coding exon, exon 2, and, hence, no protein expression. In this family, three homozygote subjects for the null mutation had no detectable plasma CBG, whereas 19 heterozygotes for the null mutation had CBG levels 50% below the normal limit. Two subjects were heterozygous for both the null and the Lyon mutations; these compound heterozygous subjects had lower CBG levels compared with null heterozygous. Fatigue was a frequent complaint in subjects with either mutation, blood pressure was reduced relative to community controls, and chronic idiopathic pain was observed in six of 24 null mutation subjects (12).
The aim of this study was to determine the combined prevalence of the null/Lyon mutations in the Calabrian village where the original null/Lyon family emigrated and comprehensively compare the clinical status of people with and without CBG mutations while blinded to their mutation status.
Subjects and Methods
Study recruitment
We systematically evaluated adults living in a geographically isolated village over 1000 m above sea level, Calabria, southern Italy. Funding allowed us to survey 495 participants (41% of the population). Recruitment was performed by the local general medical practitioner (C.M.) serving the vast majority of adults living in this area. All participants signed a written informed consent approved by the local health authority Ethics Committee (Azienda Sanitaria Locale 6 of Lamezia Terme, Catanzaro, Italy). Participants were informed that the aim of the study was to assess the effects of genetic variants of CBG. Almost all of those invited to participate agreed. No monetary compensation was offered; however, participants received information with medical interpretation about fasting glucose, lipid panel, and other biochemical data of clinical relevance.
Study geographic area and population
Due to social and cultural factors, as well as its history and rugged territory, Calabria has a very stable population with a substantial degree of consanguinity, as indicated by surname analysis coupled with analysis of genetic markers (17). Emigration to distant areas, such as North America and Australia, often involving whole families was common. The study village was founded in the mid-19th century by five families who moved from a nearby town.
Procedures
The study procedures detailed below were completed after written, informed consent was obtained.
Clinical measurements
Study personnel (nurses) were trained by physicians and undertook a quality control session before performing a standardized medical history, physical examination, and sample collections. Study participants were weighed barefoot, in light clothes on the same nondigital scale; height was measured by a stadiometer. Body mass index (BMI) was calculated as kilograms per square meter. Waist circumference was measured at the midpoint of the iliac crest and ribcage. Sitting and standing blood pressure was recorded in the nondominant arm with an aneroid sphygmomanometer, which had been validated against a mercury column, repeated 5 min later.
The Italian version of The Quality of Life Abbreviated Questionnaire (WHOQOL-Brief), a widely used standardized questionnaire, was administered to assess quality of life (18). In addition, the fatigue severity scale, a 28-item questionnaire previously validated in patients with multiple sclerosis, systemic lupus erythematosus, and other chronic conditions, was administered (19).
Biochemistry: CBG, salivary cortisol, ACTH test, and metabolic parameters
A fasting venous blood sample (20 ml) was collected in EDTA-containing tubes between 0800 and 1000 h. Total and differential blood count, erythrocytes sedimentation rate, glucose, blood urea nitrogen, creatinine, electrolytes, uric acid, total cholesterol, triglycerides, HDL, γ-glutamyl transpeptidase, transaminases, free T3, free T4, TSH, thyroglobulin, and thyroid peroxidase antibodies were determined. A saliva sample was collected into a plastic tube between 0800 and 1000 h (Becton Dickinson, Hialeah, FL).
Plasma CBG was measured using an in-house method involving a two-site noncompetitive direct ELISA for CBG, using monoclonal and polyclonal antibodies (20). ELISA plates (Falcon 3912 microtest III; Beckton Dickinson, Oxnard, CA) were coated for 5–7 d at 40 C with 100 μg/well of sheep anti-CBG serum, diluted in PBS, 30 μl of antibody in 10 ml PBS. After coating, the plates were washed and blocked with assay buffer (150 μl/well) for 1 h at 20 C. The plates were emptied by inversion, and 100 μl of either standard or patient plasma (1:1000 dilution in assay buffer) was added to each well. After overnight incubation at 4 C, the plates were washed, and supernatant 12G2 monoclonal antibody (1:20 dilution) was added for 2 h at 20 C. The plates were then washed, and 100 μl of antimouse IgG Fc-peroxidase (1:1000 dilution) was added for 1 h. After the plates were washed and substrate added, color was developed and then terminated before spectrophotometry at 450 nm. The value assigned to the standard plasma was determined by calibration with purified CBG and an assumed molecular weight of 52,000.
Salivary cortisol was measured using a commercial RIA kit (Radim Diagnostics, Rome, Italy). The reference range for morning salivary cortisol was 0.3–1.0 μg/dl. An ACTH stimulation test was performed in five Lyon heterozygotes available for the test. Starting at 1400 h (time 0), ACTH 1–24 (250 μg) iv push was given. Total plasma cortisol and aldosterone were measured at −1, 30, and 60 min. Aldosterone is not bound to CBG but responds to ACTH acutely; its measurement was included as an additional test of adrenal function.
CBG gene testing
Genomic DNA was extracted from blood buffy coats using standard phenol-chloroform procedures. Exons 2 and 5 of the CBG gene were amplified using specific previously published primers (12, 21). A total of 200 ng genomic DNA was amplified in a 50-μl reaction mixture containing 20 pmol of each primer, 0.2 mm dNTPs, 1 U Taq (Eppendorf AG, Hamburg, Germany), 1.5 mm Mg(OAc)2, and 1× buffer. PCR products were subjected to restriction fragment length polymorphism analysis using the restriction enzymes BfaI (New England Biolabs, Ipswich, MA) for exon 2 and TaqI (New England Biolabs) for exon 5, as previously described (12). The resulting fragments were separated on a 2% agarose gel and visualized on a UV transilluminator after ethidium bromide staining.
Pedigrees of the CBG mutation kindreds
Each study participant was queried about their spouse, their parents' names, and dates of birth and death, when applicable. By repeating the same procedure for each enrolled participant, we were able to reconstruct a large, albeit not exhaustive, map of family links in this small village. A computerized database was maintained, including demographic, anthropometric, clinical, and genetic information of the population sample.
Statistical analyses
All analyses were performed using SPSS software (version 11.5). A χ2 test was used to compare categorical variables and Student's t test for continuous variables. Significance was accepted at level of P value <0.05.
Results
CBG null/Lyon participants
Eighteen participants (3.6% of the population) were heterozygous for one or other CBG mutation. Six of the 18 had the null mutation (three males and three females), whereas 12 had the Lyon mutation (four males and eight females). There were no homozygotes and no compound heterozygotes. Table 1 describes the medical diagnoses of the null/Lyon heterozygotes. Pedigree reconstruction revealed that these 18 null/Lyon heterozygotes belonged to eight apparently unrelated pedigrees (Fig. 1, A and B). The largest pedigree contained eight participants carrying the Lyon mutation over three generations (Fig. 1A). The remaining participants carrying the null (n = 6) or the Lyon (n = 4) mutations belonged to seven different pedigrees (Fig. 1B). General clinical measures are compared between the null and Lyon participants in Table 2.
Table 1.
Clinical characteristics of the null and Lyon participants
| No. DNA | Age (yr)/gender | Smoking and alcohol abuse | Medical diagnoses |
|---|---|---|---|
| Lyon mutation carriers | |||
| 19 | 90/M | Smoking, alcohol abuse | Chronic liver disease |
| 81 | 60/F | None | Low back pain, osteoarthritis with associated pain, liver function test abnormalities, diabetes mellitus type II, hypertension, hypertriglyceridemia, hypercholesterolemia |
| 82 | 55/M | Alcohol abuse | Chronic pain, chronic cystitis, chronic prostatitis, liver function test abnormalities |
| 85 | 22/F | None | None |
| 103 | 43/F | None | Chronic pain, chronic cystitis, chronic headache |
| 105 | 77/F | None | Osteoarthritis with associated pain, liver function test abnormalities, urinary calculosis, hypercholesterolemia |
| 130 | 20/M | None | Congenital pulmonary stenosis |
| 176 | 84/F | Alcohol abuse | Chronic muscle and joint pain, osteoarthritis with associated pain, chronic headache, liver function test abnormalities, hypertension, mild cognitive impairment |
| 195 | 58/F | None | Chronic muscle and joint pain, chronic headache, hypertension |
| 214 | 58/M | None | Chronic muscle and joint pain, osteoarthritis with associated pain, peptic ulcer |
| 215 | 54/F | None | Chronic muscle and joint pain, chronic headache, obesity, diabetes mellitus type II, hyperthyroidism, urinary calculosis |
| 220 | 87/F | None | Chronic muscle and joint pain, osteoarthritis with associated pain, chronic headache, hypertriglyceridemia, hypercholesterolemia |
| Null mutation carriers | |||
| 192 | 70/M | None | Chronic muscle and joint pain, OSteoarthritis with associated pain, hypertriglyceridemia, hypercholesterolemia |
| 197 | 26/F | None | None |
| 218 | 34/M | None | Ulcerative colitis |
| 246 | 62/F | Alcohol abuse | Chronic muscle and joint pain, osteoarthritis with associated pain, tender nodes, chronic headache, liver function test abnormalities, hypertension, urinary calculosis, chronic headache, osteoporosis, hypercholesterolemia |
| 299 | 65/F | None | Chronic muscle and joint pain, osteoarthritis with associated pain, chronic headache, tender cervical nodes, liver function test abnormalities, diabetes mellitus type II, depression, hypertension, heart murmur, hypercholesterolemia, diverticulitis |
| 1138 | 66/M | Alcohol abuse | Osteoarthritis with associated pain, liver function test abnormalities, hypertriglyceridemia, hypertension |
F, Female; M, male; No. DNA, unique number assigned to each DNA sample.
Fig. 1.
A, Pedigree from the Tas family: eight family members of this kindred were heterozygous for the Lyon mutation, localized on exon 5; B, seven small family pedigrees representing 10 mutated subjects. Circles represent women, squares men, slashes deceased, and half-filled symbols mutated in exon 5 members. Numbers inside the symbol indicate the number of individuals.
Table 2.
Clinical and laboratory measures of the null and Lyon mutation participants and controls
| Men, n = 230 |
Women, n = 265 |
Null, n = 6 (3 M, 3 F) | Lyon, n = 12 (4 M, 8 F) | |||
|---|---|---|---|---|---|---|
| Controls, n = 223 | Null/lyon, n = 7 | Controls, n = 254 | Null/lyon, n = 11 | |||
| Age (yr) | 50.4 ± 17.5 (19–91) | 56.2 ± 23.4 (20–90) | 50.0 ± 19.7 (18–91) | 57.1 ± 22.3 (22–87) | 53.7 ± 18.9 (26–70) | 58.8 ± 23.2 (20–90) |
| BMI [% (n)] | ||||||
| Underweight (<18.5 kg/m2) and normal (18.5–25 kg/m2) | 39.2 (87) | 71.4 (5) | 39.9 (101) | 36.4 (4) | 16.7 (1) | 66.7 (8) |
| Overweight (25–30 kg/m2) and obese (>30 kg/m2) | 60.8 (135) | 28.6 (2) | 60.1 (152) | 63.6 (7) | 83.3 (5) | 33.3 (4) |
| Waist (cm) | 95.1 ± 10.6 (53–124) | 86.7 ± 11.0 (69–100), P < 0.041 vs. controls | 89.6 ± 12.5 (53–130) | 94.4 ± 16.8 (70–127) | 94.3 ± 6.2 (84–100) | 89.9 ± 17.9 (70–127) |
| Systolic blood pressure (mm Hg) | 138.8 ± 23.4 (100–200) | 133.9 ± 17.4 (120–160) | 140.4 ± 26.0 (90–240) | 146.4 ± 28.0 (90–180) | 135.0 ± 28.8 (90–170) | 144.8 ± 23 (120–180) |
| Diastolic blood pressure (mm Hg) | 78.6 ± 11.9 (50–110) | 75.0 ± 13.2 (60–100) | 78.9 ± 12.7 (50–125) | 79.1 ± 13.0 (50–100) | 79.1 ± 13.0 (50–100) | 78.7 ± 11.3 (60–100) |
| Morning salivary cortisol (μ g/dl) | 0.8 ± 0.3 (0.2–2.1) | 0.7 ± 0.4 (0.1–1.2) | 0.9 ± 0.3 (0.1–2.5) | 0.8 ± 0.3 (0.5–1.3) | 0.6 ± 0.3 (0.1–0.8) | 0.9 ± 0.3 (0.5–1.2), P < 0.011 vs. null |
| Plasma CBG concentration (nmol/liter) | 558.0 ± 153.6 (255–1027) | 406.0 ± 83.7 (305–503), P < 0.017 vs. controls | 577.2 ± 151.5 (256–1123) | 380.9 ± 83.2 (221–485), P < 0.0005 vs. controls | 297.2 ± 65.2 (221–413) | 435.3 ± 38.3 (392–503) |
| Fasting glucose (mg/dl) | 92.1 ± 17.5 (64–207) | 85.4 ± 11.8 (71–103) | 89.4 ± 21.9 (60–298) | 107.9 ± 39.0 (77–186) | 102 ± 42 (71–186) | 97.7 ± 29.1 (75–181) |
| Fasting cholesterol (mg/dl) | 208.4 ± 46.6 (102–350) | 201.9 ± 39.1 (132–241) | 217.3 ± 48.6 (109–360) | 201.6 ± 33.7 (125–248) | 209.2 ± 42.6 (125–241) | 198 ± 31.5 (132–248) |
| Fasting HDL (mg/dl) | 65.4 ± 20.8 (31–162) | 67.3 ± 15.6 (41–90) | 71.0 ± 18.0 (31–163) | 69.6 ± 14.6 (49–96) | 59.3 ± 13.1 (41–66) | 73.4 ± 13.3 (54–96), P < 0.049 vs. null |
| Fasting triglycerides (mg/dl) | 119.8 ± 117.2 (30–1248) | 92.0 ± 39.3 (47–166) | 111.4 ± 79.3 (30–641) | 106.1 ± 49.5 (44–191) | 122 ± 46.9 (44–166) | 89.9 ± 42 (47–191) |
Unless indicated otherwise, data are expressed as mean ± sd (range). Significant P values are shown. F, Female; HDL, high-density lipoprotein; M, male.
CBG and cortisol levels
Plasma CBG levels and morning salivary cortisol levels in subjects with the null, Lyon and in controls are shown in Fig. 2. Subjects carrying the Lyon or the null mutation had significantly lower plasma CBG levels than control subjects; in addition, subjects carrying the null mutation had significantly lower levels than subjects carrying the Lyon mutation (Fig 2, upper panel). There were no differences in morning salivary cortisol concentrations between subjects carrying a mutation and control subjects; salivary cortisol was, however, lower in participants with the null mutation compared with participants with the Lyon mutation (Fig 2, lower panel).
Fig. 2.
Distribution of plasma CBG concentrations (upper panel) and salivary cortisol (lower panel) in CBG null or Lyon heterozygotes and controls.
Chronic pain and fatigue
The 18 participants heterozygous for CBG mutations exhibited a high prevalence of various kinds of chronic pain (Table 1). More than half of the participants carrying either mutation complained of muscle and joint pain, tender nodes, headache, and low back pain. Of 12 Lyon participants (four males and eight females), eight (two males and six females) had chronic pain, and five had pain attributed to osteoarthritis (one male and four females). Of six null participants (three males and three females), three (one male and two females) had chronic pain and four had pain attributed to osteoarthritis (two males and two females). The prevalence of chronic pain, assessed in all participants by clinical history and WHOQOL-Brief was higher, 72 vs. 30%, in subjects carrying a mutation vs. controls. Fatigue was not a spontaneous complaint on history, and the fatigue questionnaire score in CBG mutation participants was similar to controls (1.06 ± 0.24 vs. 1.09 ± 0.43).
Cortisol responses to cosyntropin (ACTH 1–24) iv were normal, as indicated by peak cortisol levels exceeding 20 μg/dl. The aldosterone response to ACTH was also normal, with plasma levels exceeding 200 pg/ml in each subject (Fig. 3).
Fig. 3.
A, ACTH 1–24 250 μg was administered and plasma cortisol measured at time 0, 30, and 60 min after injection. The dotted line at 200 pg/ml indicates a normal response 30 and/or 60 min after injection. B, Aldosterone responses after ACTH 1–24 250 μg injection.
Analyses of the relationships between plasma CBG and various parameters in controls
There was no relationship between plasma CBG and systolic blood pressure (r = 0.008; P = 0.8), diastolic blood pressure (r = 0.036; P = 0.50), weight (r = 0.013; P = 0.81), BMI (r = 0.067; P = 0.21), waist circumference (r = 0.026; P = 0.63), salivary cortisol (r = −0.024; P = 0.66), or age (r = 0.024; P = 0.66). A range of clinical measures are compared between the CBG mutation subjects and controls in Table 2. Approximately 60% of men and women were either overweight or obese. Twenty-five percent of men and half of women had abdominal obesity. Half of the men and 25% of the women had hypertension (data not shown).
Discussion
We performed a survey of CBG mutations in the Italian village of origin of a previously reported Italian-Australian kindred (12). Of 495 subjects screened, we found 18 individuals that were heterozygous for either the null or Lyon mutation.
The combined prevalence of the two CBG mutations in this survey was 3.6%. CBG mutations are presumed to be rare, given the small numbers of reports of only four function-altering mutations. One other population screen for CBG mutations was conducted in 248 patients with chronic fatigue syndrome. No CBG mutations were detected by full exon sequencing (22). We cannot definitely account for the elevated frequency of two different CBG mutations observed in this study. CBG mutations are not known to cause a reproductive disadvantage. Isolated areas such as the one studied tend to have high levels of consanguinity and parental isonomy. We conducted a review of the phone book in this area, which revealed few surnames, an observation that has been shown to associate with genetic homogeneity (17). This phenomenon likely reduces community genetic variation, allowing rare mutations to be preserved over time. By applying a genealogical method coupled with molecular biology techniques and histopathology, Bruni et al. (23) described in Calabria two large families, with early-onset Alzheimer's disease due to a mutation of a protein called presenilin 1; the ancestry of these two large families has been reconstructed back to the 17th century over 11 generations and 138 documented affected subjects. Hence, taking into account our knowledge of the particular population, the high prevalence of these mutations may relate to a founder effect in a very stable population. These circumstances allowed study of the effects of these mutations without the possible effect of ascertainment bias arising from the detection of CBG mutations in patients who have presented with clinical features, in most reports suggestive of cortisol deficiency.
Consistent with previous reports, plasma CBG levels were reduced by approximately 50% in null heterozygotes and reduced by 20% in Lyon heterozygotes (12, 13). The explanation for 50% reduced CBG levels in null heterozygotes is expected because one allele is nonfunctional. The explanation for reduced CBG levels in those with the Lyon mutation is unclear; however, we would not expect the Lyon mutation to interfere directly with the epitope for our in-house monoclonal CBG assay. Reduced CBG levels have also been seen in Lyon subjects studied with a commercial polyclonal immunoassay method (13).
Cortisol levels were measured using morning salivary cortisol levels and an ACTH stimulation test. Slightly lower salivary cortisol levels were seen in the null than Lyon subjects; however, these levels were similar to controls. Although statistically significant (null vs. Lyon), the small numbers of CBG null subjects and the wide variation in timing of saliva sampling (0800–1000 h) reduce confidence in these findings. Circadian testing of free cortisol levels in a null homozygote revealed normal free cortisol levels (2). On the other hand, a study of a new CBG variant without cortisol binding has suggested altered cortisol pulsatility (15). ACTH stimulation tests in this study, involving five available Lyon heterozygotes, showed normal cortisol (and aldosterone) responses relative to controls.
Over 70% of the null and Lyon heterozygotes, especially women, reported chronic pain at the clinical review of systems conducted during screening, while blinded to the DNA findings. A greater prevalence of idiopathic chronic pain is generally noted in women (24). There are known interrelationships between the stress system and pain. Individuals with chronic idiopathic pain, such as fibromyalgia, have reduced cortisol levels relative to the healthy population (25). Cortisol deficiency, whether induced or spontaneous, lowers the pain threshold (26, 27).
Mutations in the CBG gene may have accounted for the increased prevalence of pain. A recent population study of seven HPA axis-related genes in 994 subjects, the EPIFUND Study, found associations between several single-nucleotide polymorphism (SNP) haplotypes of the CBG (SERPINA6) gene and chronic widespread pain (28, 29). There was also a SNP haplotype that was protective against pain and was associated with reduced risk of depression and improved sleep (29). Hence, the findings in people with severe CBG mutations, and in population studies, suggest a role for CBG in the chronic neurobehavioral response to stress.
Novel mechanisms of allosteric transition of the structure of the CBG molecule are being uncovered; such changes in spatial conformation specifically affect the hormone-binding site and, hence, the propensity of a given structural state of the CBG molecule to retaining vs. releasing cortisol (30). Subtle genetic variations uncovered by these haplotypes may alter hormone release function. Because allosteric changes are modulated by temperature, we hypothesize that these mechanisms may become operational in the face of physiological conditions such as fever-induced hyperthermia.
In this population, participants with CBG mutations did not complain of chronic fatigue, as further evidenced by the fatigue questionnaire scores, which were similar to controls. Although fatigue was the prominent symptom in the original reports of null and Lyon mutations in humans, it was not universal (12, 13). Similar to this study, six of 24 null mutation subjects exhibited single or multiregional chronic unexplained pain, such as migraines, nonspecific headaches, lumbar back pain, pelvic pain, or fibromyalgia in the Australian emigrant family and in other reported families (12, 13, 31).
Blood pressure was not different in subjects carrying a mutation vs. controls. The original family report displayed lower blood pressure compared with population norms for age and gender (12).
We found no association between CBG plasma levels and BMI in this population. CBG levels have been shown to be inversely associated with obesity and insulin resistance in a population study (8), and cytokines and adipokines that are produced by the adipose tissue, particularly IL-6, decrease CBG levels (8, 13). There has been a report of altered adipocyte function ex vivo using cells from a null homozygote (32).
We did not observe an effect of CBG mutations on functional status; each one of these 18 subjects, including a 90-yr-old man were living independently. No major disabilities were reported in the other series of subjects with the null/Lyon mutations; however, subjects did report reduced functionality due to fatigue, especially in the mornings, and difficulty completing tasks relative to their aspirations or the performance of unaffected family members (12, 13).
Recently, two animal models of CBG deficiency have suggested that the CBG gene may play an important role in the stress response. In 2006, the first genetic mouse model of CBG deficiency showed viable animals that reproduced normally (33). Total plasma corticosterone was low, as expected, but free corticosterone levels were elevated relative to controls. There was a greater mortality from the sepsis model induced by bacterial lipopolysaccharide, a finding that may relate to CBG's apparent role in cortisol release by neutrophil elastase at inflammatory sites or altered cortisol release with inflammatory stress, as modulated by hyperthermia (5, 30, 33). In addition, reduced activity levels were noted, possibly a corollary to the fatigue reported in some kindreds with CBG deficiency (12, 13, 33). There was no evidence of enhanced susceptibility to infection in the published cross-sectional family series of functional CBG mutations, nor was this evident in the current sample; however, this may be obscured by low community rates of serious infections.
A second mouse model genetically deficient in CBG revealed normal basal free corticosterone levels (34). Plasma free corticosterone responses to mild stress were reduced, but behavior was normal. Activity levels in the CBG-deficient mice appeared normal. In a severe, uncontrollable stress model (foot shock) there was a marked behavioral change in the CBG-deficient mice, with increased escape failures and an increase in the learned helplessness response, a depression-like behavioral response, as well as reduced free corticosterone responses (34).
Limitations of our study included incomplete survey of the whole population and an inability to reconstruct all pedigrees, because historical archives were not available in the study village. The findings of increased chronic pain with a prevalence much higher in the mutated subjects (72%) than expected in the local population (30%) and in the European population (20.3%) (35) are of interest and suggest that more detailed HPA axis and behavioral/mood studies will be required in this and other human populations with CBG mutations or mild functional variants. We performed restricted genetic analyses aimed at detecting the two mutations observed in family members that have migrated elsewhere. It is very unlikely that full sequencing would have revealed major CBG mutations. In this regard, CBG levels, performed in all participants, did not reveal any additional cases of severe CBG deficiency.
In summary, this study has revealed a high prevalence of CBG null and Lyon mutations (3.6%) in a geographically isolated Southern Italian population. The clinical phenotype was characterized mostly by idiopathic chronic pain, rather than the fatigue phenotype reported in isolated kindreds; it may represent an environmental influence on the manifestations of the fatigue-pain spectrum.
This study suggests that individuals with either of these two severe CBG mutations have an increased propensity to chronic pain. Because this was a blinded population study, we believe we have avoided potential ascertainment bias. Other studies of these mutations have suggested a fatigue phenotype, although pain has also been reported. It is clear that the pain/fatigue phenotype is not universal; an environmental/cultural effect may underlie the prominence of fatigue in Australia and pain in Italy. The variability of the phenotype may also relate to life experience and its effect on the stress system with its influence on behavior, arousal, and antinociceptive systems. Animal studies support a role for CBG in the neurobehavioral response to severe stress, and population genetic studies suggest that common CBG genetic variations assessed by SNP typing may influence the development of chronic pain. Life stress history may influence the manifestations of these mutations in terms of fatigue vs. pain and penetrance of the features that have been observed. More study is required to determine the influence of CBG variants on the stress system and neurobehavioral responses to life stress.
Acknowledgments
We thank all study participants. Dr. Francesco Talarico, General Manager, AS8 Vibo Valentia, and Dr. Peppino Biamonte, General Manager As 6, Lamezia Terme, Catanzaro, were instrumental in making this study possible. We thank Dr. Agostino Scardamaglio AS8 Vibo Valentia, for clinical chemistry analyses. Ms. Nicolina Boso, Mr. Massimo Ferrise, Dr. Rosalba Ienco, and Ms. Suzy Franzé, AS8 Vibo Valentia, participated in data collection. Pasqualino Natrella from the Association for Neurogenetic Research implemented data entry.
This work was supported by a Project Grant (G.R. no. 772, August 6, 2002) of the Assessorato alla Salute-Regione Calabria; the Italian Minister of Health (convention no. 42 of November 26, 2003, approved on October 11, 2004), and by the National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases Intramural Program.
Disclosure Summary: None of the authors has anything to declare.
Footnotes
- BMI
- Body mass index
- CBG
- corticosteroid-binding globulin
- HPA
- hypothalamic-pituitary-adrenal
- SNP
- single-nucleotide polymorphism.
References
- 1. Hammond GL. 1990. Molecular properties of corticosteroid binding globulin and the sex-steroid binding proteins. Endocr Rev 11:65–79 [DOI] [PubMed] [Google Scholar]
- 2. Lewis JG, Bagley CJ, Elder PA, Bachmann AW, Torpy DJ. 2005. Plasma free cortisol fraction reflects levels of functioning corticosteroid-binding globulin. Clin Chim Acta 359:189–194 [DOI] [PubMed] [Google Scholar]
- 3. Underhill DA, Hammond GL. 1989. Organization of the human corticosteroid binding globulin gene and analysis of its 5′-flanking region. Mol Endocrinol 3:1448–1454 [DOI] [PubMed] [Google Scholar]
- 4. Seralini GE, Bérubé D, Gagné R, Hammond GL. 1990. The human corticosteroid binding globulin gene is located on chromosome 14q31–q32.1 near two other serine protease inhibitor genes. Hum Genet 86:73–75 [DOI] [PubMed] [Google Scholar]
- 5. Pemberton PA, Stein PE, Pepys MB, Potter JM, Carrell RW. 1988. Hormone binding globulins undergo serpin conformational change in inflammation. Nature 336:257–258 [DOI] [PubMed] [Google Scholar]
- 6. Ho JT, Al-Musalhi H, Chapman MJ, Quach T, Thomas PD, Bagley CJ, Lewis JG, Torpy DJ. 2006. Septic shock and sepsis: a comparison of total and free plasma cortisol levels. J Clin Endocrinol Metab 91:105–114 [DOI] [PubMed] [Google Scholar]
- 7. Emptoz-Bonneton A, Crave JC, LeJeune H, Brebant C, Pugeat M. 1997. Corticosteroid-binding globulin synthesis regulation by cytokines and glucocorticoids in human hepatoblastoma-derived (HepG2) cells. J Clin Endocrinol Metab 11:3758–3762 [DOI] [PubMed] [Google Scholar]
- 8. Fernandez-Real JM, Pugeat M, Grasa M, Broch M, Vendrell J, Brun J, Ricart W. 2002. Serum corticosteroid-binding globulin concentration and insulin resistance syndrome: a population study. J Clin Endocrinol Metab 87:4686–4690 [DOI] [PubMed] [Google Scholar]
- 9. Manco M, Fernández-Real JM, Valera-Mora ME, Déchaud H, Nanni G, Tondolo V, Calvani M, Castagneto M, Pugeat M, Mingrone G. 2007. Massive weight loss decreases corticosteroid-binding globulin levels and increases free cortisol in healthy obese patients: an adaptive phenomenon? Diabetes Care 30:1494–1500 [DOI] [PubMed] [Google Scholar]
- 10. Ho JT, Keogh JB, Bornstein SR, Ehrhart-Bornstein M, Lewis JG, Clifton PM, Torpy DJ. 2007. Moderate weight loss reduces renin and aldosterone but does not influence basal or stimulated pituitary-adrenal axis function. Horm Metab Res 39:694–699 [DOI] [PubMed] [Google Scholar]
- 11. Van Baelen H, Power SG, Hammond GL. 1993. Decreased cortisol-binding affinity at transcortin Leuven is associated with an amino acid substitution at residue-93. Steroids 58:275–277 [DOI] [PubMed] [Google Scholar]
- 12. Torpy DJ, Bachmann AW, Grice JE, Fitzgerald SP, Phillips PJ, Whitworth JA, Jackson RV. 2001. Familial corticosteroid-binding globulin deficiency due to a novel null mutation: association with fatigue and relative hypotension. J Clin Endocrinol Metab 86:3692–3700 [DOI] [PubMed] [Google Scholar]
- 13. Emptoz-Bonneton A, Cousin P, Seguchi K, Avvakumov GV, Bully C, Hammond GL, Pugeat M. 2000. Novel human corticosteroid-binding globulin variant with low cortisol-binding affinity. J Clin Endocrinol Metab 85:361–367 [DOI] [PubMed] [Google Scholar]
- 14. Brunner E, Baima J, Vieira TC, Vieira JG, Abucham J. 2003. Hereditary corticosteroid-binding globulin deficiency due to a missense mutation (Asp367Asn, CBG Lyon) in a Brazilian kindred. Clin Endocrinol (Oxf) 58:756–762 [DOI] [PubMed] [Google Scholar]
- 15. Perogamvros I, Underhill C, Henley DE, Hadfield KD, Newman WG, Ray DW, Lightman SL, Hammond GL, Trainer PJ. 2010. Novel corticosteroid-binding globulin variant that lacks steroid binding activity. J Clin Endocrinol Metab 95:E142–E150 [DOI] [PubMed] [Google Scholar]
- 16. Gagliardi L, Ho JT, Torpy DJ. 2010. Corticosteroid-binding globulin: the clinical significance of altered levels and heritable mutations. Mol Cell Endocrinol 316:24–34 [DOI] [PubMed] [Google Scholar]
- 17. Presciuttini S, Paoli G, Franceschi M. 1990. Surname versus gene structure of a small isolate. Ann Hum Genet 54:79–90 [DOI] [PubMed] [Google Scholar]
- 18. De Girolamo G, Rucci P, Scocco P, Becchi A, Coppa F, D'Addario A, Darú E, De Leo D, Galassi L, Mangelli L, Marson C, Neri G, Soldani L. 2000. [Quality of life assessment: validation of the Italian version of the WHOQOL-Brief]. Epidemiol Psichiatr Soc 9:45–55 (Italian) [DOI] [PubMed] [Google Scholar]
- 19. Krupp LB, LaRocca NG, Muir-Nash J, Steinberg AD. 1989. The fatigue severity scale. Application to patients with multiple sclerosis and systemic lupus erythematosus. Arch Neurol 46:1121–1123 [DOI] [PubMed] [Google Scholar]
- 20. Lewis JG, Lewis MG, Elder PA. 2003. An enzyme-linked immunosorbent assay for corticosteroid-binding globulin using monoclonal and polyclonal antibodies: decline in CBG following synthetic ACTH. Clin Chim Acta 328:121–128 [DOI] [PubMed] [Google Scholar]
- 21. Smith CL, Power SG, Hammond GL. 1992. A Leu-His substitution at residue 93 in human corticosteroid binding globulin results in reduced affinity for cortisol. J Steroid Biochem Mol Biol 42:671–676 [DOI] [PubMed] [Google Scholar]
- 22. Torpy DJ, Bachmann AW, Gartside M, Grice JE, Harris JM, Clifton P, Easteal S, Jackson RV, Whitworth JA. 2004. Association between chronic fatigue syndrome and the corticosteroid binding globulin gene Ala-Ser 224 polymorphism. Endocr Res 30:417–429 [DOI] [PubMed] [Google Scholar]
- 23. Bruni AC, Bernardi L, Colao R, Rubino E, Smirne N, Frangipane F, Terni B, Curcio SA, Mirabelli M, Clodomiro A, Di Lorenzo R, Maletta R, Anfossi M, Gallo M, Geracitano S, Tomaino C, Muraca MG, Leotta A, Lio SG, Pinessi L, Rainero I, Sorbi S, Nee L, Milan G, Pappatà S, Postiglione A, Abbamondi N, Forloni G, St George Hyslop P, Rogaeva E, Bugiani O, Giaccone G, Foncin JF, Spillantini MG, Puccio G. 2010. Worldwide distribution of PSEN1 Met146Leu mutation: a large variability for a founder mutation. Neurology 74:798–806 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Sauer K, Kemper C, Glaeske G. 2011. Fibromyalgia syndrome: Prevalence, pharmacological and non-pharmacological interventions in outpatient health care. An analysis of statutory health insurance data. Joint Bone Spine 78:80–84 [DOI] [PubMed] [Google Scholar]
- 25. Fries E, Hesse J, Hellhammer J, Hellhammer DH. 2005. A new view on hypocortisolism. Psychoneuroendocrinology 30:1010–1016 [DOI] [PubMed] [Google Scholar]
- 26. Kaganov Y, Gattas N, Rimon D. 2000. Fibromyalgia-Like Syndrome Secondary to Addison's Disease. J Clin Rheumatol 6:27–29 [DOI] [PubMed] [Google Scholar]
- 27. Kuehl LK, Michaux GP, Richter S, Schächinger H, Anton F. 2010. Increased basal mechanical pain sensitivity but decreased perceptual wind-up in a human model of relative hypocortisolism. Pain 149:539–546 [DOI] [PubMed] [Google Scholar]
- 28. Holliday KL, Nicholl BI, Macfarlane GJ, Thomson W, Davies KA, McBeth J. 2010. Genetic variation in the hypothalamic-pituitary-adrenal stress axis influences susceptibility to musculoskeletal pain: results from the EPIFUND study. Ann Rheum Dis 69:556–560 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Holliday KL, Macfarlane GJ, Nicholl BI, Creed F, Thomson W, McBeth J. 2010. Genetic variation in neuroendocrine genes associates with somatic symptoms in the general population: results from the EPIFUND study. J Psychosom Res 68:469–474 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Qi X, Loiseau F, Chan WL, Yan Y, Wei Z, Milroy LG, Myers RM, Ley SV, Read RJ, Carrell RW, Zhou A. 2011. Allosteric modulation of hormone release from thyroxine and corticosteroid-binding globulins. J Biol Chem 286:16163–16173 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Buss C, Schuelter U, Hesse J, Moser D, Phillips DI, Hellhammer D, Meyer J. 2007. Haploinsufficiency of the SERPINA6 gene is associated with severe muscle fatigue: a de novo mutation in corticosteroid-binding globulin. J Neural Transm 114:563–569 [DOI] [PubMed] [Google Scholar]
- 32. Joyner JM, Hutley LJ, Bachmann AW, Torpy DJ, Prins JB. 2003. Greater replication and differentiation of preadipocytes in inherited corticosteroid-binding globulin deficiency. Am J Physiol Endocrinol Metab 284:E1049–E1054 [DOI] [PubMed] [Google Scholar]
- 33. Petersen HH, Andreassen TK, Breiderhoff T, Bräsen JH, Schulz H, Gross V, Gröne HJ, Nykjaer A, Willnow TE. 2006. Hyporesponsiveness to glucocorticoids in mice genetically deficient for the corticosteroid binding globulin. Mol Cell Biol 26:7236–7245 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Richard EM, Helbling JC, Tridon C, Desmedt A, Minni AM, Cador M, Pourtau L, Konsman JP, Mormède P, Moisan MP. 2010. Plasma transcortin influences endocrine and behavioral stress responses in mice. Endocrinology 151:649–659 [DOI] [PubMed] [Google Scholar]
- 35. Langley PC. 2011. The prevalence, correlates and treatment of pain in the European Union. Curr Med Res Opin 27:463–480 [DOI] [PubMed] [Google Scholar]
