Serum Haptoglobin: a Novel Marker of Adiposity in Humans

Serum Haptoglobin: A Novel Marker of Adiposity in Humans

C. CHIELLINI, F. SANTINI, A. MARSILI, P. BERTI, A. BERTACCA, C. PELOSINI, G. SCARTABELLI, E. PARDINI, J. LO´ PEZ-SORIANO, R. CENTONI, A. M. CICCARONE, L. BENZI, P. VITTI, S. DEL PRATO, A. PINCHERA, AND M. MAFFEI

Dulbecco Telethon Institute (C.C., P.B., J.L-S., M.M.) and Department of Endocrinology and Metabolism, Sections of Downloaded from https://academic.oup.com/jcem/article/89/6/2678/2870304 by guest on 28 September 2021 Endocrinology (F.S., A.M., C.P., G.S., R.C., P.V., A.P., C.C., P.B., J.L.-S., M.M.) and Metabolism (A.B., A.M.C., L.B., S.D.P.), University of Pisa, 56126 Pisa, Italy; and Azienda Ospedaliera Pisana (E.P.), 56126 Pisa, Italy

Haptoglobin (Hp) is a glycoprotein involved in the acute phase a modest percentage of the total variability. Serum Hp was response to inflammation. Our previous findings indicate that positively associated with body fat, as assessed by dual-energy Hp mRNA and protein are present in the adipose tissue of x-ray absorptiometry, both in female and in male groups. The rodents and that Hp gene expression is up-regulated in obese level of significance improved when serum Hp was analyzed models. The aim of the present study was to establish whether against fat mass adjusted for lean mass. Finally, Northern and Hp could be considered a marker of obesity in humans. In 312 Western blot analyses performed in biopsies of sc abdominal subjects, serum Hp was correlated directly with body mass fat from 20 obese individuals showed the presence of Hp index (BMI), leptin, C-reactive protein (CRP), and age. In a mRNA and protein in the human adipose tissue. multivariate stepwise regression analysis, BMI and CRP were In conclusion, serum Hp constitutes a novel marker of ad- independent determinants of serum Hp in females, with BMI iposity in humans, and the adipose tissue likely contributes to having the strongest effect. CRP and age were independent determine its levels. (J Clin Endocrinol Metab 89: 2678–2683, determinants of serum Hp in males, although explaining only 2004)

N RECENT YEARS, an intriguing theory has looked at of a number of proinflammatory molecules on Hp expression I obesity as a low-grade systemic inflammation condition was assessed in 3T3-L1 adipocytes, and TNF-␣ turned out as (1). This view is supported by several lines of evidence, the strongest inducer. Within the clinical setting, plasma Hp including the association of obesity with elevated serum is a marker of inflammation, its levels increasing significantly levels of C-reactive protein (CRP) (2), IL-6 (3), TNF-␣ (4), and during inflammation, infection, and malignancy (9, 12). leptin (5). Similarly, the acute phase reactant Pentraxin 3, a The aim of the present study was to establish whether Hp member of the pentraxin family closely related to CRP, has can be considered a marker of obesity in humans. Therefore, been found to be up-regulated in the white adipose tissue Hp serum levels were measured in a cohort of lean and (WAT) of different models of murine obesity (6). The adipose overweight/obese subjects and related with several clinical tissue, until recently considered a passive depot of triglyc- and metabolic parameters. erides, is emerging as an important endocrine organ (7) and the source for most of the above-mentioned factors also Subjects and Methods known as adipokines. Study population Along this line of findings, we recently reported that hap- Three hundred twelve (221 females and 91 males) subjects referring toglobin (Hp), a plasma glycoprotein involved in the hepatic to our Endocrinology and Metabolism Department were recruited con- acute phase response to inflammation (8), is present in mu- secutively over a 6-month period. Subjects with hypothyroidism or rine WAT, as described previously by Friedrichs et al. (9), and hyperthyroidism, overt diabetes, adrenal insufficiency or hypercorti- solism, and infectious or malignant disorders were excluded from the that its gene expression is dramatically increased in genet- 2 ically or experimentally induced obese mice (10). study. Body mass index (BMI) and age range were 17.6–57.9 kg/m and ␣ 18–71 yr, respectively. Subcutaneous adipose tissue biopsies were ob- We also identified TNF- as an important signal for the tained from 20 subjects (10 females and 10 males; BMI range, 20.5–45.2 observed obesity-dependent up-regulation. As a matter of kg/m2) at the time of elective open abdominal surgery. For this recruit- fact, obese mice deficient for TNF-␣ or TNF-␣ receptors do ment, we applied the same exclusion criteria mentioned above. The not exhibit the typical increase of Hp mRNA observed in the study protocol was approved by the local Ethical Committee (protocol no. 12327), and patients gave informed consent for their participation in obese controls (10). Moreover, in a recent study (11), the effect the study.

Abbreviations: BMI, Body mass index; CRP, C-reactive protein; CV, Laboratory analysis coefficient(s) of variation; DEXA, dual-energy x-ray absorptiometry; Blood samples were obtained by venipuncture after an overnight FM, fat mass; Hp, haptoglobin; LBM, lean body mass; WAT, white fasting, and samples were stored at Ϫ20 C until further analysis. Insulin adipose tissue. was assayed by immunoradiometric assay (Medgenics Diagnostics, JCEM is published monthly by The Endocrine Society (http://www. Fleurus, Belgium), with intraassay and interassay coefficients of varia- endo-society.org), the foremost professional society serving the en- tion (CV) of 4.5 and 10.2%, respectively. Plasma leptin concentration was docrine community. measured by a RIA kit (catalog no. HL-81K; Linco Research, St. Charles,

2678 Chiellini et al. • Haptoglobin in Human Obesity J Clin Endocrinol Metab, June 2004, 89(6):2678–2683 2679

MO) using a polyclonal antibody raised in rabbits against recombinant Spearman linear regression, stepwise regression analysis), and a signif- leptin; the interassay CV was 5–7%. Hp serum concentration was mea- icance limit of P Ͻ 0.05 was set. sured by nephelometry (Dade-Behring, Marburg, Germany), with intraassay and interassay CV of 2.6 and 6.2%, respectively. CRP serum Results concentration was measured by nephelometry (Dade-Behring), and the intraassay and interassay CV were 2.4 and 4.4%, respectively. Plasma Serum Hp was positively correlated with BMI (r ϭ 0.38; glucose was measured on fluorinated plasma samples (Gluco-quant, P Ͻ 0.001) and leptin (r ϭ 0.38; P Ͻ 0.001) in the whole Roche/Hitachi modular analyzer; Roche Diagnostics, Mannheim, population (Fig. 1). Table 1 summarizes the clinical and lab- Germany). oratory characteristics of the subjects. Males and females differed significantly for serum leptin (P Ͻ 0.001) and blood Anthropometric measurements glucose levels (P Ͻ 0.01). Given these differences, the two In a subset of the total population (88 subjects: 62 females and 26 sexes were analyzed separately, and Table 2 shows the sim- males), selected to be equally distributed in three categories of BMI (lean, Ͻ Ͻ Յ Ͻ Ն ple relationships among the various parameters in the whole 17.6 BMI 25; overweight, 25 BMI 30; obese, BMI 30), the Downloaded from https://academic.oup.com/jcem/article/89/6/2678/2870304 by guest on 28 September 2021 percentage of body fat was assessed by dual-energy x-ray absorptiom- population and in the two sexes. In females, significant cor- etry (DEXA) (see below). relations were found between serum Hp and BMI (r ϭ 0.46; DEXA (QDR 4500; Hologic Inc., Bedford, MA) measurements were P Ͻ 0.001), leptin (r ϭ 0.46; P Ͻ 0.001), CRP (r ϭ 0.48; P Ͻ performed as whole body scans, with separate assessment of the three 0.001), and age (r ϭ 0.18; P Ͻ 0.01). In males, a significant compartments: total fat mass (FM), total lean body mass (LBM), and total ϭ P Ͻ bone mineral content. To better define the incidence of adiposity for each correlation was found between Hp and age (r 0.33; subject under study, FM (in kilograms) was divided by lean mass (in 0.01) (Fig. 2). kilograms). The calculated parameter was then used for statistical anal- As expected in a representative Caucasian population, ysis. All DEXA scans were performed by the same skilled laboratory significant correlations were found between age and BMI technician. and between leptin and BMI (Table 2). In a multivariate stepwise regression analysis, BMI and Adipose tissue studies CRP were independent determinants of serum Hp in fe- 2 ϭ RNA extraction and Northern blot analysis. Subcutaneous adipose tissue males, with BMI having the strongest effect (R 0.29 and biopsies were obtained from 20 subjects at the time of elective abdominal 0.13 for BMI and CRP, respectively; total R2 ϭ 0.42). CRP and surgery. Tissues were frozen in liquid nitrogen and stored at Ϫ80 C until age were independent determinants of serum Hp in males, use. Total RNA from human adipose tissue was prepared using TriPure but the low values of the correlation coefficients (R2 ϭ 0.12 TM Isolation Reagent kit (Roche Molecular Biochemicals, Indianapolis, 2 ϭ IN) following the manufacturer’s instructions. Six micrograms of total and 0.07 for CRP and age, respectively; total R 0.19) RNA were separated by denaturing formaldehyde electrophoresis and suggest that these factors account only in part for Hp vari- transferred onto nylon membrane (13). Hybridization was performed as described previously (14). For human Hp-PCR-generated probe, we used the primers HpHF 5Ј-GGTTTCCCACCATAATCTCAC-3Ј and HpHR 5Ј-CTGGATGGAAGTCACCTTCA-3Ј, which generated a fragment of 632 bp. The filter was subsequently rehybridized to a 18S PCR-generated probe, using the primers 18SF 5Ј-TGACTCAACACGGGAAACCT- CAC-3Ј and 18SR 5Ј-CGGACATCTAAGGGCATCACAG-3Ј, which generated a fragment of 266 bp. Autoradiographs were analyzed by a densitometer (GS690; Bio-Rad, Hercules, CA) using MultyAnalyst/PC-PC software for Image Analysis Systems version 1.02 (Bio-Rad).

Protein preparation. Frozen adipose tissue was homogenized in a lysis buffer containing 20 mm Tris (pH 7.5), 150 mm NaCl, 10% glycerol, 1% Triton X-100, 10 mm EDTA, and 1 mm phenylmethylsulfonyl fluoride. The homogenate was then centrifuged at 3000 rpm for 30 min at 4 C. The top and bottom layers were discarded, and the intermediate phase was collected and assayed for the protein concentration by using the Lowry method (Dc protein assay, 500-0114 and 500-0113; Bio-Rad). Immunoblotting. Proteins were separated on 15% SDS-PAGE and transferred onto a nitrocellulose filter by electroblotting. Filters were satu- rated with a solution of 4% dry milk in Tris-buffered saline (TBS) and incubated with a polyclonal primary antibody antihuman Hp raised in goat (Sigma 5035; Sigma, St. Louis, MO), at a 1:3000 dilution in TBS ϩ 0.1% Tween with 2% dry milk. After three washes with TBS, the filter was incubated1hintheantigoat secondary antibody (horseradish peroxidase conjugated), diluted 1:2000 in TBS ϩ 0.1% Tween ϩ 2% dry milk. Specific protein expression was visualized using a chemilumines- cent assay (Amersham Pharmacia/Biotech, Piscataway, NJ), after ex- posure to x-ray film for 1–5 min.

Statistical analysis Data are expressed as means Ϯ sd. The StatView SEϩ Graphics package (version 3.1, Abacus Concepts Inc., Berkeley, CA) was used to FIG. 1. Regression of serum Hp on BMI (A) and serum leptin (B) in perform univariate and multivariate statistical analysis (Student’s t test, the total population. 2680 J Clin Endocrinol Metab, June 2004, 89(6):2678–2683 Chiellini et al. • Haptoglobin in Human Obesity

TABLE 1. Clinical and laboratory parameters of subjects The next important issue to study was whether the human included in the study (means Ϯ SD) adipose tissue might be a source for the production of Hp. Females Males P The presence of both Hp mRNA and protein was demonstrated by Northern and Western blot analysis in bioptic No. of cases 221 91 Age (yr) 44.2 Ϯ 11.7 47.1 Ϯ 13.3 NS samples of human abdominal adipose tissue from 20 ran- BMI (kg/m2) 27.7 Ϯ 6.4 28.3 Ϯ 4.2 NS domly selected subjects. Figure 4 displays the results of two Serum Hp (mg/ml) 1.39 Ϯ 0.74 1.35 Ϯ 0.68 NS representative cases. Differences in the transcript and in the Serum leptin (ng/ml) 14.8 Ϯ 10.9 6.4 Ϯ 6.2 Ͻ0.001 protein size depend on the presence of allele Hp-1 (␣-1 chain) Serum insulin (pmol/liter) 60.6 Ϯ 53.4 59.4 Ϯ 43.8 NS ␣ Serum glucose (mmol/liter) 4.76 Ϯ 0.79 5.02 Ϯ 0.8 Ͻ0.05 or allele Hp-2 ( -2 chain). The alleles differ in size for the Serum CRP (mg/liter) 4.8 Ϯ 5.1 4.3 Ϯ 3.9 NS presence of one (Hp 1) or two (Hp 2) copies of a 1700-bp-long sequence (15). These data indicate that the Hp gene is ex- NS, Not significant. pressed and that Hp protein is present in the human adipose Downloaded from https://academic.oup.com/jcem/article/89/6/2678/2870304 by guest on 28 September 2021 TABLE 2. Results of simple regression analysis for each possible tissue. pair of the various metabolic and anthropometric variables in the total population (T) and in the female (F) and male (M) groups Discussion BMI Leptin Insulin Glucose CRP Age In the present study, we report for the first time a strong Hp c positive correlation between circulating Hp and BMI. Our T 0.38 0.38c 0.08 NS 0.13a 0.40c 0.22c F 0.46c 0.46c 0.09 NS 0.08 NS 0.48c 0.18b findings are in line with those of Hannerz et al. (16), who M 0.07 NS 0.17 NS 0.08 NS 0.14 NS 0.26 NS 0.33b reported that serum Hp was moderately increased in a group BMI of 20 obese female subjects compared with controls, and c c b c c T 0.61 0.27 0.18 0.29 0.29 point to Hp as a possible novel marker of adiposity. Yet, we F 0.70c 0.29c 0.20b 0.27b 0.34c M 0.36b 0.22 NS 0.06 NS 0.43b 0.15 NS could not demonstrate a correlation between Hp and BMI in Leptin males, analyzed separately. Because this cannot be ascribed T 0.19b 0.18b 0.17a 0.08 NS to the lower number of males compared with females, an F 0.24b 0.18a 0.20a 0.13 NS explanation may likely reside in the use of BMI as the in- c M 0.05 NS 0.54 0.05 NS 0.22 NS dependent variable. A more physiological explanation for Insulin T 0.07 NS 0.03 NS 0.06 NS the different correlations in the two genders may reside in the F 0.12 NS 0.001 NS 0.06 NS limited ability of BMI to account for true adiposity. Although M 0.05 NS 0.001 NS 0.08 NS BMI is widely used to categorize body weight, it cannot Glucose dissect out LBM from adipose tissue, and skeletal muscle is T 0.06 NS 0.41c F 0.09 NS 0.39c more represented, for any given BMI, in males than in fe- M 0.06 NS 0.45c males. Consistent with this hypothesis, a positive association CRP between Hp and total body FM measured by DEXA was T 0.03 NS demonstrated in both genders. This association was strength- F 0.06 NS ened on adjustment of body FM by lean mass. The higher M 0.05 NS degree of association may be due to the fact that adjusting for Numbers show correlation coefficient (r). Level of significance: LBM (muscle, vessels, organs) allows a better definition of a P Ͻ 0.05, b P Ͻ 0.01, c P Ͻ 0.001. NS, Not significant. the distribution volume of Hp. Alternatively, it may be spec- ability in males. The two sexes showed an apparent discrep- ulated that for a given FM, Hp levels vary as a function of ancy in the relationship between BMI and serum Hp. This lean mass. Whether this implies an effect of Hp on the mod- could depend on the different number of subjects in the two ulation of LBM will require more investigation. Taken to- samples (221 females vs. 91 males). However, highly signif- gether, these results support the reliability of serum Hp as a icant correlations were observed (data not shown) also when marker of adiposity rather than body size. the regression analysis between Hp and BMI was performed A positive relationship was also found between Hp and in multiple random subsets (n Ͻ 100) of the female group, age. Although worsening of the health conditions in aging thereby excluding the size of the sample as a critical param- may represent a trigger for inflammatory response (17), other eter in this gender discrepancy. Therefore, we analyzed the mechanisms could be considered. For instance, aging has serum level of Hp against a direct and reliable parameter of been associated with a progressive decline in insulin sensi- adiposity, namely the actual FM. tivity and loss of glucose tolerance (18). A strong association To gain more direct insights in the relationship between has been documented between insulin sensitivity and in- adiposity and serum Hp levels, DEXA scans were obtained flammatory response (19) so that a direct relationship be- in 62 females and 26 males. In this case, serum Hp was tween age and serum Hp levels can be explained by changes positively associated with total body fat both in the female in insulin sensitivity. (r ϭ 0.33; P Ͻ 0.01) and in the male (r ϭ 0.42; P Ͻ 0.05) groups Data reported here allow us to add Hp to the growing list (Fig. 3, A and C). Moreover, the degree of correlation became of circulating protein raised in human obesity and somehow stronger when serum Hp was plotted as a function of body involved in inflammation and/or in immune response, such FM adjusted for lean mass (females: r ϭ 0.36, P Ͻ 0.005; as TNF-␣, IL-6, CRP, and leptin. The latter has proved to be males: r ϭ 0.64, P Ͻ 0.001) (Fig. 3, B and D). These data a key signal in the relationship between obesity and associ- suggest that serum Hp is a marker of adiposity in humans. ated complications like infertility (20), diabetes (21), or au- Chiellini et al. • Haptoglobin in Human Obesity J Clin Endocrinol Metab, June 2004, 89(6):2678–2683 2681

FIG. 2. Regression of serum Hp on BMI and age in the female (A and B, Downloaded from https://academic.oup.com/jcem/article/89/6/2678/2870304 by guest on 28 September 2021 respectively) and in the male (C and D, respectively) groups. NS, Not significant.

FIG. 3. Regression of serum Hp on body fat and body fat/lean mass (as assessed by DEXA) in the female (A and B) and in the male (C and D) groups.

toimmune diseases (22). A recent report by Sanna et al. (23) model of human multiple sclerosis. According to their stud- points to leptin as an important signal for linking the met- ies, leptin and other cytokines, which are overproduced in abolic status with the susceptibility for developing experi- obesity, affect the disease-inducing potential of the T cells. A mental autoimmune encephalomyelitis, which serves as a variety of immunomodulatory effects have been attributed to 2682 J Clin Endocrinol Metab, June 2004, 89(6):2678–2683 Chiellini et al. • Haptoglobin in Human Obesity

Acknowledgments We thank the doctors and nurses who took care of the patients participating in this study and Sandro Fornaciai for precious technical assistance during the DEXA measurements.

Received November 12, 2003. Accepted February 23, 2004. Address all correspondence and requests for reprints to: Dr. Mar- gherita Maffei, Dulbecco Telethon Institute, Department of Endocrinol- ogy and Metabolism, Ospedale di Cisanello, Via Paradisa, 2, 56126 Pisa, Italy. E-mail: [email protected]. This work was supported by Compagnia San Paolo and by Ministero dell’Universita` e della Ricerca (MIUR 2002). C.C., P.B., and J.L.-S. were supported by a Telethon Fellowship. M.M. is an Assistant Telethon

Scientist. Downloaded from https://academic.oup.com/jcem/article/89/6/2678/2870304 by guest on 28 September 2021

References

FIG. 4. Hp is present in the human WAT. A, Representative Northern 1. Das UN 2001 Is obesity an inflammatory condition? Nutrition 17:953–966 blot analysis showing the presence of Hp mRNA in the human WAT 2. Visser M, Bouter LM, McQuillan GM, Wener MH, Harris TB 1999 Elevated of two subjects. Total RNA was isolated from bioptic samples of ab- C-reactive protein levels in overweight and obese adults. JAMA 282:2131–2135 dominal adipose tissue, and 8 ␮g was loaded in each lane. Equal 3. Bastard JP, Jardel C, Delattre J, Hainque B, Bruckert E, Oberlin F 1999 32 Evidence for a link between adipose tissue interleukin-6 content and serum loading is shown by hybridization of the same filter to the P-labeled C-reactive protein concentrations in obese subjects. Circulation 99:2221–2222 18S probe. Differences in transcript size depend on Hp genotype (see 4. Zahorska-Markiewicz B, Janowska J, Olszanecka-Glinianowicz M, Zura- Results for details). B, Representative Western blot showing Hp pro- kowski A 2000 Serum concentrations of TNF-␣ and soluble TNF-␣ receptors tein expression in human WAT. Total protein extracts (20 ␮g) were in obesity. Int J Obes Relat Metab Disord 24:1392–1395 separated by SDS-PAGE and transferred onto nitrocellulose. The blot 5. Maffei M, Halaas J, Ravussin E, Pratley RE, Kim S, Lallone R, Ranganathan was probed with antibodies against human Hp and horseradish per- S, Kern P, Friedman JM 1995 Leptin levels in human and rodent: measurement oxidase-conjugated secondary antibody. Results identical to those of plasma leptin and ob RNA in obese and weight reduced subjects. Nat Med shown in A and B were obtained for all of the 20 samples analyzed. 1:1155–1161 6. Abderrahim-Ferkoune A, Bezy O, Chiellini C, Maffei M, Grimaldi P, Bonino F, Moustaid-Moussa N, Pasqualini F, Mantovani A, Ailhaud G, Amri EZ 2003 Characterization of the long pentraxin PTX3 as a TNF␣-induced secreted Hp as well, which are supposed to be partly mediated protein of adipose cells. J Lipid Res 44:994–1000 7. Guerre-Millo M 2002 Adipose tissue hormones. J Endocrinol Invest 25:855– through binding of Hp to the CD11b/CD18 receptor (24, 25). 861 Furthermore, obesity is associated with insulin resistance 8. Bensi G, Raugei G, Klefenz H, Cortese R 1985 Structure and expression of the and increased rate of cardiovascular events (26). Of note is human haptoglobin locus. EMBO J 4:119–126 9. Friedrichs WE, Navarijo-Ashbaugh AL, Bowman BH, Yang F 1995 Expression the knowledge that Hp has been related to the development and inflammatory regulation of haptoglobin gene in adipocytes. Biochem of arterial hypertension (27) and to the incidence of myo- Biophys Res Commun 209:250–256 cardial infarction and stroke (28). Thus, we can speculate that 10. Chiellini C, Bertacca A, Novelli SE, Gorgun CZ, Ciccarone A, Giordano A, Xu H, Soukas A, Costa M, Gandini D, Dimitri R, Bottone P, Cecchetti P, Hp, which is also overproduced in obesity, might constitute Pardini E, Perego L, Navalesi R, Folli F, Benzi L, Cinti S, Friedman JM, a novel link between obesity and some of its comorbidities. Hotamisligil GS, Maffei M 2002 Obesity modulates the expression of haptoglobin in the white adipose tissue via TNF␣. J Cell Physiol 190:251–258 It has already been demonstrated that most of the inflam- 11. Oller do Nascimento C, Hunter L, Trayhurn P 2004 Regulation of haptoglobin matory obesity-related proteins are directly produced by the gene expression in 3T3–L1 adipocytes by cytokines, catecholamines, and adipose tissue (29). The question raises whether the increased PPAR␥. Biochem Biophys Res Commun 313:702–708 12. Oliviero S, Morrone G, Cortese R 1987 The human haptoglobin gene: tran- serum levels of Hp in obesity depend on an altered metabolic scriptional regulation during development and acute phase induction. EMBO status affecting Hp release by the liver or on a direct pro- J 6:1905–1912 duction from the adipose tissue, as already demonstrated in 13. Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K 1999 Current protocols in molecular biology. New York: Wiley and rodents (10). In this regard, the presence of Hp mRNA and Sons protein that we observed in the human WAT strongly sug- 14. Chiellini C, Costa M, Novelli SE, Amri EZ, Benzi L, Bertacca A, Cohen P, Del Prato S, Friedman JM, Maffei M 2003 Identification of cathepsin K as a novel gests that this organ contributes to the increase in circulating marker of adiposity in white adipose tissue. J Cell Physiol 195:309–321 Hp levels found in obese individuals. Consistent with our 15. Koch W, Latz W, Eichinger M, Roguin A, Levy AP, Schomig A, Kastrati A findings, Fain et al. (30) recently reported that Hp is released 2002 Genotyping of the common haptoglobin Hp 1/2 polymorphism based on PCR. Clin Chem 48:1377–1382 by explants of human adipose tissue, and this, together with 16. Hannerz J, Greitz D, Ericson K 1995 Is there a relationship between obesity the data reported here, suggests that Hp is a novel adipokine. and intracranial hypertension? Int J Obes Relat Metab Disord 19:240–244 The simple increase in the adipose mass of obese subjects, 17. Roubenoff R 2003 Catabolism of aging: is it an inflammatory process? Curr Opin Clin Nutr Metab Care 6:295–299 due to an hypertrophy and an hyperplasia of the tissue (31), 18. Barzilai N, Gupta G 1999 Interaction between aging and syndrome X: new may account for the higher level of serum Hp, although we insights on the pathophysiology of fat distribution. Ann NY Acad Sci 892:58–72 cannot rule out the possibility that liver participates to this 19. Yudkin JS, Stehouwer CD, Emeis JJ, Coppack SW 1999 C-reactive protein in healthy subjects: associations with obesity, insulin resistance, and endothelial overproduction. dysfunction: a potential role for cytokines originating from adipose tissue? In conclusion, serum Hp constitutes a novel marker of Arterioscler Thromb Vasc Biol 19:972–978 20. Chehab FF, Lim ME, Lu R 1996 Correction of the sterility defect in homozy- adiposity in humans, and the adipose tissue likely contrib- gous obese female mice by treatment with the human recombinant leptin. Nat utes to determine its levels. Here, we speculate that Hp could Genet 12:318–320 constitute an important link between obesity and its comor- 21. Taylor SI, Barr V, Reitman M 1996 Does leptin contribute to diabetes caused by obesity? Science 274:1151–1152 bidities by mediating some of the inflammatory effects as- 22. Kuchroo V, Nichoson LB 2003 Fast and feel good? Nature 422:27–28 sociated with the obesity status. 23. Sanna V, Di Giacomo A, LaCava A, Lechler RI, Fontana S, Zappacosta S, Chiellini et al. • Haptoglobin in Human Obesity J Clin Endocrinol Metab, June 2004, 89(6):2678–2683 2683

Matarese G 2003 Leptin surge precedes onset of autoimmune encephalomy- hypertension: from the perspective of a vicious triangle. Int J Obes Relat Metab elitis and correlates with development of pathogenic T cell responses. J Clin Disord 26:S28–S38 Invest 111:241–250 28. Engstrom G, Janzon L, Berglund G, Lind P, Stavenow L, Hedblad B, 24. El Ghmati SM, Van Hoeyveld EM, Van Strijp JG, Ceuppens JL, Stevens EA Lindgarde F 2002 Blood pressure increase and incidence of hypertension in 1996 Identification of haptoglobin as an alternative ligand for CD11b/CD18. relation to inflammation-sensitive plasma proteins. Arterioscler Thromb Vasc J Immunol 156:2542–2552 Biol 22:2054–2058 25. Arredouani M, Matthijs P, Van Hoeyveld E, Kasran A, Baumann H, Ceup- 29. Trayhurn P, Beattie JH 2001 Physiological role of adipose tissue: white adipose pens JL, Stevens E 2003 Haptoglobin directly affects T cells and suppresses T tissue as an endocrine and secretory organ. Proc Nutr Soc 60:329–339 helper cell type 2 cytokine release. Immunology 108:144–145 30. Fain JN, Bahouth SW, Madan AK 2004 Haptoglobin release by human adipose 26. Uchegbu EC, Kopelman PG 2002 Cardiovascular risks in obesity. J Endocrinol tissue in primary culture. J Lipid Res 45:536–542 Invest 25:915–918 31. Hausman DB, DiGirolamo M, Bartness TJ, Hausman GJ, Martin RJ 2001 The 27. Montani JP, Antic V, Yang Z, Dulloo A 2002 Pathways from obesity to biology of white adipocyte proliferation. Obes Rev 2:239–254

JCEM is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community. Downloaded from https://academic.oup.com/jcem/article/89/6/2678/2870304 by guest on 28 September 2021