Comparative expression analysis of isolated human adipocytes and the human adipose cell lines LiSa-2 and PAZ6.

OBJECTIVE: To obtain insight in the extent to which the human cell lines LiSa-2 and PAZ6 resemble isolated primary human adipocytes. DESIGN: A combination of cDNA subtraction (representative difference analysis; RDA) and cDNA microarray analysis was used to select adipose specific genes to compare isolated (pre-)adipocytes with (un)differentiated LiSa-2 and PAZ6 cells. MEASUREMENTS: RDA was performed on adipose tissue against lung tissue. A total of 1400 isolated genes were sequenced and cDNA microarray technology was used for further adipose related gene selection. 30 genes that were found to be enriched in adipose tissue were used to compare isolated human adipocytes and LiSa-2 and PAZ6 cells in the differentiated and undifferentiated states. RESULTS: RDA and microarray analysis resulted in the identification of adipose enriched genes, but not in adipose specific genes. Of the 30 most differentially expressed genes, as expected, most were related to lipid metabolism. The second category consisted of methyltransferases, DNMT1, DNMT3a, RNMT and SHMT2, of which the expression was differentiation dependent and higher in differentiated adipocytes. Using the 30 adipose expressed genes, it was found that isolated adipocytes on one hand, and PAZ6 and LiSa-2 adipocytes on the other, differ primarily in lipid metabolism. Furthermore, LiSa-2 cells seem to be more similar to isolated adipocytes than PAZ6 cells. CONCLUSION: The LiSa-2 cell line is a good model for differentiated adipocytes, although one should keep in mind that the lipid metabolism in these cells deviates from the in vivo situation Furthermore, our results imply that methylation may have an important function in terminal adipocyte differentiation.

Low number of omental preadipocytes with high leptin and low adiponectin secretion is associated with high fasting plasma glucose levels in obese subjects.

OBJECTIVE: This study investigates whether fasting plasma glucose (FPG) levels in obese subjects are associated with the number of preadipocytes and their adipokine-secretion capabilities. DESIGN: Abdominal subcutaneous and omental adipose tissues were obtained from 10 female and four male obese subjects (age 37 +/- 8 years; BMI 48 +/- 13 kgm(2)) with a wide range of FPG (range: 4.3-10.6 mm). Stromal vascular cells (SVC) were isolated and cultured and the number of attached SVC (aSVC) per gram adipose tissue determined. The aSVCs were differentiated in vitro to become adipocytes, and the secretion of the adipokine leptin and adiponectin in the culture media was determined. Spearman rank correlation coefficients were calculated between FPG and preadipocyte number and adipokine secretion. PATIENTS: Subject-inclusion criteria: BMI >40 kg/m(2) and for severe comorbid conditions BMI >35 kg/m(2). Subject-exclusion criteria: severe cardiopulmonary pathology (ASA class 3), history of bariatric surgery, manifest psychopathology, 18 years < age > 60 years and for upper-abdominal surgery, age >50 years. All females in the study had regular menstrual periods. None of participants received glucose-lowering medication. RESULTS: No association was observed between BMI and fasting glucose levels. More than 90 +/- 20% of the cultured aSVC fraction was able to store fat droplets, indicating the presence of preadipocytes. A strong negative association was observed between omental preadipocyte number and FPG. A strong association was observed between adipokine secretion by the omental preadipocytes and FPG. No association was observed between subcutaneous preadipocyte number and adipokine secretion and FPG. CONCLUSIONS: In morbid obese subjects, low number of omental preadipocytes with high-leptin- and low-adiponectin-secretion profiles is associated with high FPG.

PPARgamma activity in subcutaneous abdominal fat tissue and fat mass gain during short-term overfeeding.

OBJECTIVE: As the peroxisome proliferator-activated receptor gamma (PPARgamma) plays a central role in fat mass regulation, we investigated whether initial subcutaneous PPARgamma activity is related to fat mass generation during overfeeding. SUBJECTS: Fourteen healthy female subjects (age 25 +/- 4 years, BMI 22.1 +/- 2.3 kg/m2). DESIGN AND MEASUREMENTS: Subjects were overfed with a diet supplying 50% more energy than baseline energy requirements for 14 days. Fasting blood samples were analyzed for leptin, insulin and glucose. Fasting subcutaneous abdominal fat biopsies were obtained for analysis of PPARgamma1, PPARgamma2, aP2 and UCP2 mRNAs. RESULTS: Initial PPARgamma1 and 2, aP2 and UCP2 mRNAs were not related to fat gain (P > 0.12). However, PPARgamma1, PPARgamma2 and aP2 mRNA changes were positively related to changes in plasma leptin (P < 0.05) and, except aP2 (P = 0.06), to fat gain (P < 0.05). PPARgamma and aP2 mRNA changes were positively related (P<0.01), indicating that PPARgamma mRNA levels reflected PPARgamma activity. CONCLUSION: These data suggest that the ability to increase PPARgamma activity might be involved in the susceptibility to gain weight during a positive energy balance.

Apolipoprotein E protects against neuropathology induced by a high-fat diet and maintains the integrity of the blood-brain barrier during aging.

SUMMARY: The present study provides evidence that chronic intake of a high-fat diet induces a dramatic extravasation of immunoglobulins, indicating alterations in blood-brain barrier (BBB) functioning, in the brains of apolipoprotein E (apoE)-knockout mice, but not of C57Bl/6 control mice. Using sodium fluorescein as a marker for the permeability of the BBB, we found additional support for age-related disturbances of BBB function in apoE-knockout mice. Behavioral analysis of apoE-knockout mice compared with C57Bl/6 mice indicated that they were also less efficient in acquiring the spatial Morris water maze task. Furthermore, apoE-knockout mice are known to develop severe atherosclerosis, which is exacerbated with a high-fat diet. We therefore compared the apoE-knockout mice with the apoE3-Leiden transgenic mice, which are known to develop atherosclerosis. However, apoE3-Leiden mice that were kept on a high-fat, high-cholesterol diet and that developed atherosclerosis to an extent similar to the apoE-knockout mice, showed no signs of BBB disturbances. These results indicate for the first time that apoE plays an essential role in the maintenance of the integrity of the BBB during aging and that it protects the brain from neuropathology induced by a high-fat diet. We therefore hypothesize that the role of apoE in the maintenance of the integrity of the BBB may be the mechanism by which apoE affects the progression of neurodegeneration, as seen in Alzheimer's disease.

Fiber type dependent upregulation of human skeletal muscle UCP2 and UCP3 mRNA expression by high-fat diet.

OBJECTIVE: To test the hypothesis that consumption of a high-fat diet leads to an increase in UCP mRNA expression in human skeletal muscle. In a group of endurance athletes, with a range in fiber type distribution, we hypothesized that the effect of the high-fat diet on UCP2 and UCP3 mRNA expression is more pronounced in muscle fibers which are known to have a high capacity to shift from carbohydrate to fat oxidation (type IIA fibers). DESIGN: Ten healthy trained athletes (five males, five females) consumed a low-fat diet (17+/-0.9 en% of fat) and high-fat diet (41.4+/-1.4 en% fat) for 4 weeks, separated by a 4 week wash-out period. Muscle biopsies were collected at the end of both dietary periods. MEASUREMENTS: Using RT-PCR, levels of UCP2 and UCP3 mRNA expression were measured and the percentage of type I, IIA and IIB fibers were determined using the myofibrillar ATPase method in all subjects. RESULTS: UCP3L mRNA expression tended to be higher on the high-fat diet, an effect which reached significance when only males were considered (P=0.037). Furthermore, diet-induced change in mRNA expression of UCP3T (r: 0.66, P=0.037), UCP3L (r: 0.61, P=0.06) and UCP2 (r: 0.70, P=0.025), but not UCP3S, correlated significantly with percentage dietary fat on the high-fat diet. Plasma FFA levels were not different during the two diets. Finally, the percentage of type IIA fibers was positively correlated with the diet-induced change in mRNA expression for UCP2 (r: 0.7, P=0.03), UCP3L (r: 0.73, P=0.016) and UCP3T (r: 0.68, P=0.03) but not with UCP3S (r: 0.06, NS). CONCLUSION: UCP2 and UCP3 mRNAs are upregulated by a high-fat diet. This upregulation is more pronounced in humans with high proportions of type IIA fibers, suggesting a role for UCPs in lipid utilization.

Binding of human urokinase-type plasminogen activator to its receptor: residues involved in species specificity and binding.

Urokinase-type plasminogen activator (UPA), particularly when bound to its receptor (UPAR), is thought to play a major role in local proteolytic processes, thus facilitating cell migration as may occur during angiogenesis, neointima and atherosclerotic plaque formation, and tumor cell invasion. To facilitate understanding of the need and function of the UPA/UPAR interaction in cell migration and vascular remodeling, we changed several amino acid residues in UPA so as to interfere with its interaction with its receptor. The receptor-binding domain of UPA has been localized to a region in the growth factor domain between residues 20 and 32. Since the binding of UPA to UPAR appears to be species specific, we used the differences in amino acid sequences in the growth factor domain of UPA between various species to construct a human UPA variant that does not bind to the human UPAR. We substituted Asn22 for its mouse equivalent Tyr by site-directed mutagenesis. This mutant UPA had similar plasminogen activator characteristics as wild-type UPA, including its specific activity and interaction with plasminogen activator inhibitor-1. However, no UPA/UPAR complexes could be observed in cross-linking experiments using DFP-treated 125I-labeled mutant UPA and lysates of various cells, including U937 histiocytic lymphoma cells, phorbol myristate acetate-treated human ECs, and mouse LB6 cells transfected with human UPAR cDNA. In direct binding experiments, DFP-treated 125I-labeled mutant UPA could not bind to phorbol myristate acetate-treated ECs, whereas wild-type UPA did bind. Furthermore, a 25-fold excess of wild-type UPA completely prevented the binding of DFP-treated 125I-labeled wild-type UPA to the human receptor on transfected LB6 cells, whereas an equal amount of mutant UPA had only a very small effect. In ligand blotting assays, very weak binding of mutant UPA to human UPAR could be observed. Changing Asn22 into the other amino acid residues alanine or glutamine had no effect on binding to UPAR on human ECs. The functional integrity of the growth factor domain in the non-receptor binding Asn22Tyr mutant is suggested by the fact that binding of this mutant to a murine UPAR can be restored after additional mutations in the growth factor domain, Asn27,His29,Trp30 to Arg27,Arg29,Arg30. We conclude that Asn22 and Asn27,His29,Trp30 in human UPA are key determinants in the species-specific binding of the enzyme to its receptor and that changing Asn22 into Tyr results in a UPA mutant with strongly reduced binding to UPAR.

Modified proenzymes as artificial substrates for proteolytic enzymes: colorimetric assay of bacterial collagenase and matrix metalloproteinase activity using modified pro-urokinase.

We describe a new principle for assessment of the activity of proteolytic enzymes of all classes and show the application of this principle for the quantitative assay of bacterial collagenase and human matrix metalloproteinases (MMPs). Central to this new principle is the presence of a proenzyme that can be activated into an active enzyme by a single proteolytic event. The regular activation sequence in the proenzyme is replaced using protein engineering by an artificial sequence recognized by the proteinase to be determined. The latter can act as an activator for the newly engineered proenzyme. In the present paper a simple colorimetric assay for the determination for MMPs is described based on this principle. With the aid of protein engineering, a modified pro-urokinase has been prepared in which the activation sequence normally recognized by plasmin (Pro-Arg-Phe-Lys upward arrowIle-Ile-Gly-Gly) has been replaced by a sequence expected to be recognized and hydrolysed by many MMPs (Arg-Pro-Leu-Gly upward arrowIle-Ile-Gly-Gly). The active urokinase resulting from activation of the modified pro-urokinase by a MMP could be measured either directly, using a specific chromogenic peptide substrate for urokinase, or indirectly via urokinase-catalysed plasminogen activation. The response of the assay to equal molar quantities of active MMPs decreases in the order MMP-2>MMP-9>MMP-1>MMP-3>MMP-7. The detection limit for MMP-9 was below 15 pM, corresponding to 3. 75x10(-15) mol per assay. Using the assay, increased MMP activity was detected in synovial tissue extracts from rheumatoid arthritis patients compared with those from osteoarthritis patients, and in stomach tumour extracts as compared with normal stomach tissue extracts.

Intrusiveness of rheumatoid arthritis on sexuality in male and female patients living with a spouse.

OBJECTIVE: To determine whether physical disability, pain, depressive mood, and criticism by the spouse are differentially related to intrusiveness of rheumatoid arthritis (RA) on sexuality in male and female patients. METHODS: Physical and psychological aspects of health were assessed in 102 male and 118 female RA patients who were living with a spouse. Patients were classified into 3 levels of intrusiveness of RA on sexuality. The data were analyzed by means of analysis of covariance and multiple regression analysis. RESULTS: Greater intrusiveness of RA on sexuality was related to greater physical disability, pain, and depression in male and female RA patients. Female patients, compared with male patients, appeared to have lower levels of mobility and self-care. Male and female patients did not differ in their level of intrusiveness of RA on sexuality. CONCLUSION: Physical disability, pain, and, to a lesser extent, depression were found to contribute to intrusiveness of RA on sexuality. It is suggested that differences in sexual motivation between men and women might have been influential in the absence of gender differences in intrusiveness.

Domain-domain interactions in hybrids of tissue-type plasminogen activator and urokinase-type plasminogen activator.

Fibrin-dependent plasminogen activation by tissue-type plasminogen activator (t-PA) is in part associated with the presence of the kringle 2 domain in t-PA. Within this kringle 2 domain a lysyl-binding site has been described. The plasminogen to plasmin conversion by urokinase-type plasminogen activator (u-PA), in contrast to that of t-PA, is not enhanced in the presence of fibrin. Within the u-PA kringle domain no lysyl-binding site is found. To study whether introduction of a lysyl-binding site in the u-PA kringle domain will make u-PA a fibrin-dependent plasminogen activator, three stretches of amino acid residues of the u-PA kringle domain (A28-Q33, D55-N57 and G67-V72) were substituted by three stretches of amino acids from the corresponding positions of the kringle 2 domain of t-PA (M28-K33, D55-D57 and N67-W72). These changes resulted in the creation of the lysyl-binding site consensus of the kringle 2 domain (K33, D55, D57, W62 and W72) in the u-PA kringle. However, the resulting u-PA mutant did not interact with lysyl-Sepharose, nor did it display fibrin-enhanced plasminogen activation in the presence of soluble fibrin mimic. When the kringle domain of u-PA was replaced by the kringle 2 domain of t-PA, similar results were obtained. The hybrid protein hardly interacted with lysyl-Sepharose and the plasminogen activation was not enhanced in the presence of fibrin mimic. However, the N-terminal fragment isolated from this hybrid molecule (consisting of growth factor domain and kringle 2 domain) did interact with lysyl-Sepharose, suggesting that in the hybrid molecule a functional lysyl-binding site is present but not operational. Indeed, lysine analogue (epsilon-amino-caproic acid) sensitive binding of isolated t-PA kringle 2 domain to u-PA could be observed. The modified u-PA kringle, the wild type u-PA kringle and the kringle 2 of the u-PA hybrid were also placed N-terminal of the protease domain of t-PA. As expected, the t-PA mutant consisting of the kringle 2 domain and the protease domain bound to lysyl-Sepharose and showed fibrin-dependent plasminogen activation. Further, the hybrid molecule consisting of the u-PA kringle placed N-terminal of the t-PA protease domain did not display these features. Introduction of the modified u-PA kringle N-terminal of the t-PA protease domain resulted in a very weak interaction with lysyl-Sepharose. Despite the high overall similarity in primary structure of the modified u-PA kringle and t-PA kringle 2 (68%), no fibrin-dependent plasminogen activation of this hybrid molecule was observed. The above-mentioned results question the concept that the structural auto-nomous domains within hybrid plasminogen activators t-PA and u-PA function as autonomous domains and suggest that interactions between the kringle and the protease domain in hybrid molecules strongly influences their functional features.

The role of the lysyl binding site of tissue-type plasminogen activator in the interaction with a forming fibrin clot.

To describe the role of the lysyl binding site in the interaction of tissue-type plasminogen activator (t-PA, FGK1K2P) with a forming fibrin clot, we performed binding experiments with domain deletion mutants GK1K2P, K2P, and the corresponding point mutants lacking the lysyl binding site in the absence and the presence of epsilon-amino caproic acid (EACA). Occupation of the lysyl binding site in the K2 domain with EACA has a pronounced effect on the binding of FGK1K2P to a fibrin clot (C50 = 77 +/- 11 nM versus 376 +/- 45 nM with EACA). Deleting the lysyl binding site in the K2 domain (substitution D236N) also impairs fibrin binding but to a lesser extent (C50 = 169 +/- 20 nM). Although the binding of K2P to a fibrin clot is weak (C50 = 1163 +/- 490 nM), it still is 2 orders of magnitude stronger than the binding of EACA to K2P. Therefore it was surprising to find that deletion of the lysyl binding site in K2P completely abolishes fibrin binding. Even when both the F domain and the lysyl binding site were deleted, considerable fibrin binding is still observed (C50 = 557 +/- 126 nM), suggesting other than F and K2-mediated interactions. The binding of FGK1K2P, FGK1K2P (D236N), GK1K2P, and GK1K2P (D236N) to fibrin could be competitively inhibited by FGK1K2P and K2P, indicating that all molecules recognize the same interaction sites on a fibrin clot. Based on these results, a new model for the interaction of t-PA with a forming fibrin clot is proposed. The fibrin binding sites in t-PA are not confined to the F and K2 domain. The main role of the lysyl binding site in the K2 domain of t-PA appears indirect rather than direct, most likely stabilizing a conformation favorable for fibrin binding.

The position of the structurally autonomous kringle 2 domain influences the functional features of tissue-type plasminogen activator.

Tissue-type plasminogen activator (t-PA) is composed of structurally autonomous domains. From the N-terminus of t-PA, a finger-like domain (F), an epidermal growth factor-like domain (G), two kringle domains (K1 and K2) and a serine protease domain (P) can be discerned. The K2 domain of t-PA is known to be involved in lysine binding, fibrin binding and fibrin-dependent plasminogen activation. To study the functional autonomy of the K2 domain in t-PA we constructed, with the aid of a cassette t-PA gene [Rehberg et al. (1989) Protein Engng, 2, 371-377], mutant t-PA genes coding for four molecules (FGK1K2P, FGK2K1P, GK1K2P and GK2K1P) in which the K2 domain was placed in two different positions in t-PA. The DNAs of wild-type t-PA and the t-PA variants were expressed in Chinese hamster ovary cells and the recombinant proteins were purified by affinity chromatography. All molecules were expressed in their single-chain form and could be converted to their two-chain form. With these molecules, lysine binding, fibrin binding and fibrin-dependent plasminogen activation were studied. All variants showed affinity for lysyl-Sepharose and aminohexyl-Sepharose. Reversal of the K domains (FGK2K1P versus FGK2K1P and GK1K2P versus GK2K1P) resulted in a 23-47% weaker interaction to both lysyl-Sepharose and aminohexyl-Sepharose. Deleting the F domain (FGK1K2P versus GK1K2P and FGK2K1P versus GK2K1P) resulted in a 20-70% improvement of the interactions lysyl-Sepharose and aminohexyl-Sepharose.(ABSTRACT TRUNCATED AT 250 WORDS)

Introduction of lysine and clot binding properties in the kringle one domain of tissue-type plasminogen activator.

Despite the high overall similarity in primary structure between kringle one (K1) and kringle two (K2) of tissue-type plasminogen activator (t-PA) there exists an enormous functional difference. It is thought that, in contrast to K1, K2 mediates lysine binding and fibrin binding and is involved in stimulation of plasminogen activation by fibrin or derivatives as CNBr fragments of fibrinogen. Hypothesizing that sequence differences are responsible for differences in function, we compared the amino acid sequences of K1 and K2 with a consensus kringle sequence. Six consecutive amino acids unique to K2 of t-PA were found, i.e. from Asn248 to Trp253. To test whether these residues are involved in lysine binding, fibrin binding, and fibrin-dependent plasminogen activation, we constructed a set of t-PA mutant proteins containing only a kringle and the protease (P) domain: K2P, K1P, and k1P. In the latter molecule the original amino acid residues Ala160-Ser165 from K1 were substituted by Asn248-Trp253 from K2. As expected, K2P showed enhanced plasminogen activation in the presence of CNBr fragments of fibrinogen, bound to lysine-Sepharose and to a forming fibrin clot. K1P did not show any of these features. In contrast, k1P could be stimulated by CNBr fragments of fibrinogen and bound to lysine-Sepharose and a forming fibrin clot. These results indicate that at least a part of the functional differences between K1 and K2 of t-PA can be localized to a stretch of 6 amino acid residues from Asn248 to Trp253 present in K2.