München, Techn. Universiẗat, Diss., 2001.

Recent epidemiological data have implicated prenatal events in the development of cardiovascular disorders. Thus low birth weight strongly predicts the later occurrence of hypertension, type II diabetes mellitus, syndrome X and deaths from ischaemic heart disease. The mechanism linking prenatal events and later disease is not clear, although maternal malnutrition has been advocated. We have advanced the hypothesis that glucocorticoids might be important as they retard foetal growth and programme offspring hypertension in rats. The foetus has been thought to be protected from the 2-10 times higher maternal glucocorticoid levels by the placental enzyme 11B-hydroxysteroid dehydrogenase (11B-HSD), which is present in many tissues and in humans catalyses the conversion of the active glucocorticoid cortisol to inert cortisone (corticosterone to 11-dehydrocorticosterone in rats). The precise role of 11B-HSD as a barrier to maternal glucocorticoids during prenatal life has not been fully characterised. The role of 11B-HSD in controlling prenatal glucocorticoid exposure in humans and animals has thus been examined. Two isoforms of 11B-HSD exist, type 1, a widespread NADP dependent reversible enzyme and type 2, a high affinity NAD dependent dehydrogenase found mainly in placenta and kidney. 11B-HSD was found in abundance in the ovary and placenta. The main site of immunohistochemical staining and expression of mRNA (11B-HSD-1) in the rat ovary was in the oocyte. 11B-HSD was oxidative, inactivating corticosterone. In both rat placenta in-vitro (11B-HSD-2), and human placenta in-vitro and ex-vivo (11B-HSD-2) the bioactivity was also predominantly oxidative. The lowest placental enzyme activity at term (and hence the greatest foetal glucocorticoid exposure) was found in the smallest rats with the largest placentas, i.e. those in human studies who would be predicted to develop the highest adult blood pressures (birth weight vs. placental 11B-HSD activity: n = 56; r = 0.46; p < 0.0005). A method to examine 11B-HSD function in fresh intact human placentas was developed (ex-vivo dual circuit cotyledon perfusion) which allows close approximation to the in-vivo situation. The majority of cortisol, from low to high nanomolar concentrations, infused through the maternal circulation was metabolised to inert cortisone by the time it reached the foetal circulation, although considerable individual variation was observed. 118-HSD was the only significant contributor to placental cortisol metabolism at physiological maternal concentrations and inhibition of 118-HSD with either the liquorice constituent glycyrrhetinic acid or its hemi-succinate, carbenoxolone, resulted in abolition of the glucocorticoid barrier, allowing maternally administered cortisol to pass unmetabolised through the placenta. In a prospective study, on 16 normal primiparous women whose placentas were studied with this technique, a positive and significant correlation was found between the effectiveness of 118-HSD and offspring birth weight (r = 0. 67; p < 0. 005). The relationship between placental 118-HSD effectiveness in-vivo and term cord blood osteocalcin (a sensitive marker of glucocorticoid exposure) was prospectively examined in 19 women. Cord blood levels of the bone specific protein osteocalcin were determined with radioimmunoassay. The lowest cord blood osteocalcin levels were found in the foetuses whose placental 118-HSD barrier function was poorest (r = 0.58; p < 0.02), (and had presumably had the greatest glucocorticoid exposure), suggesting that term cord blood osteocalcin levels might be a useful predictor of hypertension, ischaemic heart disease and possibly metabolic bone disease. The findings presented in this thesis represent direct evidence that 118-HSD is the barrier to maternal glucocorticoids, its effectiveness correlating with foetal growth in rats (in-vitro), in humans (ex-vivo), and in-vivo with human cord blood osteocalcin levels (osteocalcin may be a marker of glucocorticoid exposure). In the light of studies on pregnant rats in which administration of exogenous glucocorticoids or 118-HSD inhibitors reduces birth weight and programmes hypertension in the offspring, it is reasonable to propose that increased foetal glucocorticoid exposure consequent upon attenuated placental 118-HSD function may play a role in intrauterine programming of later hypertension.

Zugl.: Erlangen, Nürnberg, Univ., Diss., 2008

Techn. Universiẗat, Diss., 2001--München.

München, Techn. Univ., Diss., 2004.

Diss. (sammanfattning) Stockholm : Karol. inst., 2004. / Härtill 5 uppsatser.

Diss. (sammanfattning) Stockholm : Karol. inst., 2002. / Härtill 6 uppsatser.

Glucocorticoids are steroid hormones that have potent anti-inflammatory actions. Endogenous glucocorticoid action is modulated by 11β-hydroxysteroid dehydrogenase (11β-HSD) which catalyses the interconversion of active glucocorticoids (cortisol, corticosterone) and intrinsically inert forms (cortisone, 11-dehydrocorticosterone). There are 2 isozymes; 11β-HSD type 1 regenerates active glucocorticoids in vivo whereas 11β-HSD type 2 inactivates glucocorticoids. Although 11β-HSD1 is highly expressed in the lung, its role there has been little explored. In this study, the expression and localization of 11β-HSD1 mRNA in lung was confirmed by in situ hybridization. Immunohistochemical staining of mouse lung localized 11β-HSD1 to the cytoplasm of fusiform cells in alveolar walls, in a multivesicular pattern characteristic of interstitial fibroblasts. A lung fibrosis model of inflammation was used to test the role and regulation of 11β-HSD1. The results suggest that levels of 11β-HSD1 mRNA and enzyme were not changed during bleomycin-induced lung inflammation. However, 11β-HSD1-deficient mice showed a more severe inflammatory response than congenic wild-type controls, with greater inflammatory cell infiltration into the lung, and increased levels of HO-1 and iNOS mRNA 14 days following bleomycin installation into lung. Picrosirius red staining of lung sections suggested more collagen deposition in 11β-HSD1-deficient mice than in wild-type controls during the course of the lung inflammatory response. Moreover, whereas naïve 11β-HSD1-deficient mice had significantly lower collagen content in lung (84% of WT levels, p<0.05). 28d after bleomycin there was no significant difference between genotypes (KO having 94% of WT levels, p=0.42) confirming more collagen production in 11β-HSD1-deficient mice following bleomycin. Fibroblasts are critical in the regulation of inflammatory responses and are essential in the model of bleomycin-induced lung injury. Lung fibroblasts may have a different transcriptional regulation of 11β-HSD1 compared to other tissues. In the majority of tissues, 11β-HSD1 can be transcribed from 2 promoters; the P1 promoter is the main promoter used in lung, with other tissues mainly using the P2 promoter. To address the relevance of the P1 promoter in lung and to identify the cell type using the P1 promoter, mouse lungs were collagenase-digested to isolate primary fibroblast and epithelial cells. Isolated lung fibroblasts highly expressed 11β-HSD1, predominantly from the P1 promoter. During passage, primary lung fibroblasts switched promoter usage from P1 to P2. In fibroblast primary culture, treatment with TGF-β for 72h markedly decreased 11β-HSD1 expression to 38% of untreated levels, an effect which was reversed by SB431542, a TGF-β receptor antagonist. Whilst TGF-β reduced levels of mRNA initiating at the P2 promoter, initiation from the P1 promoter was completely repressed. Treatment with TGF-β receptor antagonist increased levels of P1-initiated 11β-HSD1 mRNA by 6.6-fold compared to untreated cells. These data suggest that the switch in 11β-HSD1 promoter usage may be regulated by TGF-β during an inflammatory response. Furthermore, as the P1 and P2 promoters are differentially regulated (e.g. by C/EBPβ, a cytokine-responsive transcription factor), the promoter switch may place 11β-HSD1 under a different transcriptional regulation during inflammation. Taken together, these results suggest that 11β-HSD1 deficiency worsens lung inflammation and results in greater lung fibrosis. Therefore, amplification of intracellular glucocorticoids levels, by 11β-HSD1, may represent an important mechanism to limit the inflammatory response and shape fibroblast function, limiting subsequent collagen production and fibrosis.

Glucocorticoids can reduce infarct size when given immediately after myocardial infarction (MI) but are detrimental when administration is continued into the post-infarct healing phase. A number of experimental studies have shown that reduction of infarct expansion by enhancing blood supply to the infarct border reduces remodelling and improves heart function post-MI. Previous experiments from this laboratory have shown that mice unable to locally regenerate corticosterone due to deficiency in 11β-hydroxysteroid dehydrogenase type 1 (11HSD1) have an enhanced angiogenic response during myocardial infarct healing that is associated with improved cardiac function. We hypothesized that the enhanced angiogenic response in 11HSD1 knock out (-/-) mice would be preceded by augmented inflammation. Moreover this would be associated with improved cardiac function. This thesis aimed firstly to establish that murine cardiac phenotype was not influenced by 11HSD1 deficiency. 11HSD1-/- and C57Bl6 control mice had comparable cardiac structure and function. 11HSD1 expression was localised to fibroblasts and vascular smooth muscle cells in the myocardium. The second aim of this thesis was to characterise the healing response after MI in 11HSD1-/- mice compared to C57Bl6 mice. Neutrophil infiltration peaked 2 days after MI and was significantly enhanced in the 11HSD1-/- mice relative to C57Bl6 mice, despite comparable infarct size in both groups. This was followed by increased macrophage accumulation in the infarct border. Furthermore, in the 11HSD1-/- mice a greater proportion of macrophages were of the alternatively activated phenotype. Left ventricular expression of pro-angiogenic IL-8, but not VEGF, was increased. Cellular proliferation and vessel density at 7 days were greater in 11HSD1-/- compared to C57Bl6 hearts. This was associated with improved cardiac function 7 days post-MI. The third aim of this thesis was to determine whether the enhancement in vessel density and cardiac function was maintained beyond the initial wound healing phase. 11HSD1-/- mice retained the increased vessel density compared to C57Bl6 mice and these vessels were smooth muscle coated suggesting vessel maturation. This was associated with sustained improvement in cardiac function and modification of the scar characteristics. The final aim of this thesis was to establish whether the effect of the knock out could be recapitulated by administration of a small molecule inhibitor of 11HSD1 after MI. Oral administration of the 11HSD1 inhibitor had no effect on inflammation, angiogenesis and heart function as determined at 7 days post-MI relative to vehicle treated animals. In conclusion, the data confirm the enhancement in vessel density and cardiac function in 11HSD1-/- mice and demonstrate that this was preceded by enhanced inflammation. This was not due to an underlying cardiac phenotype or modification of the infarct size. Increased infiltration of alternatively activated macrophages may have been the source of pro-angiogenic factor, IL-8, which was also increased at the time of angiogenesis. Importantly the enhanced vessel density was retained 4 weeks after MI, these vessels were mature suggesting longevity and the improvement in cardiac function was retained. While pharmacological inhibition did not recapitulate the effect of the knock out this may have been due to route of administration. The data provides compelling evidence that further development and use of small molecule inhibitors of 11HSD1 may be of benefit post-MI.

Polychlorinated biphenyls (PCBs) are persistent, man-made toxicants that are linked to adverse health effects and diseases such as endocrine disruption, diabetes, obesity, cardiovascular effects, and cancer. Since their manufacturing began in 1929 for industrial use, and was banned in 1979, they have bioaccumulated in water, sediment, food, animals, humans and more. PCBs are also found in indoor air of older buildings and as inadvertent byproducts in the manufacture of paints and pigments. The lower chlorinated PCBs, those with fewer than 5 chlorine atoms, are readily metabolized to form hydroxylated PCBs (OH-PCBs) that are further converted to PCB-sulfates in reactions catalyzed by cytosolic sulfotransferases (SULTs). Steroid sulfotransferases SULT1E1 and SULT2A1 participate in regulating the homeostasis of estrogens and androgens, respectively, through the deactivation of active hormones. The estrogen sulfotransferase (SULT1E1) is also a potential key player in adipogenesis. Recent literature has shown that downregulating expression of SULT1E1 in cells derived from humans and mice caused opposite effects, where adipogenesis was inhibited or stimulated, respectively. Adipogenesis is the maturation of preadipocytes into mature adipocytes, which is regulated by peroxisome proliferating-activator γ (PPARγ). Adipocytes are a main component of adipose tissue, which is important for energy homeostasis, organ protection, and thermoregulation. Adipose tissue also secretes various cytokines such as adiponectin. Adipose tissue dysfunction can result from adipocyte dysfunction, which can be caused by alterations in cell signaling. The objective of this dissertation research was to determine if OH-PCBs and PCB-sulfates are inhibitors of SULT1E1 and SULT2A1 and if inhibition of SULT1E1 by OH-PCBs could potentially affect adipogenesis. We hypothesized that PCB metabolites would inhibit SULT1E1 and SULT2A1 and potentially affect adipogenesis in both human and murine cell models. Using purified recombinant human SULT1E1 and SULT2A1, I found that 4’-OH-PCB 3, 4-OH-PCB 8, 4-OH-PCB 11, 4’-OH-PCB 25, and 4-OH-PCB 52 were potent inhibitors of the sulfation of representative substrates (7.0 nM estradiol for SULT1E1 and 1.0 µM dehydroepiandrosterone for SULT2A1, Figures 3-3 and 3-4, respectively). Moreover, 4-OH-PCB 11 and 4-OH-PCB 52 were the most potent inhibitors of SULT1E1 and SULT2A1 with IC50 values of 7.2 nM and 1.5 μM, tables 3-1 and 3-2, respectively. The least potent inhibitor of SULT1E1 was 4’-OH-PCB 3, with an IC50 of 1300 nM. The PCB-sulfates were not potent inhibitors for either enzyme. 4-OH-PCB 11 inhibited the sulfation of estradiol in the cytosol of both pre-adipocytes and fully differentiated adipocytes (Figure 4-9). Immortalized human adipocytes were treated with 10 µM of triclosan (a known inhibitor of SULT1E1), 4’-OH-PCB 3 and 4-OH-PCB 11. Experiments included exposure to these toxicants for 1) 72 hours to preadipocytes, 2) 72 hours to preadipocytes followed by 11-day differentiation, 3) to differentiating adipocytes and for 48 hours post-differentiation. The lipid accumulation levels remained unaffected, as determined by microscopic imaging and quantification using AdipoRed. The mRNA expression levels of prominent adipogenic markers SULT1E1, PPARγ, and AdipoQ were measured using RT-Q-PCR. Changes in SULT1E1 and PPARγ expression were unaffected upon treatment before, during and after adipogenesis when compared to controls. However, the increase in AdipoQ expression was reduced upon treatment with 4-OH-PCB 11 in differentiated adipocytes and in preadipocytes exposed for 72 hours followed by 11-day differentiation (Figure 4-14). This could be an indicator of adipocyte dysfunction that was not manifested by a change in lipid accumulation. Murine 3T3-L1 cells were also treated with 10 µM of triclosan, 4’-OH-PCB 3 and 4-OH-PCB 11 for 48 hours to preadipocytes, during 8-day differentiation and for 48 hours after differentiation. The mRNA expression levels of prominent markers of cardiovascular and adipogenesis functions, ACE2, PPARγ, FABP4, and AdipoQ were measured using RT-PCR. Compared to controls, the increase in AdipoQ expression was reduced following treatment of preadipocytes with triclosan and 4-OH-PCB 11 and subsequent differentiation (Figure 5-11). The increase in PPARγ expression remained either unchanged from controls or slightly stimulated in differentiating and differentiated adipocytes (Figures 5-11 and 5-13). Angiotensin-converting enzyme 2 (ACE2) expression was decreased compared to control values, upon treatment with 4’-OH-PCB 3 (Figure 5-12), while fatty acid binding protein 4 (FABP4) expression was stimulated to the same extent across all treatment groups in differentiating adipocytes (Figure 5-12). The results, overall, show that these OH-PCBs did not affect lipid accumulation in human adipocytes, but they may affect other signaling pathways in adipogenesis. 4-OH-PCB 11 decreased adiponectin expression compared to the increase that was seen in unexposed differentiating human and mouse adipocytes. Adiponectin is secreted from adipose tissue, and this decrease could indicate a form of dysfunction. This finding is consistent with the results of the purified SULT1E1 study, where 4-OH-PCB 11 potently inhibited SULT1E1, but 4’-OH-PCB 3 did not (Figure 3-3 and Table 3-1). Thus, there is a potential for OH-PCBs to disrupt the expression of adiponectin and perhaps other vital adipokines and this could negatively affect adipose tissue function. Future studies will be needed to determine if these effects are indeed mediated by intracellular estradiol and SULT1E1. Moreover, the potential for in vivo disruption of circulating adiponectin by OH-PCBs and other toxicants that inhibit SULTs remains to be studied.