Expression of heterochromatin protein 1 in the primary tumor of breast cancer patients in the presence or absence of occult metastatic cells in the bone marrow

Gene silencing may occur in breast cancer samples from patients presenting with occult metastatic cells in the bone marrow and one mechanism regulating gene suppression is heterochromatin formation. We have studied whether members of the heterochromatin protein 1 family Hp1(Hs alpha), Hp1(Hs beta) and Hp1(Hs gamma) which take part in chromatin packaging and gene expression regulation, were differentially expressed in tumors from patients with and without occult metastatic cells in their bone marrow. Tumor samples and bone marrow aspirates were obtained from 37 breast cancer patients. Median age was 63 years and 68% of the patients presented with clinical stage I/II disease. Presence of occult metastatic cells in bone marrow was detected through keratin-19 expression by nested RT-PCR in samples from 20 patients (54.1%). The presence of occult metastatic cells in bone marrow was not associated with node involvement, histological grade, estrogen receptor and ERBB2 immunoexpression. Relative gene expression of HP1(Hs alpha), HP1(Hs beta) and HP1(Hs gamma) was determined by real-time RT-PCR and did not vary according to the presence of occult metastatic cells in bone marrow. In addition, the combined expression of these three transcripts could not be used to classify samples according to the presence of bone marrow micrometastasis. Our work indicates that regulation of heterochromatin formation through HP1 family members may not be the sole mechanism implicated in the metastatic process to the bone marrow. (Int J Biol Markers 2008; 23: 219-24)

Silencing of LRRC49 and THAP10 genes by bidirectional promoter hypermethylation is a frequent event in breast cancer

Previously we found that levels of LRRC49 (leucine rich repeat containing 49; FLJ20156) transcripts were elevated in ER-positive breast tumors compared with ER-negative breast tumors. The LRRC49 gene is located on chromosome 15q23 in close proximity to the THAP10 (THAP domain containing 10) gene. These two genes have a bidirectional organization being arranged head-to-head on opposite strands, possibly sharing the same promoter region. Analysis of the promoter region of this gene pair revealed the presence of potential estrogen response elements (EREs), suggesting the potential of this promoter to be under the control of estrogen. We used quantitative real-time PCR (qPCR) to evaluate the expression of LRRC49 and THAP10 in a series of 72 primary breast tumors, and found reduced LRRC49 and THAP10 expression in 61 and 46% of the primary breast tumors analyzed, respectively. In addition, the occurrence of LRRC49/THAP10 promoter hypermethylation was examined by methylation specific PCR (MSP) in a sub-group of the breast tumors. Hypermethylation was observed in 57.5% of the breast tumors analyzed, and the levels of mRNA expression of both genes were inversely correlated with promoter hypermethylation. We investigated the effects of 17 beta-estradiol on LRRC49 and THAP10 expression in MCF-7 breast cancer cells and found both transcripts to be up-regulated 2- to 3-fold upon 17 beta-estradiol treatment. Our results show that the transcripts of LRRC49/THAP10 bidirectional gene pair are co-regulated by estrogen and that hypermethylation of the bidirectional promoter region simultaneously silences both genes. Further studies will be necessary to elucidate the role of LRRC49/THAP10 down-regulation in breast cancer.

The soybean concentrated extract proliferates the vagina of adult rats

Objective: The aim of this study was to evaluate changes induced on the vagina of ovariectomized rats after treatment with soybean concentrated extract or conjugated equine estrogens and the association of both drugs. Methods: We conducted an experimental study with 50 ovariectomized rats that were randomly divided into five equal groups of 10 animals: GI received vehicle, GII received soybean concentrated extract 46 mg/kg per day, GIII received soybean concentrated extract 120 mg/kg per day, GIV received conjugated equine estrogens 50 mu g/kg per day, and GV received conjugated equine estrogens 50 mu g/kg and soybean concentrated extract 46 mg/kg per day. The substances were administered by gavage during 21 consecutive days. After that, the animals were killed under anesthesia and the vagina was removed for histological and immunohistochemical analysis. Data were initially submitted to analysis of variance. Whenever a significant difference was detected, the study was complemented with the Tukey-Kramer test for multiple comparisons. Results: GII did not show any differences on the vaginal epithelium or collagen compared with GI. GIII presented an increase in vaginal epithelium and collagen amount. GIV had the highest amount of collagen and the signals of vaginal proliferation. GV did not show any additional effect compared with GIV. Conclusions: Our data suggest that a high dose of isoflavone-rich soy extract may have positive effects on the vaginal structures of ovariectomized rats, but this action is less than that of estrogen treatment on vaginal thickness. In addition, soy extract may not block the estrogen effect on vaginal tissue.

Genes Differentially Expressed in Prolactinomas Responsive and Resistant to Dopamine Agonists

Background/Aims: Prolactin (PRL) secretion and its gene expression are inhibited by dopamine. Prolactinomas are the most common secreting pituitary adenomas, and dopamine agonists (DA) are the first choice for their treatment. However, a subset of patients is resistant to DA. As the mechanisms involved in DA resistance are not fully understood, the aim of this study was to obtain new insights regarding the molecular differences between the prolactinomas that are responsive to DA and those that are resistant. Methods: Tumor tissue samples were collected from 17 patients who harbored prolactinomas, which were classified as responsive or resistant according to their clinical and laboratorial reaction to DA. The expression of 6 genes was evaluated by real-time polymerase chain reaction: dopamine receptor type 2 (DRD 2), nerve growth factor-beta (NGFB) and its receptor (NGFR), estrogen receptor-alpha (ERA), estrogen receptor-beta (ERB) and the pituitary tumor transforming gene (PTTG). Results: Median DRD 2 and NGFR expression in responsive patients was significantly higher than in resistant ones (p = 0.029 and p = 0.020, respectively). Moreover, the expressions of DRD 2 and NGFR were positively correlated with PRL decrease during treatment (r = 0.66, p = 0.005 and r = 0.57, p = 0.044, respectively). Furthermore, ERB expression was positively correlated to PTTG expression (r = 0.68, p = 0.032) and negatively correlated to NGFB expression (r = -0.75, p = 0.02). Conclusions: DRD2 and NGFR expressions are related to the responsiveness of prolactinoma to DA. However, PTTG, ERB and ERA expressions are not. Also ERB, ERA and PTTG expressions did not present a clear correlation to tumor aggressiveness. Furthermore, the response of prolactinomas to DA should be viewed as a spectrum ranging from the most responsive to the most resistant ones. Copyright (c) 2008 S. Karger AG, Basel

CXCL12 rs1801157 polymorphism and expression in peripheral blood from breast cancer patients

The role of chemokines has been extensively analyzed both in cancer risk and tumor progression. Among different cytokines, CXCR4 and its ligand CXCL12 have been recently subjected to a closer examination. The single-nucleotide polymorphism (SNP) rs1801157 (previously known as CXCL12-A/SDF1-3`A) in the CXCL12 gene and the relative expression of mRNA CXCL12 in peripheral blood were assessed in breast cancer patients, since the chemokine CXCL12 and its receptor CXCR4 regulate leukocyte trafficking and many essential biological processes, including tumor growth, angiogenesis and metastasis of different types of tumors. Genotyping was performed by PCR-RFLP (polymerase chain reaction followed by restriction fragment length polymorphism) using MspI restriction enzyme and the expression analyses by quantitative RT-PCR. No difference in GG genotype and allele A carrier frequencies were observed between breast cancer patients and healthy blood donors and nor when CXCL12 mRNA expression was assessed among patients with different tumor stages. However a significant difference was observed when CXCL12 mRNA relative expression was analyzed in breast cancer patients in accordance to the presence or absence of the CXCL12 rs1801157 allele A. Allele A breast cancer patients presented a mRNA CXCL12 expression about 2.1-fold smaller than GG breast cancer patients. Estrogen positive patients presenting CXCL12 allele A presented a significantly lower expression of CXCL12 in peripheral blood (p = 0.039) than GG hormone positive patients. Our findings demonstrated that allele A is associated with low expression of CXCL12 in the peripheral blood from ER-positive breast cancer patients, which suggests implications on breast cancer clinical outcome. (C) 2011 Elsevier Ltd. All rights reserved.

Role of N-Methyl-D-Aspartate and Non-N-Methyl-D-Aspartate Receptors in the Cardiovascular Effects of L-Glutamate Microinjection Into the Hypothalamic Paraventricular Nucleus of Unanesthetized Rats

We report on the cardiovascular effects of L-glutamate (L-glu) microinjection into the hypothalamic paraventricular nucleus (PVN) as well as the mechanisms involved in their mediation. L-glu microinjection into the PVN caused dose-related pressor and tachycardiac responses in unanesthetized rats. These responses were blocked by intravenous (i.v.) pretreatment with the ganglion blocker pentolinium (PE; 5 mg/kg), suggesting sympathetic mediation. Responses to L-glu were not affected by local microinjection of the selective non-NMDA receptor antagonist NBQX (2 nmol) or by local microinjection of the selective NMDA receptor antagonist LY235959 (LY; 2 nmol). However, the tachycardiac response was changed to a bradycardiac response after treatment with LY235959, suggesting that NMDA receptors are involved in the L-glu heart rate response. Local pretreatment with LY235959 associated with systemic PE or dTyr(CH(2))(5)(Me)AVP (50 mu g/kg) respectively potentiated or blocked the response to L-glu, suggesting that L-glu responses observed after LY235959 are vasopressin mediated. The increased pressor and bradycardiac responses observed after LY + PE was blocked by subsequent i.v. treatment with the V(1)-vasopressin receptor antagonist dTyr(CH(2))(5)(Me)AVP, suggesting vasopressin mediation. The pressor and bradycardiac response to L-glu microinjection into the PVN observed in animals pretreated with LY + PE was progressively inhibited and even blocked by additional pretreatment with increasing doses of NBQX (2, 10, and 20 nmol) microinjected into the PVN, suggesting its mediation by local non-NMDA receptors. In conclusion, results suggest the existence of two glutamatergic pressor pathways in the PVN: one sympathetic pathway that is mediated by NMDA receptors and a vasopressinergic pathway that is mediated by non-NMDA receptors. (C) 2009 Wiley-Liss, Inc.

Assessment of Vascular Effects of Tamoxifen and Its Metabolites on the Rat Perfused Hindquarter Vascular Bed

Tamoxifen has been suggested to produce beneficial cardiovascular effects, although the mechanisms for these effects are not fully known. Moreover, although tamoxifen metabolites may exhibit 30-100 times higher potency than the parent drug, no previous study has compared the effects produced by tamoxifen and its metabolites on vascular function. Here, we assessed the vascular responses to acetylcholine and sodium nitroprusside on perfused hindquarter vascular bed of rats treated with tamoxifen or its main metabolites (N-desmethyl-tamoxifen, 4-hydroxy-tamoxifen, and endoxifen) for 2 weeks. Plasma and whole-blood thiobarbituric acid reactive substances (TBARS) concentrations were determined using a fluorometric method. Plasma nitrite and NOx (nitrite + nitrate) concentrations were determined using an ozone-based chemiluminescence assay and Griess reaction, respectively. Treatment with tamoxifen reduced the responses to acetylcholine (pD(2) = 2.2 +/- 0.06 and 1.9 +/- 0.05 after vehicle and tamoxifen, respectively; P < 0.05), while its metabolites improved these responses (pD(2) = 2.5 +/- 0.04 after N-desmethyl-tamoxifen, 2.5 +/- 0.03 after 4-hydroxy-tamoxifen, and 2.6 +/- 0.08 after endoxifen; P < 0.01). Tamoxifen and its metabolites showed no effect on endothelial-independent responses to sodium nitroprusside (P > 0.05). While tamoxifen treatment resulted in significantly higher plasma and whole blood lipid peroxide levels (37% and 62%, respectively; both P < 0.05), its metabolites significantly decreased lipid peroxide levels (by approximately 50%; P < 0.05). While treatment with tamoxifen decreased the concentrations of markers of nitric oxide formation by approximately 50% (P < 0.05), tamoxifen metabolites had no effect on these parameters (P > 0.05). These results suggest that while tamoxifen produces detrimental effects, its metabolites produce counteracting beneficial effects on the vascular system and on nitric oxide/reactive oxygen species formation.

THE MEDIAL AMYGDALOID NUCLEUS MODULATES CARDIOVASCULAR RESPONSES TO ACUTE RESTRAINT IN RATS

The medial amygdaloid nucleus (MeA) modulates several physiological and behavioral processes and among them, the cardiovascular correlates of behavioral responses to stressful stimuli. Acute restraint evokes cardiovascular responses, which are characterized by both elevated blood pressure (BP) and intense heart rate (HR) increase. We presently report effects of MeA pharmacological manipulations on BP and HR responses evoked by acute restraint in rats. Bilateral microinjection of 100 nL of the unspecific synaptic blocker COCl(2) (1 mM) into the MeA increased HR response to acute restraint, without significant effect on the BP response. This result indicates an inhibitory influence of MeA on restraint-evoked HR changes. Injections of the non-selective muscarinic receptor antagonist atropine (3 nmol); the inhibitor of choline uptake hemicholinium (2 nmol) or the selective M(1)-receptor antagonist pirenzepine (6 nmol) caused effects that were similar to those caused by cobalt. These results suggest that local cholinergic neurotransmission and M(1)-receptors mediate the MeA inhibitory influence on restraint-related HR responses. Pretreatment with the M3 receptor antagonist 4-DAMP (4-Diphenylacetoxy-N-methylpiperidine methiodide-2 nmol) did not affect restraint-related cardiovascular responses, reinforcing the idea that M(1)-receptors mediate MeA-related inhibitory influence on restraint-evoked HR increase. (C) 2009 IBRO. Published by Elsevier Ltd. All rights reserved.

Teleantagonism: A pharmacodynamic property of the primary nociceptive neuron

Previous work from our group showed that intrathecal (i.t.) administration of substances such as glutamate, NMDA, or PGE(2) induced sensitization of the primary nociceptive neuron (PNN hypernociception) that was inhibited by a distal intraplantar (i.pl.) injection of either morphine or dipyrone. This pharmacodynamic phenomenon is referred to in the present work as ""teleantagonism``. We previously observed that the antinociceptive effect of i.t. morphine could be blocked by injecting inhibitors of the NO signaling pathway in the paw (i.pl.), and this effect was used to explain the mechanism of opioid-induced peripheral analgesia by i.t. administration. The objective of the present investigation was to determine whether this teleantagonism phenomenon was specific to this biochemical pathway (NO) or was a general property of the PNNs. Teleantagonism was investigated by administering test substances to the two ends of the PNN (i.e., to distal and proximal terminals; i.pl. plus i.t. or i.t. plus i.pl. injections). We found teleantagonism when: (i) inhibitors of the NO signaling pathway were injected distally during the antinociception induced by opioid agonists; (ii) a nonselective COX inhibitor was tested against PNN sensitization by IL-1 beta; (iii) selective opioid-receptor antagonists tested against antinociception induced by corresponding selective agonists. Although the dorsal root ganglion seems to be an important site for drug interactions, the teleantagonism phenomenon suggests that, in PNNs, a local sensitization spreads to the entire cell and constitutes an intriguing and not yet completely understood pharmacodynamic property of this group of neurons.

Serotonergic neurotransmission in the dorsal raphe nucleus recruits in situ 5-HT2A/2C receptors to modulate the post-ictal antinociception

The post-ictal immobility syndrome is followed by a significant increase in the nociceptive thresholds in animals and humans. The aim of this study was to assess the involvement of the dorsal raphe nucleus (DRN) in the post-ictal antinociception. The second aim was to study the role of serotonergic intrinsic mechanisms of the DRN in this hypo-algesic phenomenon. Pentylenetetrazole (PTZ), an ionophore GABA-mediated Cl- influx antagonist, was peripherally used to induce tonic-clonic seizures in Wistar rats. The nociceptive threshold was measured by the tail-flick test. Neurochemical lesions of the DRN, performed with microinjection of ibotenic acid (1.0 mu g/0.2 mu L), caused a significant decrease of tonic-clonic seizure-induced antinociception, suggesting the involvement of this nucleus in this antinociceptive Process. Microinjections of methysergide (1.0 and 5.0 mu g/0.2 mu L), a non-selective serotonergic receptor antagonist, into DRN caused a significant decrease in the post-ictal antinociception in seizing animals, compared to controls, in all post-ictal periods Presently studied. These findings were corroborated by microinjections of ketanserin (at 1.0 and 5.0 mu g/0.2 mu L) into DRN. Ketanserin is an antagonist with large affinity for 5-HT2A/2C serotonergic receptors, which, in this Case, Caused a significant decrease in the tail-flick latencies in seizing animals, compared to controls after the first 20 min following tonic-clonic convulsive reactions. These results indicate that serotonergic neurotransmission of the DRN neuronal clusters is involved in the organization of the post-ictal hypo-algesia. The 5-HT2A/2C receptors of DRN neurons seem to be critically involved in the increase of nociceptive thresholds following tonic-clonic seizures. (c) 2008 Elsevier Inc, All rights reserved.

The ventrolateral periaqueductal gray is involved in the cardiovascular response evoked by L-glutamate microinjection into the lateral hypothalamus of anesthetized rats

Microinjection Of L-glutamate (L-glu: 1, 3, 10 and 30nmol/100nL) into the lateral hypothalamus (LH) caused dose-related depressor and bradycardiac responses. The cardiovascular response to L-glu stimulation of the LH was blocked by pretreatment of the ventrolateral portion of the periaqueductal gray matter (vIPAG) with CoCl(2) (1 mM/100nL), indicating the existence of a synaptic relay of the hypotensive pathway in that area. Furthermore, the response to L-glu Was blocked by pretreatment of the vIPAG with 2 nmol/100 nL of the selective NMDA-receptor antagonist LY235959 and was not affected by pretreatment with 2 nmol/100 nL of the selective non-NMDA-receptor antagonist NBQX, suggesting a mediation of the hypotensive response by NMDA receptors in the APAG. In conclusion, our results indicate that the hypotensive pathway activated by microinjection Of L-glu into the LH involves a NMDA synaptic relay in the vIPAG. (c) 2007 Elsevier Ireland Ltd. All rights reserved.

Oestradiol Potentiates Hormone Secretion and Neuronal Activation in Response to Hypertonic Extracellular Volume Expansion in Ovariectomised Rats

Secretion of vasopressin (VP), oxytocin (OT) and atrial natriuretic peptide (ANP) is an essential mechanism for the maintenance of hydromineral homeostasis. Secretion of these hormones is modulated by several circulating factors, including oestradiol. However, it remains unclear how oestradiol exerts this modulation. In the present study we investigated the participation of oestradiol in the secretion of VP, OT and ANP and in activation of vasopressinergic and oxytocinergic neurones of the supraoptic (SON) and paraventricular (PVN) nuclei of the hypothalamus in response to extracellular volume expansion (EVE). For this purpose, ovariectomised (OVX) rats treated for 7 days with vehicle (corn oil, 0.1 ml/rat, OVX+O group) or oestradiol (oestradiol cypionate, 10 mu g/kg, OVX+E group) were subjected to either isotonic (0.15 m NaCl, 2 ml/100 g b.w., i.v.) or hypertonic (0.30 m NaCl, 2 ml/100 g b.w., i.v.) EVE. Blood samples were collected for plasma VP, OT and ANP determination. Another group of rats was subjected to cerebral perfusion, and brain sections were processed for c-Fos-VP and c-Fos-OT double-labelling immunohistochemistry. In OVX+O rats, we observed that both isotonic and hypertonic EVE increased plasma OT and ANP concentrations, although no changes were observed in VP secretion. Oestradiol replacement did not alter hormonal secretion in response to isotonic EVE, but it increased VP secretion and potentiated plasma OT and ANP concentrations in response to hypertonic EVE. Immunohistochemical data showed that, in the OVX+O group, hypertonic EVE increased the number of c-Fos-OT and c-Fos-VP double-labelled neurones in the PVN and SON. Oestradiol replacement did not alter neuronal activation in response to isotonic EVE, but it potentiated vasopressinergic and oxytocinergic neuronal activation in the medial magnocellular PVN (PaMM) and SON. Taken together, these results suggest that oestradiol increases the responsiveness of vasopressinergic and oxytocinergic magnocellular neurones in the PVN and SON in response to osmotic stimulation.

The role of tumor hypoxia in MUC1-positive breast carcinomas

Mucin 1 (MUC1) is a glycoprotein that is expressed on apical cell membranes in a variety of normal tissues. MUC1 is involved in cell signaling, inhibition of cell-cell and cell matrix adhesion, apoptosis, proliferation, and transcription. Hypoxia is an important factor that promotes cancer metastasis and stimulates angiogenesis and tumor progression. Hypoxia inducible factor 1 (HIF-1 alpha) and carbonic anhydrase IX (CAIX) are two molecules that are involved in this process. The role of hypoxia in MUC1+ invasive ductal breast carcinomas is not well established. In this study, the expression of MUC1 was correlated with the hypoxia-associated markers HIF-1 alpha and CAIX, as well as several immunohistochemical markers and clinicopathologic features of prognostic significance in 243 invasive ductal carcinomas. MUC1 was overexpressed in 37.0% of patients and correlated with the expression of estrogen receptor (p = 0.0001), progesterone receptor (p = 0.0001), HIF-1 alpha (p = 0.006), VEGF (p = 0.024), and p53 (p = 0.025). In breast cancer, MUC1 expression has been associated with increased degradation of inhibitor of NF-kappa B (I kappa B alpha), driving NF-kappa B to the nucleus and blocking apoptosis and promoting cell survival. We analyzed NF-kappa B expression in MUC1+ breast carcinoma and found a very significant relationship between these proteins (p = 0.0001). Our findings indicate that MUC1 may play a role in the regulation of hormone receptors by increasing the inactivation of p53 and targeting NF-kappa B to the nucleus. Our data also support the notion that activation of HIF-1 alpha in MUC1+ breast carcinomas may modulate VEGF expression, allowing a metabolic adaptation to hypoxia.

Significance of topoisomerase III beta expression in breast ductal carcinomas: strong associations with disease-specific survival and metastasis

Topoisomerases are ubiquitous nuclear enzymes that regulate DNA structure in eukaryotic cells. The role of topoisomerase III beta, the newest member of the topoisomerase family, in the clinical outcome of breast cancer is still poorly understood. This study aims to investigate the immunoexpression of topoisomerase III beta in breast cancer and its relationships with clinicopathologic features and immunohistochemical markers of prognostic significance in breast pathology. Using tissue microarrays containing 171 cases of primary invasive breast cancer, we analyzed the immunoexpression of topoisomerase III beta, estrogen receptor, progesterone receptor, HER-2, BRCA-1, p53, and Ki67. Immunostaining for topoisomerase III beta was found in 33.9% of breast carcinomas, and immunopositivity was correlated with distant metastasis (P = .036) and death (P = .006). Decreased expression of topoisomerase III beta correlated with low expression of Ki67 (P < .001) and negativity for HER-2 (P < .001), BRCA-1 (P = .001), and p53 (P < .001). In the multivariate analysis, topoisomerase Hip expression was a significant predictor of survival (hazard ratio, 3.006 [95% confidence interval, 1.582-5.715]; P = .001). In conclusion, topoisomerase III beta expression can be a useful marker in assessing the prognosis of patients with breast cancer and is an independent predictor of survival. (C) 2010 Elsevier Inc. All rights reserved.

Comparison of the effects of aerobic and resistance training on cardiac autonomic adaptations in ovariectomized rats

We have compared the effects of two types of physical training on the cardiac autonomic control in ovariectomized and sham-operated rats according to different approaches: double autonomic blockade (DAB) with methylatropine and propranolol; baroreflex sensibility (BRS) and spectral analysis of heart rate variability (HRV). Wistar female rats (+/- 250 g) were divided into two groups: sham-operated and ovariectomized. Each group was subdivided into three subgroups: sedentary rats, rats submitted to aerobic trained and rats submitted to resistance training. Ovariectomy did not change arterial pressure, basal heart rate (HR), DAB and BRS responses, but interfered with HRV by reducing the low-frequency oscillations (LF = 0.20-0.75 Hz) in relation to sedentary sham-operated rats. The DAB showed that both types of training promoted an increase in the predominance of vagal tonus in sham-operated rats, but HR variations due to methylatropine were decreased in the resistance trained rats compared to sedentary rats. Evaluation of BRS showed that resistance training for sham-operated and ovariectomized rats reduced the tachycardic responses in relation to aerobic training. Evaluation of HRV in trained rats showed that aerobic training reduced LF oscillations in sham-operated rats, whereas resistance training had a contrary effect. In the ovariectomized rats, aerobic training increased high frequency oscillations (HF = 0.75-2.5 Hz), whereas resistance training produced no effect. In sham-operated rats, both types of training increased the vagal autonomic tonus, but resistance training reduced HF oscillations and BRS as well. In turn, both types of training had similar results in ovariectomized rats, except for HRV, as aerobic training promoted an increase in HF oscillations. (C) 2011 Elsevier B.V. All rights reserved.