Abstract PCB 180 is a persistent and abundant non-dioxin-like PCB (NDL-PCB). We determined the developmental toxicity profile of ultrapure PCB 180 in developing offspring following in utero and lactational exposure with the focus on endocrine, metabolic and retinoid system alterations. Pregnant rats were given total doses of 0, 10, 30, 100, 300 or 1000 mg PCB 180/kg bw on gestational days 7−10 by oral gavage, and the offspring were sampled on postnatal days (PND) 7, 35 and 84. Decreased serum testosterone and triiodothyronine concentrations on PND 84, altered liver retinoid levels, increased liver weights and induced 7-pentoxyresorufin O-dealkylase (PROD) activity were the sensitive effects used for margin of exposure (MoE) calculations. Liver weights were increased together with induction of the metabolizing enzymes cytochrome P450 (CYP) 2B1, CYP3A1, and CYP1A1. Less sensitive effects included decreased serum estradiol and increased luteinizing hormone levels in females, decreased prostate and seminal vesicle weight and increased pituitary weight in males, increased cortical bone area and thickness of tibial diaphysis in females and decreased cortical bone mineral density in males. Developmental toxicity profiles were partly different in male and female offspring, males being more sensitive to increased liver weight, PROD induction and decreased thyroxine concentrations. MoE assessment indicated that the 95th percentile of current maternal PCB 180 concentrations do not exceed the estimated tolerable human lipid-based PCB 180 concentration. Although PCB 180 is much less potent than dioxin-like compounds, it shares several toxicological targets suggesting a potential for interactions.
Bone mass is often negatively affected in autoimmune diseases such as rheumatoid arthritis. It has generally been assumed that bone mass is controlled by endocrine mechanisms and/or the local bone environment. Recent findings demonstrate that central pathways are involved in the regulation of bone mass. Estrogen is involved in the regulation of bone mass and the central nervous system is also a target for estrogen actions. The aim of this study was to investigate in vivo the role of central estrogen receptor-α (ERα) expression for bone mass.
Methods
Nestin-Cre mice were crossed with ERαflox mice to generate mice lacking ERα expression specifically in nervous tissue (nestin-ERα−/−). Mice homozygous for ERα-loxP, but lacking Cre expression, were used as controls. Phenotyping of the bone was performed using pQCT, µCT, histomorphometry and three-point bending. Gene expression analyses were performed using real-time PCR and serum analyses with commercial kits.
Results
Bone mineral density was increased in both the trabecular and cortical bone compartments in nestin-ERα−/− mice compared to controls. Femoral bone strength was increased in nestin-ERα−/− mice as demonstrated by increased stiffness and maximal load of failure. The high bone mass phenotype in nestin-ERα−/− mice was mainly caused by increased bone formation. Serum leptin levels were elevated as a result of increased leptin expression in white adipose tissue (WAT) and slightly increased amount of WAT in nestin-ERα−/− mice. Leptin receptor mRNA levels were reduced in the hypothalamus but not in bone. Furthermore, nestin-ERα−/− mice displayed disturbed regulation of T cells with reduced proliferation and decreased frequency in bone marrow.
Conclusions
In summary, while peripheral ERα activation increases bone mass, the authors here demonstrate that central ERα activation decreases bone mass. Thus, the balance between peripheral stimulatory and central inhibitory ERα actions is important in the regulation of bone mass. The authors propose that the increased bone mass in nestin-ERα−/− mice is mediated via decreased central leptin sensitivity and thereby increased secretion of leptin from WAT which, in turn, results in increased peripheral leptin-induced bone formation. In addition, an altered T cell regulation might contribute to the high bone mass phenotype in the nestin-ERα−/− mice.
We consider Arnoldi-like processes to obtain symplectic subspaces for Hamiltonian systems. Large dimensional systems are locally approximated by ones living in low dimensional subspaces, and we especially consider Krylov subspaces and some of their extensions. These subspaces can be utilized in two ways: by solving numerically local small dimensional systems and then mapping back to the large dimension, or by using them for the approximation of necessary functions in exponential integrators applied to large dimensional systems. In the former case one can expect an excellent energy preservation and in the latter this is so for linear systems. We consider second order exponential integrators which solve linear systems exactly and for which these two approaches are in a certain sense equivalent. We also consider the time symmetry preservation properties of the integrators. In numerical experiments these methods combined with symplectic subspaces show promising behavior also when applied to nonlinear Hamiltonian problems.
In the arena of privacy-preserving machine learning, differentially private stochastic gradient descent (DP-SGD) has outstripped the objective perturbation mechanism in popularity and interest. Though unrivaled in versatility, DP-SGD requires a non-trivial privacy overhead (for privately tuning the model's hyperparameters) and a computational complexity which might be extravagant for simple models such as linear and logistic regression. This paper revamps the objective perturbation mechanism with tighter privacy analyses and new computational tools that boost it to perform competitively with DP-SGD on unconstrained convex generalized linear problems.
Estrogens are important regulators of bone mass and their effects are mainly mediated via estrogen receptor (ER)α. Central ERα exerts an inhibitory role on bone mass. ERα is highly expressed in the arcuate (ARC) and the ventromedial (VMN) nuclei in the hypothalamus. To test whether ERα in proopiomelanocortin (POMC) neurons, located in ARC, is involved in the regulation of bone mass, we used mice lacking ERα expression specifically in POMC neurons (POMC-ERα(-/-)). Female POMC-ERα(-/-) and control mice were ovariectomized (OVX) and treated with vehicle or estradiol (0.5 μg/d) for 6 weeks. As expected, estradiol treatment increased the cortical bone thickness in femur, the cortical bone mechanical strength in tibia and the trabecular bone volume fraction in both femur and vertebrae in OVX control mice. Importantly, the estrogenic responses were substantially increased in OVX POMC-ERα(-/-) mice compared with the estrogenic responses in OVX control mice for cortical bone thickness (+126 ± 34%, P < .01) and mechanical strength (+193 ± 38%, P < .01). To test whether ERα in VMN is involved in the regulation of bone mass, ERα was silenced using an adeno-associated viral vector. Silencing of ERα in hypothalamic VMN resulted in unchanged bone mass. In conclusion, mice lacking ERα in POMC neurons display enhanced estrogenic response on cortical bone mass and mechanical strength. We propose that the balance between inhibitory effects of central ERα activity in hypothalamic POMC neurons in ARC and stimulatory peripheral ERα-mediated effects in bone determines cortical bone mass in female mice.
Abstract A standard planetary nebula stays more than 10 000 years in the state of a photoionized nebula. As long as the timescales of the most important ionizing processes are much smaller, the ionization state can be characterized by a static photoionization model and simulated with codes like CLOUDY (Ferland et al . 1998). When the star exhibits a late helium flash, however, its ionizing flux stops within a very short period. The star then re-appears from its opaque shell after a few years (or centuries) as a cold giant star without any hard ionizing photons. Describing the physics of such behavior requires a fully time-dependent radiative transfer model. Pollacco (1999), Kerber et al . (1999) and Lechner & Kimeswenger (2004) used data of the old nebulae around V605 Aql and V4334 Sgr to derive a model of the pre-outburst state of the CSPN in a static model. Their argument was the long recombination time scale for such thin media. With regard to these models Schönberner (2008) critically raised the question whether a significant change in the ionization state (and thus the spectrum) has to be expected after a time of up to 80 years, and whether static models are applicable at all.
We consider a Krylov subspace approximation method for the symmetric differential Riccati equation $\dot{X} = AX + XA^T + Q - XSX$, $X(0)=X_0$. The method we consider is based on projecting the large scale equation onto a Krylov subspace spanned by the matrix $A$ and the low rank factors of $X_0$ and $Q$. We prove that the method is structure preserving in the sense that it preserves two important properties of the exact flow, namely the positivity of the exact flow, and also the property of monotonicity. We also provide a theoretical a priori error analysis which shows a superlinear convergence of the method. This behavior is illustrated in the numerical experiments. Moreover, we derive an efficient a posteriori error estimate as well as discuss multiple time stepping combined with a cut of the rank of the numerical solution.
In this paper, we take a quasi-Newton approach to nonlinear eigenvalue problems (NEPs) of the type M(λ)v = 0, where $M:\mathbb {C}\rightarrow \mathbb {C}^{n\times n}$ is a holomorphic function. We investigate which types of approximations of the Jacobian matrix lead to competitive algorithms, and provide convergence theory. The convergence analysis is based on theory for quasi-Newton methods and Keldysh's theorem for NEPs. We derive new algorithms and also show that several well-established methods for NEPs can be interpreted as quasi-Newton methods, and thereby, we provide insight to their convergence behavior. In particular, we establish quasi-Newton interpretations of Neumaier's residual inverse iteration and Ruhe's method of successive linear problems.
Training machine learning models with differential privacy (DP) is commonly done using first-order methods such as DP-SGD. In the non-private setting, second-order methods try to mitigate the slow convergence of first-order methods. The DP methods that use second-order information still provide faster convergence, however the existing methods cannot be easily turned into federated learning (FL) algorithms without an excessive communication cost required by the exchange of the Hessian or feature covariance information between the nodes and the server. In this paper we propose DP-FedNew, a DP method for FL that uses second-order information and results in per-iteration communication cost similar to first-order methods such as DP Federated Averaging.
We propose an algorithm for the adaptation of the learning rate for stochastic gradient descent (SGD) that avoids the need for validation set use. The idea for the adaptiveness comes from the technique of extrapolation: to get an estimate for the error against the gradient flow which underlies SGD, we compare the result obtained by one full step and two half-steps. The algorithm is applied in two separate frameworks: federated and differentially private learning. Using examples of deep neural networks we empirically show that the adaptive algorithm is competitive with manually tuned commonly used optimisation methods for differentially privately training. We also show that it works robustly in the case of federated learning unlike commonly used optimisation methods.
