Contextual semantics for sentiment analysis of Twitter

Sentiment analysis on Twitter has attracted much attention recently due to its wide applications in both, commercial and public sectors. In this paper we present SentiCircles, a lexicon-based approach for sentiment analysis on Twitter. Different from typical lexicon-based approaches, which offer a fixed and static prior sentiment polarities of words regardless of their context, SentiCircles takes into account the co-occurrence patterns of words in different contexts in tweets to capture their semantics and update their pre-assigned strength and polarity in sentiment lexicons accordingly. Our approach allows for the detection of sentiment at both entity-level and tweet-level. We evaluate our proposed approach on three Twitter datasets using three different sentiment lexicons to derive word prior sentiments. Results show that our approach significantly outperforms the baselines in accuracy and F-measure for entity-level subjectivity (neutral vs. polar) and polarity (positive vs. negative) detections. For tweet-level sentiment detection, our approach performs better than the state-of-the-art SentiStrength by 4-5% in accuracy in two datasets, but falls marginally behind by 1% in F-measure in the third dataset.

Evidence for the early clinical relevance of hallucinatory-delusional states in the general population

OBJECTIVE: To analyze, in a general population sample, clustering of delusional and hallucinatory experiences in relation to environmental exposures and clinical parameters. METHOD: General population-based household surveys of randomly selected adults between 18 and 65 years of age were carried out. SETTING: 52 countries participating in the World Health Organization's World Health Survey were included. PARTICIPANTS: 225 842 subjects (55.6% women), from nationally representative samples, with an individual response rate of 98.5% within households participated. RESULTS: Compared with isolated delusions and hallucinations, co-occurrence of the two phenomena was associated with poorer outcome including worse general health and functioning status (OR = 0.93; 95% CI: 0.92-0.93), greater severity of symptoms (OR = 2.5 95% CI: 2.0-3.0), higher probability of lifetime diagnosis of psychotic disorder (OR = 12.9; 95% CI: 11.5-14.4), lifetime treatment for psychotic disorder (OR = 19.7; 95% CI: 17.3-22.5), and depression during the last 12 months (OR = 11.6; 95% CI: 10.9-12.4). Co-occurrence was also associated with adversity and hearing problems (OR = 2.0; 95% CI: 1.8-2.3). CONCLUSION: The results suggest that the co-occurrence of hallucinations and delusions in populations is not random but instead can be seen, compared with either phenomenon in isolation, as the result of more etiologic loading leading to a more severe clinical state.

Topic extraction from microblog posts using conversation structures

Conventional topic models are ineffective for topic extraction from microblog messages since the lack of structure and context among the posts renders poor message-level word co-occurrence patterns. In this work, we organize microblog posts as conversation trees based on reposting and replying relations, which enrich context information to alleviate data sparseness. Our model generates words according to topic dependencies derived from the conversation structures. In specific, we differentiate messages as leader messages, which initiate key aspects of previously focused topics or shift the focus to different topics, and follower messages that do not introduce any new information but simply echo topics from the messages that they repost or reply. Our model captures the different extents that leader and follower messages may contain the key topical words, thus further enhances the quality of the induced topics. The results of thorough experiments demonstrate the effectiveness of our proposed model.

RGD-independent cell adhesion via a tissue transglutaminase-fibronectin matrix promotes fibronectin fibril deposition and requires syndecan-4/2 and α5β1 integrin co-signaling

Fibronectin (FN) deposition mediated by fibroblasts is an important process in matrix remodeling and wound healing. By monitoring the deposition of soluble biotinylated FN, we show that the stress-induced TG-FN matrix, a matrix complex of tissue transglutaminase (TG2) with its high affinity binding partner FN, can increase both exogenous and cellular FN deposition and also restore it when cell adhesion is interrupted via the presence of RGD-containing peptides. This mechanism does not require the transamidase activity of TG2 but is activated through an RGD-independent adhesion process requiring a heterocomplex of TG2 and FN and is mediated by a syndecan-4 and ß1 integrin co-signaling pathway. By using a5 null cells, ß1 integrin functional blocking antibody, and a a5ß1 integrin targeting peptide A5-1, we demonstrate that the a5 and ß1 integrins are essential for TG-FN to compensate RGD-induced loss of cell adhesion and FN deposition. The importance of syndecan-2 in this process was shown using targeting siRNAs, which abolished the compensation effect of TG-FN on the RGD-induced loss of cell adhesion, resulting in disruption of actin skeleton formation and FN deposition. Unlike syndecan-4, syndecan-2 does not interact directly with TG2 but acts as a downstream effector in regulating actin cytoskeleton organization through the ROCK pathway. We demonstrate that PKCa is likely to be the important link between syndecan-4 and syndecan-2 signaling and that TG2 is the functional component of the TG-FN heterocomplex in mediating cell adhesion via its direct interaction with heparan sulfate chains.

Fibronectin-tissue transglutaminase matrix rescues RGD-impaired celladhesion through syndecan-4 and β integrin co-signaling

Heterotropic association of tissue transglutaminase (TG2) with extracellular matrix-associated fibronectin (FN) can restore the adhesion of fibroblasts when the integrin-mediated direct binding to FN is impaired using RGD-containing peptide. We demonstrate that the compensatory effect of the TG-FN complex in the presence of RGD-containing peptides is mediated by TG2 binding to the heparan sulfate chains of the syndecan-4 cell surface receptor. This binding mediates activation of protein kinase Ca (PKCa) and its subsequent interaction with ß1 integrin since disruption of PKCa binding to ß1 integrins with a cell-permeant competitive peptide inhibits cell adhesion and the associated actin stress fiber formation. Cell signaling by this process leads to the activation of focal adhesion kinase and ERK1/2 mitogen-activated protein kinases. Fibroblasts deficient in Raf-1 do not respond fully to the TG-FN complex unless either the full-length kinase competent Raf-1 or the kinase-inactive domain of Raf-1 is reintroduced, indicating the involvement of the Raf-1 protein in the signaling mechanism. We propose a model for a novel RGD-independent cell adhesion process that could be important during tissue injury and/or remodeling whereby TG-FN binding to syndecan-4 activates PKCa leading to its association with ß1 integrin, reinforcement of actin-stress fiber organization, and MAPK pathway activation.

Using organoclay to promote morphology refinement and co-continuity in high-density polyethylene/polyamide 6 blends - Effect of filler content and polymer matrix composition

We investigate the gradual changes of the microstructure of two blends of high-density polyethylene (HDPE) and polyamide 6 (PA6) at opposite composition filled with increasing amounts of an organomodified clay. The filler locates preferentially inside the polyamide phase, bringing about radical alterations in the micron-scale arrangement of the polymer phases. When the host polyamide represents the major constituent, a sudden reduction of the average sizes of the polyethylene droplets was observed upon addition of even low amounts of organoclay. A morphology refinement was also noticed at low filler contents when the particles distributes inside the minor phase. In this case, however, keep increasing the organoclay content eventually results in a high degree of PA6 phase continuity. Rheological analyses reveal that the filler loading at which the polyamide assembles in a continuous network corresponds to the critical threshold for its rheological transition from a liquid- to a gel-like behaviour, which is indicative of the structuring of the filler inside the host PA6. On the basis of this finding, a schematic mechanism is proposed in which the role of the filler in driving the space arrangement of the polymer phases is discussed. Finally, we show that the synergism between the reinforcing action of the filler and its ability to affect the blend microstructure can be exploited in order to enhance relevant technological properties of the materials, such as their high temperature structural integrity.

myo-inositol pentakisphosphates. Structure, biological occurrence and phosphorylation to myo-inositol hexakisphosphate

1. Standard and high-performance anion-exchange-chromatographic techniques have been used to purify myo-[3H]inositol pentakisphosphates from various myo-[3H]inositol-prelabelled cells. Slime mould (Dictyostelium discoideum) contained 8 microM-myo-[3H]inositol 1,3,4,5,6-pentakisphosphate 16 microM-myo-[3H]inositol 1,2,3,4,6-pentakisphosphate and 36 microM-D-myo-[3H]inositol 1,2,4,5,6-pentakisphosphate [calculated intracellular concentrations; Stephens & Irvine (1990) Nature (London) 346 580-583]; germinating mung-bean (Phaseolus aureus) seedlings contained both D- and L-myo-[3H]inositol 1,2,4,5,6-pentakisphosphate (which was characterized by 31P and two-dimensional proton n.m.r.) and D- and/or L-myo-[3H]inositol 1,2,3,4,5-pentakisphosphate; HL60 cells contained myo-[3H]inositol 1,3,4,5,6-pentakisphosphate (in a 500-fold excess over the other species), myo-[3H]inositol 1,2,3,4,6-pentakisphosphate and D- and/or L-myo-[3H]inositol 1,2,4,5,6-pentakisphosphate; and NG-115-401L-C3 cells contained myo-[3H]inositol 1,3,4,5,6-pentakisphosphate (in a 100-fold excess over the other species), D- and/or L-myo-[3H]inositol 1,2,4,5,6-pentakisphosphate, myo-[3H]inositol 1,2,3,4,6-pentakisphosphate and D- and/or L-myo-[3H]inositol 1,2,3,4,5-pentakisphosphate. 2. Multiple soluble ATP-dependent myo-inositol pentakisphosphate kinase activities have been detected in slime mould, rat brain and germinating mung-bean seedling homogenates. In slime-mould cytosolic fractions, the three myo-inositol pentakisphosphates that were present in intact slime moulds could be phosphorylated to myo-[3H]inositol hexakisphosphate: the relative first-order rate constants for these reactions were, in the order listed above, 1:8:31 respectively (with first-order rate constants in the intact cell of 0.1, 0.8 and 3.1 s-1, assuming a cytosolic protein concentration of 50 mg/ml), and the Km values of the activities for their respective inositol phosphate substrates (in the presence of 5 mM-ATP) were 1.6 microM, 3.8 microM and 1.4 microM. At least two forms of myo-inositol pentakisphosphate kinase activity could be resolved from a slime-mould cytosolic fraction by both pharmacological and chromatographic criteria. Rat brain cytosol and a soluble fraction derived from germinating mung-bean seedlings could phosphorylate myo-inositol D/L-1,2,4,5,6-, D/L-1,2,3,4,5-, 1,2,3,4,6- and 1,3,4,5,6-pentakisphosphates to myo-inositol hexakisphosphate: the relative first-order rate constants were 57:27:77:1 respectively for brain cytosol (with first-order rate constants in the intact cell of 0.0041, 0.0019, 0.0056 and 0.000073 s-1 respectively, assuming a cytosolic protein concentration of 50 mg/ml) and 1:11:12:33 respectively for mung-bean cytosol (with first-order rate constants in a supernatant fraction with a protein concentration of 10 mg/ml of 0.0002, 0.0022, 0.0024 and 0.0066 s-1 respectively).

A three dimensional surface measurement and reconstruction system for respiratory function analysis

Respiration is a complex activity. If the relationship between all neurological and skeletomuscular interactions was perfectly understood, an accurate dynamic model of the respiratory system could be developed and the interaction between different inputs and outputs could be investigated in a straightforward fashion. Unfortunately, this is not the case and does not appear to be viable at this time. In addition, the provision of appropriate sensor signals for such a model would be a considerable invasive task. Useful quantitative information with respect to respiratory performance can be gained from non-invasive monitoring of chest and abdomen motion. Currently available devices are not well suited in application for spirometric measurement for ambulatory monitoring. A sensor matrix measurement technique is investigated to identify suitable sensing elements with which to base an upper body surface measurement device that monitors respiration. This thesis is divided into two main areas of investigation; model based and geometrical based surface plethysmography. In the first instance, chapter 2 deals with an array of tactile sensors that are used as progression of existing and previously investigated volumetric measurement schemes based on models of respiration. Chapter 3 details a non-model based geometrical approach to surface (and hence volumetric) profile measurement. Later sections of the thesis concentrate upon the development of a functioning prototype sensor array. To broaden the application area the study has been conducted as it would be fore a generically configured sensor array. In experimental form the system performance on group estimation compares favourably with existing system on volumetric performance. In addition provides continuous transient measurement of respiratory motion within an acceptable accuracy using approximately 20 sensing elements. Because of the potential size and complexity of the system it is possible to deploy it as a fully mobile ambulatory monitoring device, which may be used outside of the laboratory. It provides a means by which to isolate coupled physiological functions and thus allows individual contributions to be analysed separately. Thus facilitating greater understanding of respiratory physiology and diagnostic capabilities. The outcome of the study is the basis for a three-dimensional surface contour sensing system that is suitable for respiratory function monitoring and has the prospect with future development to be incorporated into a garment based clinical tool.

The importance of the tissue transglutaminase-fibronectin heterocomplex in the RGD-independent cell adhesion and fibronectin matrix deposition

Tissue transglutaminase (TG2) has been reported as a wound response protein. Once over-expressed by cells under stress such as during wound healing or following tissue damage, TG2 can be secreted and deposited into extracellular matrix, where it forms a heterocomplex (TG-FN) with the abundant matrix protein fibronectin (FN). A further cellular response elicited after tissue damage is that of matrix remodelling leading to the release of the Arg-Gly-Asp (RGD) containing matrix fragments by matrix matelloproteinases (MMPs). These peptides are able to block the interaction between integrin cell surface receptors and ECM proteins, leading to the loss of cell adhesion and ultimately Anoikis. This study provides a mechanism for TG2, as a stress-induced matrix protein, in protecting the cells from the RGD-dependent loss of cell adhesion and rescuing the cells from Anoikis. Mouse fibroblasts were used as a major model for this study, including different types of cell surface receptor knockout mouse embryonic fibroblasts (MEFs) (such as syndecan-4, a5, ß1 or ß3 integrins). In addition specific syndecan-2 targetting siRNAs, ß1 integrin and a4ß1 integrin functional blocking antibodies, and a specific targeting peptide against a5ß1 integrin A5-1 were used to investigate the involvement of these receptors in the RGD-independent cell adhesion on TG-FN. Crucial for TG-FN to compensate the RGD-independent cell adhesion and actin cytoskeleton formation is the direct interaction between the heparan sulfate chains of syndecan-4 and TG2, which elicits the inside-out signalling of a5ß1 integrin and the intracellular activation of syndecan-2 by protein kinase C a (PKCa). By using specific inhibitors, a cell-permeable inhibiting peptide and the detection of the phosphorylation sites for protein kinases and/or the translocation of PKCa via Western blotting, the activation of PKCa, focal adhesion kinase (FAK), ERK1/2 and Rho kinase (ROCK) were confirmed as downstream signalling molecules. Importantly, this study also investigated the influence of TG-FN on matrix turnover and demonstrated that TG-FN can restore the RGD-independent FN deposition process via an a5ß1 integrin and syndecan-4/2 co-signalling pathway linked by PKCa in a transamidating-independent manner. These data provide a novel function for TG2 in wound healing and matrix turnover which is a key event in a number of both physiological and pathological processes.

The suitability of biodegradable poly(dl-lactide-co-glycolide)75:25 microspheres for the sustained release of antibiotics

Post-operative infections resulting from total hip arthroplasty are caused by bacteria such as Staphylococcus aureus and Pseudomonas aeruginosa entering the wound perioperatively or by haemetogenous spread from distant loci of infection. They can endanger patient health and require expensive surgical revision procedures. Gentamicin impregnated poly (methyl methacrylate) bone cement is traditionally used for treatment but is often removed due to harbouring bacterial growth, while bacterial resistance to gentamicin is increasing. The aim of this work was to encapsulate the antibiotics vancomycin, ciprofloxacin and rifampicin within sustained release microspheres composed of the biodegradable polymer poly (dl-lactide-co-glycolide) [PLCG] 75:25. Topical administration to the wound in hydroxypropylmethylcellulose gel should achieve high local antibiotic concentrations while the two week in vivo half life of PLCG 75:25 removes the need for expensive surgical retrieval operations. Unloaded and 20% w/w antibiotic loaded PLCG 75:25 microspheres were fabricated using a Water in Oil emulsification with solvent evaporation technique. Microspheres were spherical in shape with a honeycomb-like internal matrix and showed reproducible physical properties. The kinetics of in vitro antibiotic release into newborn calf serum (NCS) and Hank's balanced salt solution (HBSS) at 37°C were measured using a radial diffusion assay. Generally, the day to day concentration of each antibiotic released into NCS over a 30 day period was in excess of that required to kill St. aureus and Ps. auruginosa. Only limited microsphere biodegradation had occurred after 30 days of in vitro incubation in NCS and HBSS at 37°C. The moderate in vitro cytotoxicity of 20% w/w antibiotic loaded microspheres to cultured 3T3-L1 cells was antibiotic induced. In conclusion, generated data indicate the potential for 20% w/w antibiotic loaded microspheres to improve the present treatment regimens for infections occurring after total hip arthroplasty such that future work should focus on gaining industrial collaboration for commercial exploitation.

Development of a co-culture model of the human lungs for toxicity testing and identification of biomarkers of inhalation toxicity

The airway epithelium is the first point of contact in the lung for inhaled material, including infectious pathogens and particulate matter, and protects against toxicity from these substances by trapping and clearance via the mucociliary escalator, presence of a protective barrier with tight junctions and initiation of a local inflammatory response. The inflammatory response involves recruitment of phagocytic cells to neutralise and remove and invading materials and is oftern modelled using rodents. However, development of valid in vitro airway epithelial models is of great importance due to the restrictions on animal studies for cosmetic compound testing implicit in the 7th amendment to the European Union Cosmetics Directive. Further, rodent innate immune responses have fundamental differences to human. Pulmonary endothelial cells and leukocytes are also involved in the innate response initiated during pulmonary inflammation. Co-culture models of the airways, in particular where epithelial cells are cultured at air liquid interface with the presence of tight junctions and differentiated mucociliary cells, offer a solution to this problem. Ideally validated models will allow for detection of early biomarkers of response to exposure and investigation into inflammatory response during exposure. This thesis describes the approaches taken towards developing an in vitro epithelial/endothelial cell model of the human airways and identification biomarkers of response to exposure to xenobiotics. The model comprised normal human primary microvascular endothelial cells and the bronchial epithelial cell line BEAS-2B or normal human bronchial epithelial cells. BEAS-2B were chosen as their characterisation at air liquid interface is limited but they are robust in culture, thereby predicted to provide a more reliable test system. Proteomics analysis was undertaken on challenged cells to investigate biomarkers of exposure. BEAS-2B morphology was characterised at air liquid interface compared with normal human bronchial epithelial cells. The results indicate that BEAS-2B cells at an air liquid interface form tight junctions as shown by expression of the tight junction protein zonula occludens-1. To this author’s knowledge this is the first time this result has been reported. The inflammatory response of BEAS-2B (measured as secretion of the inflammatory mediators interleukin-8 and -6) air liquid interface mono-cultures to Escherichia coli lipopolysaccharide or particulate matter (fine and ultrafine titanium dioxide) was comparable to published data for epithelial cells. Cells were also exposed to polymers of “commercial interest” which were in the nanoparticle range (and referred to particles hereafter). BEAS-2B mono-cultures showed an increased secretion of inflammatory mediators after challenge. Inclusion of microvascular endothelial cells resulted in protection against LPS- and particle- induced epithelial toxicity, measured as cell viability and inflammatory response, indicating the importance of co-cultures for investigations into toxicity. Two-dimensional proteomic analysis of lysates from particle-challenged cells failed to identify biomarkers of toxicity due to assay interference and experimental variability. Separately, decreased plasma concentrations of serine protease inhibitors, and the negative acute phase proteins transthyretin, histidine-rich glycoprotein and alpha2-HS glycoprotein were identified as potential biomarkers of methyl methacrylate/ethyl methacrylate/butylacrylate treatment in rats.

A novel extracellular role for tissue transglutaminase in matrix-bound VEGF-mediated angiogenesis

The importance of tissue transglutaminase (TG2) in angiogenesis is unclear and contradictory. Here we show that inhibition of extracellular TG2 protein crosslinking or downregulation of TG2 expression leads to inhibition of angiogenesis in cell culture, the aorta ring assay and in vivo models. In a human umbilical vein endothelial cell (HUVEC) co-culture model, inhibition of extracellular TG2 activity can halt the progression of angiogenesis, even when introduced after tubule formation has commenced and after addition of excess vascular endothelial growth factor (VEGF). In both cases, this leads to a significant reduction in tubule branching. Knockdown of TG2 by short hairpin (shRNA) results in inhibition of HUVEC migration and tubule formation, which can be restored by add back of wt TG2, but not by the transamidation-defective but GTP-binding mutant W241A. TG2 inhibition results in inhibition of fibronectin deposition in HUVEC monocultures with a parallel reduction in matrix-bound VEGFA, leading to a reduction in phosphorylated VEGF receptor 2 (VEGFR2) at Tyr1214 and its downstream effectors Akt and ERK1/2, and importantly its association with b1 integrin. We propose a mechanism for the involvement of matrix-bound VEGFA in angiogenesis that is dependent on extracellular TG2-related activity. © 2013 Macmillan Publishers Limited. All rights reserved.

The role of organoclay in promoting co-continuous morphology in high-density poly(ethylene)/poly(amide) 6 blends

The effect of organically modified clay on the morphology, rheology and mechanical properties of high-density polyethylene (HDPE) and polyamide 6 (PA6) blends (HDPE/PA6 = 75/25 parts) is studied. Virgin and filled blends were prepared by melt compounding the constituents using a twin-screw extruder. The influence of the organoclay on the morphology of the hybrid was deeply investigated by means of wide-angle X-ray diffractometry, transmission and scanning electron microscopies and quantitative extraction experiments. It has been found that the organoclay exclusively places inside the more hydrophilic polyamide phase during the melt compounding. The extrusion process promotes the formation of highly elongated and separated organoclay-rich PA6 domains. Despite its low volume fraction, the filled minor phase eventually merges once the extruded pellets are melted again, giving rise to a co-continuous microstructure. Remarkably, such a morphology persists for long time in the melt state. A possible compatibilizing action related to the organoclay has been investigated by comparing the morphology of the hybrid blend with that of a blend compatibilized using an ethylene–acrylic acid (EAA) copolymer as a compatibilizer precursor. The former remains phase separated, indicating that the filler does not promote the enhancement of the interfacial adhesion. The macroscopic properties of the hybrid blend were interpreted in the light of its morphology. The melt state dynamics of the materials were probed by means of linear viscoelastic measurements. Many peculiar rheological features of polymer-layered silicate nanocomposites based on single polymer matrix were detected for the hybrid blend. The results have been interpreted proposing the existence of two distinct populations of dynamical species: HDPE not interacting with the filler, and a slower species, constituted by the organoclay-rich polyamide phase, which slackened dynamics stabilize the morphology in the melt state. In the solid state, both the reinforcement effect of the filler and the co-continuous microstructure promote the enhancement of the tensile modulus. Our results demonstrate that adding nanoparticles to polymer blends allows tailoring the final properties of the hybrid, potentially leading to high-performance materials which combine the advantages of polymer blends and the merits of polymer nanocomposites.

Selective localization of organoclay and effects on the morphology and mechanical properties of LDPE/PA11 blends with distributed and co-continuous morphology

A study was made on the effect of small amounts of organically modified clay on the morphology and mechanical properties of blends of low-density polyethylene and polyamide 11 at different compositions. The influence of the filler on the blend morphology was investigated using wide angle X-ray diffractometry, scanning and transmission electron microscopy and selective extraction experiments. The filler was found to locate predominantly in the more hydrophilic polyamide phase. Although such uneven distribution does not have a significant effect on the onset of phase co-continuity of the polymer components, it brings about a drastic refinement of the microstructure for the blends both with droplets/matrix and co-continuous morphologies. In addition to the expected reinforcing action of the filler, the resulting fine microstructure plays an important role in enhancing the mechanical properties of the blends. This is essentially because of a good quality of stress transfer across the interface between the constituents, which also seems to benefit for a good interfacial adhesion promoted by the filler. Our results provide the experimental evidence for the capabilities of nanoparticles added to multiphase polymer systems to act selectively as a reinforcing agent for specific domains of the material and as a medium able to assist the refinement of the polymer phases during mixing.

Precipitation casting of drug-loaded microporous PCL matrices:incorporation of progesterone by co-dissolution

Microporous, poly(ε-caprolactone) (PCL) matrices were loaded with progesterone by precipitation casting using co-solutions of PCL and progesterone in acetone. Progesterone loadings up to 32% w/w were readily achieved by increasing the drug content of the starting PCL solution. The kinetics of steroid release in PBS at 37°C over 10 days could be described effectively by a diffusional release model although the Korsmeyer-Peppas model indicated the involvement of multiple release phenomena. The diffusion rate constant (D) increased from 8 to 24 μg/mg matrix/day0.5 as the drug loading increased from 3.6 to 12.4% w/w. A total cumulative release of 75%-95% indicates the high efficiency of steroid delivery. Increasing the matrix density from 0.22 to 0.39 g/cm3, by increasing the starting PCL solution concentration, was less effective in changing drug release kinetics. Retention of anti-proliferative activity of released steroid was confirmed using cultures of breast cancer epithelial (MCF-7) cells. Progesterone released from PCL matrices into PBS at 37°C over 14 days retarded the growth of MCF-7 cells by a factor of at least 3.5 compared with progesterone-free controls. These findings recommend further investigation of precipitation-cast PCL matrices for delivery of bioactive molecules such as anti-proliferative agents from implanted, inserted or topical devices.