Statistical methods for analyzing microarray feature data with replications.

Expression levels in oligonucleotide microarray experiments depend on a potentially large number of factors, for example, treatment conditions, different probes, different arrays, and so on. To dissect the effects of these factors on expression levels, fixed-effects ANOVA methods have previously been proposed. Because we are not necessarily interested in estimating the specific effects of different probes and arrays, we propose to treat these as random effects. Then we only need to estimate their means and variances but not the effect of each of their levels; that is, we can work with a much reduced number of parameters and, consequently, higher precision for estimating expression levels. Thus, we developed a mixed-effects ANOVA model with some random and some fixed effects. It automatically accounts for local normalization between different arrays and for background correction. The method was applied to each of the 6,584 genes investigated in a microarray experiment on two mouse cell lines, PA6/S and PA6/8, where PA6/S enhances proliferation of Pre B cells in vitro but PA6/8 does not. To detect a set of differentially expressed genes (multiple testing problem), we applied the method of controlling the false discovery rate (FDR), which successfully identified 207 genes with significantly different expression levels.     

The Effects of Arginine Vasotocin on the Calling Behavior of Male Cricket Frogs in Changing Social Contexts

We investigated the effects of the neurohypophysial peptide, arginine vasotocin (AVT), on the calling behavior of maleAcris crepitansduring and immediately following a simulated acoustic agonistic encounter. AVT did not block the aggressive response to agonistic calls, as the changes in temporal call characteristics in response to the encounter were similar to those of saline-treated males. However, AVT caused males to begin calling sooner during the agonistic encounter and to call significantly more than saline males during and after the agonistic encounter. In addition, AVT-treated males maintained a higher dominant frequency compared to saline animals during and following the agonistic encounter. Changes in temporal characteristics in the period following the agonistic encounter indicated that control males were more likely to exhibit a rebound effect which resulted in larger changes in calling parameters compared to AVT-treated animals. The results indicate that AVT causes changes in calling behavior in maleA. crepitansduring and following an agonistic encounter that are consistent with animals highly motivated to maintain vigorous active calling throughout changing social conditions.     

Van Buchem disease (hyperostosis corticalis generalisata) maps to chromosome 17q12-q21.

Van Buchem disease (hyperostosis corticalis generalisata; OMIM 239100 [http://www3.ncbi.nlm.nih. gov:80/htbin-post/Omim/dispmim?239100]) is an autosomal recessive disorder characterized by hyperostosis of the skull, mandible, clavicles, ribs, and diaphyseal cortices of the long bones. The most striking clinical features are the enlargement of the jaw and the thickness of the skull, which may lead to facial nerve palsy, hearing loss, and optic atrophy. Increased formation, by osteoblasts, of qualitatively normal bone has been proposed as the underlying pathological mechanism, but the molecular defect is unknown. We studied 11 van Buchem patients and their highly inbred family, who live in The Netherlands in a small ethnic isolate, that had a common ancestor approximately 9 generations ago. A genomewide search with highly polymorphic microsatellite markers showed linkage to marker D17S1299 on chromosome 17q12-21 (maximum LOD score of 8.82 at a recombination fraction [straight theta] of .01). Analysis of additional markers from that region delineated a candidate region of <1 cM, between markers D17S1787 and D17S934. Interestingly, the only marker not showing recombination with the disease locus was an intragenic marker of the thyroid-hormone receptor alpha1 (THRA1) gene, which generated a LOD score of 12.84 at straight theta=.00. Since thyroid hormones are known to stimulate bone resorption, the THRA1 gene might be involved in the etiology and pathogenesis of van Buchem disease. Unraveling the underlying mechanism for this disorder could contribute to the understanding of the regulatory processes conditioning bone density and the underlying pathological processes.     

Fish interactions with the sediment-water interface

In two mesocosm experiments of cross-classified design, using sixteen 900-liter containers, we measured how benthivorous, omnivorous, and planktivorous fish interact with the sediment-water interface to influence planktonic and benthic production. Experiment 1 used three fish treatments (Ictalurus punctatus, Notemigonus crysoleucas, Lepomis macrochirus) and a fishless control in the presence or absence of a natural pond sediment layer. The benthivorous Ictalurus enhanced turbidity but had no effects on dissolved oxygen, diel changes in dissolved oxygen, pH, or nutrient concentrations. All parameters measured were unaffected by the planktivorous Notemigonus. Experiment 2 compared Ictalurus nebulosus with those of other benthivorous (Cyprinus carpio) and omnivorous (Dorosoma cepedianum) fish, again in the presence or absence of a sediment layer but at a higher stocking density than experiment 1. In the second experiment, Dorosoma enhanced dissolved oxygen levels but had no effect on turbidity while Cyprinus and Ictalurus enhanced turbidity but suppressed dissolved oxygen. Nitrogen concentrations in sediment tubs were enhanced by Cyprinus and Ictalurus but nitrogen concentrations in sediment-free tubs were enhanced by Dorosoma. This would suggest that the benthivores affected nutrient levels through resuspension of sediments while omnivores affected nutrient levels through physiological processes.     

BEHAVIORAL EVIDENCE FOR HOST-RACE FORMATION IN EUROSTA SOLIDAGINIS

We report behavioral evidence that Eurosta solidaginis, a stem-galling tephritid fly, has formed host races on its two goldenrod hosts, Solidago altissima and S. gigantea. Previous work has shown that flies from each host plant differ electrophoretically at the level of host races. The two host-associated populations were truly sympatric and were frequently found on host plants of the two species growing interdigitated with each other. Each host-associated population demonstrated a strong preference for ovipuncturing its own host. The S. gigantea–associated population emerged 10 to 14 d earlier than the S. altissima–associated population, contributing to the reproductive isolation between populations. Partial reproductive isolation is also maintained by a preference for mating on the host from which the fly emerged. The populations meet the criteria established for host races, suggesting that they may be in an intermediate stage of sympatric speciation.     

Prenatal Ethanol Exposure: Changes in Regional Brain Catecholamine Content Following Stress

Abstract: Previous studies have shown that fetal ethanol exposure (FEE) may have long-term effects on the function of catecholaminergic neurons in different regions of the CNS. The present study is the first to examine the effects of FEE on regional brain catecholamine responses following acute stress (a single 60-min restraint stress), repeated stress (single periods of restraint stress on 1, 5, or 10 consecutive days), and recovery from stress (recovery for up to 60 min in the home cage following a single 60-min period of restraint stress). Both male and female offspring from FEE, pair-fed (PF), and ad libitum-fed control (C) groups were tested in adulthood to determine catecholamine content in the cortex, hypothalamus, and hippocampus. A single period of restraint reduced cortical norepinephrine (NE) content in FEE and PF animals compared with that in the cortex of C animals, and reduced hypothalamic NE content in FEE female offspring below that found in animals in all other groups. In contrast, hippo-campal NE content was higher in FEE than in C animals following a single period of restraint; PF animals had intermediate levels of hippocampus NE and did not differ significantly from either FEE or C animals. Following repeated periods of restraint, cortical NE content was lower in FEE than in C animals; PF animals once again had intermediate levels of NE. Importantly, basal (non stressed) NE content did not differ among groups in any brain area examined. In addition, several significant changes in regional brain catecholaminergic responses to acute stress were observed in animals across all treatment groups. Females generally had significantly lower cortical NE levels than males following both single and multiple exposures to restraint. In addition, the cortical NE content decreased below non-stressed levels in all groups following a single restraint period, and remained significantly below basal levels during the 60-min recovery period, whereas the hypothalamic NE content was significantly decreased immediately following the restraint period but showed some recovery toward basal levels by 60 min. There were no significant changes over time in hippocampal NE level or in cortical or hypothalamic dopamine (DA) content following a single restraint stress. Following multiple periods of restraint, hippocampal NE levels were significantly increased and hypothalamic DA levels were significantly decreased in all animals compared with basal levels. These data suggest that the brain noradrenergic response to acute stress is particularly sensitive to the effects of FEE, and that with regard to the hypothalamus, male and female offspring were differentially affected. Furthermore, nutritional effects appear to play some role in mediating the changes in regional brain catecholamine content that are observed. In addition, stress effects on brain catecholamine content across all treatment groups were found to be both region and sex specific.     

Rapamycin blocks cell cycle progression of activated T cells prior to events characteristic of the middle to late G1 phase of the cycle

The effects of rapamycin (RAP) on cell cycle progression of human T cells stimulated with PHA were examined. Cell cycle analysis showed that the RNA content of cells stimulated with PHA in the presence of RAP was similar to that of control T cells stimulated with PHA for 12–24 hr in the absence of the drug. This level was substantially higher than that seen in cells stimulated in the presence of cyclosporin A (CsA), an immunosuppressant known to block cell cycle progression at an early point in the cycle. However, the point in the cell cycle at which RAP acted appeared to be well before the G1/S transition, which occurs about 30–36 hr after stimulation with PHA. In an attempt to further localize the point in the cell cycle where arrest occurred, a set of key regulatory events leading to the G1/S boundary were examined, including p110^Rb phosphorylation, which occurred at least 6 hr prior to DNA synthesis, p34^cdc2 synthesis, and cyclin A synthesis. In control cultures, p110^Rb phosphorylation was detected within 24 hr of PHA stimulation; p34^cdc2 and cyclin A synthesis were detected within 30 hr. Addition of RAP to the cultures inhibited each of these events. In contrast, early events, including c-fos, IL-2, and IL-4 mRNAs expression, and IL-2 receptor (p55) expression, were only marginally affected, if at all, in PHA-stimulated T cells. Furthermore, the inhibition of cell proliferation by RAP could not be overcome by addition of exogenous IL-2. These results indicate that RAP blocks cell cycle progression of activated T cells after IL-2/IL-2 receptor interaction but prior to p110^Rb phosphorylation and other key regulatory events signaling G₁/S transition. © 1993 Wiley-Liss, Inc.