Listeria keratitis in a horse.

Listeria monocytogenes is ubiquitous in the environment but is rarely reported as a cause of keratitis in animals. In this case, a mare was presented with epiphora and evidence of pain in the right eye. Listeria monocytogenes was isolated from a corneal lesion, and bacteria were also seen in the cytologic evaluation. This is the first reported case of ulcerative keratitis associated with L. monocytogenes in a horse.

Cytokines promote the survival of mouse cranial sensory neurones at different developmental stages.

To investigate when the neurotrophic cytokines ciliary neurotrophic factor (CNTF), leukaemia inhibitory factor (LIF), oncostatin-M (OSM), interleukin-6 (IL-6) and cardiotrophin-1 (CT-1) act on developing sensory neurones and whether they co-operate with neurotrophins in regulating neuronal survival, we studied the in vitro trophic effects of these factors on two well-characterized populations of cranial sensory neurones at closely staged intervals throughout embryonic development. The cutaneous sensory neurones of the trigeminal ganglion, which show an early, transient survival response to BDNF and NT3 before becoming NGF-dependent, were supported by CNTF, LIF, OSM and CT-1 during the late fetal period, several days after the neurones become NGF-dependent. At this stage of development, these cytokines promoted the survival of a subset of NGF-responsive neurones. The enteroceptive neurones of the nodose ganglion, which retain dependence on BDNF throughout fetal development, were supported throughout their development by CNTF, LIF, OSM and CT-1, and displayed an additional survival response to IL-6 in the late fetal period. These findings indicate that populations of sensory neurones display different developmental patterns of cytokine responsiveness and show that embryonic trigeminal neurones pass through several phases of differing neurotrophic factor survival requirements.

Using random weights to train multilayer networks of hard-limiting units.

A gradient descent algorithm suitable for training multilayer feedforward networks of processing units with hard-limiting output functions is presented. The conventional backpropagation algorithm cannot be applied in this case because the required derivatives are not available. However, if the network weights are random variables with smooth distribution functions, the probability of a hard-limiting unit taking one of its two possible values is a continuously differentiable function. In the paper, this is used to develop an algorithm similar to backpropagation, but for the hard-limiting case. It is shown that the computational framework of this algorithm is similar to standard backpropagation, but there is an additional computational expense involved in the estimation of gradients. Upper bounds on this estimation penalty are given. Two examples which indicate that, when this algorithm is used to train networks of hard-limiting units, its performance is similar to that of conventional backpropagation applied to networks of units with sigmoidal characteristics are presented.

Evidence from neuron/lymphoma heterokaryons for a common trans-acting factor suppressing Thy-1 expression.

We have investigated the tissue specificity of a trans-acting regulatory factor observed to suppress Thy-1 expression during normal neuronal differentiation. Heterokaryons were constructed between Thy-1.1+ expressing mouse T cell lymphoma cells and Thy-1.2- mouse sensory neurons and their surface phenotypes determined by immunofluorescence histochemistry 16 h after fusion. Thy-1.1 expression was observed to be specifically suppressed in such heterokaryons whilst the expression of the lymphoma cell surface marker, Ly-1, was not altered. As the nuclei did not fuse in the heterokaryons, it appears that the developmentally regulated diffusible suppressor factor active in sensory neurons is not tissue specific and can down-regulate Thy-1 expression in lymphoma cells. We also report the suppression of the H-2Kk surface antigen in heterokaryons by a similar but independent regulatory mechanism.

Reproductive effects of alternative disinfectants.

Organohalides formed through the reaction of chlorine and organic compounds in natural and waste waters pose potential health hazards. For this reason, alternative water disinfectants that do not form organohalides are being investigated with great interest. Limited data are available on the health effects, in particular reproductive toxicity effects, of these compounds. In our laboratory, we have examined the reproductive effects of chloramine and chlorine administered by gavage in Long-Evans rats. Animals were treated for a total of 66 to 76 days. Males were treated for 56 days and females for 14 days prior to breeding and throughout the 10-day breeding period. Females were treated throughout gestation and lactation. Following breeding, the males were necropsied and evaluated for sperm parameters and reproductive tract histopathology. Adult females and some pups were necropsied at weaning on postnatal day 21. Other pups were treated postweaning until 28 or 40 days of age. These pups were evaluated for the day of vaginal patency and thyroid hormone levels. No differences were observed between control rats and those rats exposed to up to 5 mg/kg/day chlorine or 10 mg/kg/day chloramine when fertility, viability, litter size, day of eye opening, or day of vaginal patency were evaluated. No alterations in sperm count, sperm direct progressive movement (micron/sec), percent motility, or sperm morphology were observed among adult male rats. In addition, male and female reproductive organ weights were comparable to their respective control groups, and no significant histopathologic changes were observed among chlorine- or chloramine-treated male and female rats.