Dendritic spine loss deep in the neocortex and dendrite distortion with diffusion disturbances occur early in experimental pneumococcal meningitis

Introduction

Streptococcus pneumoniae (pneumococcus) meningitis is a serious disease with substantial lethality and long-term disability in survivors. Loss of synaptic staining in the superficial layers of the neocortex in rodent models and in humans, and pneumolysin (a major pneumococcal toxin)-dependent dendritic spine collapse in brain slices have been described. It remains unclear how deep in the neocortex more discrete changes are present, how soon after disease onset these changes occur, and whether other properties of dendrites are also affected.

Methods

Using a mouse model of pneumococcal meningitis, we studied changes in the neocortex shortly (3-6 h) after the onset of clinical symptoms via modified Golgi-Cox silver staining.

Results

Dendritic changes were present in areas with otherwise unchanged cell numbers and no signs of necrosis or other apparent neuronal pathology. Mature dendritic spines were reduced in the pyramidal neurons running through layers 1-5. Additionally, spine morphology changes (swelling, spine neck distortion), were also observed in the deeper layers 4 and 5 of the neocortex. Immature spines (filopodia) remained unchanged between groups, as well as the dendritic arborization of the analyzed neurons. In a third of the animals with meningitis, massive mechanical distortion of the primary dendrites of most of the pyramidal neurons through layers 1-5 was observed. This distortion was reproduced in acute brain slices after exposure to pneumolysin-containing bacterial lysates (S. pneumoniae D39 strain), but not to lysates of pneumolysin-deficient bacteria, which we explain by the tissue remodeling effect of the toxin. Experimental mechanical dendrite distortion in primary neural cultures demonstrated diminished FRAP diffusion of neuronally-expressed enhanced green fluorescent protein (eGFP), indicative of disturbed dendritic diffusion.

Discussion

Our work extends earlier knowledge of synaptic loss in the superficial cortical layers during meningitis to deeper layers. These changes occurred surprisingly early in the course of the disease, substantially limiting the effective therapeutic window. Methodologically, we demonstrate that the dendritic spine collapse readout is a highly reliable and early marker of neural damage in pneumococcal meningitis models, allowing for reduction of the total number of animals used per a group due to much lower variation among animals.

Easy to build cost-effective acute brain slice incubation system for parallel analysis of multiple treatment conditions

BACKGROUND

Acute brain slices represent a powerful tool for analysis of brain function in physiology and pathology. Commercial systems and custom-build solutions with carbogen (95% O₂/5% CO₂) aeration, but they are expensive, have a high working volume requiring large amount of substances, and only limited options for treatment in parallel are possible.

NEW METHOD

We developed a novel cost-effective incubation system using materials available in every laboratory, allowing parallel incubation of several treatment conditions, thus also reducing the number of experimental animals. Our system incubation parameters were optimized for cortical neuron observation.

RESULTS

We tested several different options using 6, 12 or 24 standard culture well plates, combining them with cell strainer baskets inside. The system was placed in a pre-warmed incubator at 37 °C. Carbogen was injected through a 22 gauge needle, positioned between the basket and the wall of the well. Best results were achieved in a 6-well plate. In 12 and 24-well plates bubbles accumulated beneath the basket, displacing it upwards, making it unsuitable for our purposes. The gas oxygenized the medium without mechanically disturbing the slices, protected within the strainer basket, but still allowing optimal diffusion through the 100 µm pores. In a 6-well plate, six simultaneous treatments were possible in parallel. LDH/Cytotoxicity tests showed an acute toxicity of less than 7%. The system lost about 2.5% per hour of the fluid through evaporation, which was replenished every 2 h. Up to 6 h after treatment, however, this evaporation was excellently tolerated by the neurons even without fluid replenishment, most probably due to the anti-swelling effect of the mildly hypertonic medium. We performed two staining procedures, working excellently with this experimental setup, namely - a modified DiI staining and a slice silver impregnation method, both confirming the intact neuronal morphology. Preserved CA3 calcium influx and removal response following KCl depolarization confirmed the normal physiology of the pyramidal neurons 6 h after exposure in the system.

COMPARISON TO EXISTING METHODS

The proposed system is much cheaper than the commercial solutions, can be constructed in any lab, allows up to 6 different treatments in parallel, which none of the existing systems allows. Antibiotic presence in the incubation medium and adequate evaporation control is required if longer incubation (> 6 h) is needed. Lower incubation volumes (3-6 ml) allow sparing expensive reagents. Our procedure was optimized for cortical neurons, further fine tuning to meet other specific requirements is possible.

CONCLUSIONS

The system we propose allows filling the gap for budget solutions for short to mid-term incubation of acute brain slices.

Plasma cyclic adenosine 3',5'-monophosphate (cAMP) levels in hyperthyroid patients

Plasma cyclic AMP levels were studied in ten subjects with hyperthyroidism before and after propylthiouracilic treatment. In untreated hyperthyroid subjects the plasma cyclic AMP levels were significantly elevated but were normalized when plasma iodothyronine levels declined in the normal range.