Richard A. Radcliffe

The Role of RNA Editing of Kainate Receptors in Synaptic Plasticity and Seizures

The ionotropic glutamate receptor subunit GluR6 undergoes developmentally and regionally regulated Q/R site RNA editing that reduces the calcium permeability of GluR6-containing kainate receptors. To investigate the functional significance of this editing in vivo, we engineered mice deficient in GluR6 Q/R site editing. In these mutant mice but not in wild types, NMDA receptor-independent long-term potentiation (LTP) could be induced at the medial perforant path-dentate gyrus synapse. This indicates that kainate receptors with unedited GluR6 subunits can mediate LTP. Behavioral analyses revealed no differences from wild types, but mutant mice were more vulnerable to kainate-induced seizures. Together, these results suggest that GluR6 Q/R site RNA editing may modulate synaptic plasticity and seizure vulnerability.

Growth inhibition and regression of lung tumors by silibinin: modulation of angiogenesis by macrophage-associated cytokines and nuclear factor-kappaB and signal transducers and activators of transcription 3.

The latency period for lung tumor progression offers a window of opportunity for therapeutic intervention. Herein, we studied the effect of oral silibinin (742 mg/kg body weight, 5 d/wk for 10 weeks) on the growth and progression of established lung adenocarcinomas in A/J mice. Silibinin strongly decreased both tumor number and tumor size, an antitumor effect that correlates with reduced antiangiogenic activity. Silibinin reduced microvessel size (50%, P < 0.01) with no change in the number of tumor microvessels and reduced (by 30%, P < 0.05) the formation of nestin-positive microvessels in tumors. Analysis of several proteins involved in new blood vessel formation showed that silibinin decreased the tumor expression of interleukin-13 (47%) and tumor necrosis factor-α (47%), and increased tissue inhibitor of metalloproteinase-1 (2-fold) and tissue inhibitor of metalloproteinase-2 (7-fold) expression, without significant changes in vascular endothelial growth factor levels. Hypoxia- inducible factor-1α expression and nuclear localization were also decreased by silibinin treatment. Cytokines secreted by tumor cells and tumor-associated macrophages regulate angiogenesis by activating nuclear factor-κB (NF-κB) and signal transducers and activators of transcription (STAT). Silibinin decreased the phosphorylation of p65NF-κB (ser276, 38%; P < 0.01) and STAT-3 (ser727, 16%; P < 0.01) in tumor cells and decreased the lung macrophage population. Angiopoietin-2 (Ang-2) and Ang-receptor tyrosine kinase (Tie-2) expression were increased by silibinin. Therapeutic efficacy of silibinin in lung tumor growth inhibition and regression by antiangiogenic mechanisms seem to be mediated by decreased tumor-associated macrophages and cytokines, inhibition of hypoxia-inducible factor-1α, NF-κB, and STAT-3 activation, and up-regulation of the angiogenic inhibitors, Ang-2 and Tie-2.

Contextual and cued fear conditioning in C57BL/6J and DBA/2J mice: context discrimination and the effects of retention interval.

Context discrimination and time course studies of contextual fear conditioning revealed strain differences between C57BL/6J (B6) and DBA/2J (D2) mice. Both strains discriminated contexts, but D2 mice exhibited less freezing in a shock-paired context. The strains did not differ immediately, or at 1 and 3 hr after contextual fear conditioning training. D2 mice showed less freezing at 15 min, 30 min, and 24 hr after training. B6 mice exhibited exaggerated generalized freezing and poor discrimination between the context and altered context 7-30 days after training. The acoustic startle response in B6 mice was also enhanced at 14 days after training. D2 mice did not show this pattern of generalized freezing. B6, but not D2, mice retained contextual memories for at least 60 days.

High-resolution mapping of quantitative trait loci for emotionality in selected strains of mice.

While a genetic contribution to many behavioral traits is not in doubt, attempts to find the genes themselves have not met with much success. One advance towards this goal has been the demonstration that it is possible to map genes that determine variation in quantitatively measured behavioral traits with crosses between inbred strains (both rodents and insects have been used). Further progress requires high-resolution mapping of quantitative trait loci (QTL). Current detection methods place QTLs within an interval of about half a chromosome, far too large for positional cloning to be a viable option, and a number of approaches have been advocated to increase resolution (Darvasi 1998). Here we report a novel, high-resolution mapping strategy that exploits the effects of artificial selection followed by inbreeding. The strategy requires prior information about the likely position of QTLs, such as would be available from an F 2 or backcross genome scan, and the localizations it provides need to be confirmed by further tests of segregation. For the latter purpose, we describe a variant of the recombinant inbred segregation test (RIST; Darvasi 1998), which can be employed for confirmation and further fine-mapping of QTLs. We have previously shown, using an F 2 genome scan, that three QTL (on Chrs 1, 12, and 15) contribute to the genetic variance of a psychological trait in mice termed emotionality. In that experiment, we used two measures of emotionality: total distance the animal traverses in 5 minutes in an open-field arena (open-field activity, OFA) and the number of fecal boli produced in the open field during the 5-min test period (open-field defecation, OFD; Flint et al. 1995). Calvin Hall devised the open-field arena in the 1930s (Hall 1934) as a way of assessing emotionality in rats primarily because strong emotion, especially fear, is known to result in defecation and urination in humans (Stouffer et al. 1949). The open-field arena is a white box or circular arena that is brightly lit and considered to be unpleasant for rats and mice. Animals that are relatively inactive (low OFA scores) and have high defecation scores are regarded as having higher emotionality scores (or being more emotionally reactive) than active animals with low defeca-tion scores. The two inbred strains in which we detected QTLs influencing emotionality are the product of an earlier selection experiment for open-field activity: BALB/cJ and C57BL/6 mice were intercrossed and offspring selected for differences in activity scores …

Fine scale mapping of a genetic locus for conditioned fear.

Fear conditioning is one of a number of models for investigating the genetic basis of individual variation in emotion and learning. Genetic mapping using crosses between strains of laboratory mice has identified a locus on chromosome one that appears to influence not only variation in conditioned fear, but also in other validated tests of fear-related behaviour, (including the open-field and the elevated-plus maze), suggesting that the rodent locus may act in ways consistent with how a locus influencing susceptibility to anxiety in humans is believed to operate. Here we use high-resolution mapping in genetically heterogeneous mice to show that a quantitative trait locus influencing conditioned fear can be separated from loci influencing open-field activity. Mapping in two different heterogeneous stocks, the Boulder and Northport HS, gave similar map locations for open-field activity at two positions on the current mouse physical map, one at 162 Mb on chromosome one (negative log P-value 5.4) the other at 173 Mb (negative log P-value 4.8), while mapping of contextual conditioned fear in the Boulder HS identified a locus at 170 Mb (negative log P-value 5.4). Estimates of the 95% confidence intervals show that the locations do not overlap. The region containing a gene or genes that influence variation in conditioned fear is approximately 1 megabase in size and contains only one gene of known function, a pre-B cell leukaemia factor.

Cued and Contextual Fear Conditioning in Mice

Contextual and cued fear conditioning is a robust form of learning in which an association is made between stimuli and their aversive consequences. Fear conditioning has been used in laboratory rodents in part because it is a highly conserved form of behavior that is exhibited in both laboratory situations and in normal environments. Training requires only a single trial and this makes it adaptable to genetic, pharmacological, and biochemical studies. Clinically, it is has relevance to human behavior in that fear conditioning can be produced in humans, and damage to the amygdala prevents fear conditioning.

Behavioural changes produced by transgenic overexpression of γ2L and γ2S subunits of the GABAA receptor

Transgenic mice overexpressing either the mouse γ2L or γ2S subunit of the GABAA receptor were generated in a C57BL/6 J × DBA/2 J mixed background and expanded into transgenic lines. Transgenic mice and littermate controls were analysed with respect to altered behaviour indicative of anxiety, motor activity and acute effects of benzodiazepines and alcohol, as well as with regard to altered responses to alcohol withdrawal and acute functional tolerance to alcohol. Biochemical tests assessed flunitrazepam‐ and ethanol‐enhanced 36Cl– flux stimulated by muscimol in cerebellar and cortical microsacs and [3H]‐flunitrazepam binding to cerebellar membranes. There were no significant differences in any of these measures between the transgenic and control mice, except in tests of acute functional tolerance to acute injection of ethanol. Compared to controls, mice carrying either the γ2L or γ2S transgene developed significantly less tolerance to the ataxic effects of ethanol. We conclude that acute functional tolerance to ethanol is very sensitive to the amount of GABAA receptor γ2 subunit available (regardless of whether it is γ2L or γ2S) but overexpression of neither subunit isoform alters other behavioural and biochemical phenotypes.

Confirmation of contextual fear conditioning QTLs by short-term selection.

A short-term selection study for contextual fear conditioning was conducted as a confirmational strategy to analyze the chromosomal locations of five previously mapped contextual fear conditioning quantitative trait loci (QTLs). The founding population was a C57BL/6 (B6) × DBA/2 (D2) F2 intercross. High and low lines were selected for three generations based on contextual fear conditioning scores. Fear conditioning was quantified as the percentage of time spent in a “frozen” posture when placed back into the chamber, where a mild footshock and a tone had been paired with exposure to the context 24 h earlier. Allele frequencies of at least three SSLP DNA markers linked to each of the five QTLs were determined in each generation. As the selection progressed, the frequency of D2 alleles decreased in the low line and increased in the high line for chromosomes 1 and 16, while the opposite was observed in chromosomes 2, 3, and 10. The direction of divergence for alleles on these five chromosomes is consistent with the original QTL mapping study. Differences between the lines in D2 allele frequencies were found to be significant for markers on chromosomes 2, 3, and 16 but did not reach significance on chromosomes 1 or 10. In general, the results are in good agreement with our original fear conditioning QTL mapping study and provide further evidence that these QTLs regulate variation in contextual fear conditioning in crosses of B6 and D2 mice.

Ethanol Sensitivity of GABAergic Currents in Cerebellar Granule Neurons Is Not Increased by a Single Amino Acid Change (R100Q) in the α6 GABAA Receptor Subunit

Cerebellar granule neurons (CGNs) extrasynaptically express GABAA receptors containing α6βxδ subunits, which mediate tonic inhibitory currents. Although it has been shown that the function of these receptors is potently and directly enhanced by ethanol, this finding has not been reproducible across different laboratories. In outbred Sprague-Dawley rats, a naturally occurring arginine (R) to glutamine (Q) mutation in position 100 of the α6 subunit was reported to increase the ethanol sensitivity of these receptors. However, we did not detect an action of this mutation in selectively bred rats (alcohol-tolerant and alcohol-nontolerant). Consequently, we reexamined the effect of the mutation on ethanol sensitivity in Sprague-Dawley rats. Using patch-clamp electrophysiological techniques in cerebellar vermis parasagittal slices, we found that 25 mM ethanol increases the tonic current amplitude, tonic current noise, and spontaneous inhibitory postsynaptic current (sIPSC) frequency to a similar extent in α6-100R/100R and α6-100Q/100Q CGNs. Exposure to 80 mM ethanol increased the tonic current amplitude to a significantly greater extent in α6-100R/100R than in α6-100Q/100Q CGNs; however, the effects of 80 mM ethanol on the tonic current noise and sIPSC frequency were not significantly different between these groups. In the presence of tetrodo-toxin, a non-N-methyl-d-aspartate receptor antagonist, exogenous GABA, and a GABA transporter inhibitor, neither 8 nor 40 mM ethanol consistently affected tonic current amplitude or noise in α6-100R/100R or α6-100Q/100Q CGNs. Thus, the α6-R100Q GABAA receptor subunit polymorphism does not in-crease the acute ethanol sensitivity of extrasynaptic receptors, lending further support to the hypothesis that ethanol modulates these currents indirectly via a presynaptic mechanism.

Tnfa and Il‐10 deficiencies have contrasting effects on lung tumor susceptibility: Gender‐dependent modulation of IL‐10 haploinsufficiency

Epidemiologic evidence suggests that pulmonary diseases with a prominent chronic inflammatory component elevate lung cancer risk. Genetic manipulations of mouse models of lung inflammation and tumorigenesis can be used to investigate this association. The genes encoding pro‐inflammatory tumor necrosis factor‐α (TNFα) and antiinflammatory IL‐10 cytokines map within quantitative trait loci that regulate susceptibility to lung tumor development in mice; sensitive A/J and resistant C57BL/6J (B6) mice have different Tnfa and Il‐10 alleles. Genetic ablation studies were performed to examine whether these genes would qualify as candidate tumor modifiers. Tnfa null (−/−) mice on a B6 background and B6.129 Il‐10−/− mice were intercrossed with A/J mice and subjected to urethane carcinogenesis; lung tumor multiplicity was determined 20 weeks later. In the absence of one copy of Tnfa, tumor number. Male Il‐10+/+ mice developed more tumors than did female mice (P < 0.001), absence of one copy of Il‐10 raised tumor number in female mice to that observed in +/+ males, but no change in multiplicity occurred in Il‐10 hemizygous males. Thus, a deficit of pro‐inflammatory TNFα decreased the number of tumors, whereas diminished gene copy number of anti‐inflammatory IL‐10 increased tumorigenesis; manifestation of an effect of Il‐10 haploinsufficiency is gender dependent. These studies support a role for inflammation in lung cancer susceptibility.