Specific patterns of neuronal firing induce changes in synaptic strength that may contribute to learning and memory. If the postsynaptic NMDA (N-methyl-D-aspartate) receptors are blocked, long-term potentiation (LTP) and long-term depression (LTD) of synaptic transmission and the learning of spatial information are prevented. The NMDA receptor can bind a protein known as postsynaptic density-95 (PSD-95), which may regulate the localization of and/or signalling by the receptor. In mutant mice lacking PSD-95, the frequency function of NMDA-dependent LTP and LTD is shifted to produce strikingly enhanced LTP at different frequencies of synaptic stimulation. In keeping with neural-network models that incorporate bidirectional learning rules, this frequency shift is accompanied by severely impaired spatial learning. Synaptic NMDA-receptor currents, subunit expression, localization and synaptic morphology are all unaffected in the mutant mice. PSD-95 thus appears to be important in coupling the NMDA receptor to pathways that control bidirectional synaptic plasticity and learning.

We have hypothesized that end-to-end chromosome fusions observed in some tumours could play a part in genetic instability associated with tumorigenesis and that fusion may result from the loss of the long stretches of G-rich repeats found at the ends of all linear chromosomes. We therefore asked whether there is telomere loss or reduction in common tumours. Here we show that in most of the colorectal carcinomas that we analysed, there is a reduction in the length of telomere repeat arrays relative to the normal colonic mucosa from the same patient. We speculate on the consequences of this loss for tumorigenesis. We also show that the telomere arrays are much smaller in colonic mucosa and blood than in fetal tissue and sperm, and that there is a reduction in average telomere length with age in blood and colon mucosa. We propose that the telomerase is inactive in somatic tissues, and that telomere length is an indicator of the number of cell divisions that it has taken to form a particular tissue and possibly to generate tumours.

Using a combination of different oligonucleotides and restriction enzymes we have examined the gross organisation of repeats within the most distal region of human chromosomes. We demonstrate here that human telomeres do not contain a pure uniform 6 base pair repeat unit but that there are at least three types of repeat. These three types of repeat are present at the ends of most or all human chromosomes. The distribution of each type of repeat appears to be non-random. Each human telomere has a similar arrangement of these repeats relative to the ends of the chromosome. This could reflect differences in the functions that they perform, or might result from the mutation and correction processes occurring at human telomeres. The number of repeat units, the repeat types and arrangement differs at mouse telomeres. Analysing the change in length of the telomeric repeat region between an individuals blood and germline DNA reveals that this is due to variable amounts of the TTAGGG repeat and not the other repeat types. This organization of repeat units at human telomeres will only be confirmed upon the isolation and sequencing of full length (10-15 kb), intact human telomeres.

Creutzfeldt-Jakob disease (CJD) is a transmissible neurodegenerative disorder characterized by the accumulation of aggregates of a cellular protein, PrP, in the brain. In both human and animals, genetic alterations to the gene encoding PrP (PRNP in human) modulate susceptibility to CJD. The recent epidemic of bovine spongi-form encephalopathy in the UK has raised the possibility of transmission from animal produce to humans. To provide a baseline against which to assess possible risk factors, we have determined the frequencies of predisposing mutations and allelic variants in PRNP and their relative contributions to disease. Systematic PRNP genotype analysis was performed on suspected CJD cases referred to the National Surveillance Unit in the UK over the period 1990-1993. Inspection of 120 candidate cases revealed 67 patients with definite and probable CJD, based on clinical and neuropathological criteria. No PRNP mutations were detected in any of the remaining 53 patients assessed as "non-CJD". A disease-associated mutation in the PRNP gene was identified in nine (13.4%) definite and probable cases of CJD, a reliable estimate of the incidence of PRNP-related inherited CJD based on a prospective epidemiological series. Within the group of sporadic CJD patients (lacking PRNP mutations), we confirmed that the genotype distribution with respect to the common methionine/valine (Met/Val) polymorphism at codon 129 within PRNP was significantly different from the normal Caucasian population. The incidence of Met homozygosity at this site was more than doubled and correlated with increased susceptibility to the development of sporadic CJD. Unlike other recent studies, Val homozygosity was also confirmed to be a significant risk factor in sporadic CJD, with the relative risks for the three genotypes Met/Met: Val/Val:Met/Val being 11:4:1.

A case of familial Creutzfeldt-Jakob disease associated with a 144 base pair insertion in the-open reading frame of the prion protein gene is described. Sequencing of the mutated allele showed an arrangement of-six octapeptide repeats, distinct from that of a recently described British family with an insertion of similar size. Thirteen years previously the brother of the proband had died from "Huntington's disease", but re-examination of his neuropathology revealed spongiform encephalopathy and antiprion protein immunocytochemistry gave a positive result. The independent evolution of at least two distinct pathological 144 base pair insertions in Britain is proposed. The importance of maintaming a high index of suspicion of inherited Creutzfeldt-Jakob disease in cases of familial neurodegenerative disease is stressed. (J Neurol Neurosurg Psychiatry 1995;58:65-69)