Telomeres in most immortal cells are maintained by the enzyme telomerase, allowing cells to divide indefinitely. Some telomerase-negative tumors and immortal cell lines maintain long heterogeneous telomeres by the ALT (alternative lengthening of telomeres) mechanism; such tumors are expected to be resistant to anti-telomerase drug therapies. Occasionally telomerase-negative Saccharomyces cerevisiae mutants survive, and 10% of them (type II survivors) have unstable telomeres. As in human ALT+ cells, short telomeres in yeast type II survivors lengthen abruptly; in yeast, this is dependent on the recombination proteins Rad52p and Rad50p. In human cells, ALT involves copying of sequence from a donor to a recipient telomere. We have characterized for the first time a class of complex telomere mutations seen only in ALT+ cells. The mutant telomeres are defined by the replacement of the progenitor telomere at a discrete point (fusion point) with a different telomere repeat array. Among 19 characterized fusion points, one occurred within the first six repeats of the telomere, indicating that these recombination-like events can occur anywhere within the telomere. One mutant telomere may have been involved in a secondary recombination-like mutation event, suggesting that these mutations are sporadic but ongoing in ALT+ cells. We also identified simple intra-allelic mutations at high frequency, which evidently contribute to telomere instability in ALT+ cells.

Alternative lengthening of telomere (ALT) tumors maintain telomeres by a telomerase-independent mechanism and are characterized by a nuclear structure called the ALT-associated PML body (APB). TRF2 is a component of a telomeric DNA/protein complex called shelterin. However, TRF2 function in ALT cells remains elusive. In telomerase-positive tumor cells, TRF2 inactivation results in telomere de-protection, activation of ATM, and consequent induction of p53-dependent apoptosis. We show that in ALT cells this sequence of events is different. First, TRF2 inactivation/silencing does not induce cell death in p53-proficient ALT cells, but rather triggers cellular senescence. Second, ATM is constitutively activated in ALT cells and colocalizes with TRF2 into APBs. However, it is only following TRF2 silencing that the ATM target p53 is activated. In this context, PML is indispensable for p53-dependent p21 induction. Finally, we find a substantial loss of telomeric DNA upon stable TRF2 knockdown in ALT cells. Overall, we provide insight into the functional consequences of shelterin alterations in ALT cells.

Amoebae from natural isolates of Physarum polycephafum, unlike the plasmodial phase, are unable to grow in axenic medium. A mutant strain of amoebae, CLd-AXE, can be cultured in the liquid, semi-defined medium used for plasmodial culture but lacks some of the properties required for studies of development and gene expression. From crosses of CLd-AXE with wildtype amoebae, new amoebal strains able to grow in axenic medium have been isolated; some of these can also undergo the reversible amoeba-flagellate transformation and apogamic plasmodium development in axenic conditions. Amoebae maintained in active growth in liquid culture for several months showed little change in their properties. Subcultures made with diluted inocula indicated that the same growth rate was achieved even when single amoebae were inoculated in liquid medium. All strains produced colonies with high efficiency when replated on. bacterial lawns. Measurements of DNA content by flow cytometry indicated that the majority of amoebae in liquid cultures were haploid. Homozygous diploid amoebae constructed from one strain grew less well than the haploid cells. Genetic analysis of crosses suggested that amoebae able to grow in liquid axenic medium fell into one major phenotypic class with respect to growth rate, and that mutation at only one or two loci was necessary to allow amoebae to grow in axenic medium. Diploid, heterozygous amoebae constructed by mating a mutant with a wildtype strain were unable to grow in axenic medium, indicating that at least one of the putative axe alleles was recessive.

Evidence is presented that in the true slime mould Physarumpolycephalum the ability of the amoebal strain CLd-AXE to grow in axenic medium is determined by mutations in two genes axeA and axeB and that the axenic growth of RSD~-AXE amoebae is also under genetic control. Mutant amoebal strains are also able to grow on bacterial lawns and the ability to grow in axenic media persists during prolonged culture on bacteria. However, some of the mutant strains grow less well on bacteria than strains of similar genetic background which are unable to grow in axenic media. CLd-AXE has the same nuclear DNA content (0.59 pg per nucleus) as a d , the strain from which it was derived. Amoebae able to grow in axenic media were also derived from strains ~2 7 and ~~8 5 8 which carry mutations for temperature sensitivity and leucine requirement expressed in the plasmodia1 phase. Tests in axenic media showed that these mutations were expressed in the amoebal phase. The elucidation of the genetic basis of axenic growth will allow the construction of a range of amoebal strains able to grow in axenic media and this will greatly facilitate the isolation and analysis of mutants in this organism.

Previous workers reported that Physarum polycephalum amoebae cultured in liquid axenic medium were induced to form cysts b y the addition of mannitol. Their criterion for encystment was the formation of detergent (Triton)-resistant cells (TRC). In this study the frequencies of TRC in suspensions of amoebae from various treatments were compared with counts of cell types identified by transmission electron microscopy. Amoebae treated with mannitol in axenic liquid culture formed 50% TRC after 17 h but no walled cysts were found. It was concluded that TRC induced by mannitol were dense, rounded cells without walls. In contrast, TRC formed after growth to stationary phase on bacterial lawns were walled cells. When resuspended in growth medium, most mannitol-induced TRC reverted to active amoebae within a f e w minutes, whereas TRC formed on bacteria remained T r i t o n resistant for many hours. It was concluded that delayed reversion of TRC was a more reliable indication of wall formation than Triton resistance alone. Transfer of amoebae from liquid culture to the surface of diluted axenic agar medium resulted in the formation of walled cysts identical in appearance with those formed on bacterial lawns. The results indicated that efficient encystment requires a solid substratum as well as nutrient deprivation.