Kin recognition and cannibalism in polyphenic salamanders

We investigated kin discrimination among larvae of Arizona tigersalamanders (Ambystoma tigrinum nebulosum) which occur as "typical"morphs that feed mostly on invertebrate prey and occasionallyon conspecifics, and as "cannibal" morphs that feed primarilyon conspecifics. When housed with smaller larvae that differedin relatedness, both cannibals and typicals preferentially consumedless-related individuals. Cannibals ate typicals much quickerwhen the choice was between nonkin and siblings than when thechoice was between nonkin and cousins, indicating that cannibalscould distinguish different categories of relatives. Cannibalswere less likely to eat a larval sibling that was a cannibalmorph than a sibling that was a typical morph. Occluding animals'nares temporarily eliminated kin discrimination, implying thatolfaction is important in recognition. Larvae from differentsibships varied considerably in their ability to discriminatekin, and the greater the probability that a larva from a givensibship would develop into a cannibal morph, the more likelythe members of that sibship were to discriminate kin. Our resultsenable us to infer the functional significance of kin recognitionin this species and to develop an evolutionary model of themechanisms underlying the joint control of kin recognition andcannibalistic polyphenism.     

Genetic construction and functional analysis of hybrid polyketide synthases containing heterologous acyl carrier proteins.

The gene that encodes the acyl carrier protein (ACP) of the actinorhodin polyketide synthase (PKS) of Streptomyces coelicolor A3(2) was replaced with homologs from the granaticin, oxytetracycline, tetracenomycin, and putative frenolicin polyketide synthase gene clusters. All of the replacements led to expression of functional synthases, and the recombinants synthesized aromatic polyketides similar in chromatographic properties to actinorhodin or to shunt products produced by mutants defective in the actinorhodin pathway. Some regions within the ACP were also shown to be interchangeable and allow production of a functional hybrid ACP. Structural analysis of the most abundant polyketide product of one of the recombinants by electrospray mass spectrometry suggested that it is identical to mutactin, a previously characterized shunt product of an actVII mutant (deficient in cyclase and dehydrase activities). Quantitative differences in the product profiles of strains that express the various hybrid synthases were observed. These can be explained, at least in part, by differences in ribosome-binding sites upstream of each ACP gene, implying either that the ACP concentration in some strains is rate limiting to overall PKS activity or that the level of ACP expression also influences the expression of another enzyme(s) encoded by a downstream gene(s) in the same operon as the actinorhodin ACP gene. These results reaffirm the idea that construction of hybrid polyketide synthases will be a useful approach for dissecting the molecular basis of the specificity of PKS-catalyzed reactions. However, they also point to the need for reducing the chemical complexity of the approach by minimizing the diversity of polyketide products synthesized in strains that produce recombinant polyketide synthases.     

Biosynthesis and function of membrane bound and secreted forms of recombinant CD11b/CD18 (Mac-1)

Full-length (membrane bound) and truncated (secreted) forms of the beta 2 integrin heterodimer, CD11b/CD18 (Mac-1), were expressed in a human kidney cell line (293) that normally does not express leukocyte adhesion molecules (Leu-CAMs). The biosynthesis of recombinant Mac-1 in 293 cells differed from that reported for leukocytes in that heterodimer formation was not required for CD11b to be exported to the cell surface. A stable cell line was constructed that constitutively secreted the recombinant, truncated Mac-1 heterodimer into growth conditioned cell culture medium. A novel monoclonal antibody that enabled an immunoaffinity method for the selective purification of recombinant Mac-1 heterodimers was identified. Sufficient protein was purified to allow the first measurement of the 50% inhibitory concentration (IC₅₀) for CD11b/CD18 and for the direct comparison of the inhibitory activity of recombinant soluble Mac-1 with that of various CD18 and CD11b specific monoclonal antibodies. Purified recombinant soluble Mac-1 inhibited the binding of neutrophils, activated by opsonized zymosan or fMet-Leu-Phe peptide, to human umbilical vein endothelial cells. Similarly, the recombinant integrin was effective in inhibiting the binding of unactivated neutrophils to tumor necrosis factor (TNF-alpha) activated endothelial cells. The availability of an abundant source of purified, biologically active Mac-1 will enable direct physical and chemical investigations into the relationship between the structure and function of this leukocyte adhesion molecule.     

Mechanisms of cellular resistance to hydrogen peroxide, hyperoxia, and 4-hydroxy-2-nonenal toxicity: the significance of increased catalase activity in H2O2-resistant fibroblasts.

An H2O2-resistant variant (OC14) of the HA1 Chinese hamster fibroblast cell line which demonstrates a 20-fold increase in catalase activity was utilized in the study of mechanisms responsible for cellular resistance to hydrogen peroxide, oxygen, and 4-hydroxy-2-nonenal toxicity. HA1 and OC14 cells were treated with 9 mM aminotriazole which resulted in a 60 to 80% reduction in catalase activity. Pretreatment with aminotriazole resulted in significant sensitization to the toxicity of 1-h exposures to exogenously applied H2O2, which was proportional to the reduction in catalase activity. Treatment with aminotriazole produced significant sensitization to the toxicity of 95% O2 after 45 h of O2 exposure but no sensitization to the toxicity of a 1-h exposure to 50 microM 4-hydroxy-2-nonenal. Inhibition of catalase activity by aminotriazole had no effect on the metabolism of 4-hydroxy-2-nonenal by either cell line tested. These results support the conclusion that in H2O2-resistant cells, catalase activity is a major determinant of cellular resistance to H2O2 toxicity, whereas catalase activity has a limited role in cellular resistance to an acute exposure to 95% O2 and is unrelated to cellular resistance to 4-hydroxy-2-nonenal.     

Enhanced stability in vivo of a thermodynamically stable mutant form of yeast iso-1-cytochrome c

Previous work has established that the N57I amino acid replacement in iso-1-cytochrome c from the yeast Saccharomyces cerevisiae causes an unprecedented increase in thermodynamic stability of the protein in vitro, whereas the N57G replacement diminishes stability. Spectrophotometric measurements of intact cells revealed that the N57I iso-l-cytochrome c is present at higher than normal levels in vivo. Although iso-1-cytochrome c turnover is negligible during aerobic growth, transfer of fully derepressed, aerobically grown cells to anaerobic growth conditions leads to reduction in the levels of all of the cytochromes. Pulsechase experiments carried out under these anaerobic conditions demonstrated that the N57I iso-l-cytochrome c has a longer half-life than the normal protein. This is the first report of enhanced stability in vivo of a mutant form of a protein that has an enhanced thermodynamic stability in vitro. Although the N57I protein concentration is higher than the normal level, reduced growth in lactate medium indicated that the specific activity of this iso-l-cytochrome c in vivo is diminished relative to wild-type. On the other hand, the level of the thermodynamically labile N57G iso-1-cytochrome c was below normal. The in vivo levels of the N57I and N57G iso-l-cytochrome c suggest that proteins in the mitochondrial intermembrane space can be subjected to degradation, and that this degradation may play a role in controlling their normal levels.     

Population dynamics of a meiotic/mitotic expansion model for the fragile X syndrome.

A model to explain the mutational process and population dynamics of the fragile X syndrome is presented. The mutational mechanism was assumed to be a multipathway, multistep process. Expansion of CGG repeats was based on an underlying biological process and was assumed to occur at two time points: meiosis and early embryonic development (mitosis). Meiotic expansion was assumed to occur equally in oogenesis and spermatogenesis, while mitotic expansion was restricted to somatic, or constitutional, alleles of maternal origin. Testable hypotheses were predicted by this meiotic/mitotic model. First, parental origin of mutation is predicted to be associated with the risk of a woman to have a full mutation child. Second, "contractions" seen in premutation male transmissions are predicted not to be true contractions in repeat size, but a consequence of the lack of mitotic expansion in paternally derived alleles. Third, a portion of full-mutation males should have full-mutation alleles in their sperm, due to the lack of complete selection against the full-mutation female. Fourth, a specific premutation-allele frequency distribution is predicted and differs from that based on models assuming only meiotic expansion. Last, it is predicted that approximately 65 generations are required to achieve equilibrium, but this depends greatly on the expansion probabilities.     

Recombination and maternal age-dependent nondisjunction: molecular studies of trisomy 16.

Trisomy 16 is the most common human trisomy, occurring in > or = 1% of all clinically recognized pregnancies. It is thought to be completely dependent on maternal age and thus provides a useful model for studying the association of increasing maternal age and nondisjunction. We have been conducting a study to determine the parent and meiotic stage of origin of trisomy 16 and the possible association of nondisjunction and aberrant recombination. In the present report, we summarize our observations on 62 spontaneous abortions with trisomy 16. All trisomies were maternally derived, and in virtually all the error occurred at meiosis I. In studies of genetic recombination, we observed a highly significant reduction in recombination in the trisomy-generating meioses by comparison with normal female meioses. However, most cases of trisomy 16 had at least one detectable crossover between the nondisjoined chromosomes, indicating that it is reduced--and not absent--recombination that is the important predisposing factor. Additionally, our data indicate an altered distribution of crossing-over in trisomy 16, as we rarely observed crossovers in the proximal long and short arms. Thus, it may be that, at least for trisomy 16, the association between maternal age and trisomy is due to diminished recombination, particularly in the proximal regions of the chromosome.