Su M. Metcalfe

The interaction of small molecules with spin-labelled erythrocyte membranes

Abstract The spectra of non-covalent spin labels have been examined in the presence of erythrocyte membranes and their separated lipid and protein components. Evidence has been obtained for the perturbations induced in the membrane structure by a range of extraneous molecules. The perturbations are detected either as changes in the spectra of the bound spin label, or as changes in the partition of the spin label into the membrane. A comparison of the perturbations detected by different spin labels provides preliminary evidence for the localisation of the perturbing agents in the membrane. With steroid spin labels it is possible to detect the onset of irreversible structural breakdown in the membrane at high concentrations of perturbing agents. This results in the exposure of new protein binding sites for the steroid spin labels which are not accessible in the intact membrane. The interpretation of the spin label experiments is supported in detail by directly comparable studies of the nuclear magnetic relaxation of the same perturbing agents in erythrocyte membranes.

Genetic control of IL-1β bioactivity through differential regulation of the IL-1 receptor antagonist

Regulation of IL‐1 bioactivity by the IL‐1 receptor antagonist (IL‐1Ra) is critical for the preservation of normal vascular structure and function in mice. In humans, IL‐1 bioactivity may be further fine‐tuned by genetic polymorphisms. We recently proposed that an intronic polymorphism of the human gene encoding the IL‐1R antagonist (IL1RN) is a genetic risk factor for the development of chronic cardiac allograft rejection. Here we aimed to establish a physiological basis for this susceptibility. Plasma and peripheral blood mononuclear cells (PBMC) were obtained from 55 healthy human volunteers, whose genotypes for four polymorphisms of IL1 family genes (IL1B, IL1R1 and IL1RN) were determined by PCR. IL‐1β and IL‐1Ra production and release in PBMC cultures were studied by flow cytometry and ELISA. Functional and genotypic data were pooled and analyzed first by multivariate and, subsequently, by univariate statistical tests. Wewere able to confirm our observed association of IL1RN genotype with chronic cardiac allograft rejection using multivariate statistics. IL1RN genotype also emerged as the principal regulator of both constitutive and stimulated IL‐1Ra and IL‐1β release. Allele IL1RN*;2 was consistently associated with higher IL‐1Ra and lower IL‐1β release, in a dosage‐dependent manner. Conversely, IL1RN*;2 carriage was associated with reduced production of the intracellular isoform of IL‐1Ra. These genotypic effects were only observed with prolonged culture prior to stimulation and werenot appreciably influenced by the presence of exogenous modulators of IL‐1β production. We conclude that IL1RN genotype is the principal determinant of IL‐1β bioactivity within theIL1 gene cluster in humans and discuss its putative role in disease.

The interaction of p53 with the nuclear matrix is mediated by F-actin and modulated by DNA damage.

The tumour suppressor protein p53 is localized in the cell nucleus where it serves to initiate cellular responses to a variety of stresses, particularly DNA damage and has the capacity to transactivate stress response genes. An emerging body of evidence indicates that its action is also exerted through direct protein–protein interactions. An approach to understanding p53 function has been to analyse its positioning in relation to nuclear structures and we have shown that p53 can associate with the nuclear matrix. A potential nuclear matrix component for this association is actin. Here we show that p53 interacts with nuclear F-actin and we map the domains involved in this interaction. Using fluorescence resonance energy transfer, we demonstrate that the partition of p53 between F-actin bound and unbound forms is not constant, but is modulated by the presence of DNA damage, which increases binding. Our results indicate that the dynamic interaction of p53 with the nuclear matrix has to be considered for a full understanding of the mechanisms of the p53-mediated cellular response to DNA damage.

Tolerance of porcine renal allografts induced by donor spleen cells and seven days' treatment with cyclosporine

Liver allografts in pigs and in rats elicit a substantial cellular immune response that can resolve spontaneously with the induction of donor-specific systemic tolerance. Self-limiting interactions between host and donor (graft)-derived leukocytes may be the basis for tolerogenesis. We have attempted to reproduce this effect of liver grafting in pigs by peroperative infusion of donor leukocytes into kidney graft recipients given an interrupted short course of CsA designed to promote donor leukocyte survival and interaction with host cells. This protocol can secure long-term kidney graft survival resistant to challenge by donor skin grafting. Donor skin is, however, rejected, but more slowly than third-party skin, indicating a degree of systemic specific unresponsiveness in these long-term kidney graft recipients.

The phosphorylation status of Ascl1 is a key determinant of neuronal differentiation and maturation in vivo and in vitro

Generation of neurons from patient fibroblasts using a combination of developmentally defined transcription factors has great potential in disease modelling, as well as ultimately for use in regeneration and repair. However, generation of physiologically mature neurons in vitro remains problematic. Here we demonstrate the cell-cycle-dependent phosphorylation of a key reprogramming transcription factor, Ascl1, on multiple serine-proline sites. This multisite phosphorylation is a crucial regulator of the ability of Ascl1 to drive neuronal differentiation and maturation in vivo in the developing embryo; a phosphomutant form of Ascl1 shows substantially enhanced neuronal induction activity in Xenopus embryos. Mechanistically, we see that this un(der)phosphorylated Ascl1 is resistant to inhibition by both cyclin-dependent kinase activity and Notch signalling, both of which normally limit its neurogenic potential. Ascl1 is a central component of reprogramming transcription factor cocktails to generate neurons from human fibroblasts; the use of phosphomutant Ascl1 in place of the wild-type protein significantly promotes neuronal maturity after human fibroblast reprogramming in vitro. These results demonstrate that cell-cycle-dependent post-translational modification of proneural proteins directly regulates neuronal differentiation in vivo during development, and that this regulatory mechanism can be harnessed to promote maturation of neurons obtained by transdifferentiation of human cells in vitro.

New rapamycin derivatives by precursor-directed biosynthesis

Rapamycin (1) is a polyketide macrolide produced by Streptomyces hygroscopicus that displays potent immunosuppressant activity. In recent years there has been great interest in the chemistry and biology of rapamycin, and of the structurally similar immunosuppressants FK506 and FK520. We now describe results that promise to broaden the scope for biosynthetic engineering of new rapamycin and FK506/FK520 derivatives. Feeding studies have confirmed the largely polyketide origin of rapamycin, and have shown that the unusual trisubstituted cyclohexane ring arises from shikimate. The entire biosynthetic gene cluster for rapamycin was sequenced by Leadlay, Staunton and co-workers in 1994. The rapamycin polyketide synthase (PKS) is a type I mixed PKS/non-ribosomal peptide synthetase (NRPS) system in which 14 polyketide chain-extension modules are housed in three giant multimodular proteins (RAPS1 ±3). An NRPS-related multienzyme (pipecolate-incorporating enzyme) inserts pipecolate and (probably) ensures closure of the macrocycle. The genes encoding the rapamycin PKS (RAPS) have proved an excellent resource for combinatorial biosynthesis, owing in part to the diverse chemical functionality of the polyketide product. DNA from the rap PKS genes encoding individual and multiple enzyme activities and even whole chain-extension modules, has been spliced into the genes for other PKSs to generate functional, hybrid PKSs. In contrast, despite the potential clinical value of the products, this technology has not yet been extensively applied to engineer the biosynthesis of novel rapamycins. We previously identified 4,5-dihydroxycyclohex-1-enecarboxylic acid (2) as the most likely starter unit for the rapamycin PKS (Scheme 1). We have proposed that this substrate is activated

LIF in the regulation of T-cell fate and as a potential therapeutic

At the heart of lineage commitment within the adaptive immune response is the intrinsic genetic plasticity of the naive peripheral T lymphocyte (T cell). Primary activation by presentation of cognate antigen is coupled to rapid T-cell cycling and progressive epigenetic changes that guide the cell down distinct T-cell lineages, either effector (Th1, Th2, Th17) or tolerogenic (Treg). Fate choice is influenced both by strength of the priming activation signal and by cues from the micro-environment that are integrated with lineage-specific gene expression profiles, eventually becoming hard-wired in the fully differentiated cell. The micro-environmental cues include cytokines, and the discovery that leukaemia inhibitory factor (LIF) and interleukin (IL)-6 counter-regulate development of the Treg and Th17 lineages places LIF within the core regulatory circuitry of T cells. I first summarise current understanding of LIF and the LIF receptor in the context of T cells. Next, the central relevance of the LIF/IL-6 axis in immune-mediated disease is set in the context of (i) a new nano-therapeutic approach for targeted delivery of LIF and (ii) MARCH-7, a novel E3-ligase discovered to have a central mechanistic role in LIF-mediated T-cell biology, functioning as a rheostat-type regulator of endogenous LIF-signalling.

Leukemia inhibitory factor is linked to regulatory transplantation tolerance.

Background. The specific regulation of allo-tolerance in vivo occurs within a complex microenvironment and involves co-operation between a small proportion of different cell types within the spleen or draining lymph node. By analyzing unmanipulated whole spleen cell populations we have aimed to mimic this in vivo situation to identify critical signaling molecules in regulatory allo-tolerance. Methods. We compared the kinetics of cytokine release and induction of signaling proteins in BALB/c-tolerantCBA, versus BALB/c-rejectedCBA, spleen cells after challenge with BALB/c antigen. Results. The distinguishing features of allo-tolerance were Foxp3 protein expression, LIF release, and increased levels of STAT3. Comparison of isogenic clones of Tr1, Th1, and Th2 cells revealed that only the regulatory Tr1 cells are characterized by both LIF and IL10 release. Conclusions. Overall, our findings demonstrate that allo-antigen driven signaling events can be detected within a whole spleen cell population and identify a role for LIF in the regulation of transplantation tolerance in vivo.

Rapamycin and p53 act on different pathways to induce G1 arrest in mammalian cells

Certain growth regulatory kinases contain a common domain related to the phospho-inositol 3 (PI-3) kinase catalytic site. These include the ATM gene product, DNA-PKcs, and the target of rapamycin (TOR in yeast; and FRAP in mammalian cells). Rapamycin inhibits growth factor signalling and induces G1 arrest in many cell types. Some growth regulatory PI-3 kinases appear functionally linked to p53 and we have explored potential links between cellular effects induced by rapamycin and p53. In p53 null cells rapamycin inhibited cell cycling but did not induce G1 arrest. In cells which showed selective G1 arrest in response to rapamycin, rapamycin had no effect on basal levels of p53 protein. Similarly p21(WAF1) protein was not induced by rapamycin. The kinetics of the cellular p53/p21(WAF1) response to ionising radiation was unaffected by rapamycin; and the ability of growth factor to protect against p53-mediated apoptosis in response to DNA damage was also unaffected by rapamycin. The ATM gene is mutated in the cancer susceptibility syndrome ataxia telangiectasia (AT) but such mutant cells showed a similar sensitivity to rapamycin compared to their normal counterparts. RKO cell lines of common genetic background, but with different levels of functional p53 protein, also responded similarly to rapamycin. Thus, although rapamycin and p53 are each able to induce G1 arrest, they appear to act through independent growth regulatory pathways.

Wild-type p53 protein shows calcium-dependent binding to F-actin.

Nuclear localization of p53 is required for p53 to detect and respond to DNA strand abnormalities and breaks following DNA damage. This leads to activation of the tumour suppressive functions of p53 resulting in either cell cycle arrest and DNA repair; or apoptosis. Critical functional changes in DNA which require strand breaks, including gene rearrangement, may transiently mimic DNA damage: here it is important not to trigger a p53 response. The fine control of p53 in these different circumstances is unknown but may include transient sequestering of p53 in the cytoplasm. Reversible nuclear-cytoplasmic shuttling is an intrinsic property of p53 (Middeler et al., 1997) associated with cell cycle-related changes in p53s subcellular distribution. Takahashi and Suzuki (1994) described p53 inactivation by shuttling to the cytoplasm and Katsumoto et al. (1995) found wild-type p53 to be closely associated with cytoplasmic actin filaments during DNA synthesis. Here we show that, in the presence of free calcium ions, p53 binds directly to F-actin with a dissocation constant of about 10 μM. Thus, part of the regulatory machinery in normal cell cycling may involve p53-actin interactions regulated by calcium fluxes and the dynamic turnover of F-actin.