Advances in vertebrate aging research 2007

Among this year's highlights in vertebrate aging research, we find a study in which, contrary to the oxidative stress hypothesis of aging, reduced expression of a major cellular antioxidant, glutathione peroxidase 4, led to a small increase in mouse lifespan. By contrast, a large comparative proteomic analysis discovered a remarkably robust and previous unsuspected inverse association between species lifespan and relative frequency of cysteine residues in mitochondrially encoded respiratory chain proteins only, which the authors attribute to cysteine's ease of oxidation. Another study evaluated more cleanly than any previous work the hypothesis that blood glucose concentration is a key mediator of aging, and concluded that it wasn't. Several new mouse longevity mutants were also reported this year, some ( PAPP-A, IRS-1 , and IRS-2 knockouts) supporting previous work on the importance of insulin/insulin-like growth factor-1 signaling and aging. However, there were inconsistencies between laboratories in some of the results, which merit further investigation. Also, somewhat inconsistent with these findings, over-expression of insulin-like growth factor-1 in heart only lengthened life. From a completely new direction, type 5 adenylyl cyclase knockout mice were observed to live more than 30% longer than controls. Finally, a new program for evaluating potential pharmaceutical interventions in aging and longevity made its appearance, and is notable at this point chiefly for the excellence of its experimental design. A similar program for the disinterested evaluation of reported longevity mutations in mice would be a service to the community of vertebrate aging researchers.     

Replication of extended lifespan phenotype in mice with deletion of insulin receptor substrate 1

We previously reported that global deletion of insulin receptor substrate protein 1 (Irs1) extends lifespan and increases resistance to several age-related pathologies in female mice. However, no effect on lifespan was observed in male Irs1 null mice. We suggested at the time that the lack of any effect in males might have been due to a sample size issue. While such lifespan studies are essential to our understanding of the aging process, they are generally based on survival curves derived from single experiments, primarily due to time and economic constraints. Consequently, the robustness of such findings as a basis for further investigation has been questioned. We have therefore measured lifespan in a second, separate cohort of Irs1 null female mice, and show that, consistent with our previous finding, global deletion of Irs1 significantly extends lifespan in female mice. In addition, an augmented and completed study demonstrates lifespan extension in male Irs1 null mice. Therefore, we show that reduced IRS1-dependent signalling is a robust mechanism through which mammalian lifespan can be modulated.     

Pregnancy-associated plasma protein-A2 (PAPP-A2), a novel insulin-like growth factor-binding protein-5 proteinase

A novel metalloproteinase with similarity to pregnancy-associated plasma protein-A (PAPP-A), which we denoted PAPP-A2, has been identified. Through expression in mammalian cells we showed that recombinant PAPP-A2 polypeptide of 1558 residues resulted from processing of a 1791-residue prepro-protein. Unlike PAPP-A, PAPP-A2 migrated as a monomer (of 220 kDa) in non-reducing SDS-polyacrylamide gel electrophoresis. The prepro-parts of PAPP-A2 and PAPP-A are not homologous, but mature PAPP-A2 shares 45% of its residues with PAPP-A. Because PAPP-A specifically cleaves insulin-like growth factor-binding protein (IGFBP)-4, one of six known modulators of IGF-I and -II, we looked for a possible PAPP-A2 substrate among the members of this family. We showed that PAPP-A2 specifically cleaved IGFBP-5 at one site, between Ser-143 and Lys-144. In contrast to the cleavage of IGFBP-4 by PAPP-A that strictly requires the presence of IGF, the cleavage of IGFBP-5 by PAPP-A2 was IGF-independent. Recent data firmly establish PAPP-A and IGFBP-4 as an important functional pair in several systems. Because of its close relationship with PAPP-A, both structurally and functionally, PAPP-A2 is a likely candidate IGFBP-5 proteinase in many tissues and conditioned media where IGFBP-5 proteolysis has been reported.     

Expression of recombinant human pregnancy-associated plasma protein-A and identification of the proform of eosinophil major basic protein as its physiological inhibitor

Pregnancy-associated plasma protein-A (PAPP-A), originally known from human pregnancy serum, has recently been demonstrated to be a metzincin superfamily metalloproteinase involved in normal and pathological insulin-like growth factor (IGF) physiology. PAPP-A specifically cleaves IGF-binding protein (IGFBP)-4, one of six antagonists of IGF action, which results in release of IGF bound to IGFBP-4. IGFBP-4 is the only known PAPP-A substrate. Its cleavage by PAPP-A uniquely depends on the presence of IGF. We here report mammalian expression and purification of recombinant 1547-residue PAPP-A (rPAPP-A). The recombinant protein is secreted as a homodimer of about 400 kDa composed of two 200-kDa disulfide-bound subunits. Antigenically and functionally, rPAPP-A behaves like the native protein. In human pregnancy, PAPP-A is known to circulate as a 500-kDa disulfide-bound 2:2 complex with the proform of eosinophil major basic protein (proMBP), PAPP-A/proMBP. A comparison between rPAPP-A and pregnancy serum PAPP-A/proMBP complex surprisingly reveals a difference greater than 100-fold in proteolytic activity, showing that proMBP functions as a proteinase inhibitor in vivo. We find that polyclonal antibodies against PAPP-A abrogate all detectable IGFBP-4 proteolytic activity in pregnancy serum, pointing at PAPP-A as the dominating, if not the only, IGFBP-4 proteinase present in the circulation. We further show that pregnancy serum and plasma contain traces (<1%) of uncomplexed PAPP-A with a much higher specific activity than the PAPP-A/proMBP complex. The measurable activity of the PAPP-A/proMBP complex probably results from the presence of a minor subpopulation of partly inhibited PAPP-A that exists in a 2:1 complex with proMBP. Inhibition of PAPP-A by proMBP represents a novel inhibitory mechanism with the enzyme irreversibly bound to its inhibitor by disulfide bonds.     

Complex of pregnancy-associated plasma protein-A and the proform of eosinophil major basic protein. Disulfide structure and carbohydrate attachment

Pregnancy-associated plasma protein-A (PAPP-A) is a metzincin superfamily metalloproteinase responsible for cleavage of insulin-like growth factor-binding protein-4, thus causing release of bound insulin-like growth factor. PAPP-A is secreted as a dimer of 400 kDa but circulates in pregnancy as a disulfide-bound 500-kDa 2:2 complex with the proform of eosinophil major basic protein (pro-MBP), recently shown to function as a proteinase inhibitor of PAPP-A. Except for PAPP-A2, PAPP-A does not share global similarity with other proteins. Three lin-notch (LNR or LIN-12) modules and five complement control protein modules (also known as SCR modules) have been identified in PAPP-A by sequence similarity with other proteins, but no data are available that allow unambiguous prediction of disulfide bonds of these modules. To establish the connectivities of cysteine residues of the PAPP-A.pro-MBP complex, biochemical analyses of peptides derived from purified protein were performed. The PAPP-A subunit contains a total of 82 cysteine residues, of which 81 have been accounted for. The pro-MBP subunit contains 12 cysteine residues, of which 10 have been accounted for. Within the 2:2 complex, PAPP-A is dimerized by a single disulfide bond; pro-MBP is dimerized by two disulfides, and each PAPP-A subunit is connected to a pro-MBP subunit by two disulfide bonds. All other disulfides are intrachain bridges. We also show that of 13 potential sites for N-linked carbohydrate substitution of the PAPP-A subunit, 11 are occupied. The large number of disulfide bonds of the PAPP-A.pro-MBP complex imposes many restraints on polypeptide folding, and knowledge of the disulfide pattern of PAPP-A will facilitate structural studies based on recombinant expression of individual, putative PAPP-A domains. Furthermore, it will allow rational experimental design of functional studies aimed at understanding the formation of the PAPP-A.pro-MBP complex, as well as the inhibitory mechanism of pro-MBP.     

Reevaluation of the effect of dietary restriction on different recombinant inbred lines of male and female mice

Dietary restriction (DR) was reported to either have no effect or reduce the lifespan of the majority of the 41‐recombinant inbred (RI) lines studied by Liao et al. (Aging Cell, 2010, 9, 92). In an appropriately power longevity study (n > 30 mice/group), we measured the lifespan of the four RI lines (115‐RI, 97‐RI, 98‐RI, and 107‐RI) that were reported to have the greatest decrease in lifespan when fed 40% DR. DR increased the median lifespan of female RI‐115, 97‐RI, and 107‐RI mice and male 115‐RI mice. DR had little effect (<4%) on the median lifespan of female and male 98‐RI mice and male 97‐RI mice and reduced the lifespan of male 107‐RI mice over 20%. While our study was unable to replicate the effect of DR on the lifespan of the RI mice (except male 107‐RI mice) reported by Liao et al. (Aging Cell, 2010, 9, 92), we found that the genotype of a mouse had a major impact on the effect of DR on lifespan, with the effect of DR ranging from a 50% increase to a 22% decrease in median lifespan. No correlation was observed between the changes in either body composition or glucose tolerance induced by DR and the changes observed in lifespan of the four RI lines of male and female mice. These four RI lines of mice give the research community a unique resource where investigators for the first time can study the anti‐aging mechanism of DR by comparing mice in which DR increases lifespan to mice where DR has either no effect or reduces lifespan.     

Inhibition of proteolysis by the proform of eosinophil major basic protein (proMBP) requires covalent binding to its target proteinase

By proteolytic cleavage of insulin-like growth factor binding proteins, the metalloproteinase pregnancy-associated plasma protein-A (PAPP-A) is able to control the biological activity of insulin-like growth factors. PAPP-A circulates in pregnancy as a proteolytically inactive complex, disulfide bound to the proform of eosinophil major basic protein (proMBP). We here demonstrate that co-transfection of mammalian cells with PAPP-A and proMBP cDNA results in the formation of a covalent PAPP-A/proMBP complex in which PAPP-A is inhibited. Formation of the complex also occurs when PAPP-A and proMBP synthesized separately are incubated. Complex formation was monitored by Western blotting, and by using an immunoassay specific for the complex. Using mutagenesis, we further demonstrate that the complex forms in a specific manner and depends on the presence of two proMBP cysteine residues. Mutated proMBP, in which Cys-51 and -169 are replaced by serine, is unable to form the covalent complex with PAPP-A. Of particular interest, such mutated proMBP further lacks the ability to inhibit PAPP-A. For the first time, this conclusively demonstrates that proMBP is a proteinase inhibitor. We further conclude that proMBP inhibits PAPP-A in an unusual manner, not paralleled by other proteinase inhibitors of our knowledge, which requires proMBP to be covalently bound to PAPP-A by disulfide bonds. ProMBP binding to PAPP-A most likely either abrogates substrate access to the active site of PAPP-A or induces a conformational change in the structure of PAPP-A, as we, by further mutagenesis, were able to exclude that the inhibitory mechanism of proMBP is based on a cysteine switch-like mechanism.     

Surface association of pregnancy-associated plasma protein-A accounts for its colocalization with activated macrophages

Intense immunostaining for pregnancy-associated plasma protein-A (PAPP-A), a newly characterized metalloproteinase in the insulin-like growth factor system, colocalizes with activated macrophages in human atherosclerotic plaque. To determine macrophage regulation of PAPP-A expression, we developed two models of human macrophages with basal and activated phenotypes. THP-1 cells and peripheral blood monocytes could be differentiated into macrophages and activated upon specific treatment regimens with phorbol myristate acetate, macrophage colony-stimulating factor, and interleukin-1beta. Activation was assessed by cell secretion of tumor necrosis factor-alpha, which increased 30- to 100-fold with activation. Activated macrophages also secreted matrix metalloproteinase-9. However, no PAPP-A mRNA or PAPP-A antigen could be detected in these cells under any condition. Upon incubation with recombinant PAPP-A, we found that activated macrophages bound and internalized more PAPP-A than unactivated macrophages or monocytes. Internalization accounted for at least 50% of macrophage-associated PAPP-A, as assessed in studies with cytochalasin B. Membrane-bound PAPP-A retained protease activity, whereas internalized PAPP-A had little or no activity. Similar experiments carried out with a mutated variant of PAPP-A, which retains functionality as a protease but is unable to bind surface-associated glycosaminoglycan, showed no macrophage association or internalization. Absence of PAPP-A expression was confirmed in activated macrophages isolated from a hypercholesterolemic rabbit model of atherosclerosis. We therefore conclude that PAPP-A is not synthesized in, but rather is bound and internalized by, macrophages. Our findings likely account for the observed intense immunostaining for PAPP-A colocalizing with activated macrophages and may have physiological significance in the development of vulnerable plaque.     

Rapamycin increases lifespan and inhibits spontaneous tumorigenesis in inbred female mice

The nutrient-sensing TO R (target of rapamycin) pathway is involved in cellular and organismal aging. Rapamycin, an inhibitor of TO R, extends lifespan in yeast, fruit flies and genetically heterogeneous mice. Here, we demonstrate that lifelong administration of rapamycin extends lifespan in female 129/Sv mice characterized by normal mean lifespan of 2 y. Importantly, rapamycin was administrated intermittently (2 weeks per month) starting from the age of 2 mo. Rapamycin inhibited age-related weight gain, decreased aging rate, increased lifespan (especially in the last survivors) and delayed spontaneous cancer. 22.9% of rapamycin-treated mice survived the age of death of the last mouse in control group. Thus we demonstrated for the first time in normal inbred mice that lifespan can be extended by rapamycin. This opens an avenue to develop optimal doses and schedules of rapamycin as an anti-aging modality.     

Cell surface targeting of pregnancy-associated plasma protein A proteolytic activity. Reversible adhesion is mediated by two neighboring short consensus repeats

The activities of insulin-like growth factor (IGF)-I and -II are regulated by IGF-binding proteins (IGFBPs). Cleavage of IGFBP-4 by the metalloproteinase pregnancy-associated plasma protein-A (PAPP-A) causes release of bound IGF and has been established in several biological systems including the human reproductive system. Using flow cytometry, we first demonstrate that PAPP-A reversibly binds to the cell surface of several cell types analyzed. Heparin and heparan sulfate, but not dermatan or chondroitin sulfate, effectively compete for PAPP-A surface binding, and because incubation of cells with heparinase abrogated PAPP-A adhesion, binding is probably mediated by a cell surface heparan sulfate proteoglycan. Furthermore, the proteolytic activity of PAPP-A is preserved while bound to cells, suggesting that adhesion functions to target its activity to the vicinity of the IGF receptor, decreasing the probability that released IGF is captured by another IGFBP molecule before receptor binding. This mechanism potentially functions in both autocrine and paracrine regulation, as PAPP-A need not be synthesized in a cell to which it adheres. A truncated PAPP-A variant without the five short consensus repeats in the C-terminal third of the 1547-residue PAPP-A subunit, lacked surface binding. We also show that PAPP-A2, a recently discovered IGFBP-5 proteinase with homology to PAPP-A, does not bind cells. This finding allowed further mapping of the PAPP-A adhesion site to short consensus repeat modules 3 and 4 by the expression and analysis of nine PAPP-A/PAPP-A2 chimeras. Interestingly, the proteolytically inactive, disulfide-bound complex of PAPP-A and the proform of eosinophil major basic protein (proMBP), PAPP-A.proMBP, shows only weak surface binding, probably because the adhesion site of PAPP-A is occupied by heparan sulfate, known to be covalently bound to proMBP. This hypothesis was further substantiated by demonstrating that heparinase treatment of PAPP-A.proMBP restores surface binding. We finally propose a model in which IGF bioactivity is regulated by reversible cell surface binding of PAPP-A, which in turn is regulated by proMBP.     

SIRT1 but not its increased expression is essential for lifespan extension in caloric‐restricted mice

The SIRT1 deacetylase is one of the best-studied putative mediators of some of the anti-aging effects of calorie restriction (CR), but its role in CR-dependent lifespan extension has not been demonstrated. We previously found that mice lacking both copies of SIRT1 displayed a shorter median lifespan than wild-type mice on an ad libitum diet. Here, we report that median lifespan extension in CR heterozygote SIRT1^+/− mice was identical (51%) to that observed in wild-type mice, but SIRT1^+/− mice displayed a higher frequency of certain pathologies. Although larger studies in additional genetic backgrounds are needed, these results provide strong initial evidence for the requirement of SIRT1 for the lifespan extension effects of CR, but suggest that its high expression is not required for CR-induced lifespan extension.     

Cell surface detachment of pregnancy-associated plasma protein-A requires the formation of intermolecular proteinase-inhibitor disulfide bonds and glycosaminoglycan covalently bound to the inhibitor

The metzincin metalloproteinase pregnancy-associated plasma protein-A (PAPP-A, pappalysin-1) promotes cell growth by proteolytic cleavage of insulin-like growth factor-binding proteins 4 and 5, causing the release of bound insulin-like growth factors. PAPP-A binds an unknown cell-surface heparan sulfate proteoglycan, suggesting that it controls insulin-like growth factor signaling spatially. In human pregnancy, the majority of PAPP-A circulates as a disulfide-bonded complex with its inhibitor, the proform of eosinophil major basic protein (proMBP). Interestingly, Ser-62 of proMBP is substituted with a glycosaminoglycan (GAG) chain, possibly a heparan sulfate type, and the PAPP-A.proMBP complex is unable to bind to the cell surface. We show here that proMBP detaches surface-bound PAPP-A in a process that depends on the proMBP GAG and also on the formation of intermolecular disulfide bonds between PAPP-A and proMBP. Unlike what was expected, we demonstrate that the GAG of proMBP is not required for PAPP-A.proMBP complex formation and that proMBP residues His-137, Ser-178, Arg-179, and Asn-181 are important for the recognition of PAPP-A. Using a mouse model, we find that the half-life of circulating PAPP-A and proMBP in complex is severalfold higher than both of the uncomplexed proteins, further suggesting that the PAPP-A.proMBP complex is formed at the cell surface in vivo rather than in the circulation. Further supporting this, we show that formation of the PAPP-A.proMBP complex at the cell surface proceeds rapidly compared with the slow rate of complex formation in solution. Because both PAPP-A and proMBP are expressed ubiquitously, this model may be applicable to many tissues in which insulin-like growth factor bioavailability is locally regulated.     

Pregnancy-associated plasma protein-A in pregnant cynomolgus monkeys (Macaca fascicularis): radioimmunoassay, normal levels, effect of RU 486, and preliminary characterization

Pregnancy-associated plasma protein-A (PAPP-A) is a human macromolecular glycoprotein produced by the trophoblast and possibly by the decidua. Its biological function is unknown, but in vitro, PAPP-A has been reported to be an inhibitor of granulocyte elastase. The present study was undertaken to see if pregnant cynomolgus monkeys could be an animal model sufficiently close to the human situation to study the physiology of PAPP-A. An antiserum to pregnant cynomolgus plasma was raised in rabbits. After adsorption with normal monkey plasma, this antiserum was used together with radioiodinated human PAPP-A to develop an heterologous radioimmunoassay for measurements of monkey PAPP-A. On polyacrylamide gel electrophoresis, it was shown that this polyspecific-antiserum bound the same molecular species of radioiodinated human PAPP-A as the available anti-human PAPP-A antiserum. The concentrations of cynomolgus PAPP-A (cPAPP-A) throughout pregnancy follow the same pattern as human PAPP-A (hPAPP-A) with an almost exponential increase up to term. The doubling time of cPAPP-A was similar to that of hPAPP-A. After RU 486-induced abortion or after spontaneous abortion, the levels of cPAPP-A decreased, with an apparent half-life of 2-3 days. Preliminary characterization of cPAPP-A revealed that although cPAPP-A was only immunologically related to hPAPP-A, it was biochemically very similar: they had the same PI and the same molecular weight, and both PAPP-As bound heparin. It is concluded that pregnant cynomolgus monkeys are a good model to study the physiology of PAPP-A.     

Studies on human pregnancy-associated plasma protein A. Purification by affinity chromatography and structural comparisons with alpha 2-macroglobulin.

Pregnancy-associated plasma protein-A (PAPP-A) has been purified by a combination of methods including antibody-affinity chromatography. The resultant protein, obtained in 16% yield from maternal serum, appeared as a single major component on non-denaturing polyacrylamide and SDS/polyacrylamide gel electrophoresis. The protein showed a single component when analysed by isoelectric focusing under denaturing conditions in the presence and absence of reduction and had a pI of 4.34 and 4.42 respectively. These pI values were indistinguishable from those of alpha 2-macroglobulin (alpha 2M). The molecular weight of the PAPP-A polypeptide as shown by SDS/polyacrylamide-gel electrophoresis was 187000, with a minor component of mol.wt. 82500 that was attributed to proteolysis. Since native PAPP-A had a molecular weight on gel chromatography very similar to that of alpha 2M (620000--820000), it was concluded that PAPP-A was a homotetramer. In the absence of reduction, a high-molecular-weight (420000) protomer of PAPP-A was found. It was deduced that PAPP-A, like alpha 2M, is a dinner, whose protomers are composed of disulphide-linked polypeptide chains. It was found that the molecular weight of the PAPP-A polypeptide exceeded that of alpha 2M by 3.3%, but that the total carbohydrate content of PAPP-A exceeded that of alpha 2M by 10% and that its neutral carbohydrate content exceeded that of alpha 2M by between 7.4 and 9.0%. The significance of the estimated molecular weights of alpha 2M (181000) and its major tryptic fragments is discussed in the light of published values. A tryptic fragment alpha 2M (82500 mol.wt.) was apparently the same size as the major tryptic fragment of PAPP-A.