Paul A. Walton

Mechanisms of Cx43 and Cx26 transport to the plasma membrane and gap junction regeneration

Previous reports have suggested that Cx26 exhibits unique intracellular transport pathways en route to the cell surface compared with other members of the connexin family. To directly examine and compare nascent and steady-state delivery of Cx43 and Cx26 to the plasma membrane and gap junction biogenesis we expressed fluorescent-protein-tagged Cx43 and Cx26 in BICR-M1Rk and NRK cells. Static and time-lapse imaging revealed that both connexins were routed through the Golgi apparatus prior to being transported to the cell surface, a process inhibited in the presence of brefeldin A (BFA) or the expression of a dominant-negative form of Sar1 GTPase. During recovery from BFA, time-lapse imaging of nascent connexin Golgi-to-plasma membrane delivery revealed many dynamic post-Golgi carriers (PGCs) originating from the distal side of the Golgi apparatus consisting of heterogeneous vesicles and long, tubular-like extensions. Vesicles and tubular extensions were also observed in HBL-100 cells expressing a human, disease-linked, Golgi-localized Cx26 mutant, D66H-GFP. A diffuse cell surface rim of fluorescent-protein-tagged wild-type connexins was observed prior to the appearance of punctate gap junctions, which suggests that random fusion of PGCs occurred with the plasma membrane followed by lateral diffusion of connexins into clusters. Fluorescence recovery after photobleaching studies revealed that Cx26-YFP was more mobile within gap junction plaques compared with Cx43-GFP. Intriguingly, Cx43-GFP delivery and gap junction regeneration was inhibited by BFA and nocodazole, whereas Cx26-GFP delivery was prevented by BFA but not nocodazole. Collectively, these studies suggest that during gap junction biogenesis two phylogenetically distinct members of the connexin family, Cx43 and Cx26, share common secretory pathways, types of transport intermediates and turnover dynamics but differ in their microtubule-dependence and mobility within the plasma membrane, which might reflect differences in binding to protein scaffolds.

Mg2-dependent phosphatidate phosphohydrolase of rat lung: development of an assay employing a defined chemical substrate which reflects the phosphohydrolase activity measured using membrane-bound substrate.

An assay of pulmonary phosphatidate phosphohydrolase activity has been developed that employs a chemically defined liposome substrate of equimolar phosphatidate and phosphatidylcholine. Enzyme assays employing this substrate resolved two distinct activities based upon their requirements for Mg2+. Assays were performed in the presence and absence of 2 mM MgCl2 and the Mg2+-dependent phosphatidate phosphohydrolase activity calculated by difference. The Mg2+-independent phosphatase activity resembled that found using aqueous dispersions of phosphatidate (PAaq). Approximately 90% of the Mg2+-dependent phosphatidate phosphohydrolase activity was recovered in the cytosol and the remainder was associated with the microsomal fraction. The Mg2+-dependent phosphatidate phosphohydrolase activity has kinetic parameters of Km = 55 microM, Vmax = 1.6 nmol/min/mg protein for the microsomal fraction, and Km = 215 microM, Vmax = 6.8 nmol/min/mg protein for the cytosolic fraction. These parameters resembled those found using the microsomal membrane-bound (PAmb) substrate. In addition, the pH optima and sensitivity to detergents and thermal inactivation are equal to those for the PAmb-dependent phosphatidate phosphohydrolase activity. In the course of these studies the microsomal and cytosolic activities were qualitatively equal, indicative of a single enzyme in two subcellular locations. In conclusion, the assay of Mg2+-dependent phosphatidate phosphohydrolase activity measured using equimolar phosphatidate and phosphatidylcholine liposomes is equivalent to that activity previously described using microsomal membrane-bound substrate. However, the chemically-defined system provides a more simplified starting point for further studies on this important enzyme.

Restoration of peroxisomal catalase import in a model of human cellular aging

Peroxisomes play an important role in human cellular metabolism by housing enzymes involved in a number of essential biochemical pathways. Many of these enzymes are oxidases that transfer hydrogen atoms to molecular oxygen forming hydrogen peroxide. The organelle also contains catalase, which readily decomposes the hydrogen peroxide, a potentially damaging oxidant. Previous work has demonstrated that aging compromises peroxisomal protein import with catalase being particularly affected. The resultant imbalance in the relative ratio of oxidases to catalase was seen as a potential contributor to cellular oxidative stress and aging. Here we report that altering the peroxisomal targeting signal of catalase to the more effective serine‐lysine‐leucine (SKL) sequence results in a catalase molecule that more strongly interacts with its receptor and is more efficiently imported in both in vitro and in vivo assays. Furthermore, catalase‐SKL monomers expressed in cells interact with endogenous catalase subunits resulting in altered trafficking of the latter molecules. A dramatic reduction in cellular hydrogen peroxide levels accompanies this increased peroxisomal import of catalase. Finally, we show that catalase‐SKL stably expressed in cells by retroviral‐mediated transduction repolarizes mitochondria and reduces the number of senescent cells in a population. These results demonstrate the utility of a catalase‐SKL therapy for the restoration of a normal oxidative state in aging cells.

Progeric effects of catalase inactivation in human cells

Peroxisomes generate hydrogen peroxide, a reactive oxygen species, as part of their normal metabolism. A number of pathological situations exist in which the organelles capacity to degrade the potentially toxic oxidant is compromised. It is the peroxidase, catalase, which largely determines the functional antioxidant capacity of the organelle, and it is this enzyme that is affected in aging, in certain diseases, and in response to exposure to specific chemical agents. To more tightly control the enzymatic activity of peroxisomal catalase and carefully document the effects of its impaired action on human cells, we employed the inhibitor 3-amino-1,2,4-triazole. We show that by chronically reducing catalase activity to approximately 38% of normal, cells respond in a dramatic manner, displaying a cascade of accelerated aging reactions. Hydrogen peroxide and related reactive oxygen species are produced, protein and DNA are oxidatively damaged, import into peroxisomes and organelle biogenesis is corrupted, and matrix metalloproteinases are hyper-secreted from cells. In addition, mitochondria are functionally impaired, losing their ability to maintain a membrane potential and synthesize reactive oxygen species themselves. These latter results suggest an important redox-regulated connection between the two organelle systems, a topic of considerable interest for future study.

Effects of peroxisomal catalase inhibition on mitochondrial function.

Peroxisomes produce hydrogen peroxide as a metabolic by-product of their many oxidase enzymes, but contain catalase that breaks down hydrogen peroxide in order to maintain the organelle’s oxidative balance. It has been previously demonstrated that, as cells age, catalase is increasingly absent from the peroxisome, and resides instead as an unimported tetrameric molecule in the cell cytosol; an alteration that is coincident with increased cellular hydrogen peroxide levels. As this process begins in middle-passage cells, we sought to determine whether peroxisomal hydrogen peroxide could contribute to the oxidative damage observed in mitochondria in late-passage cells. Early-passage human fibroblasts (Hs27) treated with aminotriazole (3-AT), an irreversible catalase inhibitor, demonstrated decreased catalase activity, increased levels of cellular hydrogen peroxide, protein carbonyls, and peroxisomal numbers. This treatment increased mitochondrial reactive oxygen species levels, and decreased the mitochondrial aconitase activity by ∼85% within 24 h. In addition, mitochondria from 3-AT treated cells show a decrease in inner membrane potential. These results demonstrate that peroxisome-derived oxidative imbalance may rapidly impair mitochondrial function, and considering that peroxisomal oxidative imbalance begins to occur in middle-passage cells, supports the hypothesis that peroxisomal oxidant release occurs upstream of, and contributes to, the mitochondrial damage observed in aging cells.

Hypocatalasemic Fibroblasts Accumulate Hydrogen Peroxide and Display Age-Associated Pathologies

Human epidemiological studies point to an association of hypocatalasemia and an increased risk of age‐related disease. Unfortunately, the cellular and molecular manifestations of hypocatalasemia are only poorly understood. In this analysis, we have extensively characterized hypocatalasemic human fibroblasts and report that they amass hydrogen peroxide and are oxidatively damaged. Protein and DNA alike are affected, as are functioning and biogenesis of peroxisomes – the subcellular organelles which normally house catalase. Despite these pathologies and their relative inability to grow, the cells do not appear to be intrinsically senescent. With the goal of restoring oxidative balance and perhaps reversing some of the accumulated damage to critical cellular components, we transduced hypocatalasemic fibroblasts with a form of catalase specifically designed to efficiently traffic to peroxisomes. We show the strategy is extremely effective, with dramatic reductions seen in cellular hydrogen peroxide levels. Future longitudinal studies aimed at examining the effects of a more continuous and long‐term protein therapy may now commence.

The role of Mg2+-dependent phosphatidate phosphohydrolase in pulmonary glycerolipid biosynthesis

Rat lung microsomes washed with increasing concentrations of NaCl show a displacement of protein from microsomes to the wash supernatant. Among the proteins removed from the microsomal surface was the Mg2+-dependent phosphatidate phosphohydrolase, while the Mg2+-independent activity remained associated with the microsomes. The Mg2+-dependent activity could be quantitatively assayed in the wash supernatant. Microsomes washed with increasing concentrations of NaCl showed a progressive impairment in the synthesis of labelled neutral lipid and phosphatidylcholine from [14C]glycerol 3-phosphate with a concomitant increase in the labelling of phosphatidic acid. The impairment was sigmoidal and correlated highly with the decrease in Mg2+-dependent phosphatidate phosphohydrolase activity. When Mg2+-dependent phosphatidate phosphohydrolase from wash supernatant was incubated with microsomes previously washed with high salt concentrations, the labelling of neutral lipid and phosphatidylcholine was returned to control levels. Labelling of neutral lipids and phosphatidylcholine could be restored upon addition of a cytosolic Mg2+-dependent phosphatidate phosphohydrolase isolated by gel filtration. Mg2+-independent phosphatidate phosphohydrolase isolated from cytosol was incapable of restoring the labelling of neutral lipids and phosphatidylcholine. These findings confirm that the Mg2+-dependent phosphatidate phosphohydrolase of rat lung is involved in pulmonary glycerolipid biosynthesis. The role of the Mg2+-independent phosphatidate phosphohydrolase activity remains unknown.

Requirement for microtubules and dynein motors in the earliest stages of peroxisome biogenesis.

Our aim was to determine the role of microtubules in the biogenesis of peroxisomes. Fusion experiments between human PEX16‐ and PEX1‐mutant cells in the presence of nocodazol implied that microtubules were not required for import of proteins into the peroxisomal matrix after cell fusion complementation. We further studied the importance of microtubules in the early stages of peroxisome biogenesis following the microinjection complementation of PEX16‐mutant cells. In the absence of nocodazol, nuclear microinjection of plasmids expressing EGFP‐SKL and Pex16p in PEX16‐mutant cells resulted in the accumulation of EGFP‐SKL into newly formed peroxisomes. However, pretreatment of the cells with nocodazol, prior to microinjection, resulted in the inhibition of complementation of the PEX16 mutant and the cytosolic location of the EGFP‐SKL. In addition, coexpression of a dominant‐negative CC1 subunit of the dynein/dynactin motor complex resulted in the inability to complement PEX16‐mutant cells. Both of these treatments resulted in the cytosolic localization of expressed Pex16p. Our results demonstrate that the formation of peroxisomes via the preperoxisomal compartment is dependent upon microtubules and minus‐end‐directed motor proteins and that the inhibition described above occurs at a step that precedes the association of Pex16p with the vesicles that would otherwise become the peroxisomes.

Protein structure and import into the peroxisomal matrix.

Proteins destined for the peroxisomal matrix are synthesized in the cytosol, and imported post‐translationally. It has been previously demonstrated that stably folded proteins are substrates for peroxisomal import. Mammalian peroxisomes do not contain endogenous chaperone molecules. Therefore, it is possible that proteins are required to fold into their stable, tertiary conformation in order to be imported into the peroxisome. These investigations were undertaken to determine whether proteins rendered incapable of folding were also substrates for import into peroxisomes. Reduction of albumin resulted in a less compact tertiary structure as measured by analytical centrifugation. Microinjection of unfolded albumin molecules bearing the PTS1 targeting signal resulted in their import into peroxisomes. Kinetic analysis indicated that native and unfolded molecules were imported into peroxisomes at comparable rates. While import was unaffected by treatment with cycloheximide, hsc70 molecules were observed to be imported along with the unfolded albumin molecules. These results indicate that proteins, which are incapable of assuming their native conformation, are substrates for peroxisomal import. When combined with previous observations demonstrating the import of stably folded proteins, these results support the model that tertiary structure has no effect on protein import into the peroxisomal matrix.

Amyloid-beta neuroprotection mediated by a targeted antioxidant

Amyloid-beta (Aβ)-induced neurotoxicity is a major contributor to the pathologies associated with Alzheimers disease (AD). The formation of reactive oxygen species (ROS), an early response induced by the peptide and oligomeric derivatives of Aβ, plays a significant role in effecting cellular pathogenesis. Here we employ particularly toxic forms of Aβ with cultured primary cortical/hippocampal neurons to elicit ROS and drive cellular dysfunction. To prevent and even reverse such effects, we utilized a cell-penetrating, peroxisome-targeted, protein biologic – called CAT-SKL. We show the recombinant enzyme enters neurons, reverses Aβ-induced oxidative stress, and increases cell viability. Dramatic restorative effects on damaged neuronal processes were also observed. In addition, we used DNA microarrays to determine Aβs effects on gene expression in neurons, as well as the ability of CAT-SKL to modify such Aβ-induced expression profiles. Our results suggest that CAT-SKL, a targeted antioxidant, may represent a new therapeutic approach for treatment of disorders, like Alzheimers disease, that are driven through oxidative stress. Preclinical testing is ongoing.