Sensors and effectors of oxidative stress : roles in metabolism and signaling. The organelle dialogue : the role of peroxisomes

Listen to audio

The third lecture (March 13, 2014) focused on the role of the peroxisome in redox signaling, the biology of this organelle, and its dialogue with other organelles or cellular compartments.

In 1966, Christian De Duve discovered the peroxisome, which he isolated from rat liver, and proceeded to characterize it biochemically (De Duve & Baudhuin, 1966). Surrounded by a simple membrane, it is filled with a granular matrix and houses an electron-dense crystalline body. The article reporting this discovery demonstrated the presence in this organelle of oxidases that produce _H2O2, and a peroxidase, catalase, that degrades _H2O2, hence the functional name given to it. Eight years later, in 1974, Christian De Duve was awarded the Nobel Prize for this discovery.

Peroxisomes, organelles devoid of nucleic acid, are revealed by the peroxidase reaction of catalase, which oxidizes 3,3'-diaminobenzidine to form an electron-dense precipitate. Their size, shape and enzymatic content vary considerably from one tissue to another. Their diameter ranges from 0.1 to 0.5 micrometers in adipocytes, where they are mainly located close to lipid vesicles. The peroxisome is therefore a plastic organelle.

Peroxisomes are present in all eukaryotic cells (plants, fungi and animals). Human genetics, by associating diseases with peroxisome damage, has revealed the importance of this organelle in cell life. Genetic analysis of yeast and multicellular fungi has deciphered the mechanisms involved in peroxisome formation and degradation. Peroxisomes are involved in lipid metabolism, through their oxidation of fatty acids, and in oxidative metabolism. They have a number of characteristics that appear to be specific: they can import fully folded and even oligomerized proteins into their matrix, but are unable to fold proteins; they can form de novo. The origin of the peroxisome remains debated: endosymbiotic for some (as are mitochondria or chloroplasts, and as proposed by Christian de Duve (De Duve, 1969), or mitochondrial for others. Two types of data can be added to this debate: on the one hand, as mentioned above, the existence of a de novo peroxisome synthesis pathway, and on the other, the fact that peroxisomes and mitochondria share certain functions (production and degradation of _H2O2, oxidation of long-chain fatty acids, role as signaling platforms in innate immunity..) and compounds essential to their biogenesis (transcription factors of the peroxisome proliferation-activated receptor [PPAR] family, compounds of the fission machinery of their respective precursors).

Peroxisomes are involved in lipid metabolism, synthesis and degradation. They degrade very long-chain fatty acids (VLCFA) by ß-oxidation, and branched-chain fatty acids (BCFA) by α-oxidation. Since 1996, it has been known that they are involved in lipid synthesis. From cholesterol, they synthesize phospholipid ethers that enter into the composition of myelin and bile acids. VLCFAs enter the peroxisome via an ABC-type transporter, BCFAs and myelin and bile acid synthesis intermediates via the PMP70 protein.

The ß-oxidation of long-chain fatty acids has long been attributed exclusively to the mitochondria. Studies in patients with Zellweger syndrome have shown that the profoundly altered biogenesis of peroxisomes (only their membranes remain) is associated with very high blood levels of long-chain and branched-chain fatty acids (Schrader & Yoon, 2007). The peroxisome handles VLCFAs, while the mitochondria handle fatty acids with shorter chain lengths. Mitochondria and peroxisomes therefore cooperate in the degradation of long-chain fatty acids. In energetic terms, the two organelles behave very differently. The mitochondrion takes advantage of the energy released by fatty acid degradation to synthesize ATP by coupling to oxidative phosphorylation, while the peroxisome dissipates this energy in the form of heat.

PPAR-type transcription factors (Lodhi & Semenkovich, 2014) are nuclear receptors; they form heterodimers with the retinoic acid receptor. The genes whose expression they control are involved in adipocyte formation, lipid metabolism, carbohydrate homeostasis, and peroxisome formation itself; in particular, they control the expression of PEX genes, which code for peroxynes, proteins essential to peroxisome biogenesis. The promoter sequences that bind PPAR factors have been defined.

Molecular oxygen is consumed in three cellular compartments: the mitochondria, the peroxisome and the endoplasmic reticulum. Mammalian peroxisomal oxidases all lead to _H2O2 formation by transferring hydrogen from their respective molecular substrates to molecular oxygen; xanthine oxidase additionally leads to ^O2- synthesis (Schrader & Fahimi, 2006). _O2 consumption by the peroxisome is considerable. It is estimated to account for 20% of cellular _O2 consumption. _H2O2 production by the peroxisome accounts for around 35% of total cellular production. Some even suggest that the peroxisome is the major source of _H2O2 and ROS in cells. Among the enzymes of the peroxisome, those involved in the ß-oxidation of fatty acids appear to be the most important in terms of _H2O2 production. Among the oxidases discovered by Christian de Duve are "D" amino acid oxidases; damage to peroxisomal D-amino acid oxidase has recently been reported to be responsible for a form of amyotrophic lateral sclerosis; the mouse model created by inactivation of the corresponding gene shows, like the patients, motor neuron degeneration; D-serine accumulation in the spinal cord has been demonstrated (Sasabe et al., 2012). The NMDA receptor for glutamate, the main neurotransmitter in pyramidal neurons, has D-serine as its co-agonist, and an increase in its concentration would therefore lead to hyperexcitability of motor neurons, resulting in their degeneration.

_H2O2 degradation in the peroxisome relies, in addition to catalase, on Trx-dependent Prx (Prx1 and Prx5), GPx and thiol peroxidase (PMP20). As _O2 concentration rises, peroxisomes increase in size and concentration of _{H2O2-lowering} enzymes. Thus, when CHO cells are incubated in the presence of high _O2 concentration, peroxisome volume doubles, and the concentration of catalase, GPx and SOD increases by a factor of two to four. One pathway of peroxisome formation, via elongation of a pre-existing peroxisome, is dependent on the redox status of the cell (Smith & Aitchison, 2013).

The response of peroxisomes to ROS is less well understood. Transcriptional regulation of the genes encoding peroxisome-associated antioxidant enzymes is also poorly understood. Depending on the gene, it may involve the transcription factor PPAR-α or FOXO3a(Forkhead box class O).

On the other hand, the addressing of proteins to the peroxisome has been extensively studied. A distinction must be made between matrix proteins and membrane proteins (Nagotu et al., 2012; Lodhi & Semenkovich, 2014). Peroxynes are involved in the former. Matrix proteins are recognized by the peroxisome targeting sequence or PTS tag they carry: the PTS1 sequence is a tripeptide located at their C-terminus, while the PTS2 sequence is located in their N-terminal region. Peroxyn 5 recognizes PTS1, and peroxyn 7, PTS2; they take on the corresponding proteins (Smith & Aitchison, 2013). Other peroxynes are also involved in this matrix protein import machinery, including peroxyne 14, which plays a major role. Attached to peroxyne 5 (or 7), the matrix protein enters the peroxisome via a pore composed of other peroxynes. Peroxyn 5 releases its cargo inside the matrix, then returns to the cytoplasm where it either undergoes ubiquitination and is degraded in the proteasome, or enters a new import cycle. Today, over thirty peroxynes are known. Peroxyn 5 activity is dependent on redox status (see below).

The long-prevailing model for the synthesis of peroxisome membrane proteins stipulated that, synthesized on free ribosomes, they were then addressed to the peroxisome (Dimitrov et al., 2013). It was subsequently discovered that some of these proteins transit via the endoplasmic reticulum, a pathway that some now wonder might not be the only one possible. In favor of the existence of a dual pathway, some peroxisome membrane proteins carry a label, which is not fully defined, but is known to be recognized by peroxyn 19, which addresses these proteins to the peroxisome (Lodhi & Semenkovich, 2014).

The physiological role of ROS and their abnormally high peroxisomal production in a pathological context was illustrated by the discussion of an article concerning the role of melanocortin in food intake (Diano et al., 2011). Among the neurons of the arcuate nucleus of the hypothalamus, those that synthesize pro-opiomelanocortin (POMC) and those that synthesize the neuropeptide NPY are involved in eating behavior. Stimulation of the former induces a sensation of satiety, leading to a reduction in food intake, while stimulation of the latter is orexigenic (causes food intake). The activity of both types of neuron is ROS-dependent (observations based on the injection of a compound (honokiol) that traps ROS in the cerebral ventricles). Decreased ROS reduces the activity of POMC neurons and stimulates that of NPY neurons. These changes in activity lead to an increase in food intake. Local hyperproduction of ROS by intra-ventricular injection of _H2O2 increases the activity of POMC neurons, and food intake decreases. ROS were then studied in DIO(diet-induced obesity) mice, which exhibit obesity due to leptin resistance. While ROS and leptin levels normally vary in parallel, in DIO mice leptin levels are very high, but ROS levels in POMC neurons are not as high as expected. The possibility of a failure of ROS induction in DIO mice was examined. Mitochondria and peroxisomes were studied. A comparison of wild-type and DIO mice, both overfed, showed no abnormalities in mitochondrial numbers, whereas peroxisome numbers in POMC neurons increased threefold. This proliferation of peroxisomes could be related to the increase in transcripts coding for PPAR-γ. The effect of PPAR-γ was validated using an antagonist of this transcription factor, GW9662. When this antagonist was injected into the brain ventricles, peroxisome proliferation decreased, ROS increased and food intake fell. When _H2O2_isinjected intra-ventricularly into these mice, ROS increases and food intake decreases. The electrical activity of POMC neurons in DIO mice increases as a result of the decrease in PPAR-γ. The overall results are interpreted as follows: DIO mice exhibit elevated PPAR-γ, which leads to peroxisome proliferation, resulting in decreased ROS, which, by reducing the activity of POMC neurons, leads to increased food intake.

Finally, ROS are classically considered agents of aging. Experimental evidence supporting a direct role for ROS remains tenuous. For some, the weakness of the experimental arguments can be explained by the complexity of the system and the variety of its buffer substances. However, two studies using modulation of catalase expression, one in the nematode C. elegans and the other in mice, have produced results that argue in favor of a role for ROS in aging. The nematode C. elegans possesses three catalases (Petriv & Rachubinski, 2004). Catalase 2, which is expressed in the peroxisome, plays a key role in the reduction of_H2O2_to_H2O. In a mutant animal, in which the expression of this catalase is affected, lifespan falls from 17 to around 13 days. A reduction in the number of eggs laid, also observed during normal aging, is detected. Their peroxisomes are numerous, but clustered in aggregates, indicating that their function is probably impaired. The other study involved the overproduction of catalase in various cellular compartments: nucleus, peroxisome and mitochondria (Schriner et al., 2005). Overproduction of catalase in the peroxisome or nucleus has no effect on longevity. However, when this overproduction occurs in the mitochondria, a modest but significant increase in longevity is observed, accompanied by a protective effect against the deleterious effects of _H2O2.