Histochemistry and Cell Biology

, Volume 131, Issue 4, pp 459–463

Peroxisomes and reactive oxygen species, a lasting challenge


  • Sabine Angermüller
    • Department of Anatomy and Cell Biology IIUniversity of Heidelberg
  • Markus Islinger
    • Department of Anatomy and Cell Biology IIUniversity of Heidelberg
    • Department of Anatomy and Cell Biology IIUniversity of Heidelberg

DOI: 10.1007/s00418-009-0563-7

Cite this article as:
Angermüller, S., Islinger, M. & Völkl, A. Histochem Cell Biol (2009) 131: 459. doi:10.1007/s00418-009-0563-7


Oxidases generating and enzymes scavenging H2O2 predestine peroxisomes (PO) to a pivotal organelle in oxygen metabolism. Catalase, the classical marker enzyme of PO, exhibits both catalatic and peroxidatic activity. The latter is responsible for the staining with 3,3′-diamino-benzidine, which greatly facilitated the visualization of the organelle and promoted further studies on PO. d-Amino acid oxidase catalyzes with strict stereospecificity the oxidative deamination of d-amino acids. The oxidase is significantly more active in the kidney than in liver and more in periportal than pericentral rat hepatocytes. Peroxisomes in these tissues differ in their enzyme activity and protein concentration not only in adjacent cells but even within the same one. Moreover, the enzyme appears preferentially concentrated in the central region of the peroxisomal matrix compartment. Urate oxidase, a cuproprotein catalyzing the oxidation of urate to allantoin, is confined to the peroxisomal core, yet is lacking in human PO. Recent experiments revealed that cores in rat hepatocytes appear in close association with the peroxisomal membrane releasing H2O2 generated by urate oxidase to the surrounding cytoplasma. Xanthine oxidase is exclusively located to cores, oxidizes xanthine thereby generating H2O2 and O2 radicals. The latter are converted to O2 and H2O2 by CuZn superoxide dismutase, which has been shown recently to be a bona fide peroxisomal protein.


PeroxisomeOxidasesHydrogen peroxideReactive oxygen species





Reactive oxygen species


d-amino acid oxidase


Urate oxidase


Xanthine oxidase


CuZn superoxide dismutase

It is less complicated to sum up the pivotal functions of mitochondria or lysosomes than of peroxisomes (PO). The former are notably the powerhouse of a cell providing it with ATP, and contribute to the initiation of apoptosis when this vital task becomes severely impaired. Lysosomes are the rubbish chute of a cell degrading both external metabolites taken up by endocytosis but also cellular constituents like organelles or protein complexes internalized by autophagy. The physiological role of PO appears less striking, though they represent a nearly ubiquitous subcellular organelle. Recalling, however, the fatal heritable peroxisomal disorders like the Zellweger syndrome, the indispensability of PO becomes instantly evident.

Peroxisomes are multi-purpose organelles, integrated both into anabolic as well as catabolic reactions (Wanders and Waterham 2006). They contribute to the biosynthesis of compounds as diverse as ether lipids and bile acids, but also house enzymes termed oxidases, catalyzing the breakdown of a bulk of distinct substrates, e.g. purines, fatty acids, d-amino- and α-hydroxy-acids. A common feature of all the latter reactions is the direct transfer of hydrogen, extracted from the substrate, to O2. This results in the generation of H2O2, which is subsequently converted to H2O + O2 by catalase, the most prominent enzyme of PO. The side by side localization of oxidases and catalase and their enzymatic interactions notably prompted De Duve about four decades ago (De Duve 1965) to coin the term PO for an organelle named microbody before.

The production of H2O2, which amounts to about 35% of total hepatic H2O2 formation (Boveris et al. 1972), and accounts for ~20% of total O2 consumption (Reddy and Mannaerts 1994), and last but not least the generation of reactive oxygen species (ROS) such as O2•, •OH and •NO (for references see Schrader and Fahimi 2006), predestine PO to a source of oxidative stress. The set of antioxidant enzymes degrading such ROS simultaneously points out, however, that they are also powerful scavengers of ROS and defenders against oxidative stress, even when produced outside the organelle. The selection of enzymes compiled in Table 1 reflects the diversity of ROS reactions proceeding in PO.
Table 1

Peroxisomal enzymes generating and degrading ROS





 d-amino oxidase

d-amino acids


 Urate oxidase

Uric acid


 Xanthine oxidase


H2O2; O2

 Acyl-CoA oxidase

Fatty acids


ROS degrading enzymes







 Epoxid hydrolase



Catalase is the most abundant protein and the classical marker enzyme of PO. Its localization to PO has been established in biochemical and electron microscopic studies more than 40 years ago (Baudhuin et al. 1965; Leighton et al. 1968). It is, however, not only confined to the peroxisomal matrix compartment but localizes also outside the organelle, e.g. in the cytoplasm or the nucleus (Yamamoto et al. 1988). Catalase is a bifunctional enzyme, disproportionating on the one hand H2O2 (catalatic function), metabolising on the other a variety of substrates (peroxidatic activity) e.g. ethanol, methanol, phenol or nitrites (Oshino et al. 1973). It was the visualization of PO by means of DAB (3,3′-diamino-benzidine) described four decades ago in three independent papers (Novikoff and Goldfischer 1968; Hirai 1968; Fahimi 1968), which greatly facilitated and promoted further studies on PO. DAB was originally used to localize horseradish peroxidase by EM (Graham and Karnovsky 1966) and indeed, it is the peroxidatic activity of catalase, which is responsible for the staining of PO with DAB as was clearly demonstrated by Fahimi (1969) (see Brief report Fahimi, this issue).

d-Amino acid oxidase (DAAOx) is a FAD-dependent enzyme and catalyzes with strict stereospecificity the oxidative deamination of d-amino acids to α-keto acids and ammonium. It was discovered more than 70 years ago by Krebs (1935), and as one of the first peroxisomal oxidases reported to be localized in PO (De Duve and Baudhuin 1966). The distribution pattern of DAAOx in mammals is unique in manifold respect. DAAOx is expressed more or less in each species of the mammalian taxonomic class, mainly in kidney, liver and brain (Pollegioni et al. 2007). Tissue expression appears to be species- and gender-specific and is characterized by regional differences. Thus, the enzyme is not found in the adult mouse liver (Konno et al. 1997), is significantly more active in the renal proximal tubule epithelium of male than of female rats, and shows lower levels of activity in pericentral than in periportal rat hepatocytes (Angermüller 1989). A regional and cellular heterogeneity of DAAOx activity has been also reported in the CNS of rats, mice and humans (Katagiri et al. 1991; Horiike et al. 1994) with astrocytes and radial glia cells mostly enriched in DAAOx (Moreno et al. 1999). Last but not least, in experiments aimed to detect the flavoprotein either by means of the cerium technique, adopted and refined by Angermüller and Fahimi (1988a) or immunocytochemically using the cognate antibody (Yokota et al. 1987), a remarkably distinct labeling of hepatic and renal PO was observed (Angermüller and Fahimi 1988b; Usuda et al. 1986). Results of a recent study on the rat kidney (Fig. 1), confirm these former findings. PO with heavy and light electron-dense reaction product were identified by cerium labeling not only in adjacent cells (Fig. 1a) but even within the same cell (Fig. 1b). Moreover, the reaction product was preferentially concentrated in the central region of the peroxisomal matrix compartment (Fig. 1a, b). As reported earlier DAAOx protein is indeed confined to this particular sub-compartment (Usuda et al. 1986; Yokota et al. 1987). Recently, peroxisomal subpopulations were purified from rat liver and analyzed by electron microscopy, immunoblotting and mass spectrometry. They were found to differ in their enzyme activity and protein concentration of DAAOx (unpublished observations).
Fig. 1

EM localization of D-AAOx in rat kidney peroxisomes (PO) by means of the cerium method. Sections (80 μm) of perfused rat kidney fixed with 0.25% glutaraldehyde, 2% sucrose in 0.1 Pipes buffer, pH 7.4 and preincubated for 30 min at 37°C without substrate, were incubated for 60 min in freshly prepared medium: 3 mM CeCl3, 100 mM NaN3, 10 mM d-alanine in 0.1 M Tris–maleate buffer, pH 7.8. Sections were postfixed with reduced OsO4, dehydrated in graded ethanol and embedded in Epon 812. Ultrathin sections were examined without counterstaining in a Zeiss transmission electron microscope LEO 906 E. PO in adjacent cells are shown (Fig. 1a) with heavy and light (↓) electron-dense reaction product, and at higher magnification also within the same cell (Fig. 1b). It should also be noted that the reaction product is preferentially concentrated in the central region of the peroxisomal matrix compartment (Fig. 1a, b: P)

Like DAAOx, urate oxidase (UOx) is one of the oxidases, which have led De Duve (1965) to rename the microbodies to PO. Urate oxidase is a cuproprotein and catalyzes the oxidation of urate to allantoin. The tetrameric holoenzyme in most vertebrate species tends to aggregate giving rise to crystalline inclusions in the matrix compartment of PO termed cores or nucleoids. However, this does not apply to PO in fishes and amphibia, which have no cores albeit they exhibit UOx activity (Kramar et al. 1974). No cores at all, yet also no UOx are present e.g. in rat renal PO (Fig. 1), and more importantly also in human hepatic PO, implicating that uric acid can not be degraded in those particles. The absence of urate oxidase in humans and other primates is considered an evolutionary advantage, since the antioxidant properties of urate might have a protective effect against oxidative stress (Ames et al. 1981). Cross sections of cores isolated by centrifugation from rat liver exhibit a polytubular structure with ten small primary tubules surrounding a larger secondary one (Völkl et al. 1988). Enzyme cytochemistry and immunoelectron microscopy have assigned urate oxidase solely to the primary tubules (Angermüller and Fahimi 1986).

In serial sections of rat liver PO, cores do not seem to be clearly positioned; they rather appear scattered in the matrix compartment. Recent experiments, however, using an ultrasensitive luminometric method for assaying H2O2 in real time provided evidence that intact PO are inefficient at degrading H2O2 generated by core-localized UOx (Fritz et al. 2007). This prompted further experiments aimed to re-examine in more detail, whether cores have a preferred position inside PO. Cerium staining and immunoelectron microscopy revealed indeed, that in most isolated core-containing particles the nucleoids appear in close association with the PO membrane with the tubular structures perpendicularly oriented towards it (Fritz et al. 2007). These findings could be corroborated as is exemplified by Fig. 2, which demonstrates a core stained by cerium, in the periphery of the matrix compartment right next to the membrane of the particle. Moreover, it should also be noted that nearly the entire cytoplasmic surface of the PO is also strongly positive, suggesting diffusion of H2O2 from the matrix compartment to the adjacent cytoplasm.
Fig. 2

Ultrastructural localization of urate oxidase (UOx) in peroxisomes of rat liver with the cerium technique. For detection of UOx activity livers were fixed for 5 min by perfusion via the portal vein with 0.25% glutaraldehyde, 2% sucrose in 0.1 Pipes buffer, pH 7.8. Vibratome sections were preincubated for 30 min without the substrate, followed by incubation at 37°C for 60 min with 0.1 mM urate and 3 mM cerium chloride in Pipes buffer, pH 7.8 (Angermüller and Fahimi 1986). The electron-dense reaction product cerium perhydroxide is strictly confined to the core, which is located in the periphery of the matrix compartment right next to the membrane (arrow) of the particle. It should be also noted that electron-dense reaction product (arrow-head) is also seen around the outer surface of the PO suggesting diffusion of H2O2 out of the particle to the adjacent cytoplasm

The observations just outlined have consequences: (i) due to UOx in the cores, rat liver PO might become a direct source of H2O2; (ii) the release of this signaling molecule and potential toxin could interfere with the regulation of cellular functions such as certain signal cascades, cell cycle progression or apoptosis.

Xanthine oxidase (XOx) is a molybdenum-containing dimeric flavoenzyme involved in the catabolism of purines. It exists in two functionally distinct forms: the NAD+-dependent D- or dehydrogenase form and the O-type, which reduces O2 and hence has to be considered an oxidase. Proteolysis, heating or ischemia facilitates the transformation of the dehydrogenase into the oxidase (Engerson et al. 1987). Reducing O2, the latter gives not only rise to H2O2 but also generate toxic O2 superoxide radicals, which are highly reactive, potentially causing severe tissue injuries.

XOx activity was exclusively assayed in the core fraction of purified rat hepatic PO and verified by the cerium method (Angermüller et al. 1987). The end product of the cerium reaction proved to be deposited in the lumen of the primary tubules, which suggests that both enzymes of the purine catabolism share the same core subcompartment.

The generation of H2O2 and O2 radicals by peroxisomal XOx but also of hydroxyl radicals (OH) in the Fenton reaction catalyzed by metal ions which are abundant in PO, makes intraperoxisomal scavenging mechanisms indispensable. CuZn super-oxide dismutase (SOD1) is a powerful superoxide scavenger being able to catalyze the conversion of O2 radicals into O2 and H2O2, the substrate of catalase. SOD1 has been supposed for many years to be a bona fide peroxisomal protein due to cell fraction experiments (Dhaunsi et al. 1992), and indeed this was verified recently by mass spectrometric analysis of highly purified rat hepatic PO (Islinger et al. 2007). Yet, its import into the organelle remained a matter of discussion, since its amino acid sequence lacks the appropriate peroxisomal targeting signals required. Convincing evidence, however, could be provided in a recent study that SOD1 is imported into PO using its physiological interaction partner copper chaperone of SOD1 (CCS) as a shuttle (Islinger et al., submitted).

Summarizing the data presented, and considering further the implication of PO in a variety of physiological and pathological processes (e.g. inflammation, aging, apoptosis), an organelle once termed a fossil proves to be highly vivid, and is still a challenging and demanding subject of scientific research.

Copyright information

© Springer-Verlag 2009