Differential distribution of the 5-alpha-reductase in the central nervous system of the rat and the mouse: are the white matter structures of the brain target tissue for testosterone action?

In the brain of several animal species testosterone is converted into a series of 5-alpha-reduced metabolites, and especially into 17-beta-hydroxy-5-alpha-androstan-3-one (DHT), by the action of the enzyme 5-alpha-reductase. The formation of DHT has never been evaluated in the white matter structures of the brain, which are composed mainly of myelinated axons. The experiments here described were performed in order to study, in the rat and the mouse, the DHT forming activity of several white matter structures, in comparison with that of the cerebral cortex and of the hypothalamus. Two sampling techniques were used in the rat: microdissection under a stereo-microscope from frozen brain sections of fragments of corpus callosum, optic chiasm and cerebral cortex; fresh tissue macrodissection of subcortical white matter, cerebral cortex and hypothalamus. Only macrodissection was used in the mice. The data show that, independently from the sampling technique used, there are considerable quantitative differences in the distribution pattern of the 5-alpha-reductase activity within different brain structures. Both in the rat and in the mouse, the enzyme appears to be present in higher concentrations in the white matter structures, than in the cerebral cortex and in the hypothalamus. The present results clearly show that the subcortical white matter and the corpus callosum are at least three times as potent as the cerebral cortex in converting testosterone into DHT. An even higher 5-alpha-reductase activity has been found in the optic chiasm. Further work is needed in order to understand the possible physiological role of DHT formation in the white matter structures.     

Testosterone metabolism in peripheral nerves: presence of the 5 alpha-reductase-3 alpha-hydroxysteroid-dehydrogenase enzymatic system in the sciatic nerve of adult and aged rats

Previous reports from this laboratory indicate that the 5 alpha-reductase, the enzyme which converts testosterone into its "active" metabolite 5 alpha-androstan-17 beta-ol-3-one (dihydrotestosterone, DHT) is highly concentrated in the white matter structures of the CNS, which are mainly composed of myelinated fibers. No studies have been performed up to now, in order to evaluate the possible presence of the 5 alpha-reductase activity in peripheral myelinated nerves. To this purpose the 5 alpha-reductase activity has been evaluated in the sciatic nerve of the rat and compared to that present in the cerebral cortex and in the subcortical white matter, a central structure mainly composed of myelinated fibers. The study has been performed in normal adult male rats (60-90-day-old) and in aged (20-month-old) animals. The data obtained in 60-90-day-old animals indicate the presence of an active metabolism of testosterone at the level of the sciatic nerve. In this structure, testosterone is actively transformed into DHT and 5 alpha-androstan-3 alpha, 17 beta-diol (3 alpha-diol); in the sciatic nerve, the formation of DHT is equal to that found in the subcortical white matter and higher than that found in the cerebral cortex. Moreover, at variance with what happens in CNS structures, where 3 alpha-diol is produced only in small amounts, in the sciatic nerve this metabolite is produced in amounts similar to those of DHT. The study in aged rats has shown that in the sciatic nerve, the formation of DHT and particularly that of 3 alpha-diol are much lower than in younger animals. No age-related variations in the 5 alpha-reductase activity in the cerebral cortex and in the subcortical white matter have been observed.     

Kinetic properties of the 5 alpha-reductase of testosterone in the purified myelin, in the subcortical white matter and in the cerebral cortex of the male rat brain

The 5 alpha-reductase, the enzyme which converts testosterone into dihydrotestosterone (DHT), is present in several CNS structures of the rat. Recent reports from this laboratory indicate that the subcortical white matter and the myelin possess a 5 alpha-reductase activity several times higher than that present in the cerebral cortex. Moreover, previous ontogenetic observations indicate that in all cerebral tissues examined (including the myelin) the 5 alpha-reductase has a higher activity in immature animals. This study was performed in order to verify whether the differences in the 5 alpha-reductase activity on the various brain components might be due to the presence of different concentrations of the same enzyme or to different isoenzymes. To this purpose, the kinetic properties Km and Vmax were measured in the purified myelin as well as in homogenates of the subcortical white matter and of the cerebral cortex, obtained from the brain of adult (60-90-day-old), immature (23-day-old), and aged (greater than 20-month-old) male rats. The results indicate that the enzymes present in the myelin, in the subcortical white matter and in the cerebral cortex of adult male rats possess a very similar apparent Km (1.93 +/- 0.2, 2.72 +/- 0.73 and 3.83 +/- 0.49 microM respectively). On the contrary, the Vmax values obtained in the myelin (34.40 +/- 5.54), in the white matter (19.57 +/- 2.36) and in the cerebral cortex (6.47 +/- 1.03 ng/h/mg protein) of adult animals have been found to be consistently different. Very similar Km values were found in the myelin obtained from the brain of immature and very old rats (2.14 +/- 0.11 and 3.39 +/- 0.75 microM respectively). The Vmax measured in the myelin purified from the immature rat brain (62.25 +/- 4.52) showed a value which was much higher than that found in the myelin of adult animals (34.40 +/- 5.54); a Vmax (34.31 +/- 9.41) almost identical to that of adult animals was found in the myelin prepared from the brain of aged rats.     

Testosterone 5 alpha-reductase activity in the rat brain is highly concentrated in white matter structures and in purified myelin sheaths of axons

Previous results obtained in this laboratory indicate that in the rat brain the 5 alpha-reductase, the enzymatic activity involved in metabolizing testosterone into 5 alpha-androstan-17 beta-ol-3-one (dihydrotestosterone), is particularly concentrated in the white matter. In the present experiments, this enzymatic activity was studied in the following white matter structures, which were microdissected using the punch technique of Palkovits: anterior commissure (CA), fornix (FX), habenulo-interpeduncular tract (HP), corpus callosum (CC), stria medullaris (SM), optic chiasm (CO), fimbria of the hippocampus (FI), cerebral peduncle (PC), pontine fibers (FP), cerebellar medulla (CMD) and corticospinal tract (TCS). Moreover brain myelin was isolated and purified by sucrose density gradient ultracentrifugation. The results obtained confirm that, in the rat brain, the enzymes involved in testosterone 5 alpha-reduction are preferentially localized in the white matter. However, clearcut differences in the metabolic activity exist between the different structures examined so far. DHT formation increases rostro-caudally, so that the highest activity has been recorded in the white matter structures punched at the level of pons (FP), medulla oblungata (TCS) and cerebellum (CMD). The high metabolic activity associated with the white matter structures appears to be linked to the presence of myelin, since the specific activity of the enzyme is particularly elevated in purified preparations of myelin sheaths.     

Testosterone 5 alpha-reductase in discrete hypothalamic nuclear areas in the rat: effect of castration

The conversion of testosterone into 5 alpha-dihydrotestosterone (DHT) has been studied in different hypothalamic nuclear areas and in the superficial layers of the cerebral cortex of normal and castrated male rats. The tissue fragments utilized in each incubation have been punched from frozen brain sections utilizing calibrated needles. Castration has been performed 12 (short term) and 180 (long term) days before sacrifice. The nuclear areas studied include: the medial preoptic nucleus (MPN), the lateral preoptic nucleus (LPN), the anterior hypothalamic nucleus (AHN), the lateral hypothalamic nucleus (LHN), the posterior hypothalamic nucleus (PHN), the nucleus ventromedialis (HVM), the arcuate nucleus (AR), the median eminence (ME), the nucleus paraventricularis (HPV), the supraoptic nucleus (SO) and the suprachiasmatic nucleus (SC). The possible effect of castration on the 5 alpha-reductase, were assessed in the MPN,LPN,AHN,LHN,PHN and in the cerebral cortex. The results indicate that, in the male rat: 1) the lateral preoptic(LPN) and the lateral hypothalamic nuclei(LHN) possess a 5 alpha-reductase activity higher than that present in the cerebral cortex and in the other hypothalamic nuclei considered; 2)the suprachiasmatic nucleus (SC) apparently possesses a testosterone metabolizing activity lower than that found in any other nervous structures studied so far; 3) castration does not seem to influence the 5 alpha-reductase activity either in the hypothalamic nuclear structures considered or in the cerebral cortex.     

Neural uptake and metabolism of testosterone and dihydrotestosterone in the guinea pig

Neural tissues from adult, castrated male guinea pigs were examined for their capability to concentrate and metabolize [1,2-³H]testosterone (T) and [1,2-³H]dihydrotestosterone (DHT), both in vitro and in vivo. In vitro uptake of DHT and T was greater in the hypothalamus and anterior pituitary than in the cerebral cortex. With DHT as the substrate, the 800×g particulate concentration of this compound was highest in the hypothalamus, although in this tissue, particulate concentration was less than that of the cytoplasm. In the cerebral cortex 5α-androstane-3,17-dione was the most abundant metabolite, whereas 5α-androstane-3,17-dione, 5α-androstane-3α,17β-diol, and 5α-androstane-3β,17β-diol were all present in equivalent amounts in the hypothalamus and pituitary. Incubation with T resulted in the formation of DHT, 4-androstane-3,17-dione, and a compound with the mobility of 5α-(or 5β-)androstane-3,17-7-dione. The radioactivity associated with DHT was the most prevalent in the pituitary (1.3%), and least prevalent in the cerebral cortex (0.6%), and in all cases cytoplasmic concentration of this compound exceeded the concentration in the particulate fraction. Recrystallization failed to confirm the presence of estradiol-17β. Although there were no apparent tissue differences in the uptake of DHT or T 1 hour after their injection, intracellular distribution varied. In all tissues examined, that percentage of total radioactivity attributable to DHT was greatest in the 800×g particulate preparations, particularly in the hypothalamus. Thus neural tissues in the guinea pig, as in other species, exhibit differential uptake and metabolism of androgen through which physiological and behavioral effects may be mediated.     

THE DISTRIBUTION AND PROPERTIES OF THE NONSPECIFIC N-METHYLTRANSFERASE IN BRAIN

Abstract— The distribution and properties of a nonspecific N -methyltransferase in the rat brain are described. The enzyme N -methylates tryptamine and N -methyltryptamine as well as β-phenylethylamine, phenylethanolamine, tyramine and octopamine. The enzyme exhibits a pH optimum of 7·9 with phosphate buffer and has a K_m for tryptamine of 28 μM. There are potent inhibitors to the enzyme that can be removed by dialysis. Enzymatic activity is present in the brains of a number of species including man, rat, mouse, guinea-pig and frog. Its activity is unevenly distributed in the brain with the highest activity in the cerebral cortex and striaturn of the rat and in the subcortical white matter in man. Studies of its subcellular distribution indicate that most of the N -methylation activity is released into the soluble fraction. Enzymatic activity is also present in a number of peripheral tissues of the rat.     

Effect of ethanol on androgen receptors in the anterior pituitary, hypothalamus and brain cortex in rats

The purpose of this study was to investigate ethanol-induced changes in androgen receptor sites in the anterior pituitary, hypothalamus, and brain cortex. Young adult male King-Holtzman rats were fed for 5 months a nutritionally complete liquid diet, with ethanol or isocaloric sucrose constituting 36% of the total calories. Androgen receptor sites were measured by sucrose density gradient and charcoal assay using tritiated dihydrotestosterone (DHT). Scatchard plot analysis of the data revealed that apparent dissociation constants of DHT-receptor complex for the anterior pituitary, hypothalamus, and brain cortex from alcohol-fed animals were estimated to be 0.7 +/- 0.13, 0.6 +/- 0.16 and 0.9 +/- 0.15 nM, respectively. These values are identical to those of their isocaloric controls. The concentrations of cytosol androgen receptors of the pituitary, hypothalamus, and brain cortex from alcohol-fed rats were 8.0 +/- 1.2, 6.2 +/- 1.0 and 4.9 +/- 0.7 fmol/mg protein, respectively. This represents about a 34, 24, and 22% reduction when compared to the values of the isocaloric control animals. In contrast to control rats, neither castration nor androgen or LHRH replacement to castrated alcohol-fed rats altered an alcohol-induced reduction of androgen receptor contents. Serum LH and testosterone levels were significantly decreased in alcohol-fed rats but these hormone levels were increased by administration of LHRH or norepinephrine. Such reduction of androgen receptors, serum LH and testosterone, but enhancement of these hormone levels by treatment with neurohormone and neurotransmitter in these animals suggests that ethanol exerts an adverse effect on the hypothalamic-pituitary unit and the neurotransmitter-hypothalamic hormone relationship, resulting in impairment of the androgen-induced sexual events and a suppression of the pituitary gonadotropin secretion.     

Regional and subcellular distributions of brain neurotensin.

The regional and subcellular distribution of neurotensin were determined using a newly developed radioimmunoassay for this central nervous system tridecapeptide. Neurotensin immunoreactivity in calf brain is high in the hypothalamus and basal ganglia, unevenly distributed through the cerebral cortex, and low in cerebellar cortex and cerebral white matter. Subcellular fractionation of rat hypothalamus reveals a strong association of neurotensin immunoreactivity with synaptosomal and microsomal fractions. These data, taken along with previously described high affinity selective brain membrane receptor binding, are consistent with a neurotransmitter candidate role for neurotensin in the brain.     

Steroid control of sexual behavior and brain aromatase in the dove: effects of nonaromatizable androgens, methyltrienolone (R1881), and 5 alpha-dihydrotestosterone

This study examines the effects of nonaromatizable androgens, methyltrienolone (R1881) and 5 alpha-dihydrotestosterone (DHT) on aggressive courtship and vocal behavior in the male ring dove. Since androgens may influence behavior by increasing the formation of estrogen in the brain, the effects of R1881 and DHT on brain aromatase activity were also studied using an in vitro microassay. Under conditions in which testosterone induced aggressive courtship patterns, the nonaromatizable androgens were ineffective. But DHT and R1881 induced vocal behavior with equal efficiency, indicating that androgens can influence mechanisms of vocal behavior without conversion to estrogens. The behavioral effectiveness of both hormones was reduced (approximately 50%) when the period between castration and treatment was doubled. Testosterone propionate increased formation of E2 from 3H-testosterone in both the preoptic (POA) and anterior hypothalamic areas. Neither of the nonaromatizable androgens affected POA aromatase activity. The results suggest that only the aromatizable androgen, testosterone, which is also required specifically for male courtship, increases preoptic formation of estrogen.     

Cytotoxic and complement-binding activity of rabbit antisera against the cortex and cerebral white matter of rats and humans]

The cytotoxicity and complement-fixation activity of rabbit antisera against rat and human brain cortex and white matter was tested against mouse and rat thymocytes and bone marrow cells. The cytotoxic test proved to be more sensitive and accurate. The cytotoxins to rodent thymocytes were found in the antisera against human brain cortex only. At the same time cytotoxic antibodies were revealed both in the antisera rat brain cortex and white matter; but the former contained much more cytotoxic antibodies than the letter. After absorption with the same antigen the antisera against rat brain cortex lost their cytotoxic effect, but retained it in case of absorption with the white matter.     

Identification of Cortexin: A Novel, Neuron-Specific, 82-Residue Membrane Protein Enriched in Rodent Cerebral Cortex

Abstract: Nucleotide sequence analysis of a cDNA clone of a rat cortex-enriched mRNA identifies a novel integral membrane protein of 82 amino acids. The encoded protein is named cortexin to reflect its enriched expression in cortex. The amino acid sequence of rat cortexin and its mouse homologue are nearly identical (98% similarity), and both contain a conserved single membrane-spanning region in the middle of each sequence. Northern blot analysis shows that cortexin mRNA is brain-specific, cortexenriched, and present at significant levels in fetal brain, with peak expression in postnatal rodent brain. In situ hybridization studies detect cortexin mRNA primarily in neurons of rodent cerebral cortex, but not in cells of the hindbrain or white matter regions. The function of cortexin may be particularly important to neurons of both the developing and adult cerebral cortex.     

Characterization of the androgen receptors in the hypothalamus, preoptic area and brain cortex of the rat.

The specific androgen receptors for testosterone (T) (1) and 5alpha-dihydrotestosterone (DHT) in the cytosol fraction of the hypothalamus, preoptic area and brain cortex of the rat have been characterized using electrophoresis and isoelectric focusing in polyacrylamide gels. After labeling of the cytosol fractions in vivo and in vitro we were able to demonstrate androgen-receptor complexes moving with an electrophoretic mobility (R(f) of 0.5 in 3.25% acrylamide gels containing 0.5% agarose and 10% glycerol. Polyacrylamide gel electrophoresis was used as a quantitative assay for androgen receptors in the tissues. The hypothalamus, preoptic area and brain cortex were found to possess a single class of high affinity binding sites for androgens and the dissociation constants (K(D) were estimated to be 3.4, 4.3 and 2.6 X 10 (-10M) respectively. The binding capacities were 3.7 (hypothalamus), 3.5 (preoptic area) and 1.8 X 10 (-15) (brain cortex) moles of high affinity binding sites per mg protein. Like other androgen-receptor complexes, the testosterone-receptor complexes of the hypothalamus, preoptic area and brain cortex were temperature labile, sulfhydryl dependent and revealed a very slow rate of dissociation at o degrees C (t1/2 greater than 36 hr). The receptors in all the tissues had an isoelectric point of 5.8. The steroid specificity of the cytoplasmic androgen receptors was tested in vitro by the competing efficiency of different unlabeled steroids for (3H)-testosterone binding. In the three tissues in investigation the following order of affinity was found: DHT greater than T greater than Cyproterone acetate greater than progesterone greater than androstenedione greater than 17beta-estradiol. Cortisol did not effect androgen binding significantly. Thus, the physiochemical characteristics of the cytoplasmic androgen receptors of the hypothalamus, preoptic area and brain cortex are very similar, if not identical, to those of the androgen receptors described in the anterior pituitary, ventral prostate, epididymis and testis.     

Developmental and Regional Studies of the Metabolism of Inositol 1,4,5-Trisphosphate in Rat Brain

Coupling of CNS receptors to phosphoinositide turnover has previously been found to vary with both age and brain region. To determine whether the metabolism of the second messenger inositol 1,4,5-trisphosphate also displays such variations, activities of inositol 1,4,5-trisphosphate 5'-phosphatase and 3'-kinase were measured in developing rat cerebral cortex and adult rat brain regions. The 5'-phosphatase activity was relatively high at birth (approximately 50% of adult values) and increased to adult levels by 2 weeks postnatal. In contrast, the 3'-kinase activity was low at birth and reached approximately 50% of adult levels by 2 weeks postnatal. In the adult rat, activities of the 3'-kinase were comparable in the cerebral cortex, hippocampus, and cerebellum, whereas much lower activities were found in hypothalamus and pons/medulla. The 5'-phosphatase activities were similar in cerebral cortex, hippocampus, hypothalamus, and pons/medulla, whereas 5- to 10-fold higher activity was present in the cerebellum. The cerebellum is estimated to contain 50-60% of the total inositol 1,4,5-trisphosphate 5'-phosphatase activity present in whole adult rat brain. The localization of the enriched 5'-phosphatase activity within the cerebellum was examined. Application of a histochemical lead-trapping technique for phosphatase indicated a concentration of inositol 1,4,5-trisphosphate 5'-phosphatase activity in the cerebellar molecular layer. Further support for this conclusion was obtained from studies of Purkinje cell-deficient mutant mice, in which a marked decrement of cerebellar 5'-phosphatase was observed. These results suggest that the metabolic fate of inositol 1,4,5-trisphosphate depends on both brain region and stage of development.     

Response of the epididymis, ductus deferens & accessory glands of the castrated prepubertal rhesus monkey to exogenous administration of testosterone or 5alpha-dihydrotestosterone.

The response of the epididymis, ductus deferens, and accessory glands of the castrated prepubertal rhesus monkey to exogenous administration of testosterone or 5alpha-dihydrotestosterone (DHT) was investigated. 200 or 800 mcg of either steroid/day were administered for 60 days beginning on the day after castration. Castration caused a marked regression of the weight of and secretory function of the reproductive organs; testosterone/DHT stimulated their growth and secretory activity which were maintained at the level of the controls. The weight of the caput epididymides however, was unaffected by testosterone but was stimulated by DHT. DHT caused a greater stimulation of the growth and secretory activity of the reproductive organs than testosterone and also caused a hyperstimulation of secretion by the seminal vesicles. The data, analyzed statistically, show that the accessory organs of the prepubertal rhesus monkey are affected by castration and vary in their response to stimulation by exogenous androgens.     

Effect of antibodies to rat and human cerebral cortex and white matter on heterologous T- and B-cells]

A study was made of the cytotoxic and complement-fixation activity of the antisera to the cortex and white matter of the rat and human brain upon the mouse and rat thymus and bone marrow cells. The cytotoxicity test proved to be more sensitive and precise. Cytotoxins to rodent thymocytes were revealed on ly in the antisera against the human brain cortex; at the same time they were revealed both in the antisera against the cortex and against the white matter of the rat brain (much more was found in the former). The sera against the rat brain cortex lost their cytotoxicity after the exhaustion with the same antigen, but retained it when the exhaustion was carried out by the white matter.     

Androgen metabolism by rat epididymis. 3. Effect of castration and anti-androgens.

The in vivo and in vitro metabolism of 3H-testosterone by rat epididymis and the changes in epididymal weight have been studied after castration and treatment with anti-androgens. The utilization of 3H-testosterone was greatly reduced after castration as was the formation of 5alpha-reduced 17 beta-hydroxy metabolites. The formation of the 17 -keto metabolites was unaffected. Castration had no effect on the ratio between water and ether soluble radioactivity. Administration of testosterone propionate, necessary for giving normal stimulated prostate weight (150 mug/day), restored the metabolism of testosterone to approximately normal values. Estradiol benzoate and progesterone inhibited metabolism of testosterone in vitro and greatly reduced the formation of DHT (17 beta-hydroxy-5alpha-androstan-3-one) and 3 alpha-diol(5 alpha-androstane-3 alpha-17 beta-diol) by experiments both in vivo and in vitro. No effect of cyproterone acetate could be demonstrated on either the in vitro or in vivo metabolism of testosterone. Castration for 14 days reduced the epididymal weight to about 30% of that found in intact animals. Administration of testosterone propionate restored the epididymal weight to about 80% of normal. Estradiol benzoate and cyproterone acetate given to intact rats led to a decrease in the epididymal weight. Progesterone had no such effect. In 14 days castrated rats receiving testosterone propionate all three anti-androgens reduced the weight of the epididymis. In conclusion, our results show that the metabolic conversion of testosterone in epididymis to DHT and 3 alpha-diol is dramatically dependent on the hormonal status of the animal; castration or treatment with anti-androgens causes a reduced formation of the "active" androgens whilst testosterone replacement treatment restores the metabolism of testosterone to normal.     

Effect of beta-endorphin on catecholamine levels in the rat hypothalamus and cerebral cortex

The present paper deals with the effect of beta-endorphin on catecholamine content in the hypothalamus and cerebral cortex of male rats. beta-endorphin was found to decrease catecholamine content in the rat brain, with the degree of reduction depending on the brain topography and the time following the peptide administration. 5 min later no changes in catecholamine content were observed either in the hypothalamus or in the cerebral cortex. 20 min later beta-endorphin induced a statistically significant fall of catecholamine concentration in the hypothalamus. A tendency towards its decrease was also observed in the cerebral cortex. 60 min later beta-endorphin produced an insignificant decrease in catecholamine level in both brain areas under study. It may be therefore suggested that beta-endorphin-induced decrease of catecholamine content in the hypothalamus and cerebral cortex represents one of the mechanisms underlying beta-endorphin stimulating action on a number of trophic functions of the hypophysis.     

Postnatal ontogeny of uridine kinase in the cerebellum,hypothalamus, and cerebral cortex of the rat

Postnatal developmental patterns of uridine kinase were determined in crude subcellular fractions of the rat cerebellum, hypothalamus and cerebral cortex at ages 3 through 60 days. The highest specific activity and predominant distribution of enzyme was in the 105,000g supernatant of the 3 brain regions. Enzyme activity in hypothalamus and cerebral cortex was maximum at 3 days and decreased with age; in cerebellum it increased through 13 days and decreased thereafter. Thus, the pattern of activity in hypothalamus and cerebral cortex paralleled changes in DNA and RNA synthesis through age 60 days; in cerebellum, it more closely approximated changes in DNA synthesis during early development. Changes inK _m with aging suggest that the brain regions contain more than one form of enzyme. The highest particulate activity was in the microsomal fraction of the cerebellum and hypothalamus at all ages and in the cortex at 35 and 60 days. Relative specific activity for microsomal fractions of the brain regions at 60 days indicate a concentration of the enzyme which may be relevant in the maintenance of RNA activity in adult brain.     

Phosphorylase and glucosyltransferases at the level of reactive astrocytes. A histochemical study]

Implantation of cobalt powder in the cerebral cortex of rat determines an epileptogenic focus where two types of reactive astrocytes are observed. The first type is mostly represented in the subcortical white matter but it does exist in the cortex around the implant. Phosphorylase and branching enzyme are both very active in these cells which are filled with glycogen. The second type is limited to the cortex and phosphorylase activity leads to an unbranched polysaccharid. These cells correspond to the "activated astrocytes" described by the authors in a previous paper and observed round irritative lesions which, in the cerebral cortex, produce epileptogenic foci.