Glial architecture of the ghost shark (Callorhinchus milii, Holocephali, Chondrichthyes) as revealed by different immunohistochemical markers

This article presents the first study on the glial architecture of a representative species of Holocephali, Callorhinchus milii (ghost shark). Holocephali are a small subclass of Chondrichthyes, with only a few extant genera, and those are considered to have a brain organization more similar to squalomorph sharks than to galeomorph sharks, skates, and rays. Three different astroglial markers--glial fibrillary acidic protein, S-100 protein, and glutamine synthetase (GS)--were investigated by immunohistochemical methods, applying both diaminobenzidine (DAB) and fluorescent techniques. They revealed similar glial structures, although most of them were detected by immunohistochemical reaction against GS and visualized by DAB. The predominant elements were radial ependymoglia spanning the area between the ventricular and meningeal surfaces, as in squalomorph sharks. Other similar features were the light appearance of myelinated neural tracts devoid of immunoreactivity, and the glial architecture of the reticular formation of the brain stem, cerebellum, and tectum, the latter with recognizable layers. The immunoreactivity of the vascular walls was similar; however, it is believed that different cell types form the blood-brain barrier in chimeras and in elasmobranchs. Some glial structures, however, resembled those of skates, rays, and galeomorph sharks. In C. milii astrocyte-like elements were observed in the telencephalon, using GS and S-100, although typical astrocyte-rich regions were not found. In some areas, especially the telencephalon, not only endfeet but also cell bodies were observed to be attached to the meningeal surface, with processes extending into the brain substance.     

Immunohistochemical distribution of neuropeptide Y in the mesencephalon and rhombencephalon of carp, Cyprinus carpio L. (Cyprinidae: Teleostei)

The localization of neuropeptide Y (NPY)-immunoreactive elements was investigated in the mesencephalon and rhombencephalon of carp, Cyprinus carpio, by using antisera raised against porcine NPY and the immunoperoxidase technique. Concurrently, to identify the distribution of NPY-immunoreactivity, we developed an atlas of the studied areas based on Nissl-stained sections. The NPY-immunoreactive (NPY-ir) elements were located in many zones of the mesencephalon and rhombencephalon. In the mesencephalon, positive fibers were the most abundant elements while neurons were scarce. The rhombencephalon rostral part was characterized by a low to moderate fiber density, distributed in the ventro-medial and ventro-lateral region. Differently the caudal part of the rhombencephalon exhibited several NPY-ir elements. In particular, a high density of immunoreactivity was located in the gustatory area at the level of the nucleus (n.) originis nervi glossopharyngei, in the n. nervi vagi, and in the vagal lobe. The latter can be considered a valid neuroanatomical model for the study of gustatory signal processing in vertebrates. Our results regarding the primary gustatory centers give neuroanatomical support to the view that NPY may act as a neurotransmitter and/or a neuromodulator in a wide neural network for feeding behavior control.     

Immunohistochemical localisation of natriuretic peptides in the brains and hearts of the spiny dogfishSqualus acanthias and the Atlantic hagfishMyxine glutinosa

Summary The avidin-biotin peroxidase technique was used to determine the distribution of natriuretic peptides in the hearts and brains of the dogfishSqualus acanthias and the Atlantic hagfishMyxine glutinosa. Three antisera were used: one raised against porcine brain natriuretic peptide which cross-reacts with atrial natriuretic and C-type natriuretic peptides (termed natriuretic peptide-like immunoreactivity); the second raised against porcine brain natriuretic peptide which cross-reacts with C-type natriuretic peptide, but not with atrial natriuretic peptide (termed porcine brain natriuretic peptide-like immunoreactivity); and the third raised against rat atrial natriuretic peptide (termed rat atrial natriuretic peptide-like immunoreactivity). Only natriuretic peptide-like immunoreactivity was observed in the heart ofS. acanthias which was most likely due to the antiserum cross-reacting with C-type natriuretic peptide. No immunoreactivity was found in theM. glutinosa heart. In the brain ofS. acanthias, natriuretic peptide-like immunoreactive fibres were located in many areas of the telencephalon, diencephalon, mesencephalon, rhombencephalon, and spinal cord. Extensive immunoreactivity was observed in the hypothalamo-hypophyseal tract and the neurointermediate lobe of the hypophysis. Natriuretic peptide-like immunoreactive perikarya were found in ventromedial regions of the telencephalon and in the nucleus preopticus. Most perikarya had short, thick processes which extended toward the ventricle. Another group of perikarya was observed in the rhombencephalon. Porcine brain natriuretic peptide-like immunoreactive fibres were observed in the telencephalon, diencephalon, mesencephalon, and rhombencephalon, but perikarya were only present in the preoptic area. In theM. glutinosa brain, natriuretic peptide-like immunoreactive fibres were present in all regions. Immunoreactive perikarya were observed in the pallium, primordium hippocampi, pars ventralis thalami, pars dorsalis thalami, nucleus diffusus hypothalami, nucleus profundus, nucleus tuberculi posterioris, and nucleus ventralis tegmenti. Procine brain natriuretic peptide-like immunoreactive perikarya and fibres had a similar, but less abundant distribution than natriuretic peptide-like immunoreactive structures. Although the chemical structures of natriuretic peptides in the brains of dogfish and hagfish are unknown, these observations show that a component of the natriuretic peptide complement is similar to porcine brain natriuretic peptide or porcine C-type natriuretic peptide. The presence of natriuretic peptides in the brain suggest they could be important neuromodulators and/or neurotransmitters. Furthermore, there appears to be divergence in the structural forms of natriuretic peptides in the hearts and brains of dogfish and hagfish.     

Changes in Glial Fibrillary Acidic Protein (GFAP) Immonureactivity Reflect Neuronal States

The astroglial marker, glial fibrillary acidic protein (GFAP) was investigated by immunohistochemistry in various brain areas in order to see its fluctuations in various functional states. Different neuronal states were either experimentally induced or studied under physiological conditions. To produce experimental alterations the visual system was chosen as a model. Upon lesioning of the lateral geniculate body with the stereotaxic injection of ibotenic acid an increase in GFAP immunoreactivity could be induced in layers III and IV of the ipsilateral visual cortex where geniculo-cortical fibres terminate. Electron microscopy has revealed a synchronous degeneration of synaptic terminals and the hypertrophy of perisynaptic astrocyte processes. To study changes in the intact animal the effect of illumination was observed. In the lateral geniculate body the dorsal subnucleus was found immunonegative when studied at day and positive at night. Similar changes were observed in the suprachiasmatic nucleus. As to more generalized influences, the effect of gonadal steroids on the GFAP-reaction interpeduncular nucleus, an area not involved in hormonal regulatory mechanisms was studied. In males only castration could reduce constantly high GFAP immonoreactivity, whereas in females GFAP showed wide-range sexual cycle-related fluctuations. It was concluded that changes in GFAP immunoreactivity can indicate synaptic events whithin a circumscribed area of the brain. Special issue article in honor of Dr. Ricardo Tapia.     

Glial fibrillary acidic protein and vimentin immunoreactivity of astroglial cells in the central nervous system of the African lungfish, Protopterus annectens (Dipnoi: Lepidosirenidae)

The distribution of glial intermediate filament molecular markers, glial fibrillary acidic protein (GFAP), and vimentin, in the brain and spinal cord of the African lungfish, Protopterus annectens, was examined by light microscopy immunoperoxidase cytochemistry. Glial fibrillary acidic protein immunoreactivity is clear and is evident in a radial glial system. It consists of fibers of different lengths and thicknesses that are arranged in a regular radial pattern throughout the central nervous system (CNS). They emerge from generally immunopositive radial ependymoglia (tanycytes), lining the ventricular surface, and are directed from the ventricular wall to the meningeal surface. These fibers give rise to endfeet that are apposed to the subpial surface and to blood vessel walls forming the glia limitans externa and the perivascular glial layer, respectively. GFAP-immunopositive star-shaped astrocytes were not found in P. annectens CNS. In the gray matter of the spinal cord, cell bodies of immunopositive radial glia are displaced from the ependymal layer. Vimentin-immunopositive structures are represented by thin fibers mostly localized in the peripheral zones of the brain and the spinal cord. While a few stained fibers appear in the gray matter, the ependymal layer shows no antivimentin immunostaining. In P. annectens the immunocytochemical response of the astroglial intermediate filaments is typical of a mature astroglia cell lineage, since they primarily express GFAP immunoreactivity. This immunocytochemical study shows that the glial pattern of the African lungfish resembles that found in tetrapods such as urodeles and reptiles. The glial pattern of lungfishes is comparable to that of urodeles and reptiles but is not as complex as that of teleosts, birds, and mammals.     

The secondary vascular system of Actinopterygii: interspecific variation in origins and investment

Vascular casts of 3 species of Chondrichthyes, 1 of Dipnoi, 1 of Chondrostei and 14 species of the Teleostei were examined by light and scanning electron microscopy in order to give a qualitative and quantitative analysis of interarterial anastomoses (iaas) that indicate the presence (or absence) of a secondary vascular system (SVS). Anastomoses were found to originate from a variety of different primary blood vessels, many of which have not been previously identified as giving rise to secondary vessels. Segmental arteries derived from the dorsal aorta and supplying body musculature were major sites of origin of the SVS, although there was considerable variation in where, in the hierarchy of arterial branching, the anastomoses occurred. The degree of investment in a SVS was species specific, with more active species having a higher degree of secondary vascularisation. This difference was quantified using an absolute count of iaas between Anguilla reinhardtii and Trachinotus baillonii. A range of general features of the SVS is also described. No evidence of iaas was found on the coeliac, mesenteric or renal circulation in any species. Evidence of iaas was lacking in the dipnoan and chondrichthyan species examined, suggesting that a SVS is restricted to Actinopterygii. The presence and distribution of a SVS does not appear to be exclusively linked to phylogenetic position, but rather to the physiological adaptation of the species.The online version of the original article can be found at     

The secondary vascular system of Actinopterygii: interspecific variation in origins and investment

Vascular casts of 3 species of Chondrichthyes, 1 of Dipnoi, 1 of Chondrostei and 14 species of the Teleostei were examined by light and scanning electron microscopy in order to give a qualitative and quantitative analysis of interarterial anastomoses (iaas) that indicate the presence (or absence) of a secondary vascular system (SVS). Anastomoses were found to originate from a variety of different primary blood vessels, many of which have not been previously identified as giving rise to secondary vessels. Segmental arteries derived from the dorsal aorta and supplying body musculature were major sites of origin of the SVS, although there was considerable variation in where, in the hierarchy of arterial branching, the anastomoses occurred. The degree of investment in a SVS was species specific, with more active species having a higher degree of secondary vascularisation. This difference was quantified using an absolute count of iaas between Anguilla reinhardtii and Trachinotus baillonii. A range of general features of the SVS is also described. No evidence of iaas was found on the coeliac, mesenteric or renal circulation in any species. Evidence of iaas was lacking in the dipnoan and chondrichthyan species examined, suggesting that a SVS is restricted to Actinopterygii. The presence and distribution of a SVS does not appear to be exclusively linked to phylogenetic position, but rather to the physiological adaptation of the species.     

Astroglial cells in the central nervous system of the adult brown anole lizard, Anolis sagrei, revealed by intermediate filament immunohistochemistry

We analyzed the distribution of intermediate filament molecular markers, glial fibrillary acidic protein (GFAP), and vimentin in the brain and spinal cord of the adult brown anole lizard, Anolis sagrei. The GFAP immunoreactivity is strong and the positive structures are basically represented by fibers of different lengths and thicknesses which are arranged in a regular radial pattern throughout the central nervous system. In the brain regions that have a thicker neural wall, the radial orientation is not so evident as in the thinner areas. These fibers emerge from radial ependymoglia (tanycytes) whose cell bodies are generally GFAP-immunopositive. The glial fibers give rise to endfeet that are apposed to the subpial surface and to blood vessel walls. In the spinal cord, the optic tectum and the lateroventral regions of the mesencephalon and medulla oblongata, star-shaped astrocytes coexist with radial structures. Vimentin-immunoreactive structures are absent in the brain and spinal cord. In A. sagrei the immunohistochemical response of the astroglial intermediate filaments appears typical of a mature astroglial cell lineage, since they fundamentally express GFAP immunoreactivity. A Western-blot analysis reveals a GFAP-positive single band, common to the different nervous areas. This immunohistochemical study shows that the star-shaped astrocytes have a different distribution in saurians and while the glial pattern of A. sagrei is more evolved than in urodeles it remains immature as compared with crocodilians, avians, and mammals. This condition suggests that reptiles represent a fundamental step in the phylogenetic evolution of the vertebrate glial cells.     

The renin-angiotensin system in the rat brain. Immunocytochemical localization of angiotensinogen in glial cells and neurons

The distribution of angiotensinogen containing cells was determined in the brain of rats using immunocytochemistry. Specific angiotensinogen immunoreactivity is demonstrated both in glial cells and neurons throughout the brain, except the neocortical and cerebellar territories. Positive neurons are easily and invariably detected in female brains, and haphazardly in male brain (sex hormone dependent). Angiotensinogen immunoreactivity in male brain neurons can be induced by water deprivation or binephrectomy in some areas and particularly in paraventricular nuclei. Finally, the highest concentrations of positive neurons are found in the anterior and lateral hypothalamus, preoptic area, amygdala and some well known nuclei of the mesencephalon and the brainstem. Our results confirm the wide distribution of angiotensinogen mRNA in the brain reported recently by Lynch et al. (1987). Thus the demonstration of angiotensinogen in neurons and glial cells allows a greater understanding of the biochemical and physiological data in accordance with multiple brain renin angiotensin systems.     

Immunohistochemical co-localization of glycogen phosphorylase with the astroglial markers glial fibrillary acidic protein and S-100 protein in rat brain sections.

Immunofluorescence double-labelling and immunoenzyme double-staining methods were used to examine the location of glycogen phosphorylase brain isozyme with the astrocyte markers glial fibrillary acidic protein (GFAP) and S-100 protein in formaldehyde-fixed, paraffin-embedded slices from adult rat brain. Astrocytes in the cerebellum and the hippocampus, which express GFAP or S-100 protein immunoreactivity, show glycogen phosphorylase immunoreactivity. Regional intensity and intracellular distribution of the three antigens vary characteristically. In ependymal cells, glycogen phosphorylase immunoreactivity is co-localized with S-100 protein immunoreactivity, but not with GFAP immunoreactivity. These findings confirm that glycogen phosphorylase in the rat brain is exclusively localized in astrocytes and ependymal cells. All astrocytes, as far as they express GFAP or S-100 protein, do contain glycogen phosphorylase.     

Glial cells positive for glial fibrillary acidic protein in the neurohypophysis of the Djungarian hamster (Phodopus sungorus)

Summary The presence and distribution of the glial fibrillary acidic protein (GFAP; an astrocytic marker protein associated with glial filaments) in the neurohypophysis of the Djungarian hamster (Phodopus sungorus) were investigated immunohistochemically. Our study revealed characteristic GFAP-staining patterns within the median eminence, infundibular stem and neural lobe. In the whole neurohypophysis, few glial cells showed immunoreactivity. In the neural lobe, immunopositive pituicytes appeared preferentially in the periphery. At the ultrastructural level, we found some pituicytes containing filaments, most notably in their processes. We thus demonstrated that, in contrast to the GFAP-immunoreactivity of cultured pituicytes, pituicytic GFAP-expression in vivo coincides with the presence of electron-microscopically detectable filaments.     

Immunohistochemical co-localization of glycogen phosphorylase with the astroglial markers glial fibrillary acidic protein and S-100 protein in rat brain sections

Summary Immunofluorescence double-labelling and immunoenzyme double-staining methods were used to examine the location of glycogen phosphorylase brain isozyme with the astrocyte markers glial fibrillary acidic protein (GFAP) and S-100 protein in formaldehydefixed, paraffin-embedded slices from adult rat brain. Astrocytes in the cerebellum and the hippocampus, which express GFAP or S-100 protein immunoreactivity, show glycogen phosphorylase immunoreactivity. Regional intensity and intracellular distribution of the three antigens vary characteristically. In ependymal cells, glycogen phosphorylase immunoreactivity is co-localized with S-100 protein immunoreactivity, but not with GFAP immunoreactivity. These findings confirm that glycogen phosphorylase in the rat brain is exclusively localized in astrocytes and ependymal cells. All astrocytes, as far as they express GFAP or S-100 protein, do contain glycogen phosphorylase.     

Fibronectin is expressed by astrocytes cultured from embryonic and early postnatal rat brain

In early primary cultures from newborn rat brain, few glial fibrillary acidic protein (GFAP)-positive glial cells expressed intracytoplasmic immunoreactivity for fibronectin. After the second week in culture, however, fibronectin was expressed by a distinct population of GFAP-positive flat astrocytes, irrespective of which brain region was studied. In cerebellar cultures, these cells were more abundant than in cortical or neostriatal cultures and often formed a major population of the GFAP-positive cells. The difference in fibronectin expression between cerebellum and the other areas studied was statistically significant. When cultures were started from 9-day-old postnatal rat brain, fibronectin-positive astrocytes appeared earlier than in those from newborn animals, in all areas studied. Further, especially in the case of cerebellum, the number of fibronectin-positive astrocytes increased as a function of time in culture. In cultures started from whole brains of 12-day-old rat embryos, fibronectin was expressed within 24 h in culture by all the cells with morphology of flat astrocytes, positive for vimentin but negative for GFAP. These results indicate that astrocytes cultured from newborn and early postnatal rat brain are a heterogeneous population of cells: depending on the brain region studied and also depending on the age of brain tissue or the time in culture, less than 1-60% of the GFAP-positive flat astrocytes expressed fibronectin. This, together with the fact that fibronectin was present in early embryonic brain cells in culture, suggests that fibronectin may be a prerequisite for the development or interactions of brain cells.     

Expression of the small conductance Ca<Superscript>2+</Superscript>-activated K<Superscript>+</Superscript> channel,SK3, in the olfactory ensheathing glial cells of rat brain

Olfactory sensory neurons are wrapped by ensheathing glial cells in the olfactory nerve layer (ONL). Neither functional roles nor electrical properties of ensheathing glial cells have been, as yet, fully clarified. Four subunits (SK1–4) of small conductance Ca²⁺-activated K⁺ (SK) channels have been cloned. In the present study, immunohistochemical analyses showed that SK3 channels are expressed in ensheathing glial cells in the rat olfactory bulb, in addition to neuronal cells in other regions. Western blotting analysis demonstrated that SK3 was predominantly expressed in the olfactory bulb, thalamus, moderately in the hippocampus and cerebellum and modestly in the cerebral cortex of the rat brain. SK3 immunoreactivity was detected in the ONL of the olfactory bulb, neural cell body and fibers of the substantia nigra and hypothalamus. SK3 immunoreactivity was quite intense in the outer (superficial) part of the ONL. SK3-immunoreactive structures were overlapped with glial fibrillary acidic protein (GFAP), but not with vimentin, markers for glial cells and olfactory sensory axons, respectively. Immunoelectron microscopy showed that SK3 immunoreactivity was localized in thin processes that enfolded fascicles of immunonegative olfactory nerve axons. These results indicate that SK3 is expressed specifically in the olfactory ensheathing glial cells in olfactory regions.This work was supported in part by a Grant-in-Aid to A.F. for Scientific Research from the Ministry of Education, Science, Sports and Culture of Japan, and by scholarship from Ono Pharmaceutical Company, and by Narishige Neuroscience Research Foundation.     

The renin-angiotensin system in the rat brain

Summary The distribution of angiotensinogen containing cells was determined in the brain of rats using immunocytochemistry. Specific angiotensinogen immunoreactivity is demonstrated both in glial cells and neurons throughout the brain, except the neocortical and cerebellar territories. Positive neurons are easily and invariably detected in female brains, and haphazardly in male brain (sex hormone dependent). Angiotensinogen immunoreactivity in male brain neurons can be induced by water deprivation or binephrectomy in some areas and particularly in paraventricular nuclei. Finally, the highest concentrations of positive neurons are found in the anterior and lateral hypothalamus, preoptic area, amygdala and some well known nuclei of the mesencephalon and the brainstem.Our results confirm the wide distribution of angiotensinogen mRNA in the brain reported recently by Lynch et al. (1987). Thus the demonstration of angiotensinogen in neurons and glial cells allows a greater understanding of the biochemical and physiological data in accordance with multiple brain renin angiotensin systems.     

Laminin is induced in astrocytes of adult brain by injury.

Laminin is a high mol. wt. non-collagenous matrix glycoprotein, confined in adult tissues to basement membranes. In normal rat brain we found laminin mainly in vessel walls but, after injury, induced by stereotaxic injection of a neurotoxin, laminin immunoreactivity appeared also in reactive astrocytes, which are characteristically positive for the glial fibrillary acidic protein (GFAP). Laminin was first detected in GFAP-immunoreactive glial cells 24 h after injury. Four days later the majority of reactive astrocytes in the gray matter were positive for laminin and the laminin immunoreactivity, but not that of GFAP, gradually subsided within a month. Fibronectin, the other major matrix glycoprotein, was found only in capillary structures both in normal and lesioned brain tissue. The results indicate that mature astrocytes have the potential to produce laminin and suggest a role for this glycoprotein in brain regeneration.     

Melatonin reduces glial reactivity in the hippocampus, cortex, and cerebellum of streptozotocin-induced diabetic rats

Hyperglycemia plays a critical role in the development and progression of diabetic neuropathy. One of the mechanisms by which hyperglycemia causes neural degeneration is via the increased oxidative stress that accompanies diabetes. Metabolic and oxidative insults often cause rapid changes in glial cells. Key indicators of this response are increased synthesis of glial fibrillary acidic protein (GFAP) and S100B, both astrocytic markers. In the present study, we examined glial reactivity in hippocampus, cortex, and cerebellum of streptozotocin (STZ)-induced diabetic rats by determining the expression of GFAP and S-100B and we evaluated the effect of melatonin on the glial response. Western blot measurement of contents in brain regions after 6 weeks of STZ-induced diabetes indicated significant increases in these constituents compared with those in nondiabetic controls. Administration of melatonin prevented the upregulation of GFAP in all brain regions of diabetic rats. Using GFAP immunohistochemistry, we observed an increase in GFAP immunostaining in the hippocampus of STZ-diabetic rats relative to levels in the control brains. Treatment with melatonin resulted in an obvious reduction of GFAP-immunoreactive astrocytes in hippocampus. Like GFAP, S100B levels also were increased in all three brain areas of diabetic rats, an effect also reduced by melatonin treatment. Finally, the levels of lipid peroxidation products were elevated as a consequence of diabetes, with this change also being prevented by melatonin. These results suggest that diabetes causes increased glial reactivity possibly due to elevated oxidative stress, and administration of melatonin represents an achievable adjunct therapy for preventing gliosis.     

The CNS syndrome. The characteristics of the intermediate filaments of the rat brain

A study was made of the effect of ionizing radiation on the content and polypeptide composition of filamentous and soluble glial fibrillary acidic protein (GFAP) in different regions of rat brain. Ionizing radiation was shown to decrease considerably the level of soluble GFAP in cerebral cortex, cerebellum, middle brain and hippocampus. Polypeptide composition of soluble GFAP detected by the immunoblot method was found to be changed considerably in different brain areas of irradiated animals.     

Characterization of glial fibrillary acidic protein and astroglial architecture in the brain of a continuously growing fish, the rainbow trout

Unlike mammals, some fish, including carp and trout, have a continuously growing brain. The glial architecture of teleost brain has been intensively studied in the carp and few data exist on trout brain. In this study, using immunoblotting we characterized the topographic distribution of glial fibrillary acidic protein (GFAP) in larval and adult rainbow trout brain and studied by immunohistochemistry the distribution and morphology of GFAP-immunoreactive cell systems in the rainbow trout hindbrain and spinal cord. Immunoblotting yielded a double band with an apparent molecular weight of 50-52 kDa in the spinal cord homogenate in the trout larval and adult stages. In the adult hindbrain and forebrain, our antibody cross reacted also with a second band at a higher molecular weight (90 kDa). Because the forebrain contained this band alone the two brain regions might contain two distinct isoforms. Conversely, the larval total brain homogenate contained the heavy 90 kDa band alone. Hence the heavy band might be a GFAP protein dimer or vimentin/GFAP copolymer reflecting nerve fiber growth and elongation, or the two isoforms might indicate two distinct astroglial cell types as recently proposed in the zebrafish. In sections from trout hindbrain and spinal cord the antibody detected a GFAP-immunoreactive glial fiber system observed in the raphe and in the glial septa separating the nerve tracts. These radial glia fibers thickened toward the pial surface, where they formed glial end feet. The antibody also labeled perivascular glia around blood vessels in the white matter, and the ependymoglial plexus surrounding the ventricular surface in the grey matter. Last, it labeled round astrocytes. The GFAP-immunoreactive glial systems had similar distribution patterns in the adult and larval spinal cord suggesting early differentiation.