Apoptogenic effect of the lipophilic o-naphthoquinone CG 10-248 on rat hepatocytes: light and electron microscopy studies

CG 10-248 (3,4-dihydro-2,2-dimethyl-9-chloro-2H-naphtho[1,2b]pyran-5,6-dione; CG-NQ), a beta-lapachone analogue, modified the ultrastructure of rat hepatocytes, as demonstrated by light and electron microscopy. After 4 h incubation with 100 microM CG-NQ, the following effects were observed: (a) nuclear chromatin condensation; (b) chromatin fragmentation; (c) displacement of mitochondria, concentrated around the nucleus; (d) disruption or expansion of mitochondrial outer or inner membranes, respectively; (e) displacement and alteration of endoplasmic reticulum (rough and smooth); (f) decrease of microvilli; (g) blebbing of plasma membrane and production of apoptotic bodies formed by folding of plasma membrane fragments around mitochondria or peroxysomes; and (h) production of hydrogen peroxide. Expression of such effects varied according to hepatocyte samples and taken together strongly support an apoptotic action of CG-NQ dependent on reactive oxygen species. (AU)

Apoptogenic effect of the lipophilic o-naphthoquinone CG 10-248 on rat hepatocytes: light and electron microscopy studies

CG 10-248 (3,4-dihydro-2,2-dimethyl-9-chloro-2H-naphtho[1,2b]pyran-5,6-dione; CG-NQ), a beta-lapachone analogue, modified the ultrastructure of rat hepatocytes, as demonstrated by light and electron microscopy. After 4 h incubation with 100 microM CG-NQ, the following effects were observed: (a) nuclear chromatin condensation; (b) chromatin fragmentation; (c) displacement of mitochondria, concentrated around the nucleus; (d) disruption or expansion of mitochondrial outer or inner membranes, respectively; (e) displacement and alteration of endoplasmic reticulum (rough and smooth); (f) decrease of microvilli; (g) blebbing of plasma membrane and production of apoptotic bodies formed by folding of plasma membrane fragments around mitochondria or peroxysomes; and (h) production of hydrogen peroxide. Expression of such effects varied according to hepatocyte samples and taken together strongly support an apoptotic action of CG-NQ dependent on reactive oxygen species.

Effect of dyskinetoplastic agents on ultrastructure and oxidative phosphorylation in Crithidia fasciculata.

Ethidium bromide (EB) is an intercalating agent which binds specifically to the kinetoplast (mitochondrial) DNA (kDNA) of trypanosomatids. Accordingly, EB inhibits DNA replication, thus inducing dyskinetoplasty. Since in eukariotic organisms mitochondrial DNA encodes the genetic information for cytochromes b, aa3 and F0F1 ATPase, it seemed of interest to establish whether a similar effect occurs in Crithidia fasciculata, a trypanosomatid used for assay of potential trypanocidal drugs. Culturing of C. fasciculata in the presence of EB inhibited growth and induced dyskinetoplasty, as confirmed by electron microscopy. The kinetoplast of EB-cultured crithidia lost its characteristic arc shape, it was misplaced in the cell cytoplasm its matrix structure and membrane differentiation were specifically modified. Dyskinetoplasty decreased crithidia respiration and oxidative phosphorylation, as indicated by the lower ATP level, ATP/ADP ratio and adenylate energy charge. The interference of EB with kinetoplastic constituents synthesis was confirmed by the lack of action of EB on crithidia in the stationary phase of growth, that ruled out direct inhibition of oxidative phosphorylation enzymes. The lipophilic o-naphthoquinone beta-lapachone produced structural alterations in kinetoplast membranes, that correlated with inhibition of oxidative phosphorylation. These latter effects involved free radicals since they were prevented by free radical scavengers.

Effect of dyskinetoplastic agents on ultrastructure and oxidative phosphorylation in Crithidia fasciculata.

Ethidium bromide (EB) is an intercalating agent which binds specifically to the kinetoplast (mitochondrial) DNA (kDNA) of trypanosomatids. Accordingly, EB inhibits DNA replication, thus inducing dyskinetoplasty. Since in eukariotic organisms mitochondrial DNA encodes the genetic information for cytochromes b, aa3 and F0F1 ATPase, it seemed of interest to establish whether a similar effect occurs in Crithidia fasciculata, a trypanosomatid used for assay of potential trypanocidal drugs. Culturing of C. fasciculata in the presence of EB inhibited growth and induced dyskinetoplasty, as confirmed by electron microscopy. The kinetoplast of EB-cultured crithidia lost its characteristic arc shape, it was misplaced in the cell cytoplasm its matrix structure and membrane differentiation were specifically modified. Dyskinetoplasty decreased crithidia respiration and oxidative phosphorylation, as indicated by the lower ATP level, ATP/ADP ratio and adenylate energy charge. The interference of EB with kinetoplastic constituents synthesis was confirmed by the lack of action of EB on crithidia in the stationary phase of growth, that ruled out direct inhibition of oxidative phosphorylation enzymes. The lipophilic o-naphthoquinone beta-lapachone produced structural alterations in kinetoplast membranes, that correlated with inhibition of oxidative phosphorylation. These latter effects involved free radicals since they were prevented by free radical scavengers.

Cyclic AMP-responsive element binding protein in brain mitochondria.

Cyclic AMP-responsive element binding protein (CREB) is critically involved in many important brain functions, including the formation of long-term memory. CREB is the best characterized member of a family of transcription factors (CREB/ATF family) recognized to be important nuclear targets for intracellular signal transduction systems. Here we show, by using different approaches, that CREB is unexpectedly localized to mitochondria of the rat brain. Controlled subcellular fractionation of hippocampus and cerebral cortex showed that both synaptic and nonsynaptic mitochondria exhibited immunoreactivity to the phosphorylated form of CREB (pCREB). Moreover, CREB extracted from synaptic mitochondria is able to be phosphorylated by the catalytic subunit of protein kinase A and dephosphorylated by protein phosphatase 1 or 2B. DNA mobility shift assays showed the presence of binding activity to the calcium-cyclic AMP-responsive element in mitochondrial extracts from hippocampus; this binding complex was specifically supershifted by an anti-CREB antibody. Immunoelectron microscopic analysis of hippocampal subcellular fractions revealed that pCREB immunoreactivity is localized in close association with the inner mitochondrial membrane. These results, together with recent findings describing the presence and phosphorylation of CREB in developing dendrites, suggest that CREB may participate in different mechanisms involved in the communication between extracellular signals and the expression of genes.

Atrial natriuretic factor in two kidney--two clip renovascular hypertension in the rat.

Hig levels of circulating atrial natriuretic factor (ANF) have been reported in several physiopathologic conditions like hypertension, heart and renal failure, pregnancy and high sodium intake. Nevertheless, neither relationships with water-sodium space regulation nor the role of an ANF vascular relaxant effect have been yet defined. The aim of present experiments was to characterize the contribution of circulating ANF and its vascular relaxing effects in the two kidney-two clip (2K2C) experimental model of renovascular hypertension. Complementary, plasma metabolites nitrite/nitrate of nitric oxide (NO) was examined because of mediation for both (NO an ANF) through cGMP. Three results showed (two-four weeks after surgery): indirect systolic blood pressure (mmHg), 186 +/- 4 in HT and 122 +/- 1 in SH (p < 0.001); a significant increase of plasma ANF (fmol/ml) in HT (n = 7, 1221 +/- 253) vs. SH (n = 9, 476 +/- 82; p < 0.02). Nitrate/nitrite plasma concentrations (mumol/l) were mpt different between SH and. The relaxant effect of ANF (10(-9), 10(-8) and 10(-7) M) on phenylephrine (3,5 x 10(-6) M) contracted rings from HT rats was smaller than SH rats (10(-8) M, p < 0.05). Contractions to phorbol 12, 13-dibutyrate (seven weeks after surgery) were significantly higher in rings from HT rats (p < 0.001). We conclude: 1) in addition to decreased granularity in atrial myocardiocytes, high circulating values of ANF here described suggest an increased turnover of the peptide in 2K2C hypertensive rats; 2) lower significant vascular relaxant effects in HT rats would indicate down regulation of ANF receptors in this model; the latter would derive from high plasma ANF concentration and, tentatively, because of greater activity of protein kinase C in the vascular wall; 39 similar values of plasma nitrite/nitrate in SH and HT rats would indicate a comparable NO circulating availability in both groups.

Differential staining of two subpopulations of Purkinje neurons in rat cerebellum with acid dyes.

We present a new method that stains differently two subpopulations of Purkinje cells in the adult rat. Deparaffinized sections of cerebella, fixed by perfusion with buffered glutaraldehyde or Bouin's fluid were stained with 0.5% light green in 50% ethanol (10-30 min). The excess dye was removed with saturated aqueous picric acid (10-30 min). At this point some Purkinje cells appeared as lightly stained neurons, while others were strongly stained. Slides were immersed in 0.5% aqueous acid fuchsin for approximately 1 min until the lightly stained neurons acquired a red color. Following immersion in 1% phosphotungstic acid, slides were rapidly dehydrated in ethanol, passed to xylene and mounted in Canada balsam. Two subpopulations of Purkinje cells differing in their protein content in somata and proximal dendrites stained differentially by this method. They occurred in all coronal and sagittal sections and in patches or stripes. Their relative proportion varied from lobule to lobule. A second staining method used potassium permanganate as the sole staining reagent. The staining reagent can be used on sections previously stained with the acid dyes. Purkinje cells appeared as subsets of brownish to deep brown stained neurons, the latter ones corresponding to green stained cells in the dichromic method. The results obtained indicated that the subpopulations reflect real differences among individual neurons and are not artifacts. The technique holds promise for identifying and localizing sub-sets of Purkinje cells differing in their protein content under normal and experimental conditions and for their further characterization by combined staining and histochemical procedures.

Mg2+/Ca(2+)-ATPase activity is not enriched in synaptic vesicles isolated from rat cerebral cortex.

Neuronal ATPases comprise a wide variety of enzymes which are not uniformly distributed in different membrane preparations. Since purified vesicle fractions have Mg2+/Ca(2+)-ATPase, the purpose of the present study was to know whether such enzyme activities have a preferential concentration in a synaptic vesicle fraction in order to be used as markers for these organelles. Resorting to a procedure developed in this Institute, we fractionated the rat cerebral cortex by differential centrifugation following osmotic shock of a crude mitochondrial fraction and separated a purified synaptic vesicle fraction over discontinuous sucrose gradients. Mg2+/Ca(2+)-ATPase activities and ultrastructural studies of isolated fractions were carried out. It was observed that similar specific activities for Mg2+/Ca(2+)-ATPases were found in all fractions studied which contain synaptic vesicles and/or membranes. Although the present results confirm the presence of Mg2+ and Ca(2+)-ATPase activities in synaptic vesicles preparations, they do not favor the contention that Mg2+/Ca(2+)-ATPase is a good marker for synaptic vesicles.

Selective axonal argentaffin staining in rat central nervous system after protein mercuration.

The mercury-silver (Hg-Ag) argentaffin technique, known to stain specifically proteins in the lateral components of triads/diads in striated muscle cells, was applied to the central nervous system of adult rats. Following fixation in glutaraldehyde, axons in white and gray matter were selectively stained, but not perikarya or their proximal axon and dendrites. Neural tissues were postfixed 24 hr in 5% (w/v) mercuric acetate in 2% (v/v) acetic acid in distilled water, stained for 12-24 hr in darkness at 37-43 C with ammoniacal silver nitrate solution, freshly prepared by adding concentrated ammonia to 60% (w/v) silver nitrate solution until a small amount of silver oxide precipitate remained undissolved. Samples were then washed with freshly prepared 5% (w/v) sodium sulfite and distilled water. All steps were carried out using dark-colored glass flasks. Samples were dehydrated with ethanol and embedded in Paraplast or Poly Bed. Electron microscopy showed the silver-reducing protein inside the axons. Methylation abolished Hg-Ag axonal reactivity indicating that carboxyl groups were necessary for silver staining. Proteins with solubility properties characteristic of neurofilament proteins were involved in Hg-Ag staining. In the cerebellum the plexus of parallel fibers in the molecular layer were not stained, while basket cell axonal processes reacted intensely. The method appears to distinguish neuronal protein variants related to cytotypic differences in cytoskeletal neurofilaments.

Effect of pyridine and its methylated derivatives on storage by rat pineal nerves of monoaminergic neurotransmitter.

In previous work of our laboratory, it was demonstrated that collidine (2-4-6-trimethylpyridine) applied briefly to fresh tissues extracted noradrenaline or closely related compound/s of neuronal origin. This effect gave rise to the abolition of osmiophilia and chromaffin reaction of electron-dense cores of monoaminergic synaptic vesicles and to the extraction of radioactive compounds in tissues that had previously taken up tritiated noradrenaline. In this work, the role of the pyridine ring and its progressive methylation on the monoamine releasing effect was investigated. For this purpose, the effect of pyridine, two picolines (2 and 4 monomethylpyridines) and a lutidine (2-6-dimethylpyridine) was compared to the effect of collidine. It was found that pyridine has a much smaller effect than collidine on the histochemical reactivity of monoaminergic synaptic vesicles and on the extraction of tritiated compounds and that its extent was dependent upon the number of methyl groups incorporated in the pyridine ring.

Ultrastructural study of atrial specific granules in experimental renovascular hypertension elicited by the constriction of both renal arteries.

The heart has an endocrine activity which depends on the secretion of a natriuretic, diuretic and hypotensive factor contained in osmophilic, secretory granules localized in the myocardiocytes and called "atrial specific granules" (the atrial natriuretic factor, ANF). In this paper, the relationship between these specific granules and renovascular hypertension elicited by the constriction of both renal arteries was investigated at the electron microscope level during the acute, subacute and chronic phases of hypertension. Male Wistar CHbb THOM rats were divided in three groups: 1) clipped rats; 2) sham operated rats; 3) ether anesthesia as unique manoeuver 48 h before decapitation. Blood pressure increased progressively after the constriction of both renal arteries. The atrial specific granules were not affected by ether anesthesia alone; 48-72 h after clipping the granules almost disappeared and this situation persisted up to the 6th week. In sham operated rats the picture was very similar to the clip rats 48 and 72 h after surgery (severe granule disappearance); in contrast, at one, two and six weeks after surgery, the granularity of cardiomyocytes in sham rats was absolutely restored. It is concluded that: 1) similarities in morphology of atrial specific granules in sham and clip rats 48 and 72 h after surgery would suggest that stress plays a primary role in determining the observed images; 2) thereafter, the contrast between sham and clip rats 1, 2 and 6 weeks after surgery would indicate that the ANF is linked to the subacute and chronic regulation of renovascular hypertension.

Ultrastructural study of atrial specific granules in experimental renovascular hypertension elicited by the constriction of both renal arteries.

The heart has an endocrine activity which depends on the secretion of a natriuretic, diuretic and hypotensive factor contained in osmophilic, secretory granules localized in the myocardiocytes and called [quot ]atrial specific granules[quot ] (the atrial natriuretic factor, ANF). In this paper, the relationship between these specific granules and renovascular hypertension elicited by the constriction of both renal arteries was investigated at the electron microscope level during the acute, subacute and chronic phases of hypertension. Male Wistar CHbb THOM rats were divided in three groups: 1) clipped rats; 2) sham operated rats; 3) ether anesthesia as unique manoeuver 48 h before decapitation. Blood pressure increased progressively after the constriction of both renal arteries. The atrial specific granules were not affected by ether anesthesia alone; 48-72 h after clipping the granules almost disappeared and this situation persisted up to the 6th week. In sham operated rats the picture was very similar to the clip rats 48 and 72 h after surgery (severe granule disappearance); in contrast, at one, two and six weeks after surgery, the granularity of cardiomyocytes in sham rats was absolutely restored. It is concluded that: 1) similarities in morphology of atrial specific granules in sham and clip rats 48 and 72 h after surgery would suggest that stress plays a primary role in determining the observed images; 2) thereafter, the contrast between sham and clip rats 1, 2 and 6 weeks after surgery would indicate that the ANF is linked to the subacute and chronic regulation of renovascular hypertension.

Compartmentalization of monoaminergic synaptic vesicles in the storage and release of neurotransmitter.

Monoaminergic nerves are characterized by the presence of a population of small synaptic vesicles (40-60 nm in diameter) containing a few large vesicles (80-90 nm in diameter). Thus, although both types of vesicles contain monoamines, the small vesicles must be considered as the organoid responsible for the storage and release of the neurotransmitter, whereas the large ones possibly are involved in the modulation of the process. The small vesicles are electron-lucent or have an osmiophilic electron-dense core that is always linked to the vesicle membrane. Considering morphological and histochemical evidence under different experimental conditions, we proposed the existence of two compartments in the small vesicles: the core and the matrix, corresponding respectively to the electron-dense core and the electron-lucent space between the core and the vesicle membrane in osmium tetroxide fixations. The sizes of both compartments are inversely related, i.e., the smaller the core, the larger the matrix and vice versa. The core even disappears, giving way to a small electron-lucent vesicle made exclusively by the matrix. Thus, the matrix is a constant component of the vesicle, whereas the core is a transient one. Each compartment has a different pool of amine: a loosely bound, easily releasable pool in the matrix and a tightly bound, more resistant pool in the core. These two pools subserve, respectively, a tonic or phasic release of the neurotransmitter, correlated with a tonic or phasic stimulation of the receptor. The core may be considered as a storage or reserve pool. Experimental evidence from our laboratory supports the concept that different mechanisms are operative in both compartments in the release of the neurotransmitter. For instance, a Ca2(+)-independent release would be primarily concerned with the neurotransmitter contained in the matrix, and a Ca2(+)-dependent efflux would be primarily related with the neurotransmitter stored in the core. However, it still must be established that a simple relationship exists between each kind of stimulus and each vesicle compartment, rather than both compartments being integrated in a dynamic functional unit.

The release of catecholamines by an endogenous factor that inhibits neuronal Na+,K(+)-ATPase.

It is known that the inhibition of Na+,K(+)-ATPase induces neurotransmitter release in several experimental models. In this laboratory it was previously observed that a brain soluble fraction separated by Sephadex G-50 (peak II) is able to inhibit Na+,K(+)-ATPase but not other membrane-bound enzymes. The object of the present study was to test the effect of brain peak II fraction on neurotransmitter content of the pineal nerves synaptic vesicles. Uninjected rats and rats injected 30 min before with 5-hydroxydopamine (30 mg per kg,i.p.) were used. 5-hydroxydopamine produces a false neurotransmitter whose presence in the synaptic vesicles is visualized after fixation with glutaraldehyde-osmium as an electron dense material with the electron microscope which fills totally or partially the vesicles. In uninjected rats the osmiophilia and the chromaffin reaction of the electron dense core were studied. The pineal glands were incubated in Tyrode solution without calcium in the presence and absence of peak II at room temperature and processed for electron microscopy. When the glands from rats pretreated with 5-hydroxydopamine were incubated with peak II a significant decrease in the number of vesicles totally stained was observed. This indicates a reduction in false neurotransmitter content, specially in the matrix of the synaptic vesicles. In the presence of an aged peak II, which does not inhibit Na+,K(+)-ATPase, this effect on the synaptic vesicles was not observed. When the glands from uninjected rats were incubated with peak II no changes in the osmiophilia and the chromaffin reaction of the synaptic vesicles were found. The osmiophilia and the chromaffin reaction of the cores marks the monoamines storage site (catechol and indoleamines in the pineal nerves). These results are coherent with the idea of a relationship between the inhibition of Na+,K(+)-ATPase activity and the release of a pool of neurotransmitter stored in the nerve endings.

The release of catecholamines by an endogenous factor that inhibits neuronal Na+,K(+)-ATPase.

It is known that the inhibition of Na+,K(+)-ATPase induces neurotransmitter release in several experimental models. In this laboratory it was previously observed that a brain soluble fraction separated by Sephadex G-50 (peak II) is able to inhibit Na+,K(+)-ATPase but not other membrane-bound enzymes. The object of the present study was to test the effect of brain peak II fraction on neurotransmitter content of the pineal nerves synaptic vesicles. Uninjected rats and rats injected 30 min before with 5-hydroxydopamine (30 mg per kg,i.p.) were used. 5-hydroxydopamine produces a false neurotransmitter whose presence in the synaptic vesicles is visualized after fixation with glutaraldehyde-osmium as an electron dense material with the electron microscope which fills totally or partially the vesicles. In uninjected rats the osmiophilia and the chromaffin reaction of the electron dense core were studied. The pineal glands were incubated in Tyrode solution without calcium in the presence and absence of peak II at room temperature and processed for electron microscopy. When the glands from rats pretreated with 5-hydroxydopamine were incubated with peak II a significant decrease in the number of vesicles totally stained was observed. This indicates a reduction in false neurotransmitter content, specially in the matrix of the synaptic vesicles. In the presence of an aged peak II, which does not inhibit Na+,K(+)-ATPase, this effect on the synaptic vesicles was not observed. When the glands from uninjected rats were incubated with peak II no changes in the osmiophilia and the chromaffin reaction of the synaptic vesicles were found. The osmiophilia and the chromaffin reaction of the cores marks the monoamines storage site (catechol and indoleamines in the pineal nerves). These results are coherent with the idea of a relationship between the inhibition of Na+,K(+)-ATPase activity and the release of a pool of neurotransmitter stored in the nerve endings.

Possible involvement of Na+,K(+)-ATPase inhibition in neurotransmitter release induced by collidine.

Previous histochemical studies have demonstrated that collidine abolished the osmiophilia and the chromaffin reaction of the synaptic vesicles, an effect which was ascribed to the release of stored neurotransmitters. Other studies indicated a relationship between the inhibition of Na+,K(+)-ATPase and the release of neurotransmitters. In the present study it is shown that collidine inhibits cation-stimulated ATPase activities of the brain synaptosomal membranes. This favours the idea that neurotransmitter release induced by collidine could be due to the inhibition of Na+,K(+)-ATPase activity.

Immunocytochemical localization of serotonin in the pineal gland.

The localization of serotonin (5HT) in the rat pineal gland during daylight conditions was investigated by light and electron microscopy using a polyclonal antibody to 5HT and the immunocytochemical method of peroxidase-antiperoxidase (PAP). Thin varicose 5HT fibers were found distributed in the gland at light microscopy level. Ultrastructural observations revealed the presence of terminal endings containing 5HT and localized mainly in the perivascular spaces, labeled with dense diaminobenzidine (DAB) deposits.

A silver-reducing component in rat striated muscle. I. Selective localization at the level of the terminal cistern/transverse tubule system. Light and electron microscope studies with a new histochemical procedure.

This study reports the presence of a silver-reducing constituent in rat striated muscle fiber located selectively at the level of the terminal cistern/transverse tubule system. It is related to the T tubule network at or near sites that participate in junctions with terminal cisternae, i.e., at both sides of the T tubule in skeletal muscle (triad) and, predominantly, at one side in the ventricle (dyad). Little reactivity is present in the auricle due to the scarcity of those membrane systems. The longitudinal sarcoplasmic reticulum, the sarcolemma, mitochondria and myofibrils are not outlined by the reaction product. Extraction of low molecular weight substances, nucleic acids and lipids did not suppress the chemical reaction. A new argentaffin (Hg--Ag) technique is described. Ethanol or aldehyde fixed muscles were passed to water, postfixed 6-24 h with mercuric acetate (5% w/v in 1% acetic acid), washed with 1% acetic acid and distilled water, stained 12-24 h at 43 degrees C with ammoniacal silver nitrate (60% w/v) and washed in 10% sodium sulfite (three changes) and water. All steps were carried out in darkness. Postfixation with mercuric acetate proved to be essential for immobilizing the argentaffin component without interfering with its strong argentaffinity. The procedure also provides a simple method for tracing the pathway of transversally oriented membrane systems in skeletal and cardiac muscle cells.

A silver-reducing component in rat striated muscle. II. Isolated sarcoplasmic reticulum vesicles.

In previous work on rat striated muscle cells a silver-reducing component was found selectively localized at the terminal cistern/transverse tubule system (Tandler and Pellegrino de Iraldi 1989). To further investigate that problem we performed the Hg-Ag argentaffin reaction on a sarcoplasmic reticulum fraction from rat skeletal muscle. Circular profiles corresponding to vesicular structures were found outlined by silver grains. The number of silver "stained" vesicles were less than the total number vesicles stained by conventional procedures. The correlation between argentaffinities in the intact muscle fiber and their subcellular organelles indicated that the Hg-Ag reactive vesicles must be those derived from the terminal cisternae of the sarcoplasmic reticulum. The silver-reducing constituent aggregates in the presence of 1 mM CaCl2 or 0.5 M K cacodylate. The state of aggregation induced by Ca2+ was not affected by incubation with 0.5% Triton X-100 or by 2 mM EDTA, thus suggesting a localization at or near the membrane of the terminal cistern vesicle facing the junctional gap. In Laemmli SDS-acrylamide gels the Hg-Ag reaction stained all proteins in a manner similar to Coomasie blue. It is suggested that the selective histochemical staining is the result of differential reactivities due to steric requirements of the chemical reaction.

Immunocytochemical localization of serotonin in the pineal gland.

The localization of serotonin (5HT) in the rat pineal gland during daylight conditions was investigated by light and electron microscopy using a polyclonal antibody to 5HT and the immunocytochemical method of peroxidase-antiperoxidase (PAP). Thin varicose 5HT fibers were found distributed in the gland at light microscopy level. Ultrastructural observations revealed the presence of terminal endings containing 5HT and localized mainly in the perivascular spaces, labeled with dense diaminobenzidine (DAB) deposits.