Cell Tiss. Res. 178, 323
332 (1977)
Cell and Tissue Research ,~: by Springer-Verlag 1977
The Effects of Different Methods of Fixation on Central Nervous System Synaptic Pinocytotic Vesicles M. Paula-Barbosa* Institute of Anatomy, Medical School, University of Oporto, Oporto, Portugal
9M.A. Sobrinho-Sim6es* Laboratory of Pathology, Medical School, University of Oporto, Oporto, Portugal
E.G. Gray* Department of Anatomy and Embryology, University College of London, London, England
Summary. Synaptic pinocytotic vesicles (invaginating from the surface membrane) and coated vesicles inside rat mossy fiber endings were counted after the use of different kinds of fixatives. Significantly greater numbers of pinocytotic vesicles and coated pinocytotic vesicles per unit length of membrane were found when osmium was used as the first fixative. A high positive correlation was found between these values and the number of coated vesicles per unit area of mossy fiber ending profiles. These results emphasize the need for caution when considering the theory that in vivo synaptic vesicle recycling involves a coated vesicle invagination of the surface membrane followed by internalisation and loss of coat of the vesicle.
Key words: Pinocytotic vesicles - Nervous system - Methods of fixation - Origin of synaptic vesicles.
Introduction Since Gray (1961) first noticed the existence of coated vesicles in the mossy fiber endings of rat cerebellum, several hypotheses have been advanced concerning their origin and function. Evidence pointing to the pinocytotic origin of coated vesicles was first presented by Roth and Porter (1962, 1964). These authors showed that the omega-shaped figures that resulted from micropinocytotic activity of the plasmalemma of Aedes aegypti oocyte were surrounded by a bristle-like Send of[print requests to." Dr. M. Paula-Barbosa, Institute of Anatomy, Oporto Medical School,
Oporto, Portugal * The authors are indebted to Mrs. M.L Brito and M.M. Pacheco and Mr. L.B. Nunes for technical assistance. This work has been supported by I.A.C. (Lisbon)
324
M. Paula-Barb0saet al.
material; furthermore, they suggested that the mechanical properties of this cytoplasmic material might contribute to the deformation of the plasmalemma. Andr6s (1964) suggested that coated vesicles may be related to reabsorption and transport of macromolecules, namely of synaptic transmitters. In 1965 Westrum reinforced the idea of the micropinocytotic origin of coated vesicles and proposed that they could give rise to the agranular synaptic vesicles. Later Kanaseki and Kadota (1969) pointed out that a flat coat of cytoplasmic material formed of hexagons could deform the plasmalemma if a rearrangement of these hexagons by interspersion of pentagons took place, in a way comparable to the formation of geodesic domes. They postulated that coated vesicles, once detached from the plasma membrane, would give origin to agranular synaptic vesicles by discarding their coats. Recent observations of Gray (1972, 1973) threw doubt on this theory, suggesting that the coats of coated vesicles could be artifactual condensations of the proteinaceous material which lies around the organelles of the presynaptic bag. More recently we have shown that in mossy fiber terminals the percentage of coated vesicles varies considerably with different methods of fixation (Paula-Barbosa and Gray, 1974). The purpose of this work was to determine whether different conditions of fixation also influence pinocytotic activity of the neuronal membrane and if so, whether there is any correlation between the number of pinocytotic vesicledeformations of the plasmalemma and the number of the coated vesicles, in an attempt to contribute to the explanation of the controversial origin of synaptic vesicles.
Materials and Methods Observations were made on adult male albino rats and confined to the mossy fiber endings of the vermis of the cerebellar cortex (lobules 4-6 of Larsell, 1952). Fixation and staining methods included : MethodA1. Formaldehyde-Glutaraldehyde-Osmium. A transcardial perfusion with a duration of 15 min was carried out with 4% formaldehyde prepared from paraformaldehyde and 0.5% glutaraldehyde in Sorensen's phosphate buffer with an osmolarity of 1550 m O s m , at pH 7.2. The appropriate lobules were excised and diced on a wax block into small pieces less than 1 m m in diameter and immersed in 1% osmium tetroxide, buffered with veronal acetate at pH 7.2, for two hours. Pieces were rinsed in 25% ethanol and dehydrated in a graded series of ethanols. Block staining took place at the 70% stage, using 1% uranyl acetate. After passage through propylene oxide and propylene oxide/Aratdite the tissue was embedded in Araldite. The sections were stained with lead citrate. Method A2. Small pieces were removed from the anesthetised animal, cut up and directly immersed in the pH 7.2 aldehyde fixative for 2 h. Subsequent processing followed method A l . Method A3. As in method A2, but the time of immersion in aldehyde was 20 h. Method A4. As in method A2 except that the fixative was reduced to pH 4 by the addition of approximately 5 ml of 1N HC1 to 200 ml of the fixative. Method B. Osmium-PTA. Small pieces were immersed in 1% o s m i u m tetroxide buffered with veronal acetate for 2 h and block stained at the absolute ethanol stage with 1% phosphotungstic acid.
Methods of Fixation and Pinocytotic Vesicles
325
Method C. 4% Unbuffered Osmium-Aldehyde: method of Kanaseki and Kadota (1969). Diced pieces were immersed for 2 h in 4% unbuffered osmium tetroxide, followed by 12% unbuffered glutaraldehyde. Blocks were stained for 2 h with a solution of 2% aqueous uranyl acetate. The sections were stained with lead citrate. Method D. 4% Unbuffered Osmium. As for method C except that the glutaraldehyde stage was omitted. To obtain unbiased sampling, mossy endings were selected on the viewing screen at low magnification ( x 4,000) at which it is impossible to distinguish the different vesicular populations. Profiles that just fill the plate area of the viewing screen were selected and then the magnification was increased to ( x 20,000) and the photographic exposure made. The total length of terminal membrane was measured using a mapmeter, while the surface of the terminal profiles was measured with a planimeter. All the vesicles of each profile were counted. A structure was considered to be a pinocytotic vesicle when an omega-shaped figure was observed, i.e., only when an uninterrupted connection between plasmalemma and vesicular membrane was clearly recognized. A vesicle was counted as a coated vesicle or as a pinocytotic coated vesicle when over half of its profile was coated. Data from each group were averaged, and the standard deviation and standard error were calculated. In order to compare the results, Student's two-sided test was used. Two means were considered significantly different if the probability of error (p) was smaller than 0.05 and suggestively different if probability was smaller than 0.1.
Results
In all groups it was possible to observe pinocytotic vesicles with or without cytoplasmic coats (Figs. 1 3). The morphological characteristics of these vesicles are similar to those described by Bunt (1969) and their coats clearly resembled those described in coated vesicles by Gray (1961). The numbers of coated vesicles and of total plain and coated vesicles per square micron of the mossy fiber ending profiles is expressed in Table 1,t together with the percentage of coated vesicles in relation to the total number of plain and coated vesicles. The number of coated vesicles per unit area of the mossy fiber profiles is closely related to the percentage of coated vesicles since the total number of plain and coated vesicles per unit area of the mossy terminals is quite similar in all the groups with the exception of group A1 (Table 1). As was previously stressed (Paula-Barbosa and Gray, 1974) the most important differences are found when initial aldehyde fixation is compared with initial osmium fixation, but within the groups having osmium as the first fixative (B, C and D) significant differences were also found. The number of coated vesicles per square micron of mossy fiber ending profiles was significantly (p 150 .u 125
/
C
OB
o
oD
Bo --
Co
375
o 10o
.~%~
~ 30.01
~"/
o225--
0.75
Al~
/ ~
o
oo0.50
150-
/
OA2
o
"~ 0.25 0
z
0
/ "
AL AI O
O
7.5-
oA2 OA31
i
,
]
[~_
0
,/"
~
/
AL O
A3 l
10
1
,
I
1
0 025 050 075 1.00 125 1.50 1.75 200 2.25 250 2.75 3.00 Percentage of coated vesicles (%)
0 0.25 0.50 0.75 1.00 125 1.50 1.75 2.00 225 Number of coated vesicles (n~ Fig. 4. Graphic expression of the correlation between coated pinocytotic vesicles and coated vesicles
that was used for fixation, upon the duration of fixation and finally upon the pH of the fixative. Concerning the method of fixation it must be stressed that group A1 (perfusion) showed significantly smaller numbers of total pinocytotic vesicles and uncoated pinocytotic vesicles per unit length of membrane (0.40 +_0.13 and 0,33 + 0.13, respectively) when compared with the other aldehyde groups (immersion). The remarkable interindividual variation that was found in group Al probably depends upon the relative success of perfusion that could be achieved and makes difficult its comparison with the other aldehyde groups. The difference of the duration of fixation, on the other hand, was not a significative one, for the results obtained in group A2 (2 h in aldehyde, pH 7.2) were only slightly smaller than those of group A3 (20 h in aldehyde pH 7.2) with the exception of the number of uncoated pinocytotic vesicles per 100 microns of membrane that was suggestively smaller in group A2 (0.96+_0.12) than in group A3 (1.34+0.16). Finally concerning the pH of the fixative a significantly greater value of uncoated pinoctotic vesicles per 100 microns of membrane was found in group A2 (1.47+-0.09 and 0.96+-0.12, respectively). A high positive correlation was found in the comparison of the number of coated vesicles per square micron of mossy fiber ending profiles with the number of pinocytotic and coated pinocytotic vesicles per 100 microns of membrane (r= 0.634 - p < 0.2 and r = 0 . 7 9 7 - p < 0 . 0 5 , respectively). On the other hand, similar positive correlation were found in the comparison of the percentage of coated vesicles in relation to the total number of synaptic vesicles with the number of total pinocytotic and coated pinocytotic vesicles per unit length of membrane (r=0.793-p
332 (1977)
Cell and Tissue Research ,~: by Springer-Verlag 1977
The Effects of Different Methods of Fixation on Central Nervous System Synaptic Pinocytotic Vesicles M. Paula-Barbosa* Institute of Anatomy, Medical School, University of Oporto, Oporto, Portugal
9M.A. Sobrinho-Sim6es* Laboratory of Pathology, Medical School, University of Oporto, Oporto, Portugal
E.G. Gray* Department of Anatomy and Embryology, University College of London, London, England
Summary. Synaptic pinocytotic vesicles (invaginating from the surface membrane) and coated vesicles inside rat mossy fiber endings were counted after the use of different kinds of fixatives. Significantly greater numbers of pinocytotic vesicles and coated pinocytotic vesicles per unit length of membrane were found when osmium was used as the first fixative. A high positive correlation was found between these values and the number of coated vesicles per unit area of mossy fiber ending profiles. These results emphasize the need for caution when considering the theory that in vivo synaptic vesicle recycling involves a coated vesicle invagination of the surface membrane followed by internalisation and loss of coat of the vesicle.
Key words: Pinocytotic vesicles - Nervous system - Methods of fixation - Origin of synaptic vesicles.
Introduction Since Gray (1961) first noticed the existence of coated vesicles in the mossy fiber endings of rat cerebellum, several hypotheses have been advanced concerning their origin and function. Evidence pointing to the pinocytotic origin of coated vesicles was first presented by Roth and Porter (1962, 1964). These authors showed that the omega-shaped figures that resulted from micropinocytotic activity of the plasmalemma of Aedes aegypti oocyte were surrounded by a bristle-like Send of[print requests to." Dr. M. Paula-Barbosa, Institute of Anatomy, Oporto Medical School,
Oporto, Portugal * The authors are indebted to Mrs. M.L Brito and M.M. Pacheco and Mr. L.B. Nunes for technical assistance. This work has been supported by I.A.C. (Lisbon)
324
M. Paula-Barb0saet al.
material; furthermore, they suggested that the mechanical properties of this cytoplasmic material might contribute to the deformation of the plasmalemma. Andr6s (1964) suggested that coated vesicles may be related to reabsorption and transport of macromolecules, namely of synaptic transmitters. In 1965 Westrum reinforced the idea of the micropinocytotic origin of coated vesicles and proposed that they could give rise to the agranular synaptic vesicles. Later Kanaseki and Kadota (1969) pointed out that a flat coat of cytoplasmic material formed of hexagons could deform the plasmalemma if a rearrangement of these hexagons by interspersion of pentagons took place, in a way comparable to the formation of geodesic domes. They postulated that coated vesicles, once detached from the plasma membrane, would give origin to agranular synaptic vesicles by discarding their coats. Recent observations of Gray (1972, 1973) threw doubt on this theory, suggesting that the coats of coated vesicles could be artifactual condensations of the proteinaceous material which lies around the organelles of the presynaptic bag. More recently we have shown that in mossy fiber terminals the percentage of coated vesicles varies considerably with different methods of fixation (Paula-Barbosa and Gray, 1974). The purpose of this work was to determine whether different conditions of fixation also influence pinocytotic activity of the neuronal membrane and if so, whether there is any correlation between the number of pinocytotic vesicledeformations of the plasmalemma and the number of the coated vesicles, in an attempt to contribute to the explanation of the controversial origin of synaptic vesicles.
Materials and Methods Observations were made on adult male albino rats and confined to the mossy fiber endings of the vermis of the cerebellar cortex (lobules 4-6 of Larsell, 1952). Fixation and staining methods included : MethodA1. Formaldehyde-Glutaraldehyde-Osmium. A transcardial perfusion with a duration of 15 min was carried out with 4% formaldehyde prepared from paraformaldehyde and 0.5% glutaraldehyde in Sorensen's phosphate buffer with an osmolarity of 1550 m O s m , at pH 7.2. The appropriate lobules were excised and diced on a wax block into small pieces less than 1 m m in diameter and immersed in 1% osmium tetroxide, buffered with veronal acetate at pH 7.2, for two hours. Pieces were rinsed in 25% ethanol and dehydrated in a graded series of ethanols. Block staining took place at the 70% stage, using 1% uranyl acetate. After passage through propylene oxide and propylene oxide/Aratdite the tissue was embedded in Araldite. The sections were stained with lead citrate. Method A2. Small pieces were removed from the anesthetised animal, cut up and directly immersed in the pH 7.2 aldehyde fixative for 2 h. Subsequent processing followed method A l . Method A3. As in method A2, but the time of immersion in aldehyde was 20 h. Method A4. As in method A2 except that the fixative was reduced to pH 4 by the addition of approximately 5 ml of 1N HC1 to 200 ml of the fixative. Method B. Osmium-PTA. Small pieces were immersed in 1% o s m i u m tetroxide buffered with veronal acetate for 2 h and block stained at the absolute ethanol stage with 1% phosphotungstic acid.
Methods of Fixation and Pinocytotic Vesicles
325
Method C. 4% Unbuffered Osmium-Aldehyde: method of Kanaseki and Kadota (1969). Diced pieces were immersed for 2 h in 4% unbuffered osmium tetroxide, followed by 12% unbuffered glutaraldehyde. Blocks were stained for 2 h with a solution of 2% aqueous uranyl acetate. The sections were stained with lead citrate. Method D. 4% Unbuffered Osmium. As for method C except that the glutaraldehyde stage was omitted. To obtain unbiased sampling, mossy endings were selected on the viewing screen at low magnification ( x 4,000) at which it is impossible to distinguish the different vesicular populations. Profiles that just fill the plate area of the viewing screen were selected and then the magnification was increased to ( x 20,000) and the photographic exposure made. The total length of terminal membrane was measured using a mapmeter, while the surface of the terminal profiles was measured with a planimeter. All the vesicles of each profile were counted. A structure was considered to be a pinocytotic vesicle when an omega-shaped figure was observed, i.e., only when an uninterrupted connection between plasmalemma and vesicular membrane was clearly recognized. A vesicle was counted as a coated vesicle or as a pinocytotic coated vesicle when over half of its profile was coated. Data from each group were averaged, and the standard deviation and standard error were calculated. In order to compare the results, Student's two-sided test was used. Two means were considered significantly different if the probability of error (p) was smaller than 0.05 and suggestively different if probability was smaller than 0.1.
Results
In all groups it was possible to observe pinocytotic vesicles with or without cytoplasmic coats (Figs. 1 3). The morphological characteristics of these vesicles are similar to those described by Bunt (1969) and their coats clearly resembled those described in coated vesicles by Gray (1961). The numbers of coated vesicles and of total plain and coated vesicles per square micron of the mossy fiber ending profiles is expressed in Table 1,t together with the percentage of coated vesicles in relation to the total number of plain and coated vesicles. The number of coated vesicles per unit area of the mossy fiber profiles is closely related to the percentage of coated vesicles since the total number of plain and coated vesicles per unit area of the mossy terminals is quite similar in all the groups with the exception of group A1 (Table 1). As was previously stressed (Paula-Barbosa and Gray, 1974) the most important differences are found when initial aldehyde fixation is compared with initial osmium fixation, but within the groups having osmium as the first fixative (B, C and D) significant differences were also found. The number of coated vesicles per square micron of mossy fiber ending profiles was significantly (p 150 .u 125
/
C
OB
o
oD
Bo --
Co
375
o 10o
.~%~
~ 30.01
~"/
o225--
0.75
Al~
/ ~
o
oo0.50
150-
/
OA2
o
"~ 0.25 0
z
0
/ "
AL AI O
O
7.5-
oA2 OA31
i
,
]
[~_
0
,/"
~
/
AL O
A3 l
10
1
,
I
1
0 025 050 075 1.00 125 1.50 1.75 200 2.25 250 2.75 3.00 Percentage of coated vesicles (%)
0 0.25 0.50 0.75 1.00 125 1.50 1.75 2.00 225 Number of coated vesicles (n~ Fig. 4. Graphic expression of the correlation between coated pinocytotic vesicles and coated vesicles
that was used for fixation, upon the duration of fixation and finally upon the pH of the fixative. Concerning the method of fixation it must be stressed that group A1 (perfusion) showed significantly smaller numbers of total pinocytotic vesicles and uncoated pinocytotic vesicles per unit length of membrane (0.40 +_0.13 and 0,33 + 0.13, respectively) when compared with the other aldehyde groups (immersion). The remarkable interindividual variation that was found in group Al probably depends upon the relative success of perfusion that could be achieved and makes difficult its comparison with the other aldehyde groups. The difference of the duration of fixation, on the other hand, was not a significative one, for the results obtained in group A2 (2 h in aldehyde, pH 7.2) were only slightly smaller than those of group A3 (20 h in aldehyde pH 7.2) with the exception of the number of uncoated pinocytotic vesicles per 100 microns of membrane that was suggestively smaller in group A2 (0.96+_0.12) than in group A3 (1.34+0.16). Finally concerning the pH of the fixative a significantly greater value of uncoated pinoctotic vesicles per 100 microns of membrane was found in group A2 (1.47+-0.09 and 0.96+-0.12, respectively). A high positive correlation was found in the comparison of the number of coated vesicles per square micron of mossy fiber ending profiles with the number of pinocytotic and coated pinocytotic vesicles per 100 microns of membrane (r= 0.634 - p < 0.2 and r = 0 . 7 9 7 - p < 0 . 0 5 , respectively). On the other hand, similar positive correlation were found in the comparison of the percentage of coated vesicles in relation to the total number of synaptic vesicles with the number of total pinocytotic and coated pinocytotic vesicles per unit length of membrane (r=0.793-p
