- Email: [email protected]
Relationship between lipofuscin granules and polyunsaturated fatty acids in the housefly, Musca domestica
Mechanisms of Ageing and Development, 25 (1984) 355-363
355
Elsevier ScientificPublishers Ireland Ltd.
RELATIONSHIP BETWEEN LIPOFUSCIN GRANULES AND POLYUNSATURATED FATTY ACIDS IN THE HOUSEFLY, MUSCA DOMESTICA
R.S. SOHAL Department of Biology, Southern Methodist University, Dallas, Texas 75275 (U.&A.) R.G. BRIDGES and E.A. HOWES ARC Unit of Invertebrate Chemistry and Physiology, Department of Zoology, University of Cambridge, Cambridge, CB2 3EJ (Great Britainj
(Received September 13th, 1983) SUMMARY The fractional volume occupied by lipofusein granules in epithelial cells of the midgnt or oenocytes of abdominal fat body of 3.day-old and 13-day-old male houseflies was determined in two groups of flies by electron microscopic morphometry. One group had developed from larvae reared on diets containing no added polyunsaturated fatty acids and the second from larvae reared on diets containing added linoleic acid. No polyunsaturated fatty acids could be detected in the lipids of the first group of flies using a method which would have detected their presence in amounts greater than 0.1% of the total esterified fatty acids. The second group contained at least two hundred times more than this minimal level. The volume of lipofuscin granules increased significantly (p < 0.01) (about threefold for the fat body and twofold in midgut cells) between 3 days and 13 days of age but no statistically significant difference was seen between the two groups of flies at the same age. The results show that if lipofuscin formation depends on the oxidation of polyunsaturated fatty acids in the housefly, then extremely small amounts of the acids are involved which lie below the detection limit of the methods employed. The age-associated small increase of extractable fluorescence seen previously in the linoleic acid group of flies is not associated with an increase in the lipofuscin granules. Key words." Insects;Ageing; Age pigment; Cellular ageing; Lipofuscin
INTRODUCTION One of the most consistent manifestations of ageing in the long-rived, non-dividing cells of multicelluhr organisms of widely divergent phylogeny is the progressive accumu0047-6374/84/$03.00 Printed and Publishedin Ireland
© 1984 ElsevierScientificPublishers Ireland Ltd.
356 lation of colored, autofluorescent bodies seen in tissue section, which are generally referred to as lipofuscin granules. Such granules have been shown to contain various lysosomal enzymes, and morphological-cytochemical studies on a variety of cell types suggest that the granules contain undigested residues resulting from cytoplasmic autophagocytosis involving lysosomal lyric activity [1,2]. The fluorescence of compounds extractable from tissues using non-polar solvents and having properties similar to the fluorescence seen in tissue section has been shown to increase with age [3,4]. There has been a tendency to assume that this material is derived from the lipofuscin granules detected morphologically in spite of the fact that there is little evidence that extraction with organic solvents has a detectable effect on the fluorescence of these granules [5-8]. These extracted fluorescent compounds have spectral properties typical of Schiff bases and it has been postulated that such compounds could result from the peroxidation of polyunsaturated long-chain fatty acids which gives rise to malondialdehyde. This could react with free amino groups of proteins, nucleic acids and phospholipids [3,4,9]. However, the supportive evidence for this postulate is based primarily on in vitro studies [7,10] and it is not at all certain that the fluorescent material extracted by non-polar solvents really does represent that seen in tissue sections. Our earlier work [1 1] has shown that in houseflies such extractable material is a complex mixture of compounds most of which is formed and increases with age in the absence of any source of polyunsaturated fatty acids. Statistical analysis of these data did suggest that when linoleic acid was available there was a small increase in the age-related extractable fluorescence and it was poss~le that this increase might have been associated with the lipofuscin granules. The present paper is an extension of these earlier studies to see if there is evidence to support such an interpretation. Studies by one of us have shown the age-associated accumulation of certain structures in fat body oenocytes and in midgut epithelium of the housefly. These structures have been identified as lysosomal, autofluorescent, osmiophilic and are similar in morphology to the lipofuscin granules seen in vertebrates [5,12,13]. Houseflies are unable to synthesize polyunsaturated fatty acids [14]. By rearing larvae on defined diets, it is possible to obtain adult houseflies which contain either no detectable levels of polysaturated fatty acids, or high levels of linoleic acid esterified in their lipids [1 1]. Using a morphometric technique for estimating the content of lipofuscin granules in the fat body and midgut tissues from such houseflies, the dependence of the formation and accumulation of the age-associated granules on the presence of polyunsaturated fatty acids was investigated. MATERIALSAND METHODS Rearing of insects Housefly larvae were raised aseptically on a basic diet containing casein, choline, salt mixture, r~onucleic acid, agar, methyl-p-hydroxybenzoate, vitamin solution, and
357 cholesterol according to the procedure described previously [15]. When included, linoleic acid (c/s-9-c/s-12-octadecadienoic acid, approximately 99% pure, Sigma, London) was converted to its sodium salt and added to the larval nutrient mixture at a concentration of 100 #mol/g casein which was then sterilized in a 250 ml flask at 120°C for 15 min. Approximately 200 eggs, surface-sterilized in 2.5% formalin, were placed in each flask and maintained at 28°C. Larvae were grown on these diets for 6 days, removed from the medium and washed in distilled water and allowed to pupate in covered petri dishes at 28°C. After eclosion, the adult flies were housed in 0.03 cubic meter cages at 25°C and fed on glucose and water.
Electron microscopy Lipofuscin volume was measured morphometrically in the oenocytes in the abdominal fat body and epithelial cells of the midgut. The middle part of the midgut, and fat body from the posterodorsal region of the abdomen from male flies was fixed in 2.5% phosphate-buffered glutaraldehyde. Following post fixation in 1% phosphate-buffered osmium tetroxide, tissues were dehydrated in ethyl alcohol and embedded in Maraglas. Silver-grey sections of three randomly selected tissue blocks belonging to six different flies, were stained with uranyl acetate and lead citrate, and were randomly photographed at a magnification of about 10 000 by a Hitachi HUllB electron microscope. The negatives were further enlarged 2.5 times. The fractional area of the cytoplasm occupied by lipofuscin was measured by point counting method [16] from 25-30 electron micrographs from each group. Student's t test was used to determine the significance of the differences between the groups. Differences were considered significant when p < 0.05.
Detection of trace amounts of polyunsaturated fatty acids (PUFAsJ Standard methods for methylation and separation of methylated fatty acids by gasliquid chromatography (GLC) were used [14]. As previously reported [11], no detectable amounts of any PUFAs were found in houseflies reared from larvae fed on diets to which no fatty acids had been added. The addition of a range of amounts of linoleic, ^f-linolenic and arachidonic acids (Sigma, London) to aliquots of lipid extract of a similar size to that routinely used showed that the mimimum amounts of the three PUFAs added that could be detected was between 0.05% to 0.1% of the fatty acid content of the sample. Samples of casein, RNA, and agar were extracted with 2 : l(v/v) chloroform-methanol (1 g per 10 ml) and the washed extract was used for methylation and GLC analysis. Cholesterol was refluxed with 1 M KOH in ethanol (1 g cholesterol per 6 ml) for 1.5 h and then partitioned between water and diethyl ether. The water layer was acidified with 6 M HC1 and extracted with diethyl ether. The diethyl ether extract was methylated and used for GLC analysis. Washed and surface-sterilized housefly eggs [15] were dried on filter paper and weighed. These were homogenized in 2 : 1 (v/v) chloroform-methanol (1 g per 18 ml),
358 centrifuged and the extract was washed [15]. Samples of eggs were counted and weighed to give a mean weight of 70/~g/egg. This figure was used to calculate the number of eggs extracted. Aliquots of the washed extract were used for analysis of fatty acids by GLC. RESULTS
Presence o f trace amounts o f PUFAs in fly eggs and dietary components Small amounts of PUFAs were detected in the fly egg extracts and in the dietary components. Table I shows the results of these analyses expressed as a percentage of the esterified fatty acids present in the 3-day-old adult male housefly. It was assumed that the PUFAs found per egg were carried through the larval and pupal stages to the adult without loss. The small amounts of PUFAs detected in the larval dietary constituents were assumed to be completely taken up and divided between the 200 larvae and that they were then taken through to the adult without loss. Such figures must represent the upper limit of PUFA contamination in the adult fly and are below the limit of detection established for the method used (0.05-0.10% of total fatty acid in the sample). Thus the linoleic acid present in the flies from larvae fed on diets containing added linoleic acid [11] represents at least two hundred times the maximum possible concentration in the adults from larvae fed on diets to which no PUFA had been added.
TABLE I POLYUNSATURATED FATTY ACIDS DETECTED IN HOUSEFLY EGGS AND IN VARIOUS COMPONENTSOF THE LARVALDIET Results are expressed as a percentage of the total esterified fatty acids present in the 3-day-oldmale housefly (195 ~g fatty acid per insect).
Source
Percentage of fatty acid in adult Linoleic acid
"r-Ltnolenic acid
Arachidonic acid
Eggsa Caseinb Agar b
0.0136 0.0135 0.0002
0.0038 0.0198 0.0001
0,0076 0 0
RNAb Cholesterolb Total
0.0007
0.0005
0
0
0
0
0.0280
0.0242
0.0076
aIt was assumed that the PUFAs detected in the egg were carried through to the adult fly without loss. bThe PUFAs detected in the various dietary components were assumed to be taken up completelyby the larvae roared on the diet (200) and that they were taken through to the adult fly without loss.
359
L~ofuscin volume Fat body. The abdominal fat body in Musca domestica contains two different cell types the fat cells and the oenocytes [12]. The cytoplasm of the oenocytes is rich in free polysomes and mitochondria and contains appreciable amounts of both agranular and granular endoplasmic reticulum. In both groups, a comparison between 3-day-old and 13-day-old flies indicated an age-associated increase in content of lipofuscin granules (Fig. 1). Morphologically, the granules were easily identifiable as membrane-bounded structures containing electron-dense material intermingled with pleomorphic membranous component. Occasionally, organelles such as ribosomes and cisternae of endoplasmic reticulum were identifiable within the confines of the lipofuscin granule. That these granules in the oenocytes are lysosomal in nature was demonstrated previously by the positive localization of the reaction product of fl-glycerophosphatase activity [12]. A quantitative comparison of lipofuscin in the oenocytes of the two groups of flies -
Fig. 1. An electron micrograph of a section through the abdominal fat body of a 13-day-oldlinoleic acid-free housefly showing lipofuscin granules (L) in the oenocytes, x 18 920. Bar = 1.0 #m.
360 TABLE II THE PERCENTAGE OF CYTOPLASMICVOLUME OCCUPIED BY LIPOFUSCINGRANULES IN OENOCYTES OF THE FAT BODY AND EPITHELIALCELLS OF THE MIDGUTOF 3-DAY-OLD AND 13-DAY-OLDMALEHOUSEFLIES Group 1 : linoleic acid-containing adult houseflies. Group 2: insects containing leasthan 0.1% of their fatty acids as polyunsaturates. Values axe mean ± S.E.M. Measurements were made on sections from three randomly selected blocks from six flies.
Group
1 2
Age (days after emergence of adult) 3 13 3 13
Lipofuscin {percentagecytoplasmic volume) Fat body
Midgut
2.13 ± 0.16 6.23 ± 0.80 1.88 ± 0.23 6.19 ± 0.70
2.87 ± 0.48 6.00 ± 1.10 3.61 + 0.46 6.10 ± 0.74
(Table II) indicated that 13-day-old flies in both groups contained significantly greater amounts of lipofuscin in their eytopla~n than the 3.day-old flies (p < 0.01); however, the two groups did not exhibit any statistically significant differences at either of these ages. Thus the increase in the percentage volume of cell cytoplasm occupied by the granules that occurred between 3 and 13 days of age was similar in both groups. Midgut. The midgut comprises the longest component of the alimentary tract in the housefly. The epithelial cells form a single cell thick layer which is separated from the lumen of the gut by a peritrophic membrane. The epithelial cells pt~rsist throughout the lifespan of the adult fly and exh~ited no manifestations of cell turnover [17]. The structural organization of the epithelial cells in the region of the midgut examined was quite uniform. The cytoplasm is rich in granular endoplasmic reticulum, free polysomes and numerous elongated mitochondria. The Golgi complex and the associated primary lysosomes are well developed. The percentage of the cytoplasm occupied by the lipofuscAn granules of the cells increased significantly (p < 0.01) from 3 to 13 days of age in both groups (Fig, 2, Table II). There were, however, no significant differences either in the age-associated increase or th percentage volume occupied by the granules at these ages between the two groups of flies. No major differences in the morphological appearance of the granules were noticed between either group of flies.
361
Fig. 2. Section through an epithelial cell of the midgut of a 13-day-old linoleic add-free housefly showing lipofuscin granules (L). × 18 920. Bar = 1.0 urn.
DISCUSSION The present results based on the estimation of lipofuscin granule volume by morphometric techniques clearly show that the formation and the age-associated accumulation of such granules in two cell types of the housefly occur even when the level of any PUFA is below that which can be detected using the methods used (<0.1% of the total esterified fatty acids). They also show that the accumulation is not enhanced by the presence of the linoleic acid in the fly. At Present two different methods are used for the quantification of lipofuscin granules. One, the older method, employs the technique used in this study, while the other depends on the estimation of levels of fluorescence present in washed chloroformmethanol extracts. The second method was originally used by Tappel [3,4] and was used by us in a previous study on houseflies [1 1]. In the case of the housefly both methods can be interpreted as giving essentially the same result, in that the age-associated
362 accumulation of granules and the increase of extractable fluorescent material with age occur in the absence of detectable levels of PUFAs in the organism. However, the maall increase in the amount of extractable fluorescence which occurs with age when linoleic acid is available [ 11 ] is not accompanied by an increase in the percentage of lipofuscin granules in the cytoplasm. If the peroxidation of PUFAs is involved in the formation of lipofuscin granules as has been outlined by Tappel [3,4], then the present results indicate that only very small amounts of PUFAs are involved and these lie below the present methods of detection. It would seem unlikely that any trace amounts of PUFA.~ which might have been carried over from the ~.~,g or derived from the larval diet would be adequate to maintain the continuing formation of fluorescent material with increasing age of the fly. Such a conclusion is supported by the findings of Gutteridge [18] that PUFAs are not the only source of malondialdehyde, which can also be derived from amino acids, nucleic acids and sugars. It also seems unlikely that the peroxidation of PUFAs is the main or only pathway for the formation of the fluorescent material extracted with chloroform-methanol from adult houseflies. Much of this material must be derived from non-lipid sources or, if from a lipid source, then from a monounsaturated fatty acid. However, this latter possibility is unlikely, as it has been shown that oleic acid does not produce malondialdehyde on peroxidation [19]. Whether the age-associated increase in extractable fluorescence results from the actual increase in fluorescent granules or whether it is derived from other sources cannot be clearly decided at this stage of the research. ACKNOWLEDGEMENTS This research was partially supported by grants to R,S.S. from the National Institutes of Health, National Institute on Aging (R01 AG00171) and the Glenn Foundation for Medical Research.
REFERENCES 1 D. Brandes, Lysosomes and age pigment. In P.L. Krohn (ed.), Topics in the Biology of,4gln~, Interscience, New York, 1966, pp. 149-158. 2 C. DeDuve and R. Wattiaux, Functions of lysosomes. Annu. Rev. Physiol., 28 (1966) 435-491. 3 A.L. Tappel, Lipid peroxidation and fluorescent molecular damage to membranes. In B.F. Trump and A.V. Arstila (eds.), Pathobloiogy o f Cell Membranes, Vol. I, Academic Press, New York, 1975, pp. 145-170. 4 A.L. Tappel, Measurement and protection from in rive lipid peroxidation. In W.A. Pryor (ed.), Free Radicals in Biology, Vol. IV, Academic Press, New York, pp. 1-47. 5 H. Dormto and R.S. Sohal, Age-related changes in lipofuscin associated fluorescent substances in the adult male housefly, Musca domestica. Exp. Gerontol., 12 (1978) 171-179. 6 B.L. Strehler, On the histochemistry .and ultrastructure of age pigment. Adv. Gerontol. Res., 1 (1964) 343-384.
363 7 M. Elleder, So-caUed neuronal ceroid-lipofuscinosis. Histochemical study with evidence of extractability of the stored material. Acta Neuropathol., 38 (1977) 117-122. 8 M. EUeder, Chemical characterization of age pigments. In R.S. Sohal (ed.), Age Pigments, Elsevier/ North-Holland, Amsterdam, pp. 203-241. 9 J. Miquel, J. Oro, T.G. Bensch and J.E. Johnson, Lipofuscin: fine structural and biochemical studies. In W.A. Pryor (ed.), Free Radicals in Biology, Vol. III, Academic Press, New York, 1975, pp. 133-182. 10 H. Donato, Lipid peroxidation, cross-linking reactions and aging. In R.S. Sohal (ed.), Age Pigments, Elsevier/North-Holland, Amsterdam, 1981, pp. 63-81. I 1 R.G. Bridges and R.S. Sohal, Relationship between age-associated fluorescence and linoleic acid in the housefly, Musca domestica. Insect Bioche~, 10 (1980) 557-562. 12 R.S. Sohal, Fine structural alterations with age in the fat body of the adult male housefly, Musca domestica. Z. Zellforsch. Microsk. Anat., 140 (1973) 169-175. 13 R.S. Sohal, Relationship between metabolic rate, lipofuscin accumulation and lysosomal enzyme activity during aging in the adult housefly, Musca domestica. Exp. Gerontol., 16 (1981) 347-355. 14 R.G. Bridges, Incorporation of fatty acids into the lipids of the housefly, Musca domestica. J. Insect Physiol., 17 (1971) 881-895. 15 R.G. Bridges, J. Rickets and J.T. Cox, The replacement of lipid-bound choline by other bases in the phospholipids of the housefly, Musca domestica. J. Insect Physiol., 11 (1965) 225-236. 16 E.R. Weibel, Stereological principles for morphometry in electron microscopic cytology. Int. Rev. Cytol., 126 (1969) 235-302. 17 R.S. Sohal, P.D. Peters and T.H. Hall, Origin, structure, composition and age-dependence of mineralized dense bodies (concretions) in the midgut epithelium of the adult housefly, Musca domestica. Tissue Cell, 9 (1977) 87-102. 18 J.M.C. Gutteridge, Free radical damage to lipids, amino acids, carbohydrates and nucleic acids determined by thiobarbituric acid reactivity. Int. J. Biochem., 245 (1970) 4641-4646. 19 M.C. McArthur and R.S. Sohal, Relationships between metabolic rate, aging, lipid peroxidation and fluorescent age pigment in milkweed bug, Oncopdtus fasciatus (Hemiptera). J. Gerontol., 37 (198D 268-274.
355
Elsevier ScientificPublishers Ireland Ltd.
RELATIONSHIP BETWEEN LIPOFUSCIN GRANULES AND POLYUNSATURATED FATTY ACIDS IN THE HOUSEFLY, MUSCA DOMESTICA
R.S. SOHAL Department of Biology, Southern Methodist University, Dallas, Texas 75275 (U.&A.) R.G. BRIDGES and E.A. HOWES ARC Unit of Invertebrate Chemistry and Physiology, Department of Zoology, University of Cambridge, Cambridge, CB2 3EJ (Great Britainj
(Received September 13th, 1983) SUMMARY The fractional volume occupied by lipofusein granules in epithelial cells of the midgnt or oenocytes of abdominal fat body of 3.day-old and 13-day-old male houseflies was determined in two groups of flies by electron microscopic morphometry. One group had developed from larvae reared on diets containing no added polyunsaturated fatty acids and the second from larvae reared on diets containing added linoleic acid. No polyunsaturated fatty acids could be detected in the lipids of the first group of flies using a method which would have detected their presence in amounts greater than 0.1% of the total esterified fatty acids. The second group contained at least two hundred times more than this minimal level. The volume of lipofuscin granules increased significantly (p < 0.01) (about threefold for the fat body and twofold in midgut cells) between 3 days and 13 days of age but no statistically significant difference was seen between the two groups of flies at the same age. The results show that if lipofuscin formation depends on the oxidation of polyunsaturated fatty acids in the housefly, then extremely small amounts of the acids are involved which lie below the detection limit of the methods employed. The age-associated small increase of extractable fluorescence seen previously in the linoleic acid group of flies is not associated with an increase in the lipofuscin granules. Key words." Insects;Ageing; Age pigment; Cellular ageing; Lipofuscin
INTRODUCTION One of the most consistent manifestations of ageing in the long-rived, non-dividing cells of multicelluhr organisms of widely divergent phylogeny is the progressive accumu0047-6374/84/$03.00 Printed and Publishedin Ireland
© 1984 ElsevierScientificPublishers Ireland Ltd.
356 lation of colored, autofluorescent bodies seen in tissue section, which are generally referred to as lipofuscin granules. Such granules have been shown to contain various lysosomal enzymes, and morphological-cytochemical studies on a variety of cell types suggest that the granules contain undigested residues resulting from cytoplasmic autophagocytosis involving lysosomal lyric activity [1,2]. The fluorescence of compounds extractable from tissues using non-polar solvents and having properties similar to the fluorescence seen in tissue section has been shown to increase with age [3,4]. There has been a tendency to assume that this material is derived from the lipofuscin granules detected morphologically in spite of the fact that there is little evidence that extraction with organic solvents has a detectable effect on the fluorescence of these granules [5-8]. These extracted fluorescent compounds have spectral properties typical of Schiff bases and it has been postulated that such compounds could result from the peroxidation of polyunsaturated long-chain fatty acids which gives rise to malondialdehyde. This could react with free amino groups of proteins, nucleic acids and phospholipids [3,4,9]. However, the supportive evidence for this postulate is based primarily on in vitro studies [7,10] and it is not at all certain that the fluorescent material extracted by non-polar solvents really does represent that seen in tissue sections. Our earlier work [1 1] has shown that in houseflies such extractable material is a complex mixture of compounds most of which is formed and increases with age in the absence of any source of polyunsaturated fatty acids. Statistical analysis of these data did suggest that when linoleic acid was available there was a small increase in the age-related extractable fluorescence and it was poss~le that this increase might have been associated with the lipofuscin granules. The present paper is an extension of these earlier studies to see if there is evidence to support such an interpretation. Studies by one of us have shown the age-associated accumulation of certain structures in fat body oenocytes and in midgut epithelium of the housefly. These structures have been identified as lysosomal, autofluorescent, osmiophilic and are similar in morphology to the lipofuscin granules seen in vertebrates [5,12,13]. Houseflies are unable to synthesize polyunsaturated fatty acids [14]. By rearing larvae on defined diets, it is possible to obtain adult houseflies which contain either no detectable levels of polysaturated fatty acids, or high levels of linoleic acid esterified in their lipids [1 1]. Using a morphometric technique for estimating the content of lipofuscin granules in the fat body and midgut tissues from such houseflies, the dependence of the formation and accumulation of the age-associated granules on the presence of polyunsaturated fatty acids was investigated. MATERIALSAND METHODS Rearing of insects Housefly larvae were raised aseptically on a basic diet containing casein, choline, salt mixture, r~onucleic acid, agar, methyl-p-hydroxybenzoate, vitamin solution, and
357 cholesterol according to the procedure described previously [15]. When included, linoleic acid (c/s-9-c/s-12-octadecadienoic acid, approximately 99% pure, Sigma, London) was converted to its sodium salt and added to the larval nutrient mixture at a concentration of 100 #mol/g casein which was then sterilized in a 250 ml flask at 120°C for 15 min. Approximately 200 eggs, surface-sterilized in 2.5% formalin, were placed in each flask and maintained at 28°C. Larvae were grown on these diets for 6 days, removed from the medium and washed in distilled water and allowed to pupate in covered petri dishes at 28°C. After eclosion, the adult flies were housed in 0.03 cubic meter cages at 25°C and fed on glucose and water.
Electron microscopy Lipofuscin volume was measured morphometrically in the oenocytes in the abdominal fat body and epithelial cells of the midgut. The middle part of the midgut, and fat body from the posterodorsal region of the abdomen from male flies was fixed in 2.5% phosphate-buffered glutaraldehyde. Following post fixation in 1% phosphate-buffered osmium tetroxide, tissues were dehydrated in ethyl alcohol and embedded in Maraglas. Silver-grey sections of three randomly selected tissue blocks belonging to six different flies, were stained with uranyl acetate and lead citrate, and were randomly photographed at a magnification of about 10 000 by a Hitachi HUllB electron microscope. The negatives were further enlarged 2.5 times. The fractional area of the cytoplasm occupied by lipofuscin was measured by point counting method [16] from 25-30 electron micrographs from each group. Student's t test was used to determine the significance of the differences between the groups. Differences were considered significant when p < 0.05.
Detection of trace amounts of polyunsaturated fatty acids (PUFAsJ Standard methods for methylation and separation of methylated fatty acids by gasliquid chromatography (GLC) were used [14]. As previously reported [11], no detectable amounts of any PUFAs were found in houseflies reared from larvae fed on diets to which no fatty acids had been added. The addition of a range of amounts of linoleic, ^f-linolenic and arachidonic acids (Sigma, London) to aliquots of lipid extract of a similar size to that routinely used showed that the mimimum amounts of the three PUFAs added that could be detected was between 0.05% to 0.1% of the fatty acid content of the sample. Samples of casein, RNA, and agar were extracted with 2 : l(v/v) chloroform-methanol (1 g per 10 ml) and the washed extract was used for methylation and GLC analysis. Cholesterol was refluxed with 1 M KOH in ethanol (1 g cholesterol per 6 ml) for 1.5 h and then partitioned between water and diethyl ether. The water layer was acidified with 6 M HC1 and extracted with diethyl ether. The diethyl ether extract was methylated and used for GLC analysis. Washed and surface-sterilized housefly eggs [15] were dried on filter paper and weighed. These were homogenized in 2 : 1 (v/v) chloroform-methanol (1 g per 18 ml),
358 centrifuged and the extract was washed [15]. Samples of eggs were counted and weighed to give a mean weight of 70/~g/egg. This figure was used to calculate the number of eggs extracted. Aliquots of the washed extract were used for analysis of fatty acids by GLC. RESULTS
Presence o f trace amounts o f PUFAs in fly eggs and dietary components Small amounts of PUFAs were detected in the fly egg extracts and in the dietary components. Table I shows the results of these analyses expressed as a percentage of the esterified fatty acids present in the 3-day-old adult male housefly. It was assumed that the PUFAs found per egg were carried through the larval and pupal stages to the adult without loss. The small amounts of PUFAs detected in the larval dietary constituents were assumed to be completely taken up and divided between the 200 larvae and that they were then taken through to the adult without loss. Such figures must represent the upper limit of PUFA contamination in the adult fly and are below the limit of detection established for the method used (0.05-0.10% of total fatty acid in the sample). Thus the linoleic acid present in the flies from larvae fed on diets containing added linoleic acid [11] represents at least two hundred times the maximum possible concentration in the adults from larvae fed on diets to which no PUFA had been added.
TABLE I POLYUNSATURATED FATTY ACIDS DETECTED IN HOUSEFLY EGGS AND IN VARIOUS COMPONENTSOF THE LARVALDIET Results are expressed as a percentage of the total esterified fatty acids present in the 3-day-oldmale housefly (195 ~g fatty acid per insect).
Source
Percentage of fatty acid in adult Linoleic acid
"r-Ltnolenic acid
Arachidonic acid
Eggsa Caseinb Agar b
0.0136 0.0135 0.0002
0.0038 0.0198 0.0001
0,0076 0 0
RNAb Cholesterolb Total
0.0007
0.0005
0
0
0
0
0.0280
0.0242
0.0076
aIt was assumed that the PUFAs detected in the egg were carried through to the adult fly without loss. bThe PUFAs detected in the various dietary components were assumed to be taken up completelyby the larvae roared on the diet (200) and that they were taken through to the adult fly without loss.
359
L~ofuscin volume Fat body. The abdominal fat body in Musca domestica contains two different cell types the fat cells and the oenocytes [12]. The cytoplasm of the oenocytes is rich in free polysomes and mitochondria and contains appreciable amounts of both agranular and granular endoplasmic reticulum. In both groups, a comparison between 3-day-old and 13-day-old flies indicated an age-associated increase in content of lipofuscin granules (Fig. 1). Morphologically, the granules were easily identifiable as membrane-bounded structures containing electron-dense material intermingled with pleomorphic membranous component. Occasionally, organelles such as ribosomes and cisternae of endoplasmic reticulum were identifiable within the confines of the lipofuscin granule. That these granules in the oenocytes are lysosomal in nature was demonstrated previously by the positive localization of the reaction product of fl-glycerophosphatase activity [12]. A quantitative comparison of lipofuscin in the oenocytes of the two groups of flies -
Fig. 1. An electron micrograph of a section through the abdominal fat body of a 13-day-oldlinoleic acid-free housefly showing lipofuscin granules (L) in the oenocytes, x 18 920. Bar = 1.0 #m.
360 TABLE II THE PERCENTAGE OF CYTOPLASMICVOLUME OCCUPIED BY LIPOFUSCINGRANULES IN OENOCYTES OF THE FAT BODY AND EPITHELIALCELLS OF THE MIDGUTOF 3-DAY-OLD AND 13-DAY-OLDMALEHOUSEFLIES Group 1 : linoleic acid-containing adult houseflies. Group 2: insects containing leasthan 0.1% of their fatty acids as polyunsaturates. Values axe mean ± S.E.M. Measurements were made on sections from three randomly selected blocks from six flies.
Group
1 2
Age (days after emergence of adult) 3 13 3 13
Lipofuscin {percentagecytoplasmic volume) Fat body
Midgut
2.13 ± 0.16 6.23 ± 0.80 1.88 ± 0.23 6.19 ± 0.70
2.87 ± 0.48 6.00 ± 1.10 3.61 + 0.46 6.10 ± 0.74
(Table II) indicated that 13-day-old flies in both groups contained significantly greater amounts of lipofuscin in their eytopla~n than the 3.day-old flies (p < 0.01); however, the two groups did not exhibit any statistically significant differences at either of these ages. Thus the increase in the percentage volume of cell cytoplasm occupied by the granules that occurred between 3 and 13 days of age was similar in both groups. Midgut. The midgut comprises the longest component of the alimentary tract in the housefly. The epithelial cells form a single cell thick layer which is separated from the lumen of the gut by a peritrophic membrane. The epithelial cells pt~rsist throughout the lifespan of the adult fly and exh~ited no manifestations of cell turnover [17]. The structural organization of the epithelial cells in the region of the midgut examined was quite uniform. The cytoplasm is rich in granular endoplasmic reticulum, free polysomes and numerous elongated mitochondria. The Golgi complex and the associated primary lysosomes are well developed. The percentage of the cytoplasm occupied by the lipofuscAn granules of the cells increased significantly (p < 0.01) from 3 to 13 days of age in both groups (Fig, 2, Table II). There were, however, no significant differences either in the age-associated increase or th percentage volume occupied by the granules at these ages between the two groups of flies. No major differences in the morphological appearance of the granules were noticed between either group of flies.
361
Fig. 2. Section through an epithelial cell of the midgut of a 13-day-old linoleic add-free housefly showing lipofuscin granules (L). × 18 920. Bar = 1.0 urn.
DISCUSSION The present results based on the estimation of lipofuscin granule volume by morphometric techniques clearly show that the formation and the age-associated accumulation of such granules in two cell types of the housefly occur even when the level of any PUFA is below that which can be detected using the methods used (<0.1% of the total esterified fatty acids). They also show that the accumulation is not enhanced by the presence of the linoleic acid in the fly. At Present two different methods are used for the quantification of lipofuscin granules. One, the older method, employs the technique used in this study, while the other depends on the estimation of levels of fluorescence present in washed chloroformmethanol extracts. The second method was originally used by Tappel [3,4] and was used by us in a previous study on houseflies [1 1]. In the case of the housefly both methods can be interpreted as giving essentially the same result, in that the age-associated
362 accumulation of granules and the increase of extractable fluorescent material with age occur in the absence of detectable levels of PUFAs in the organism. However, the maall increase in the amount of extractable fluorescence which occurs with age when linoleic acid is available [ 11 ] is not accompanied by an increase in the percentage of lipofuscin granules in the cytoplasm. If the peroxidation of PUFAs is involved in the formation of lipofuscin granules as has been outlined by Tappel [3,4], then the present results indicate that only very small amounts of PUFAs are involved and these lie below the present methods of detection. It would seem unlikely that any trace amounts of PUFA.~ which might have been carried over from the ~.~,g or derived from the larval diet would be adequate to maintain the continuing formation of fluorescent material with increasing age of the fly. Such a conclusion is supported by the findings of Gutteridge [18] that PUFAs are not the only source of malondialdehyde, which can also be derived from amino acids, nucleic acids and sugars. It also seems unlikely that the peroxidation of PUFAs is the main or only pathway for the formation of the fluorescent material extracted with chloroform-methanol from adult houseflies. Much of this material must be derived from non-lipid sources or, if from a lipid source, then from a monounsaturated fatty acid. However, this latter possibility is unlikely, as it has been shown that oleic acid does not produce malondialdehyde on peroxidation [19]. Whether the age-associated increase in extractable fluorescence results from the actual increase in fluorescent granules or whether it is derived from other sources cannot be clearly decided at this stage of the research. ACKNOWLEDGEMENTS This research was partially supported by grants to R,S.S. from the National Institutes of Health, National Institute on Aging (R01 AG00171) and the Glenn Foundation for Medical Research.
REFERENCES 1 D. Brandes, Lysosomes and age pigment. In P.L. Krohn (ed.), Topics in the Biology of,4gln~, Interscience, New York, 1966, pp. 149-158. 2 C. DeDuve and R. Wattiaux, Functions of lysosomes. Annu. Rev. Physiol., 28 (1966) 435-491. 3 A.L. Tappel, Lipid peroxidation and fluorescent molecular damage to membranes. In B.F. Trump and A.V. Arstila (eds.), Pathobloiogy o f Cell Membranes, Vol. I, Academic Press, New York, 1975, pp. 145-170. 4 A.L. Tappel, Measurement and protection from in rive lipid peroxidation. In W.A. Pryor (ed.), Free Radicals in Biology, Vol. IV, Academic Press, New York, pp. 1-47. 5 H. Dormto and R.S. Sohal, Age-related changes in lipofuscin associated fluorescent substances in the adult male housefly, Musca domestica. Exp. Gerontol., 12 (1978) 171-179. 6 B.L. Strehler, On the histochemistry .and ultrastructure of age pigment. Adv. Gerontol. Res., 1 (1964) 343-384.
363 7 M. Elleder, So-caUed neuronal ceroid-lipofuscinosis. Histochemical study with evidence of extractability of the stored material. Acta Neuropathol., 38 (1977) 117-122. 8 M. EUeder, Chemical characterization of age pigments. In R.S. Sohal (ed.), Age Pigments, Elsevier/ North-Holland, Amsterdam, pp. 203-241. 9 J. Miquel, J. Oro, T.G. Bensch and J.E. Johnson, Lipofuscin: fine structural and biochemical studies. In W.A. Pryor (ed.), Free Radicals in Biology, Vol. III, Academic Press, New York, 1975, pp. 133-182. 10 H. Donato, Lipid peroxidation, cross-linking reactions and aging. In R.S. Sohal (ed.), Age Pigments, Elsevier/North-Holland, Amsterdam, 1981, pp. 63-81. I 1 R.G. Bridges and R.S. Sohal, Relationship between age-associated fluorescence and linoleic acid in the housefly, Musca domestica. Insect Bioche~, 10 (1980) 557-562. 12 R.S. Sohal, Fine structural alterations with age in the fat body of the adult male housefly, Musca domestica. Z. Zellforsch. Microsk. Anat., 140 (1973) 169-175. 13 R.S. Sohal, Relationship between metabolic rate, lipofuscin accumulation and lysosomal enzyme activity during aging in the adult housefly, Musca domestica. Exp. Gerontol., 16 (1981) 347-355. 14 R.G. Bridges, Incorporation of fatty acids into the lipids of the housefly, Musca domestica. J. Insect Physiol., 17 (1971) 881-895. 15 R.G. Bridges, J. Rickets and J.T. Cox, The replacement of lipid-bound choline by other bases in the phospholipids of the housefly, Musca domestica. J. Insect Physiol., 11 (1965) 225-236. 16 E.R. Weibel, Stereological principles for morphometry in electron microscopic cytology. Int. Rev. Cytol., 126 (1969) 235-302. 17 R.S. Sohal, P.D. Peters and T.H. Hall, Origin, structure, composition and age-dependence of mineralized dense bodies (concretions) in the midgut epithelium of the adult housefly, Musca domestica. Tissue Cell, 9 (1977) 87-102. 18 J.M.C. Gutteridge, Free radical damage to lipids, amino acids, carbohydrates and nucleic acids determined by thiobarbituric acid reactivity. Int. J. Biochem., 245 (1970) 4641-4646. 19 M.C. McArthur and R.S. Sohal, Relationships between metabolic rate, aging, lipid peroxidation and fluorescent age pigment in milkweed bug, Oncopdtus fasciatus (Hemiptera). J. Gerontol., 37 (198D 268-274.