The secretory pathway of vitellogenin in the fat body of the stick insect bacillus rossius: An ultrastructural and immunocytochemical study

TISSUE & CELL @ 1989 Longman

21 (4) 535-541 Group UK Ltd

MASSIMO

MAZZINI*,

ANNA

BURRlNlt

and FRANC0

GIORGIS

THE SECRETORY PATHWAY OF VITELLOGENIN IN THE FAT BODY OF THE STICK INSECT BACILLUS ROSS/US: AN ULTRASTRUCTURAL AND IMMUNOCYTOCHEMICAL STUDY Keywords:

Bncillw rossiw, fat body, Vitellogenin

ABSTRACT. The fat body of the adult female stick insect Bacillus rossiur was examined ultrastructurally with a view to clarifying the secretory pathway. The absence of lipid storage in the tissue allowed visualization of a polarized distribution of all organelles in the cell cytoplasm Composite granules were distributed along the baswapical axis of the cell according to progressive stages of maturation. At their final stage of maturation, these granules possess two distinct compartments, an electron-translucent compartment and a more electron-dense one. The origin of each of the two compartments was traced back to other organelles in the basal cytoplasm of the fat body cell. The differential origin of the two compartments contributing to the composite granules was further investigated by cytochemical analyses. Vitellogenin was detected both in the electrondense compartment of the composite granules and in the Golgi apparatus. The electron-translucent compartment of the composite granules appeared to consist mainly of urate crystals. Such enzyme activities as acid phosphatase, peroxidase and catalase were also detected in this latter compartment. The observations support the interpretation that secretion in the fat body of B. rossiusentails fusion of Golgi-derived vesicles with a specialized kind of multivesicular body. While Golgiderived vesicles convey their load of newly synthesized vitellogenin to the electron-dense compartment, the multivesicular body develops the urate crystals of the electron-translucent compartment.

Introduction The fat body plays a key role in the intermediary metabolism of insects (Wyatt, 1980). Depending on the stage of development, the fat body may either act as the main storage tissue sequestering proteins from the hemolymph as in the moulting larva (Locke, 1980; Marx, 1983) or behave as a protein factory that synthesizes many of the proteins appearlFacolt;l di Scienze, Universit.4 della Tuscia, Viterbo. tktituto di Biologia Generale, Univenita di Siena. -$Dipartimento di Fisiologia e Biochimica, Laboratorio di Biologia Cellulare e dello Sviluppo, Univenit.3 di Piss, Italy. Received 28 June 1988 Revised 15 April 1989.

ing in the hemolymph of the adult female (Wyatt and Pan, 1978). Much of our present knowledge about insect fat body is mainly centered on the metabolism of several secretory proteins including larval serum proteins (Wolfe et al., 1977), vitellogenin (Hagedorn and Kunkel, 1979) lipophorin (Gilbert et al. ,1977) or even hemoglobin (Bergstrom et al., 1976). It is generally recognized that the secretory pathway followed by these proteins entails synthesis in the rough endoplasmic reticulum (Lauverjat, 1977; Stoppie et al., 1981; Raikhe1 and Lea, 1983), post-translational modifications in the Golgi apparatus (Thomsen et al., 1980; Minoo and Postlethwait, 1985) and packaging into secretory granules that later

MAZZINI

fuse with the plasma membrane (Locke, 1984). However, much of what we know about the physiology of the adult fat body remains uncomplemented by equally significant ultrastructural findings. This limitation is largely due to the fact that, unlike many other secretory tissues, insect fat body is a lipid-storing tissue (Butterworth et al., 1965; Butter-worth and Bodenstein, 1968). This is a condition that obscures any cytoplasmic polarity in the tissue and thus forces a certain arbitrariness upon studies aimed at documenting a sequence of morphological transformations along a synthetic pathway. In this paper, we show that the stick insect Bacillus rossius is exceptional in this respect in that negligible amounts of lipids are retained in the fat body throughout adult development. This condition allowed the fat body cells in this species to manifest their cytoplasmic polarity and thus provided a condition favorable for studying their secretory pathway(s). A well-known problem to all cell morphologists is that micrographs lack chronological information. In this study, we tried to circumvent this difficulty by studying the secretory pathway of the fat body in B. rossius using two additional complementary approaches. The first made use of a number of cytochemical tests to establish the nature and origin of the material eventually compartmentalized inside the composite granules. The second employed autoradiography with a radioactively labeled precursor to visualize the sequence of events leading to maturation of these granules. In this first paper we report on the chemical composition of composite granules and validate our interpretation about the secretory pathway as based on structural data.

Material and Methods Rearing Bacillus rossius (Rossi) (Insecta, Phasmatodea) were maintained in a rearing room at 20°C with a 12D:12L photoperiod and fed on bramble leaves ad Zibitum. Thin section analysis Fat body was dissected from egg-laying females and immediately fixed for 1 hr at 4°C in O-1 M cacodylate buffer at pH 7.2, containing 5% glutaraldehyde and 4% formaldehyde. The tissue was then washed overnight

ETAL.

in the same buffer and post-fixed for 1 hr in 1% osmium tetroxide in 0.1 M cacodylate buffer at pH 7.2. After dehydration in a graded series of alcohols, fat body was embedded in an Epon-Araldite mixture and polymerized at 60°C for 3 days. For immunocytochemical purposes, fat body was fixed at 4°C in 1% glutaraldehyde-1% formaldehyde in 0.1 M phosphate buffer at pH 7.2, dehydrated in ethanols at -20°C and embedded in Lowicryl K4M resins at -30°C. according to the procedure of Carlemalm et al. (1985). Thin sections were stained with uranyl acetate and lead citrate and examined in a Philips 301 electron microscope. Freeze fracture analysis Fat body was fixed in glutaraldehydeformaldehyde as above and gradually infiltrated with glycerol up to a final concentration of 30%. The fat body was then rapidly frozen in Freon 22 cooled to liquid nitrogen temperature. Fracture and platinum-carbon coating were carried out in a Balzers BAF 301 freeze etching unit with a temperature stage set at - 115°C. Replicas were eventually digested with chlorox, picked up with formvar coated copper grids and examined in a Philips EM 301 electron microscope. Scanning electron microscopy Fat body was fixed in glutaraldehydeformaldehyde fixative as above, postfixed in osmium tetroxide and dehydrated in a graded series of alcohols. Subsequently, the specimens were dried by the critical point method in a Bomar apparatus equipped with a liquid CO, inlet. They were then attached to specimen holders using a silver conducting paint, coated with gold in a Balzers evaporator and observed with a Philips 505 scanning electron microscope. Antiserum preparation Yolk was purified from newly laid eggs of B. rossius as previously reported (Giorgi et al., 1982). Antiserum against yolk was raised in New Zealand white rabbits following a 2month immunization schedule. The antigen was emulsified with complete Freund’s adjuvant and injected subcutaneously three times at alternate weeks. After the last injection, blood was collected by heart puncture and the resulting serum frozen in 1 ml aliquots for further analysis. Antiserum was tested for

VITELLOGENIN

IN THE FAT BODY OF THE STICK INSECT B. ROSSIUS

its specificity towards yolk antigens by double immunodiffusion in 1% agar plates and crossed immunoelectrophoresis in 1.5% agarose plates. By both these analyses the antiserum could be shown to give rise to two distinct precipitation arcs corresponding to the two egg vitellins present in this species (Masetti , 1985). Immunocytochemistry

Silver sections of Lowicryl-embedded fat body were collected on nickel grids and hydrated in phosphate buffered saline (PBS) containing 0.1% bovine serum albumin (BSA). Hydrated sections were then incubated for 3 hr in a moist chamber containing an antiyolk serum diluted 1:2 or 1:4 with PBS-BSA. Following a prolonged rinse in PBS-BSA (three changes of 30 min each) on a nutating table, grids were treated for 3 hr at room temperature in a moist chamber with colloidal gold-conjugated protein A (10nm) (E-Y Labs) diluted 1:200 with PBS-BSA. After a final wash in PBS-BSA followed by distilled water, grids were stained in uranyl acetate and lead citrate. Localization of urate inclusions

Fat body was fixed in glutaraldehydeformaldehyde as above, rinsed twice in buffer, and transferred to a silver methenamine solution (Gomori, 1952) for 4 hr at room temperature. After a final rinse in 5% sodium thiosulphate, the tissue was dehydrated and processed for embedment in Epon-Araldite resins (Buckner et al., 1985). Localization of peroxidase and catalase activities

Peroxidase and catalase activities were localized on fat body following the procedure of Angermuller and Fahimi (1981). Small fragments of fat body strips were fixed for 30 min at 4°C in 2% glutaraldehyde in 0.1 M cacodylate buffer at pH 7.2 with 2% sucrose added. Following a quick rinse in cacodylate buffer, they were incubated for 1 hr at room temperature in one of the following solutions: (1) O.lM Tris-HCl buffer at pH 6.8 containing 5 mM diaminobenzidine-HCl (DAB) and 0.15% hydrogen peroxide for peroxidase or (2) O-1 M glycine-Tris buffer at pH 10.5 containing 5mM diaminobenzidine-HCl (DAB) and 0.15% hydrogen peroxide for catalase. At the end

591

of the incubation period, fat body fragments were rinsed in buffer, post-fixed in 2% aqueous osmium tetroxide, hence dehydrated and embedded in Epon-Araldite. Controls were done by omitting hydrogen peroxide from the incubation media. Localization of acidphosphatase activity

Acid phosphatase activity was detected in fat body following the Cerium-based method of Robinson and Kamovsky (1983). Fragments of fat body were fixed for 1 hr in 2% glutaraldehyde in cacodylate buffer at pH 7.2, rinsed in the buffer and then incubated for 45 min at 37°C in 0.1 M sodium acetate buffer at pH 5.0 containing 1 mM P-glycerophosphate, 2mM CeC& with 0.001% saponine added. Following a prolonged rinse in the buffer fragments of fat body were post-fixed in osmium tetroxide, dehydrated and embedded in Epon-Araldite. Sections were examined in the electron microscope either with or without further staining. OZIfiation

Fragments of fat body were fixed for 20 min in 5% glutaraldehyde-4% formaldehyde in 0.1 M cacodylate buffer at pH 7.2 and then thoroughly washed in the same buffer. They were then post-fixed for either 9 or 16 hr in 0.4% osmium tetroxide containing 20 mM of zinc iodide buffered at pH 7.0 (Maillet, 1963). After fixation, fragments of fat body were rinsed in the buffer, dehydrated in alcohols and embedded in Epon-Araldite. Sections were examined either with or without further staining. Results General morphology

The fat body of adult female Bacillus rossius is a thin multicellular sheet closely apposed to the abdominal wall and uniformly bathed by the hemolymph (Fig. 1). Only one cell type is apparent in the tissue, and no lipid droplets form in these cells during adult development. Cells are polarized such that their main roundish cell bodies are deeply lodged in the tissue and their peripheral processes are tightly anchored to the external basement lamina (Figs 2 and 3). Cytoplasmic organelles appear to have a differential distribution along the major cell axis with some ovoid granules being preferentially displaced to the

Fig. 1. Scanning electron micrograph (SEM) the stick insect BaciNusrossiw. x250.

of a fat body strip from

an adult

female of

Fig. 2. SEM view of a fractured fat body strip showing the basement lamina (bl) and a few granules (arrows). x4200. Fig. 3. Light micrograph of an Epon-embedded section of a fat b&y stnp. (g), granules; (n), nuclei. x3500. Fig. 4. Plasma membrane reticular system (pmrs) of an apical cell as seen by a freeze-fracture plane perpendicular to the fat body sheet. Two granules are visible (asterisks). x25,CNl. Fig. 5. Plasma membrane reticular system of an apical cell as seen by a freeze-fracture plane tangential to the fat body sheet. (p) endocytic pit. x41.Mx). Figs 6-7. SEM micrographs showing thethree-dimensional extension of the basement lamina (6) and the apical end of a fat body cell after removal of the basement lamina (7). x 15,ooO (6). x3tXKl(7). Fig. 8. Plasma membrane reticular system (pmrs) of the fat body showing one cell process a figure-eight-shaped composite granule (cg). An electron-dense and two electron-

containing

translucent compartments are visible inside the granule. (bl). basement lamina; (p), endocytic pit; (m), mitochondria. The inset shorn all membranes enclosing composite granules. (pm), plasma membrane; (lm), limiting membrane: (mc), membranous coat x41,oM). Inset

x l(w,iXK!.

594

MAZZINI ETAL..

cell folds that protrude towards the basement lamina. Partial removal of the basement lamina reveals that the top surface of the fat body is highly folded, forming a plasma membrane reticular system. Such a differentiation of the cell surface is most apparent either in freeze-fracture replicas (Figs 4 and 5) or in scanning electron microscope preparations (Figs 6 and 7). Apically in the cell periphery, there are composite granules that cause the membranes of the reticular system to bulge out of the cell folds and protrude towards the extracellular spaces (Figs 4 and 8). When examined in thin sections, these granules appeared to consist of two major compartments, an electron-dense peripheral part and a more electron-translucent one that is usually centrally located within the granule (Fig. 8). Occasionally, the electron-dense compartment combines with two distinct electron-translucent compartments to give rise to a figure-eight-shaped organelle within a larger electron-dense ellipsoid. Each granule is bounded by a limiting membrane that is closely apposed to the external plasma membrane. Only a 25 nm wide gap of cytoplasm separates the two membranes. An additional membranous coat encircles the electron-translucent compartment, separating it from the more electron-dense one (Fig. 8 and inset).

differential distribution of most cytoplasmic organelles. In general, the rough endoplasmic reticulum delimits a central area in the cell that is also occupied by mitochondria, various-sized vesicles and several Golgi apparatuses (Fig. 9). Each apparatus consists mainly of vesicular elements some of which are characterized by an electron-dense content (Fig. 10). A constant feature of this cell region is the presence of multivesicular bodies, preferentially located in the proximity of the Golgi apparatus. These organelles are partially filled with some fibrous material that may either be bound to the limiting membrane or amidst a group of smaller vesicles (Fig. 11). Moving from the basal towards the apical end of the cell one sees increasingly more instances of close association between the above multivesicular body and electron dense vesicles emerging from the Golgi complex (Fig. 12). Upon coming into close contact, the two types of organelles become enclosed within the same limiting membrane (see diagram in Fig. 13). During this transition, the vacuole develops some internal fibrous material that eventually gives rise to a central core with a crystalline morphology radiating towards the periphery. At this stage of their maturation granules appeared to be made of two compartments differing in electron density.

Origin of the composite granules

Composite granules contain vitellogenin associated with a urate granule

Due to the orientation of the fat body cell relative to the hemolymph, sections parallel to the major axis of the cell disclose a clear

In an attempt to validate the scheme outlined above (see Fig. 13), the chemical nature of

Fig. 9. The basal end of a fat body cell showing several cisternae of the rough endoplasmic reticulum (rer), vesicles (v) of varying sizes and content and mitochondria (m). (cg). composite granules. x8400. Fig. 10. Enlargement of a cell region similar to the one depicted in Fig. 9 showing the presence of three Golgi apparatus (pa) amidst cisternae of the rough endoplasmic reticulum (rer) and a multivesicular body (mvb). (m), mitochondria. ~24,000. Fig. 11. A multivesicular body (mvb) lying close to a Golgi-derived vesicle (gv) in the central cytoplasm of a fat body cell. (n), nucleus. ~21,000. Fig. 12. A Golgi-derived

x40,OtXl.

vesicle (gv) lying close to a transformed

multivesicular

body (mvb).

MAZZINI ETA L.

596

a

Fig. 13. A schematic drawing showing: (A), a fat body cell and (B) the proposed sequence of maturation events leading to formation of composite granules (cg) by way of fusion between an electron-dense vesicle (gv) originated from the Golgi apparatus (Ga) and a transformed multivesicular body (mvb). (Ly), lysosomes; (pe). peroxisomes: (rer), cisternae of the rough endoplasmic reticulum.

Fig. 14. Composite granules from a Lowicryl-embedded fat body cell. Note that protein A gold-conjugated particles bind exclusively to the electron-dense compartment (circle) of the composite granules. (asterisk), electron-translucent compartment. ~40,000. Fig. 15. Control preparation from a Lowicryl-embedded fat body cell run in the absence of immunospecilic antiserum showing an unlabeled composite granule. x40.000. Fig. 16. A composite granule from a fat body cell treated with a silver methenamine solution to stain specifically the urate inclusions. Note that silver deposits are exclusively bound to the electron-translucent compartment (asterisk) of the granule. (circle), electron-dense compartment. x50,000. Fig. 17. Control preparation from a fat body cell showing a composite granule counterstained with sodium thiosulphate run in the absence of silver methenamine. ~50,000. Fig. 18. Portion of a Lowicryl-embedded fat body cell showing gold-conjugated A particles over several cisternae of the Golgi apparatus (pa). x64.00@ Fig. 19. Control preparation an immunospecific

antiserum

of a Lowicryl-embedded

showing an unlabeled

protein

fat body cell run in the absence Golgi apparatus (ga). x64.OLW.

of

MAZZINI

the material compartmentalized in the composite granules was identified. Accordingly, the fat body from adult females of B. rossius was treated with an anti-yolk serum and its binding sites revealed by gold-conjugated protein A. The results of this immunocytochemical test showed that gold-positive sites were restricted to the electron-dense compartment of the composite granule, suggesting that a vitellogenin precursor is confined to this part of the organelle (Fig. 14). No gold-positive sites could be detected in control preparations run in the absence of the anti-yolk serum (Fig. 15). The chemical nature of the electron-translucent compartment was tested experimentally using the Gomori’s procedure for the selective reduction of silver methenamine, as recently modified for the electron microscope by Buckner et al. (1985). After this treatment, silver deposits appeared only over the electrontranslucent compartment of the composite granules (Fig. 16). In the absence of any silver methenamine treatment, no such deposits could ever be seen over the electrontranslucent compartment of these granules (Fig. 17). Based on the above observations, we conclude that the composite granules detected in the fat body cells of B. rossius are secretory granules that contain vitellogenin as a major component of the electron-dense compartment, and urate as one of the constituents present in the crystalline inclusions of the electron-translucent compartment. The immunocytochemical detection of anti-yolk

ETAL.

positive sites on the Golgi apparatus of the fat body from adult females gives further support to this conclusion (Figs 18 and 19). In fact, the observation that the electron-dense compartment of the composite granules in the female fat body shares an immunogenic reactivity with the Golgi apparatus suggests, but does not prove, that both these organelles are along the same secretory pathway. Composite granules contain peroxidatic and catalatic activities in association with their crystalline inclusion

In order to determine the origin of the electron-translucent compartment of the composite granules, we made use of cytochemical tests designed to detect peroxidase and catalase activities. Peroxidase activity could be revealed by treatment of fat body with diaminobenzidine-HCl at pH 6.8 (Fig. 20). Under these conditions, only the electrontranslucent compartment of the composite granules reacted positively, the electrondense compartment being largely devoid of reaction products. A number of other cell organelles were also positively stained by this test. These are in all likelihood to be identified as peroxisomes (Fig. 21). Changing the pH of the incubation medium to 10.5, a condition that is known to be specific for catalase (Angermuller and Fahimi, 1981), caused the reaction products to appear clearly on the multivesicular body (Fig. 22) as well as on the crystalline inclusion of the composite granules (Figs 22 and 23). The observations that organelles assumed to

Fig. 20. Apical region of fat body cell showing several composite granules reacting for peroxidase following exposure to hydrogen peroxide in the presence of diaminobenzidine at pH 6.8. A number of smaller-size vesicles are also visible in the cell (see arrows). X20,~. Fig. 21. Catalase activity as evidenced after treatment with hydrogen peroxide in the presence of diaminobenzidine at pH 10.5. Note the presence of reaction products over a multivesicular body (mvb) close to a Golgi-derived vesicle (gv) in the cytoplasm of a fat body cell. x28,lWO. Fig. 22. A composite granule showing heavy deposits (arrows) of catalase activity over the electron-translucent compartment after exposure to hydrogen peroxide in the presence of diaminobenzidine at pH 10.5 (pa), Golgi apparatus; (circle), electron-dense compartment. X38,040. Fig. 23. Control to diaminobenzidine

preparation from a fat body cell showing a composite granule at pH 10.5 in the absence of hydrogen peroxide. x38.000.

exposed

MAZZINI ETAL.

be in a developmental sequence on a sole structural basis, come also to share the same enzyme make-up when analyzed cytochemitally, led us to conclude that the electrontranslucent compartment of the composite secretory granules is likely to represent a later transformation of the multivesicular body. Maturationof the composite granules entails associationwitha lysosomal compartment Because composite granules in B. rossius fat body form by fusion of Golgi-derived vesicles with a multivesicular body, we wondered whether lysosomal enzymes were somehow involved in the process. Accordingly, fat bodies were processed to evidence acid phosphatase in all organelles potentially identifiable along the secretory pathway. The results showed that the multivesicular body lying in the immediate vicinity of the Golgi apparatus was highly enriched in reaction products (Fig. 24). Since no such precipitates were present in control preparations run in the absence of the substrate, this reaction appeared to be a valid indicator for the presence of acid phosphatase (Fig. 25). Similar reaction products were also seen over vesicles lying close to the Golgi apparatus and the rough endo-

plasmic reticulum. These we interpret to be primary lysosomes. Some reaction products were also confined to the electron-translucent compartment of the composite granules (Fig. 26). No cerium precipitates were retained over the crystalline inclusions when fat body was incubated in a substrate-free medium, suggesting that the reaction products seen at this site represent true enzyme activity (Fig. 27). To determine how acid phosphatase is transferred to the multivesicular body, we made use of an osmium-zinc iodide (OZI) complex to specifically stain the region of the Golgi apparatus. After 16 hr fixation in OZI, nearly the entire cytoplasmic area occupied by the Golgi apparatus was stained (Fig. 28). OZI-positive spots were also visible over a nearby multivesicular body. However, the composite granule itself was never found positively stained after OZI fixation (Fig. 29). We conclude that transformation of the multivesicular body into the electron-translucent compartment of the composite granules requires intervention of lysosomal enzymes and that these are conveyed to the vacuole via vesicles emerging from the Golgi apparatus. These vesicles are then presumably lost as the organelle develops some crystalline inclusions.

Fig. 24. Acid phosphatase. associated with a multivesicular body (mvb) and primary lysosomes (ly) in the central cytoplasm of a fat body cell. (rer), rough endoplasmic reticulum; (m), mitochondria. ~14.000. Fig 25. Enlargement of the central cytoplasm in a fat body cell showing reaction products for acid phosphatase on the internal matrix of the multivesicular body (mvb). (rer), rough endoplasmic reticulum; (m), mitochondria. ~2g,GU0. Fig. 26. Composite granules heavily labeled after treatment with cerium chloride to evidence acid phosphatase activity. ~60,000. Fig. 27. Control preparation from a fat body cell treated with cerium chloride in the absence of pglycerophosphate showing a composite granule. x60,OtM. Fig. 28. The central cytoplasm of a fat body cell after hxation with Zinc Iodide-Osmium tetroxide (021) showing a heavily labeled Golgi apparatus (ga) and a nearby multivesicular body (mvb). x50,ooO. Fig. 29. The apical cytoplasm of a fat body cell after fixation with OZI, showing a heavily labeled Golgi apparatus (ga) along with a composite granule (cg). Note that no OZI deposits are present over the granule itself. (m), mitochondria. x4O,OCU!

MAZZlNl

Discussion

The results of the present study indicate that secretory granules in the fat body of the adult female stick insect B. rossius are composite organelles comprised of two distinct compartments. Organelles structurally similar have already been reported by others (Thomsen and Thomsen, 1974; Dortland and Esch, 1979; Thomsen et al., 1980; Han and Bordereau, 1982), but their origin not determined. A major outcome of our morphological analysis is that composite granules in B. rossius fat body arise by fusion of Golgi-derived vesicles with a multivesicular body that later develops some internal crystalline inclusions. The validity of this interpretation rests primarily on the observation that, in the absence of a lipid storage, fat body cells manifest a clear polarized distribution of most cytoplasmic organelles. This makes it likely that granules spatially displaced along the basalapical axis of the cell represent successive stages of a continuous maturation process. Further support to this interpretation has come from the cytochemical detection of common reactivities between the electrondense compartment of the composite granules and the Golgi apparatus, on one hand, and the electron-translucent compartment and the multivesicular body, on the other. The localization of a vitellogenin precursor both in the electron-dense compartment of the composite granules and in the Golgi apparatus is consonant with the view that they are subcellular components of the same secretory pathway of B. rossius fat body. A similar interpretation applies also to the fat body of the mosquito Aedes aegypti for which a panel of specific monoclonal antibodies has been used to identify the secretory pathway for two vitellogenin polypeptides (Raikhel et al., 1986; Raikhel, 1987). But unlike the secretory granules in B. rossius, those in the mosquito fat body were reported to be exocytosed immediately after being released from the Golgi apparatus (Raikhel and Lea, 1983). Thus, the co-packaging of vitellogenin into composite secretory granules may not be a phenomenon common to all insect fat bodies. A possibility that still needs to be explored is whether co-packaging constitutes an additional processing for vitellogenin (Thomsen et al., 1980) in those instances in which it appears to be an obligatory step

ETAL.

along the secretory pathway. Perhaps, fat body secretion differs from species to species depending on such factors as the number of different cell types in the tissue, the heterogeneity of secretory proteins to be released by the cell and the complexity of the metabolic activities carried out by the tissue at any given developmental time (Locke, 1984). The association of vitellogenin with an electron-translucent compartment in the composite granules provokes additional questions as to the origin and identity of this cell organelle. As described earlier in this study, the electron-translucent compartment appeared to originate from a vacuole close to the Golgi apparatus that contains some fibrous material along with a number of internal vesicles. These morphological features make the vacuole in question to look like a multivesicular body, perhaps formed by small pinocytic vesicles carrying protein from the hemolymph (Locke and Collins, 1967; Locke, 1984). This is consonant with the view that fat body is a tissue capable of incorporating hemolymph proteins for either temporary storage or intracellular digestion (Locke and Collins, 1968; Collins and Downe, 1970). This interpretation is further corroborated by the present findings that acid phosphatase and an OZI reactivity can both be localized inside the multivesicular body, suggesting that lysosomes are to fuse with this organelle to contribute to its enzymatic make-up. The electron-translucent compartment of the composite granules was also shown to react specifically with silver methenamine, a test now believed to be specific for uric acid (Buckner et al., 1985). Although the crystalline inclusions developing inside this organelle could be chemically more heterogenous than suggested by this simple reaction, it is nevertheless true that uric acid is by far the most common form of nitrogen storage in insects (Keeley, 1985; Wigglesworth, 1987) and that fat body is the tissue where such a storage is known to occur preferentially (Cochran, 1985). Moreover, the appearance of uric acid in such a cell compartment as the multivesicular body, that is devoted to intracellular digestion, agrees with earlier observations by Dean et al. (1985) on the origin of urate granules in Calpodes fat body. The actual storage form for uric acid in insects is not known, but there is circumstantial evidence to suggest that this occurs

VITELLOGENIN

IN THE FAT BODY OF THE STICK INSECT B. ROSSIUS

as urate complexed with either sodium or potassium (Mullins and Cochran, 1974; Hyatt and Marshall, 1985a). A feature common to all urate crystals described so far in insect fat body is the presence of a cortex or dense wall, termed the membranous coat by us, around the fibrous material comprised in the crystalline inclusion. Although the true nature of this structure remains to be worked out in detail (Cochran et al., 1979), it is believed to play a role in regulating either mobilization of the stored urates and/or the ionic balance in the hemolymph (Mullins and Cochran, 1974, 1976; Cochran, 1985). If proved correct, this could mean that mobilization of uric acid from the fat body may be achieved by regulating permeability across the membranous coat (Hyatt and Marshall, 1985b; Cochran, 1985). The observation that catalase occurs in the electron-translucent compartment of the composite granules in B. rossius fat body raises the question as to where along the secretory pathway does the enzyme become associated with the granules themselves. Catalase is currently believed to be a reliable marker for microbodies in most animal cells

603

(Lazarow and Fujiki, 1985). These are organelles that undergo typical morphological changes during fat body development and are ultimately destroyed by autophagy when segregated into secondary lysosomes (Locke and McMahon, 1971). This raises the interesting possibility that catalase may come to reside in the composite granules after having been brought to the multivesicular body through isolation membranes (Locke and McMahon, 1971). Alternatively, it might be specifically transferred to the forming composite granules through fusion with newly formed peroxisomes (Locke, 1984). In summary, the developmental sequence reported in this study is consonant with the view that VG-carrying vesicles in Bacillus fat body come to fuse with a multivesicular body that later transforms into a urate-carrying compartment. Both ultrastructural and cytochemical observations point to the composite granules as the cell organelle where vesicles contributed by the Golgi apparatus and multivesicular bodies convey their load of proteins and crystalline inclusions in preparation for secretion.

References Angermuller, S. and Fahimi, H. D. 1981. Selective cytochemical localization of peroxidase, cytochrome oxidase and catalase in rat liver with 3,3’-diaminobenzidine. Histochem., 71,3&M. Bergstrom, G., Laufer, H. and Rogers, R. 1976. Fat body: a site of hemoglobin synthesis in Chironomus thumni (Diptera). I. Cell Biol., 69,26&274. Buckner, J. S., Caldwell, J. M. and Knoper, J. A. 1985. Subcellular localization of uric acid storage in the fat body of Manduca sexta during the larval-pupal transformation. J. Insect Physiol., 31,741-753. Butterworth, F. M., Bodenstein, D. and King, R. C. 1965. Adipose tissue of Drosophila melanogoster. I. An experimental study of the larval fat body. J. exp. 2001.. 158,141-154. Butterworth, F. M. and Bodenstein, D. 1968. Adipose tissue of Drosophila melanogoster. The effect of the ovary on cell growth and the storage of lipid and glycogen in the adult tissue. J. exp. Zool., 167,207-218. Carlemalm, E., Villiger, W., Hobot, J. A., Acetarin, J. D. and Kellenberger, E. 1985. Low temperature embedding with Lowicryl resins: two new formulations and some applications. J. Microscopy, 140,5563. Cixhran, D. G. 1985. Nitrogen excretion in cockroaches. Ann. Rev. Entomol., 30,2%49. Cochran, D. G., Mullins, D. E., Mullins, K. J. 1979. Cytological changes in the fat body of the American cockroach in relation to dietary nitrogen level. Ann. Entomol. Sot. Am., 72,197-205. Collins, J. V. and Downe, A. E. R. 1970. Selective accumulation of haemolymph proteins by the fat body of Galleria mellonella. J. Insect Physiol.,

16, 1697-1708.

Dean, R. L., Locke, M. and Collins, J. V. (1985). Structure of the fat body. In Comprehensive Physiology, Biochemistry and Pharmacology. (eds G. H. Kerkut and L. I. Gilbert), Vol. 3, pp. 155-210. Pergamon Press, New York. Dortland, J. F. and Esch, T. H. 1979. A fine structural survey of the development of the adult fat body of Leptinotarsa decemlineata. Cell Tissue Res., 201,423-430. Gilbert, L. I., Goodman, W. and Bollenbacher, W. E. 1977. Biochemistry of regulatory lipids and sterols in insects. In Biochembtry ofLipids II. (ad. T. W. Goodwin). Int. Rev. Eiochem., 14,1-50.

MAZZINI ETAL.

604

Giorgi, F., Baldini, G., Simonini, A. L. and Mengheri, M. 1982. Vitellogenesis in the stick insect Crzrouri~~~moros~c~. II. Purification and biochemical characterization of two vltellins from eggs. Insect Biochem., l&553-562. Gomori, G. 1952. Microscopic Histochemlstry: Principles and Practice. University of Chicago Press. Chicago, pp. 58-119. Hagedorn, H. H. and Kunkel, J. G. 1979. Vitellogenin and vitellin in insects. Ann. Rev. Entomol., 24,475-505. Han, S. H. and Bordereau, C. 1982. Origin and formation of the royal fat body of the higher termite queens. 1. Morph., 173,17-28. Hyatt, A. D. and Marshall, A. T. 1985a. Ultrastructure and cytochemistry of the fat body of Periplaneta americana (Dictyoptera: Blattldae). hr. J. Insect Morphol. & Embryol., 14,131-141. Hyatt, A. D. and Marshall, A. T. 198Sb. X-ray analysis of cockroach fat body in relation to ion and water regulation. J. Insect Physiol., 31,49.5-508. Keeley, L. L. 1985. Physiology and biochemistry of the fat body. In Comprehetuive Insecr Physiology Biochemistry and Pharmacology (eds G. H. Kerkut and L. I. Gilbert), Vol. 3, pp. 211-248. Pergamon Press, New York. Lauverjat, S. 1977. L’evolution post-imaginale du tissu adipeux femelle de Locwm migruroria et son control endocrine. Gen. Camp. Endocrinol., 33,13-34. Lazarow, P. B. and Fujiki, Y. 198s. Biogenesis of peroxisomes. Ann. Rev. Cell Biol., 1,489-530. Locke, M. 1980. The cell biology of fat body development. In Insect Biology in Ihe Furure. (eds M. Locke and D. S. Smith), pp. 227-252. Locke, M. 1984. The structure and development of the vacuolar system in the fat body of insects. in Inrecr Wrasrructure. (eds R. C. King and H. Akai), Vol. II, pp. 151-197. Locke. M. and Collins, J. V. 1967. Protein uptake in multivesicular bodies in the molt/intermolt cycle of an insect. Science (Washington), 155,467. Locke, M. and Collins, J. V. 1968. Protein uptake into multivesicular bodies and storage granules in the fat body of an insect. J. Cell Biol., 36,4X3-483. Locke, M. and McMahon, .I. T. 1971. The origin and fate of microbodies in the fat body of an insect. J. Cell Biol., 48,61-78. Maillet, M. 1%3. Le reactif au tetraoxyde d’osmium de zinc. 2. Mikrosk. Anal. Forsch., 70,397-425. Marx, R. 1983. Ultrastructural aspects of protein synthesis and protein transport in larvae of Calliphora vicina. III The larval serum proteins ofimecls. (ed. K. Scheller), pp. 50-60. Thieme-Stratton. New York.

Masetti, M. 1985. A comparative analysis of vitellins in stick insects. In (eds M. Mazzini and V. Scali). 1” Int. Symp. Stick Insects. University of Siena, Italy. Minoo, P. and Postlethwait, J. H. 1985. Biosynthesis of Drosophila yolk polypeptides. Archj. Insect Biochem. Physiol., 2,7-27.

Mullins, D. E. and Cochran, D. G. 1974. Nitrogen metabolism in the American cockroach: an examination of whole body and fat body regulation of cations in response to nitrogen balance. J. Exp. Biol., 61,557-570. Mullins, D. E. and Cochran. D. G. 1976. A comparative study of nitrogen excretion in twenty-three cockroach species. Camp. Biochem. Physiol. (A), 53,393-399. Raikhel, A. S. 1987. Monoclonal antibodies as probes for processing of the mosquito yolk protein; A high-resolution immunolocalization of secretory and accumulative pathways. TLrsue & Cell, 19,515-529. Raikhel, A. S. and Lea, A. 0. 1983. Previtellogenic development and vitellogenin synthesis in the fat body of a mosquito: an ultrastructural and immunocytochemical study. Tissue & Cell, 15,281-300. Raikhel, A. S., Pratt, L. H. and Lea, A. 0. 1986. Monoclonal antibodies as probes for processing of yolk protein in the mosquito; production and characterization. J. In.wctPhysiol., J&879-890. Robinson, J. M. and Karnovsky, M. J. 1983. Ultrastructural localization of several phosphatases with Cerium. J. Hisrochem. Cyrochem., 31,1197-1208.

Stoppie, P., Briers, T., Huybrechts, R. and De Loof, A. 1981. Moulting hormone, juvenile hormone and the ultrastructure of the fat body of adult Sarcophagu bullara (Diptera). Cell Ti.wue Res., 221,233-244. Thomsen, E. and Thomsen. M. 1974. Fine structure of the fat body of the female of Colliphora erylhrocephola during the first egg maturation cycle. Cell Ticsue Res., lS2,193-217. Thomsen, E., Hansen, B. L., Hansen, G. N. and Jensen, P. V. 1980. Ultrastructural immunocytochcmical localization of vitellogenin in the fat body of the blowfly, Calliphora vicina Rob.-Desv. (erythrocephalo Meig.) by use of the unlabelled antibody-enzyme method. Cell Tissue Res.. 208,445-455. Wigglesworth, V. B. 1987. Histochemical studies of uric acid in some insects. I. Storage in the fat body of Periplanera omericona and the action of the symbiotic bacteria. Tissue & Cell, 19,8f91. Wolfe, J., Akam, M. E. and Roberts, D. B. 1977. Biochemical and immunological studies on larval serum protein 1, the major hemolymph protein of Drosophila melanogaster third instar larvae. Eur. J. Biochem., 79,47-53. Wyatt, G. R. 1980. The fat body as a protein factory. In Insect Bio/ogy in the Furwe (eds M. Locke and D. S. Smith), pp. 201-225. New York, Academic Press. Wyatt, G. R. and Pan, M. L. 1978. Insect plasma proteins. Ann. Rev. Biochem.. 47,779-817.