Exchange Of Macromolecules Research Articles

Toward a more complete 3-D structure of the nuclear pore complex

The nuclear pore complex (NPC) is a large supramolecular assembly embedded in the double-membraned nuclear envelope (NE) that plays a pivotal role in the exchange of macromolecules and particles between the nucleus and the cytoplasm. Applying various methods of sample preparation to Xenopus laevis whole nuclei and isolated NEs in combination with conventional transmission electron microscopy and digital image processing, we have characterized several distinct components of the NPC, including massive cytoplasmic and more tenuous nuclear rings, NPCs devoid of their cytoplasmic or both rings, and prominent “knobs” that protrude from the periphery of the NPC proper into the lumen of the NE. Moreover, by quick freezing/freeze drying/rotary metal shadowing isolated NEs, we have visualized two distinct types of NPC-associated filaments: (1) eight short, highly twisted filaments that project from the cytoplasmic ring and sometimes collapse into short cylinders; and (2) eight long, thin filaments that protrude from the nuclear ring and whose ends join to form a distal ring centered above the NPC such that the assembly resembles a “fishtrap.” These nuclear fishtraps are sensitive to divalent cations: removal unfolds them and addition reforms them. The significance of these various structural components in terms of current NPC models is discussed, and the emerging asymmetry of the NPC relative to its nuclear and cytoplasmic face is stressed.

The endothelial vesicle system in cryofixed frog mesenteric capillaries analysed by ultrathin serial sectioning

Conventional EM sections of chemically fixed capillary endothelial cells reveal numerous apparently free smooth plasmalemmal vesicles. However, the method of ultrathin (less than 150 A) serial sectioning has shown that the smooth vesicle profiles arise merely as a result of the EM thin sectioning of two sets of complex vesicular invaginations from the luminal and abluminal cell surfaces, which end blindly in the cytoplasm. While 50-70% of the total population of vesicular profiles appear to lack connections to the cell surface in conventional (500-700 A thick) EM thin sections less than 1% truly free vesicles can be found by the ultrathin serial section analyses. In the present study it is examined whether similar conclusions apply to endothelial cells which were directly frozen by slam-freezing and subsequently freeze-substituted. The three-dimensional organization of the plasmalemmal vesicular system was analyzed in four series of 19, 18, 13, and 10 ultrathin sections (approximately 110 A thick) of capillaries from frog mesenteries quickly excised from decapitated frogs (Rana pipiens). None of 920 vesicular profiles (diameter 500-1,200 A) which appeared free in individual thin sections of the series represented free vesicles; all profiles either communicated with other vesicles, the cell surface, or in rare cases turned out to be part of cytoplasmic tubular membrane structures. It is concluded that free smooth plasmalemmal vesicles are very rare in rapidly frozen as well as in directly fixed frog capillary endothelium. The volume density of profiles (13-15%), the proportion of apparently free vesicle profiles (70%), and interconnected profiles (20%) were similar to the picture previously found in single EM sections of frog mesenteric capillaries. No transendothelial channels were found in the four series of ultrathin sections of capillaries. However, continuities between the luminal and abluminal cell surfaces were seen in the endothelium of venules. Furthermore, in the ultrathin series of the capillaries, vesicular units belonging to the two sets of invaginations and cytoplasmic tubular membrane structures were in more cases found in very close contact-as fused to share one unit membrane. If this finding is representative for the in vivo situation, it may reflect that the vesicular system represents a highly dynamic system with possibilities for mixing of membranes, cellular traffic of lipid, membrane proteins, and receptors between internal compartments and the cell surfaces, as well as occasional exchange of macromolecules between blood and tissue through rare temporary connections between the two sets of surface invaginations, without actually moving vesicles.

The role of capillarization in hepatic failure: Studies in carbon tetrachloride-induced cirrhosis

During the cirrhotic process, the hepatic microvascular phenotype is transformed from sinusoids (discontinuous capillaries) into continuous capillaries. This transformation has been termed capillarization. Many hepatic functions depend on the rapid, bidirectional exchange of macromolecules between plasma and hepatocytes. To determine whether capillarization contributes to hepatic failure in cirrhosis, we decided to study the plasma clearance (125I) and hepatocyte uptake (electron microscopy) of three tracers in normal and cirrhotic rats. The tracers chosen were a hemeundecapeptide with peroxidatic activity (fluid-phase pinocytosis), asialofetuin (receptor-mediated endocytosis of a medium size protein) and ferritin (receptor-mediated endocytosis of a large size protein). The results demonstrate a decreased hepatocyte uptake of hemeundecapeptide; a significant delay in plasma clearance of asialofetuin; and a minor delay in plasma clearance of ferritin, but a striking trapping of ferritin in the cirrhotic capillary basement membrane. The delayed plasma clearance in cirrhosis cannot be ascribed to a decreased number of surface receptors because, in isolated hepatocytes, the number of molecules bound per cell was equivalent in normal and cirrhotic livers. These findings support the concept of capillarization, with the formation of continuous diffusion and filtration barriers between plasma and hepatocytes, representing a significant hindrance to the bidirectional macromolecular exchange normally taking place between these two compartments. Furthermore, at least in the case of ferritin, the capillary basement membrane of cirrhotic livers seems to be the major filtration barrier. This hindrance to hepatocyte uptake, and presumably also to secretion, may be the cause (or at least a major determinant) of the hepatic failure characteristic of cirrhosis.

Understanding the coronary circulation through studies at the microvascular level.

Studies of the coronary circulation have divided vascular resistances into three large components: large vessels, small resistance vessels, and veins. Studies of the epicardial microcirculation in the beating heart using stroboscopic illumination have suggested that resistance is more precisely controlled in different segments of the circulation. Measurements of coronary pressure in different sized arteries and arterioles have indicated that under normal conditions, 45-50% of total coronary vascular resistance resides in vessels larger than 100 microns. This distribution of vascular resistance can be altered in a nonuniform manner by a variety of physiological (autoregulation, increases in myocardial oxygen consumption, sympathetic stimulation) and pharmacological stimuli (norepinephrine, papaverine, dipyridamole, serotonin, vasopressin, nitroglycerin, adenosine, and endothelin). Studies of exchange of macromolecules in the microcirculation using fluorescent-labeled dextrans have also identified the size of the small pore (35-50 A) in coronary microvessels that can be altered by myocardial ischemia. Studies of the coronary microcirculation have demonstrated that the control of vascular resistance is extremely complex, and mechanisms responsible for these heterogeneous responses need further examination.

EM visualization of nucleocytoplasmic transport processes

The nuclear envelope is strategically located between the nucleoplasm and cytoplasm, and, as such, can play a major role in controlling cellular activity by regulating the exchange of macromolecules between these two compartments. The nuclear pore complexes, which are located within circular areas formed by fusion of the inner and outer membranes of the envelope, represent the primary, if not the exclusive, exchange sites. Individual pores are able to function in both protein import and RNA efflux from the nucleus. Translocation of macromolecules occurs by either passive diffusion or facilitated transport through central channels within the pores. The functional size of the diffusion channel is approximately 9 to over 12 nm in diameter depending on the cell type. The width of the transport channel varies as a function of the number and effectiveness of the specific nuclear targeting signals contained within the permeant molecule. The maximum diameter of the channel can be over 26 nm. Nucleocytoplasmic exchanges can be regulated either by (1) differences in the properties of the transported molecule (molecular size and signal content) or (2) changes in the properties of the pore complexes, which can effect both diffusion and transport.

Nuclear Envelope Dynamics During Mitosis

The nuclear envelope is a complex membrane structure that forms the boundary of the nuclear compartment in eukaryotes. It regulates the passage of macromolecules between the two compartments and may be important for organizing interphase chromosome architecture. In interphase animal cells it forms a remarkably stable structure consisting of a double membrane ouerlying a protein meshwork or lamina and penetrated by nuclear pore complexes. The latter form the channels for nucleocytoplasmic exchange of macromolecules, At the onset of mitosis, however, it rapidly disassembles, the membranes fragment to yield small vesicles and the lamina, which is composed of predominantly three polypeptides, lamins R, B and C (MW approx. 74, 68 and 65 kDa respectiuely), breaks down. Lamins B and C are dispersed as monomers throughout the mitotic cytoplasm, while lamin B remains associated with the nuclear membrane vesicles.

Heterogeneity in basement membranes of skin--electrostatic properties of basal lamina-associated anionic sites.

AbstractIn order to evaluate the charge‐selective filter function of basement membranes (BMs), electrostatic properties of anionic sites on the BMs of dermo‐epidermal junction were compared with those of the nuclear (the thick capillary wall) and the attenuated (the thin capillary wall) portions of capillary and other small blood vessels by determining salt concentrations required to remove ruthenium red from these BMs of the rat skin previously perfused with the cationic dye. The concentration of NaCl which was required to remove ruthenium red was more than 1.4M for the attenuated portion of capillary and only 1.2M for the other BMs, indicating that the former is more highly negatively charged than the latter. It is suggested that the BM of the attenuated portion of the capillary expedites the charge‐selective exchange of macromolecules more effectively than the other BMs.

Dose-related effects of adenosine and bradykinin on microvascular permselectivity to macromolecules in the hamster cheek pouch.

The hamster cheek pouch preparation was used to assess microvascular permselectivity responses to three vasodilating agents: bradykinin, adenosine, and papaverine. Fluorescein isothiocyanate-dextran 150 was injected intravenously as a macromolecular tracer. To quantify changes in permeability, we calculated fluorochrome clearance values from the ratio of suffusate to plasma fluorescein isothiocyanate-dextran 150 concentration. The microcirculation was recorded on videotape, using epifluorescence and bright-field light microscopy. Topical application of bradykinin elicited dose-dependent increases in macromolecular permeability. Adenosine also augmented permeability in a dose-dependent fashion. The increases in tracer clearance, relative to control, were 9.4 nl/min for 10(-5) M adenosine and 39.4 nl/min for 10(-4) M adenosine. The standard error for these doses was 1.5 nl/min. Adenosine, 10(-6) M, did not alter permeability. The increment in clearance induced by 10(-4) M was comparable to that of bradykinin, 8 X 10(-7) M. Pretreatment with phenidone had no effect on the permeability response mediated by 10(-5) M adenosine. Topical application of papaverine enhanced the transvascular exchange of macromolecules in one-half of the preparations examined. Comparable doses of adenosine were approximately three times as effective. This study indicated that adenosine, like bradykinin, is capable of modifying microvascular permeability responses in the hamster cheek pouch. This modulatory effect appears to be due to a direct action on the postcapillary microvascular membrane.

Distribution and characterization of anionic sites in trophoblast and capillary basal laminas of human placental villi.

The distribution of anionic sites was studied in the trophoblastic and fetal capillary basal laminas of developing human placental villi with the cationic stain ruthenium red. At 7-12 weeks of gestation the trophoblastic basal lamina (TBL) contained ruthenium red-positive granules in a quasi-regular array throughout the lamina densa or sometimes concentrated at the interstitial surface of the lamina densa. The capillary basal lamina (CBL) (and anionic sites) were not present at this age. Anionic sites were also associated with collagen or reticular fibrils. At term, the TBL was largely devoid of anionic sites except for some distributed along its interstitial surface. The CBL was present in later gestation and sometimes had arrays of anionic sites. In order to characterize the anionic sites, minced pieces of villi were incubated in the presence or absence of either chondroitinase ABC, heparitinase, neuraminidase, or Streptomyces hyaluronidase in appropriate buffer systems. Incubation of early villi with heparitinase resulted in the disappearance of the TBL-associated sites. Chondroitinase ABC appeared to reduce staining of collagen-associated sites. In term villi, heparitinase removed those few sites still associated with the TBL but did not affect sites associated with the CBL or collagen. Chondroitinase ABC resulted in the disappearance of all anionic sites. In later gestation, a number of developmentally important macromolecules are transported across the trophoblast and enter the fetal capillaries. We conclude that the absence of an array of polyanionic sites from the term placenta TBL and the reduction in the amount of extracellular matrix intervening between the trophoblast and capillaries are adaptations to enhance the exchange of macromolecules across the placenta.

Exchange of macromolecules between peritoneal cavity and plasma.

The exchange of fluorescein isothiocyanate-labeled dextrans ranging in weight-averaged molecular weight from 19,400 to 160,000 and 125I-bovine serum albumin (BSA) between dialysis fluid (5% BSA in Krebs-Ringer solution) in the peritoneal cavity and the plasma was studied in anesthetized female Sprague-Dawley rats. Plasma and peritoneal samples were collected for 3-4 h after either 1) an intraperitoneal injection of dialysis fluid with tracer or 2) an intravenous injection of tracer material simultaneously with an intraperitoneal injection of dialysis solution without tracer. Analysis of the data by means of a mathematical model of the transport process suggests a functional asymmetry in transport of large molecules across the blood capillary wall. Substances injected intravenously have a net transport from the blood capillaries to the peritoneal cavity. Substances of molecular weight greater than or equal to 39,000 transport from the cavity to the plasma via peritoneal lymphatics; 19,400 molecular-weight dextran transports from the cavity to the plasma primarily via lymphatics with some blood capillary uptake. Tissue diffusivities and capillary mass transport coefficients are derived for the substances tested.

The plasmalemmal vesicular system in striated muscle capillaries and in pericytes

By ultrathin serial sectioning of frog mesenteric capillaries it was recently demonstrated that the many apparently free vesicles in electron microscope (EM) sections of endothelial cells may be artefacts due to conventional (500–700 Å thick) sectioning (Frøkjaer-Jensen, 1980). The vesicles were found to be part of two sets of invaginations of the cell surfaces; one set connected to the lumen, the other to the interstitium. The present study extends this view to comprise the vesicle organization in frog striated muscle capillaries. By analysis of the three-dimensional organization of the plasmalemmal vesicles in 21 ultrathin serial sections (120–150 Å) of two muscle capillaries it is demonstrated that less than 1% of the about 70% apparently free vesicles seen in conventional thin sections of the same capillaries in fact represent truly free vesicular units. By analysis of 15 conventional EM cross-sections of capillaries from the frog cutaneous-pectoris muscle containing plasmaproteins in high concentration it is furthermore demonstrated that 48% of the total vesicle population connect to the lumen at the time of fixation. This organization of the vesicular system seems incompatible with the concept that macromolecules are transferred across the capillary wall by vesicular transport or by a series of fusions and fissions between individual cytoplasmic vesicles but is compatible with the notion that macromolecules exchange across capillary walls by means of passive processes such as diffusion and convection through rare ‘large pores’. The study emphasizes that any attempts to classify vesicles in conventional thin sections as ‘luminal’, ‘cytoplasmic’ and ‘abluminal’ is impossible and may lead to erroneous interpretations of vesicle involvement in transcapillary exchange of macromolecules. The rare occurrence of transendothelial channels compared to the number of vesicle invaginations suggests that the main function of the vesicular system relates to functions other than transport.

Separate protein body compartments of the worker honeybee ( Apis mellifera L.)

1. 1. Protein of adult worker bees are secreted from their exocrine glands in royal jelly and in venom or in eggs laid by laying workers in queenless colonies. 2. 2. The present report examines the protein composition of head and thoracic glands, venom glands, fat body and ovaries, the relation of these proteins to haemolymph proteins and the possible exchange of macromolecules between glandular compartments, the haemocoele and the organs located in the haemocoele. 3. 3. The identity of proteins from various compartments was established by double-diffusion, immunoelectrophoresis, gradient acrylamide slab gel electrophoresis and isoelectric focusing. 4. 4. Royal jelly is composed of proteins secreted from hypopharyngeal, mandibular and postcerebral, but not from thoracic glands. 5. 5. No immunological identity was found between the proteins of royal jelly, venom and haemolymph. The glands producing the two secretions seem to be separate compartments from the haemocoele with regards to macromolecular traffic. 6. 6. In contrast to the exocrine glands many of haemolymph proteins were shared by the fat body and the ovaries. 7. 7. Some aspects of macromolecular traffic between the body compartments in insects are discussed.

Interstitial fluid: exchange of macromolecules between plasma and skin interstitium.

Fluid was collected from blisters developed by suction on the abdominal skin in ten normal volunteers. During the same study the transcapillary escape rate of intravenously injected 131I-albumin and 125I-IgG together with the accumulation of these tracers in the blister fluid was measured. The ratios between the concentrations of endogenous albumin, transferrin, IgG and alpha 2-macroglobulin in the blister fluid (Cb) and serum (Cs) correlated directly with the free diffusion coefficients indicating that the sieving properties of the vascular endothelium remained intact during the suction. The ratios of intravascular to total masses of the four endogenous proteins calculated from plasma volume, Cs, Cb and extracellular fluid volume determined with inulin and Cb were very similar to those obtained in long-term metabolic turnover studies. The ratio between the accumulated IgG and albumin tracers in the blister fluid (0.80) was identical with the Cb/Cs ratio between endogenous IgG and albumin (0.78) but higher than the ratio between the IgG and albumin escape rates from the total microvasculature (0.58). From the protein tracer determinations it cannot be definitely excluded that suction increases the microvascular escape rates. If the distribution volume of inulin equates that of albumin in steady-state it can, however, be concluded from the data given on the endogenous plasma protein, that blister fluid from the abdominal skin corresponds to average interstitial tissue fluid.

Exchange of Macromolecules between Plasma and Skin Interstitium in Extensive Skin Disease

The concentrations of albumin, transferrin, IgG and alpha 2-macroglobulin were measured in serum (C8) and in blister fluid (Cb) from lesional skin obtained by suction in 11 patients with extensive skin disease. The results were compared with those of 10 matched control subjects. In the patients the Cb/C8 ratios and the distribution ratios (i.e., intravascular to total masses) of the 4 proteins and that of inulin were correlated to the corresponding molecular weights. The distribution ratios of the proteins were calculated from plasma volume (PV), C8, Cb and extracellular fluid volume (ECV) determined with inulin, and that of inulin as PV/ECV. The distribution ratio of albumin was similar to that obtained from a long-term metabolic turnover study. It is concluded that the sieving properties of the dermal vascular endothelium is maintained during suction also in patients with extensive skin disease and that the blister fluid equals "average" interstitial fluid. The transcapillary escape rate (TER) of albumin and IgG (i.e., fractions of intravascular masses of albumin and IgG passing to the extravascular space per unit time) and the accumulation rate of i.v. injected 131I human serum albumin and 125I-IgG in the blister fluid was significantly increased in the patients. The latter was increased 2 to 3-fold. It is shown, that the increased extravasation of plasma proteins from the skin microvasculature accounts fully for the increased escape rates as measured in whole man, of albumin as well as IgG.

Solute permeability of the alveolar epithelium in alloxan edema in dogs.

The permeability of the alveolar epithelium following alloxan challenge was studied in dogs by determining transfer of radiolabeled solutes between alveolus and blood. Two days after injection of 131-Ialbumin into the blood, anesthetized dogs had the air space of part of one lung isolated by a balloon catheter lodged in a bronchus. We infused the atelectatis-isolated area with normal saline containing trace amounts of Blue Dextran, 125Ialbumin, and 57Co-cyanocobalamin; challenged six animals with intravenous alloxan, and six animals with alloxan added to the alveolar saline. During the pulmonary edema, 57Co-cyanocabalamin and 125I-albumin appeared in the blood and 131I-albumin entered the alveolar saline. The animals challenged by alveolar instillation showed a greater permeability change (P less than 0.05). The bidirectional transfer of macromolecules indicates that alloxan produces a change in the permeability of the alveolar epithelium, allowing diffusional exchange of macromolecules. Since alveolar flooding in hemodynamic edema does not show a similar change in the permeability of the epithelial lining, alveolar flooding in alloxan edema is not due solely to an effect on the endothelial membrane, but also to a direct effect on the epithelial membrane.

The structural basis of the blood-ocular barriers

In the mammalian eye, endothelial barriers are localized in the vessels of (i) the retina, (ii) optic nerve, (iii) ciliary muscle and (iv) iris. The endothelial barrier of the vessels in the retina is insensitive to drugs or autocoids, whereas the vessels of the iris may respond to pharmacological manipulations with increased permeability. Epithelial barriers are present in (i) the pigment epithelium of the retina, (ii) the nonpigmented layer of the ciliary epithelium, and (iii) the posterior layer of the iridial epithelium. The blood-retina barrier has two components, the endothelium of retinal vessels and the pigment epithelium. The blood-aqueous barrier also consists of two components: one is localized in the ciliary and iridial epithelia and protects the posterior chamber from circulating macromolecules; the other is an endothelial barrier which prevents movement of macromolecules from the lumen of the iris vessels into the iridial stroma. The walls of the anterior chamber, unlike those of the posterior chamber, are highly permeable and permit free exchange of macromolecules between the aqueous humor and the connective tissue spaces of the iris, cornea and limbus. Two factors are essential to maintain the normal composition of the aqueous humor: (i) junctional barriers along the walls of the posterior chamber and the endothelium of the iris vessels; (ii) continuous, unidirectional flow of the aqueous humor from its site of origin in the posterior chamber to its site of reabsorption at the sclero-corneal angle. When the intraocular pressure becomes lower than the episcleral venous pressure, the direction of flow of the ocular fluids is reversed and large amounts of plasma proteins enter the anterior chamber from the blood-filled Schlemm canal.

Fungal fimbriae. II. Their role in conjugation in Ustilago violacea.

During conjugation in the anther smut fungus Ustilago violacea cells of opposite mating type first pair tightly and then develop a conjugation tube or bridge between them. The cells of both mating types are covered in long fine hairs or fimbriae, some of which appear to end in knobs. Experiments involving enzyme treatments of the cell surface indicate that these fimbriae do not play an essential role in cell pairing, instead pairing seems to be initiated when one or both mating types produce amorphous masses of alpha-amylase-sensitive material. Electron micrographs, enzyme and inhibitor studies, and experiments using restrictive temperatures suggest, however, that fimbriae may be essential for the later stages of conjugation i.e. development of the conjugation tube. If so, it is suggested that they may permit the exchange of macromolecules between the conjugating cells, initiating localized wall-softening and wall-breakdown.

Nucleocytoplasmic exchange of macromolecules

Quantitative measurements of the cytoplasm-to-nucleus exchange of specific protein tracers were correlated with known physical properties (size and electrical charge) of the proteins. Tracers differing in their molecular parameters were produced by fluorescence labelling of wellcharacterized proteins (bovine serum albumin, mol. wt 67 500; ovalbumin, mol. wt 45 000; myoglobin, mol. wt 17 500; lysozyme, mol. wt 14 500; and cytochrome c, mol. wt 13 000) with fluorescein isothiocyanate. The labelled proteins were microinjected into the cytoplasm of living cells, and their uptake into the nucleus was followed by quantitative fluorescence microscopy. In addition, the distribution of cytoplasmically injected ferritin (mol. wt 465 000) was observed with the electron microscope. Five hours after injection, ferritin was localized almost entirely in the cytoplasm and serum albumin remained heavily concentrated in the cytoplasm. Ovalbumin showed significant nuclear uptake within 15 min, but remained more concentrated in the cytoplasm for up to 5 h. Cytochrome c was approximately evenly distributed between cytoplasm and nucleus within 5 min and remained so up to 5 h. Myoglobin and lysozyme each showed significant nuclear concentration above the cytoplasmic level within 3 min. Thus, the cytoplasm-to-nucleus transfer of proteins was limited by their molecular size. Small proteins, neutrally and positively charged, showed rapid nuclear incorporation. Nuclear concentration occurred with some, but not all, small proteins, and was not based simply on the size or charge of the penetrant.

THE ORIGIN, DISTRIBUTION AND FATE OF THE MOLTING FLUID PROTEINS OF THE CECROPIA SILKWORM.

1. Molting in insects is always accompanied by the production of a molting fluid which fills the exuvial space between the new and the old cuticle and digests the inner layers of the old cuticle. In Hyalophora cecropia, molting fluid is secreted at the outset of adult development and persists until two days before eclosion, whereupon it is absorbed. 2. The present report examines the protein composition of the molting fluid of Cecropia, the origin of the molting fluid proteins, the relation of these proteins to blood proteins and the exchange of macromolecules between the molting fluid and the blood. It also examines the sites of absorption of molting fluid. 3. Disc electrophoresis on acrylamide gels reveals that the molting fluid of Cecropia contains about fifteen protein bands which can be resolved at pH 8.6. Some of these protein bands are detected in the molting fluid at all stages, whereas others appear only at specific times. About ten of the bands are peculiar to molting fluid and are not detected in the blood. About five bands are detectable in both blood and molting fluid, but none of these common bands appears to be a major component of the molting fluid, and only one is a major blood protein. In contrast, the epidermis contains most of the major protein bands found in molting fluid but lacks all but one of the major protein bands present in the blood. 4. Immunological analysis reveals that blood and molting fluid share five antigens. At least four of these common antigens also occur in the epidermis which appears to secrete these antigens into both the molting fluid and the blood. 5. Native and foreign proteins do not penetrate from the exuvial space into the blood or vice versa. Apparently the epidermis and cuticle act as a barrier to the exchange of most macromolecules between the blood and molting fluid. The exuvial space is clearly a separate fluid compartment. 6. In addition the exuvial space itself is compartmentalized and the fluids in the compartments do not admix several days before eclosion. 7. Absorption of molting fluid during the final two days of adult development occurs most readily through particular regions of the integument. In the abdomen the principal sites of absorption are pits which represent the points through which tonofibrils make attachment to the old cuticle. Two days before ecdysis, the attachments between the tonofibrils and the pupal cuticle rupture, exposing the points of attachment on the new cuticle. It is through these exposed surfaces that much of the molting fluid is absorbed. Molting fluid is also absorbed in the head and thorax through various flexible membranes at the bases of the appendages.

Nuclear pore flow rate. A characteristic for nucleocytoplasmic exchange of macromolecules and particles.

Nuclear pore flow rate. A characteristic for nucleocytoplasmic exchange of macromolecules and particles.