Accounting for the shapes and size distributions of miniature endplate currents.

The current model does not account adequately for the characteristics of miniature endplate currents (MEPCs). We do not understand their relatively slow rise, the shape of their rise, their variable and sometimes prolonged decay, and the correlation between amplitude and decay time. If we assume that ACh is released from the vesicle through a pore and that the vesicle enlarges as it takes on additional transmitter, the predictions are more like MEPCs. However, previous measurements showed that after quantal size was increased the vesicles in the terminal were not enlarged. This need not be a problem, because some of the ACh is added to vesicles positioned at the active zones, a process known as second-stage loading. By using the false transmitter precursor monoethylcholine we provide additional evidence for second-stage loading. The distribution of quantal sizes at the junction usually does not follow a normal probability distribution; it is skewed to the right. The skew can be accounted for by a model incorporating second-stage loading in which the vesicles are released randomly, without regard to their ACh content. If the vesicles increase in size when they contain more transmitter, only vesicles at the active zone need swell.     

Non-quantal secretion of acetylcholine from motor nerve endings: Molecular mechanisms, physiological role, and regulation

Results of biochemical and electrophysiological experiments allowing researchers to identify non-quantal release of acetylcholine (ACh), in addition to quantal release, from motor nerve endings, are discussed in the lecture. Based on the analysis of our own data and publications of other experimenters on the dependence of non-quantal secretion on the composition of the ion milieu, sensitivity of this phenomenon to different physiologically active compounds, and peculiarities of its temperature dependence, the authors conclude that non-quantal secretion of the transmitter is an active transport process, and not a passive leakage of ACh from the cytoplasm of the nerve terminal. It is hypothesized that a high-affinity system of choline uptake can play the role of the ACh carrier through the presynaptic membrane. The involvement of non-quantal release in the control of electrogenesis in the muscle fiber and the relation between processes of quantal and non-quantal secretion of the transmitter providing adequate functioning of the nerve terminal in the resting state and in the course of long-lasting high-frequency rhythmic activity are described. Data on the ability of glutamate and nitric oxide to selectively modulate this type of neurosecretion are analyzed. The possible role of non-quantal secretion of ACh in pathogenesis of intoxications of ACh esterase is discussed. Neirofiziologiya/Neurophysiology, Vol. 39, Nos. 4/5, pp. 352–363, July–October, 2007.     

The release of acetylcholine: from a cellular towards a molecular mechanism

The isolation of synaptic vesicles rich in acetylcholine (ACh) from the electric organ of Torpedo has indeed strengthened the hypothesis of transmitter exocytosis, but soon after it was found that non-vesicular free ACh was released and renewed upon stimulation. In contrast, vesicular ACh and the number of vesicles remained stable during physiological stimulations. In addition free ACh variations (representing the cytoplasmic pool) were correlated to the release kinetics as measured by the electroplaque discharge. Consequently, the mechanism releasing ACh from the cytoplasm in a packet form was searched at the presynaptic membrane itself. With synaptosomes isolated from the electric organ of Torpedo, it became possible to freeze them rapidly at the peak of ACh release and study their membrane and contents after cryofracture. A statistical analysis showed that the main structural change was the occurrence of large intramembrane particles at the peak of ACh release and under all release conditions. This impressive change contrasted with the stability in the number of vesicles. Another role for the vesicle was envisaged during intense stimulations when the cytoplasmic ACh and ATP pools become exhausted. The decrease in ATP leads to an increase in calcium and protons in the cytoplasm; this signals the depletion of vesicular ACh and ATP stores in the cytoplasm. Release can go on, while ATP promotes the uptake of calcium by vesicles. At the end of its cycle the vesicle will be full of calcium and will perhaps release it. As far as the mechanism of ACh release is concerned it probably depends on a membrane component (perhaps the large particles) activated by calcium and able to translocate ACh in a quantal or subquantal form. In most recent work we showed that if a lyophilized presynaptic membrane was used to make proteoliposomes filled with ACh, they released ACh upon calcium action.     

Three types of transmitter release from embryonic neurons

Prior to the contact with their target muscle cells in culture, growth cones of many isolated Xenopus embryonic neurons release acetylcholine (ACh) spontaneously. Using patch clamp techniques, this release can be detected by an outside-out patch of muscle membrane placed near the growth cone. Intracellular recording from innervated muscle cells showed spontaneous miniature endplate potentials (MEPPs) of varying amplitudes. Amplitude histograms showed a skewed distribution with multiple peaks, suggesting the existence of subunits in either the quantal packages of ACh released by the nerve terminal or in the postsynaptic muscle response. In addition to the quantal ACh release reflected by MEPPs, nerve terminal also release a large amount of ACh in a non-quantal fashion. This non-quantal ACh release is revealed by the hyperpolarization of the muscle membrane following extracellular application of curare or alpha-bungarotoxin, as well as by denervation of the muscle cell.     

A method for incorporating macromolecules into adherent cells

We describe a simple method for loading exogenous macromolecules into the cytoplasm of mammalian cells adherent to tissue culture dishes. Culture medium was replaced with a thin layer of fluorescently labeled macromolecules, the cells were harvested from the substrate by scraping with a rubber policeman, transferred immediately to ice cold media, washed, and then replated for culture. We refer to the method as "scrape-loading." Viability of cells was 50-60% immediately after scrape-loading and was 90% for those cells remaining after 24 h of culture. About 40% of adherent, well-spread fibroblasts contained fluorescent molecules 18 h after scrape-loading of labeled dextrans, ovalbumin, or immunoglobulin-G. On average, 10(7) dextran molecules (70,000-mol wt) were incorporated into each fibroblast by scrape-loading in 10 mg/ml dextran. The extent of loading depended on the concentration and molecular weight of the dextrans used. A fluorescent analog of actin could also be loaded into fibroblasts where it labeled stress fibers. HeLa cells, a macrophage-like cell line, 1774A.1, and human neutrophils were all successfully loaded with dextran by scraping. The method of scrape-loading should be applicable to a broad range of adherent cell types, and useful for loading of diverse kinds of macromolecules.     

Quantal size is dependent on stimulation frequency and calcium entry in calf chromaffin cells

To what extent the quantal hypothesis of transmitter release applies to dense-core vesicle (DCV) secretion is unknown. We determined the characteristics of individual secretory events in calf chromaffin cells using catecholamine amperometry combined with different patterns of stimulation. Raising the frequency of action potential trains from 0.25-10 Hz in 2 mM [Ca(2+)]o or [Ca(2+)]o from 0.25-7 mM at 7 Hz elevated the amount released per event (quantal size). With increased stimulation, quantal size rose continuously, not abruptly, suggesting that release efficiency from a single population of DCVs rather than recruitment of different-sized vesicles contributed to the effect. These results suggest that catecholamine secretion does not conform to the quantal model. Inhibition of rapid endocytosis damped secretion in successive episodes, implying an essential role for this process in the recycling of vesicles needed for continuous secretion.     

Evoked Acetylcholine Release by Immortalized Brain Endothelial Cells Genetically Modified to Express Choline Acetyltransferase and/or the Vesicular Acetylcholine Transporter

Immortalized rat brain endothelial RBE4 cells do not express choline acetyltransferase (ChAT), but they do express an endogenous machinery that enables them to release specifically acetylcholine (ACh) on calcium entry when they have been passively loaded with the neurotransmitter. Indeed, we have previously reported that these cells do not release glutamate or GABA after loading with these transmitters. The present study was set up to engineer stable cell lines producing ACh by transfecting them with an expression vector construct containing the rat ChAT. ChAT transfectants expressed a high level of ChAT activity and accumulated endogenous ACh. We examined evoked ACh release from RBE4 cells using two parallel approaches. First, Ca2+-dependent ACh release induced by a calcium ionophore was followed with a chemiluminescent procedure. We showed that ChAT-transfected cells released the transmitter they had synthesized and accumulated in the presence of an esterase inhibitor. Second, ACh released on an electrical depolarization was detected in real time by a whole-cell voltage-clamped Xenopus myocyte in contact with the cell. Whether cells synthesized ACh or whether they were passively loaded with ACh, electrical stimulation elicited the release of ACh quanta detected as inward synaptic-like currents in the myocyte. Repetitive stimulation elicited a continuous train of responses of decreasing amplitudes, with rare failures. Amplitude analysis showed that the currents peaked at preferential levels, as if they were multiples of an elementary component. Furthermore, we selected an RBE4 transgenic clone exhibiting a high level of ChAT activity to introduce the Torpedo vesicular ACh transporter (VAChT) gene. However, as the expression of ChAT was inactivated in stable VAChT transfectants, the potential influence of VAChT on evoked ACh release could only be studied on cells passively loaded with ACh. VAChT expression modified the pattern of ACh delivery on repetitive electrical stimulation. Stimulation trains evoked several groups of responses interrupted by many failures. The total amount of released ACh and the mean quantal size were not modified. As brain endothelial cells are known as suitable cellular vectors for delivering gene products to the brain, the present results suggest that RBE4 cells genetically modified to produce ACh and intrinsically able to support evoked ACh release may provide a useful tool for improving altered cholinergic function in the CNS.     

Two-photon microscopic analysis of acetylcholine-induced mucus secretion in guinea pig nasal glands

The spatiotemporal changes in intracellular free Ca(2+) concentration ([Ca(2+)](i)) as well as fluid secretion and exocytosis induced by acetylcholine (ACh) in intact acini of guinea pig nasal glands were investigated by two-photon excitation imaging. Cross-sectional images of acini loaded with the fluorescent Ca(2+) indicator fura-2 revealed that the ACh-evoked increase in [Ca(2+)](i) was immediate and spread from the apical region (the secretory pole) of acinar cells to the basal region. Immersion of acini in a solution containing a fluorescent polar tracer, sulforhodamine B (SRB), revealed that fluid secretion, detected as a rapid disappearance of SRB fluorescence from the extracellular space, occurred exclusively in the luminal region and was accompanied by a reduction in acinar cell volume. Individual exocytic events were also visualized with SRB as the formation of Omega-shaped profiles at the apical membrane. In contrast to the rapidity of fluid secretion, exocytosis of secretory granules occurred with a delay of approximately 70s relative to the increase in [Ca(2+)](i). Exocytic events also occurred deep within the cytoplasm in a sequential manner with the latency of secondary exocytosis being greatly reduced compared with that of primary exocytosis. The delay in sequential compound exocytosis relative to fluid secretion may be important for release of the viscous contents of secretory granules into the nasal cavity.     

Inhibition of packing of acetylcholine into quanta by ammonium

Soaking frog motor nerve terminals in a hypertonic solution produces an increase in the size of miniature end plate potentials (mepp's) and miniature end plate currents (mepc's) after the preparations are returned to normal Ringer's solution. There is substantial evidence that the size increase occurs because additional acetylcholine (ACh+) is incorporated into the quanta. It has been proposed that ACh+ loading into synaptic vesicles requires a proton gradient. As a step in testing this hypothesis the effects of millimolar concentrations of NH4+, methylamine+, or trimethylamine+ in the extracellular solution on the increase in quantal size were measured. These substances would be expected to accumulate in acid intracellular compartments, which would diminish the acidity. The increase in quantal size was blocked by these substances, in agreement with the idea that the proton gradient is involved in ACh+ accumulation. Tetanic stimulation in solutions containing 5 mM NH4Cl also produces a decrease in quantal size, not seen in controls in NH4+-free solution. The inhibition of transmitter packaging by ammonia may play a role in the neural sequelae of hepatic failure.     

The Role of Endogenous Calcitonin Gene-Related Peptide in the Neurotransmitter Quantal Size Increase in Mouse Neuromuscular Junctions

Changes in parameters of spontaneous acetylcholine (ACh) quantal secretion caused by prolonged high-frequency burst activity of neuromuscular junctions and possible involvement of endogenous calcitonin gene-related peptide (CGRP) and its receptors in these changes were studied. With this purpose, miniature endplate potentials (MEPPs) were recorded using standard microelectrode technique in isolated neuromuscular preparations of m. EDL–n. peroneus after a prolonged high-frequency nerve stimulation (30 Hz for 2 min). An increase in the MEPP amplitudes and time course was observed in the postactivation period that reached maximum 20–30 min after nerve stimulation and progressively faded in the following 30 min of recording. Inhibition of vesicular ACh transporter with vesamicol (1 μM) fully prevented this “wave” of the MEPP enhancement. This indicates the presynaptic origin of the MEPP amplitude increase, possibly mediated via intensification of synaptic vesicle loading with ACh and subsequent increase of the quantal size. Competitive antagonist of the CGRP receptor, truncated peptide isoform CGRP_8–37 (1 μM), had no effect on spontaneous secretion parameters by itself but was able to prevent the appearance of enhanced MEPPs in the postactivation period. This suggests the involvement of endogenous CGRP and its receptors in the observed MEPP enhancement after an intensive nerve stimulation. Ryanodine in high concentration (1 μM) that blocks ryanodine receptors and stored calcium release did not influence spontaneous ACh secretion but prevented the increase of the MEPP parameters in the postactivation period. Altogether, the data indicate that an intensive nerve stimulation, which activates neuromuscular junctions and muscle contractions, leads to a release of endogenous CGRP into synaptic cleft and this release strongly depends on the efflux of stored calcium. The released endogenous CGRP is able to exert an acute presynaptic effect on nerve terminals, which involves its specific receptor action and intracellular cascades leading to intensification of ACh loading into synaptic vesicles and an increase in the ACh quantal size.     

Procathepsin L self-association as a mechanism for selective secretion

The lysosomal cysteine pro-protease procathepsin L was enriched in dense vesicles detectable when microsomes prepared from wild-type or transformed mouse fibroblasts were resolved on sucrose gradients. These dense vesicles did not comigrate with proteins characteristic of the endoplasmic reticulum, Golgi, endosomes or lysosomes. When gradient fraction vesicles were lysed at acidic pH in the presence of excess mannose 6-phosphate to prevent binding to mannose phosphate receptors, the majority of the procathepsin L was associated with the membrane, not the soluble, fraction. Immunogold labeling of procathepsin L in thin sections of cells or gradient fractions, using antibodies directed against the propeptide to avoid detection of the mature enzyme in dense lysosomes, revealed that the proenzyme was concentrated in dense cores localized in small vesicles near the plasma membrane and in multivesicular bodies. Consistent with the density of the gradient fraction and the electron density of the cores, yeast two-hybrid assays indicated the proenzyme could bind itself but could not interact with the aspartic proprotease procathepsin D. The data suggest that in mouse fibroblasts procathepsin L may self-associate into aggregates, initiating the formation of dense vesicles that could mediate the selective secretion of procathepsin L independent of mannose phosphate receptors.     

In Vivo Production and Release of Acetylcholine from Primary Fibroblasts Genetically Modified to Express Choline Acetyltransferase

Abstract: Primary rat fibroblasts genetically modified to express Drosophila choline acetyltransferase (dChAT) synthesize and release acetylcholine (ACh) in vitro. The ACh produced from the transduced fibroblasts was found to be enhanced by increasing amounts of choline chloride in the culture media. These dChAT-expressing cells were then implanted into the intact hippocampus of adult rats and in vivo microdialysis was performed 7–10 days after grafting to assess the ability of the cells to produce ACh and respond to exogenous choline in vivo. Samples collected from anesthetized rats revealed fourfold higher levels of ACh around dChAT grafts than from either non-grafted or control-grafted hippocampi. Localized choline infusion (200 μ) through the dialysis probes was found to induce a selective twofold increase in ACh release only from the dChAT-expressing fibroblasts. These results indicate not only that dChAT-expressing fibroblasts continue to synthesize and secrete ACh for at least 10 days after intracerebral grafting, but that the levels of ACh can be manipulated in vivo. The ability to regulate products within genetically modified cells in vivo may provide a powerful avenue for exploring the role of discrete substances within the CNS.     

Electron microscopic research on the mechanism of intestinal secretion

Electron microscopic investigation of the rat small intestine revealed a great number of vesicles 50-75 nm in diameter with enterocyte microvilli. The number of vesicles increased with the increase of digestive activity in the small intestine. Vesicles were formed by gemmation of enterocyte microvilli from the lateral membrane in contraction of microvillous actin skeleton. Simultaneously with the production of exocytotic vesicles, the formation of pinocytotic vesicles in the base of microvilli was observed. There is a supposition that the vesicle gemmation is a natural process of the intestinal secretion to fulfil numerous important function: it promotes the penetration of enterocyte hydrolases into the parietal layer; equilibrates an increase in the enterocyte volume during absorption. This is a possible way of translocation of synthesized enzymes into the cytoplasm and of transport proteins on the apical surface of epithelial cells.     

Effects of Choline and Quiescence on Drosophila Choline Acetyltransferase Expression and Acetylcholine Production by Transduced Rat Fibroblasts

Rat-1 fibroblasts were transduced to express Drosophila choline acetyltransferase. The presence of an active enzyme in these cells (Rat-1/dChAT) was confirmed using various methods. Rat-1/dChAT fibroblasts released acetylcholine (ACh) into the culture medium. Moreover, intra- and extracellular levels of ACh could be increased by adding exogenous choline chloride. In addition, serum starvation or confluence-induced quiescence caused an 80% decrease in recombinant choline acetyltransferase activity (compared with actively growing cells). ACh release was also repressed in quiescent fibroblast cultures. Exogenous choline could mitigate the decrease in ACh release. These results indicate that Rat-1 fibroblasts can be genetically modified to produce ACh and that ACh release can be controlled by introducing choline into the culture medium. Furthermore, these data demonstrate that the expression of the retroviral promoter used in this study decreases with the onset of quiescence; however, exogenous choline can increase the amount of ACh released by quiescent fibroblasts.     

Potentiating effect of allatostatin on transmitter quantal secretion in the mouse nerve-muscle synapse

Action of allatostatin on the spontaneous and evoked quantal acetylcholine secretion was studied for the first time in the mouse nerve-muscle synapse. End plate miniature potentials (EPMP) and miniature currents (EPMC) as well as end plate single evoked currents (SEP) were recorded in mouse semidiaphragm. Allatostatin (1 nm–1 μM) produced a dose-dependent increase of the EPMP amplitude (that reached 209% of control at 1 μM of peptide), but without affecting statistically significantly the EPMP frequency and membrane potential of muscle fibers throughout the entire range of its concentrations. The potentiating action of 1 μM peptide on the EPMP amplitude was accompanied by a rise of time of the EPMP ascent and semidrop (by 17 and 13%, respectively). Allatostatin (1 μM) caused a twofold rise of EPMC amplitude, but the time parameters of miniature postsynaptic currents did not change statistically significantly. Amplitude of SEP also increased more than twice under effect of 1 μM peptide, but the SEP quantal composition remained at the control level. On the background of allatostatin there were revealed no rise of the postsynaptic membrane input resistance (on the contrary, it decreased by 25%) and no changes of the EPMC potential-dependent amplitude and of the droptime constant that characterize cholinoreceptor conductivity. The potentiating allatostatin effect on EPMP amplitude was prevented by vesamicol (1 μM), a blocker of transport of acetylcholine into synaptic vesicles. Preliminary treatment of the nerve-muscle preparation with the inhibitor of protein kinase A (PKA) H-89 (50 nM) prevented the allatostatin-evoked EPMC amplitude increment. The obtained data allow us to suggest that allatostatin in the mouse nerve-muscle synapse acts at the presynaptic level by producing an increase of the acetylcholine quantum size due to an intraterminal cascade of reaction with participation of PKA.     

Fura-2 secretion and sequestration in macrophages. A blocker of organic anion transport reveals that these processes occur via a membrane transport system for organic anions

Fura-2, loaded into J774.2 macrophages as the acetoxymethyl ester, is sequestered into intracellular vacuoles within 90 min after the beginning of the loading at 37 degrees C. The dye is also efficiently secreted from the cells. Sequestration and secretion of fura-2 reduce the accuracy of measurements of cytosolic free Ca2+ concentration in this cell line. Fura-2 is also sequestered and secreted by J774.2 when the dye is loaded into the cytoplasm as the pentapotassium salt by reversible permeabilization of the plasma membrane. Regardless of the mechanism by which fura-2 is loaded into the cytoplasm, both sequestration and secretion are prevented by 2.5 mM probenecid, a blocker of organic anion transport. Probenecid has no effect on resting or stimulated cytosolic free Ca2+ levels or on FcR-mediated phagocytosis. These findings suggest that macrophages express a transport mechanism for the anionic form of fura-2. This transport system is responsible for the clearance of fura-2 from the cytoplasm of this cell type. Furthermore we suggest that use of probenecid to block secretion and intracellular sequestration of fura-2 may overcome problems arising in the application of this Ca2+ indicator to macrophages and perhaps to other cell types.     

硝苯吡啶对腹腔神经节突触传递的阻断作用

本文应用细胞内记录技术,观察了钙通道阻滞剂硝苯吡啶（nifedipine)对离体豚鼠腹腔神经节突触传递的影响,硝苯吡啶（0.1-10umol/L)不影响所检细胞的静息膜电位,膜电阻及细胞内刺激引起的动作电位,但能显著阻断N-型胆碱能的突触传递,并且这种作用可被低钙模拟、高钙拮抗,硝苯吡啶（10umol/L)也不影响突触后膜对乙酰胆碱（ACh)的敏感性;但在高钾克氏液中,能减少微小兴奋性突触后电位（mEPSPs)的频率;在低钙和高镁克氏液中,能减少量子含量,而对量子大小无影响。结果表明,治疗量的硝苯吡啶(0.1umol/L)通过阻滞突触前膜钙内流及ACh的量子性释放,产生突触阻断作用。这可能是硝苯吡啶降压机理的一个组成部分。     

Signaling pathway underlying stimulation of L-type Ca2+ channels in rabbit portal vein myocytes by recombinant Gbetagamma subunits

In previous studies, we (Callaghan B, Koh SD, and Keef KD, Circ Res 94: 626-633, 2004) have shown that voltage-dependent L-type Ca(2+) channels (Cav) in portal vein myocytes are enhanced when muscarinic M2 receptors are activated with ACh. Current stimulation was coupled to the G protein subunit Gbetagamma along with the downstream mediators phosphatidylinositol-3-kinase (PI3K), protein kinase C (PKC), and c-Src. The present study was designed to determine whether the same second messenger pathway could be identified when exogenous recombinant Gbetagamma subunits are introduced into cells. Smooth muscle myocytes were freshly isolated from rabbit portal vein, and Cav currents were recorded by using the patch-clamp technique. Dialysis of cells with recombinant Gbetagamma (50 nM) significantly increased Cav currents (141%). Nifedipine (1 microM) reduced both control and stimulated currents by approximately 90%. The enhancement of current by Gbetagamma was equivalent to that produced by ACh (142%), whereas the PKC activator phorbol 12,13-dibutyrate (PdBu) gave rise to greater current stimulation (192%). Current stimulation with Gbetagamma, ACh, and PdBu were not associated with changes in the voltage dependence of activation or inactivation. The PI3K inhibitor LY-294002 (20 microM) reduced peak currents by 32% in cells dialyzed with Gbetagamma, whereas the inactive analog LY-303511 resulted in a small but significant reduction in current (12%). The c-Src inhibitor PP2 (1 microM) also significantly reduced currents (34%), whereas the inactive analog PP3 was without effect. These data provide further evidence for the hypothesis that Gbetagamma leads to stimulation of Cav currents in rabbit portal vein myocytes via a signaling pathway that includes PI3K, PKC, and c-Src.     

Ca2+-dependent activator protein for secretion 1 is critical for constitutive and regulated exocytosis but not for loading of transmitters into dense core vesicles

Although CAPS1 was originally identified as a soluble factor that reconstitutes Ca(2+)-dependent secretion from permeabilized neuroendocrine cells, its exact function in intact mammalian cells remains controversial. Here we investigate the role for CAPS1 by generating stable cell lines in which CAPS1 is strongly down-regulated. In these cells, Ca(2+)-dependent secretion was strongly reduced not only of catecholamine but also of a transfected neuropeptide. These secretion defects were rescued by infusion of CAPS1-containing brain cytosol or by transfection-mediated expression of CAPS1. Whole cell patch clamp recording revealed significant reductions in slow burst and sustained release components of exocytosis in the knockdown cells. Unexpectedly, they also accumulated higher amounts of endogenous and exogenous transmitters, which were attributable to reductions in constitutive secretion. Electron microscopy did not reveal abnormalities in the number or docking of dense core vesicles. Our results indicate that CAPS1 plays critical roles not only in Ca(2+)-dependent, regulated exocytosis but also in constitutive exocytosis downstream of vesicle docking. However, they do not support the role for CAPS1 in loading transmitters into dense core vesicles.     

Organic secretion by striated ducts

In addition to their role in electrolyte homeostasis, striated ducts in the parotid and submandibular glands of many mammalian species engage in secretion of organic products. This secretion usually is in the form of serous-like granules that lack substructure. Usually, the granules are in the 2.0-2.5 mm range, but granules smaller than 0.1 mm and larger than 12 mm have been observed. In mustelids, striated duct cells contain rhomboidal crystalloids in their apical cytoplasm; in dogs and at least two bat species, the apical plasmalemma is festooned with perpendicularly-oriented rods. Rather than granules, the supranuclear cytoplasm of duct cells in a number of species contains spherical or oblong vesicles. These may convey glycoproteins to the luminal surface where they are incorporated into the glycocalyx or the plasma membrane. Certain vesicles appear to be involved in the uptake of foreign proteins introduced retrogradely into the main excretory duct or of altered proteins produced by acinar cells in streptozotocin-induced diabetes.