Workshop on algal polysaccharides

Summary This workshop was an impromptu event, but the fact that a number of interesting problems were identified by the participants from the floor may indicate that it is worth repeating. If it is to be repeated, however, it is important that notice be given, and that the scope of the workshop be defined in advance, so that participants can be better informed and bring supporting data. The best approach would be to identify a convenor and define the scope of the workshop prior to the first circular for the next Seaweed Symposium; invitations could then be issued by the convenor for specific topics and data, and a very brief program could be issued with the third circular. Such a procedure may very well allow the identification of new areas for research.     

Contraction of filaments of Escherichia coli after disruption of cell membrane by detergent.

The osmotic pressure within a living bacterium creates stresses in the peptidoglycan that stretch the sacculus. We measured the amount of stretch by monitoring the shrinkage of growing cells of Escherichia coli after removal of the osmotic pressure by disruption of the phospholipid membranes with sodium dodecyl sulfate. Because the rods of the wild type are so short, length changes of filaments of longer than 7 microns were measured on phase-contrast micrographs. The filaments were prepared by growing ftsA and ftsI strains under permissive conditions in rich medium and then shifting them to 42 degrees C for 40 to 180 min. During this time, the mutant cells became elongated but did not divide. The growing filaments were mounted on a glass surface that had been treated with poly-L-lysine or RNase. The filaments were photographed before being treated with sodium dodecyl sulfate. The filaments were rephotographed at the time when the first change in phase contrast was noted. Some filaments were also measured at 10-min time intervals from 0 to 60 min. The reduction in phase contrast signaled the leakage of solutes and the loss of turgor pressure. The average length of the filaments decreased 17%. If the circumference were stretched to the same degree, then the surface area in vivo would be 45% greater than in the relaxed state. For comparison, a fully cross-linked monolayer of E. coli peptidoglycan in its most compact conformation could stretch up to 300% in achieving the most extended conformation possible without splitting covalent bonds.     

Transfer of phosphoethanolamine residues from phosphatidylethanolamine to the membrane-derived oligosaccharides of Escherichia coli.

The membrane-derived oligosaccharides (MDO) of Escherichia coli are periplasmic constituents composed of glucose residues linked by beta-1,2 and beta-1,6 glycosidic bonds. MDO are substituted with phosphoglycerol, phosphoethanolamine, and succinic acid moieties. The phosphoglycerol residues present on MDO are derived from phosphatidylglycerol (B. J. Jackson and E. P. Kennedy, J. Biol. Chem. 258:2394-2398, 1983), but evidence as to the source of the phosphoethanolamine residues has been lacking. We now report that phosphatidylethanolamine, exogenously added to intact cells of E. coli, provides a source of phosphoethanolamine residues that are transferred to MDO. The biosynthesis of phosphoethanolamine-labeled MDO is osmotically regulated, with maximum synthesis occurring during growth in medium of low osmolarity.     

Biosynthesis and stereochemical configuration of N5-(1-carboxyethyl)ornithine. An unusual amino acid produced by Streptococcus lactis

In a recent communication (Thompson, J., Curtis, M. A., and Miller, S.P.F. (1986) J. Bacteriol. 167, 522-529) we described the purification and characterization of N5-(1-carboxyethyl)ornithine from cells of Streptococcus lactis 133. This unusual amino acid has not previously been found in nature. Radiotracer experiments presented here reveal that exogenous [14C]ornithine serves as the precursor for biosynthesis of [14C]arginine, [14C]N5-(1-carboxyethyl)ornithine, and [14C]N5-acetylornithine by cells of S. lactis K1 during growth in a defined medium lacking arginine. In the absence of both arginine and ornithine, cells of S. lactis K1 can also generate intracellular [14C]N5-(1-carboxyethyl)ornithine from exogenous [14C]glutamic acid. Previously we showed that the properties of N5-(1-carboxyethyl)ornithine prepared from S. lactis were identical to one of the two diastereomers [2S, 7S) or (2S, 7R] present in a synthetic preparation of (2S, 7RS)-N5-(1-carboxyethyl)ornithine. The two diastereomers have now been unambiguously synthesized by an Abderhalden-Haase condensation between (2S)-N2-t-butoxycarbonyl-ornithine and the chiral (2S)-, and (2R)-bromopropionates. By 13C-NMR spectroscopy it has been established that the preparation from S. lactis is exclusively (2S, 7S)-N5-(1-carboxyethyl)ornithine. has been demonstrated in a cell-free extract of S. lactis 133. The requirements for ornithine, pyruvic acid, and NAD(P)H suggest that biosynthesis of N5-(1-carboxyethyl)ornithine occurs via a reductive condensation mechanism. A general survey revealed that N5-(1-carboxyethyl)ornithine was produced only by certain strains of Group N streptococci. These findings may indicate a plasmid locus for the gene(s) encoding the enzyme(s) for N5-(1-carboxyethyl)ornithine biosynthesis.     

Acquisition of antigens characteristic of adult pericentral hepatocytes by differentiating fetal hepatoblasts in vitro

Antigens specific to pericentral hepatocytes have been studied in adult mouse liver, during fetal development, and in cultured fetal hepatoblasts. Antibody reactive with glutamine synthetase stained all fetal liver cells but almost all cells lost this antigen after birth; only a single layer of pericentral cells retained it in adulthood. In contrast, monoclonal antibodies to major urinary protein (MUP) did not detect the antigen until approximately 3 wk after birth, after which time the cells within 6-10 cell diameters of the central veins were positive. Cultured fetal liver cells from embryos at 13 +/- 1 d of gestation were capable of differentiating in vitro to mimic events that would occur had the cells remained in the animal. About 10-20% of the explanted cells grew into clusters of hepatocyte-like cells, all of which stained with albumin antibodies. MUP monoclonals were reactive with one-half of the differentiated fetal hepatocytes. Glutamine synthetase was present in all hepatocytes after several days in culture and gradually decreased and remained in only occasional cells, all of which also contained the MUP antigen. These findings suggest that a sequence of gene controls characterizes expression of specific genes in developing liver, and that differentiating fetal hepatoblasts are capable of undergoing similar patterns of gene activity in culture.     

Single Na+ channels activated by veratridine and batrachotoxin

Voltage-sensitive Na+ channels from rat skeletal muscle plasma membrane vesicles were inserted into planar lipid bilayers in the presence of either of the alkaloid toxins veratridine (VT) or batrachotoxin (BTX). Both of these toxins are known to cause persistent activation of Na+ channels. With BTX as the channel activator, single channels remain open nearly all the time. Channels activated with VT open and close on a time scale of 1-10 s. Increasing the VT concentration enhances the probability of channel opening, primarily by increasing the rate constant of opening. The kinetics and voltage dependence of channel block by 21-sulfo-11-alpha-hydroxysaxitoxin are identical for VT and BTX, as is the ionic selectivity sequence determined by bi-ionic reversal potential (Na+ approximately Li+ greater than K+ greater than Rb+ greater than Cs+). However, there are striking quantitative differences in open channel conduction for channels in the presence of the two activators. Under symmetrical solution conditions, the single channel conductance for Na+ is about twice as high with BTX as with VT. Furthermore, the symmetrical solution single channel conductances show a different selectivity for BTX (Na+ greater than Li+ greater than K+) than for VT (Na+ greater than K+ greater than Li+). Open channel current-voltage curves in symmetrical Na+ and Li+ are roughly linear, while those in symmetrical K+ are inwardly rectifying. Na+ currents are blocked asymmetrically by K+ with both BTX and VT, but the voltage dependence of K+ block is stronger with BTX than with VT. The results show that the alkaloid neurotoxins not only alter the gating process of the Na+ channel, but also affect the structure of the open channel. We further conclude that the rate-determining step for conduction by Na+ does not occur at the channel's "selectivity filter," where poorly permeating ions like K+ are excluded.     

Stimulation of T lymphocyte proliferation by monoclonal antibodies against GD3 ganglioside

Cell-surface gangliosides are presumed to play a role in cell growth and differentiation. With the use of monoclonal antibodies directed against GD3, a disialoganglioside expressed predominantly by cells of neuroectodermal origin, we have found that GD3 is expressed by a subpopulation of cells of the immune system including: 1) fetal thymocytes in subcortical regions and near vessels, 2) lymph node lymphocytes in interfollicular areas and near vessels, and 3) a small subset of T cells in the peripheral blood. Mouse monoclonal antibodies (two IgGs, one IgM, and F(ab')2 fragments) reacting with GD3 were found to stimulate proliferation of T cells derived from peripheral blood. Proliferation of T cells was observed even in cultures depleted of macrophages, suggesting that activation by anti-GD3 was not dependent on the presence of accessory cells. T cell proliferation was maximum between days 5 and 7 of stimulation and was preceded by expression of interleukin 2 receptors. No stimulation was observed with control antibodies of the identical isotype or with monoclonal antibodies recognizing the gangliosides GD2 or GM2. During stimulation by anti-GD3 monoclonal antibodies, there was an expansion of the GD3+ pool of T cells, but depletion of GD3+ T cells prior to stimulation abrogated the response. Proliferation induced by binding to GD3 could be augmented by exogenous interleukin 2 and phytohemagglutinin. Anti-CD3 (T3) monoclonal antibodies had little or no effect. These results demonstrate that binding to GD3 on the surface of T cells can elicit signals for T cell proliferation.     

Coupling of voltage-dependent gating and Ba++ block in the high- conductance, Ca++-activated K+ channel

Voltage-dependent Ca++-activated K+ channels from rat skeletal muscle were reconstituted into planar lipid bilayers, and the kinetics of block of single channels by Ba++ were studied. The Ba++ association rate varies linearly with the probability of the channel being open, while the dissociation rate follows a rectangular hyperbolic relationship with open-state probability. Ba ions can be occluded within the channel by closing the channel with a strongly hyperpolarizing voltage applied during a Ba++-blocked interval. Occluded Ba ions cannot dissociate from the blocking site until after the channel opens. The ability of the closed channel to occlude Ba++ is used as an assay to study the channel's gating equilibrium in the blocked state. The blocked channel opens and closes in a voltage-dependent process similar to that of the unblocked channel. The presence of a Ba ion destabilizes the closed state of the blocked channel, however, by 1.5 kcal/mol. The results confirm that Ba ions block this channel by binding in the K+-conduction pathway. They further show that the blocking site is inaccessible to Ba++ from both the cytoplasmic and external solutions when the channel is closed.