Depression of gustatory receptor potential in frog taste cell by parasympathetic nerve-induced slow hyperpolarizing potential

Parasympathetic nerve (PSN) innervates taste cells of the frog taste disk, and electrical stimulation of PSN elicited a slow hyperpolarizing potential (HP) in taste cells. Here we report that gustatory receptor potentials in frog taste cells are depressed by PSN-induced slow HPs. When PSN was stimulated at 30 Hz during generation of taste cell responses, the large amplitude of depolarizing receptor potential for 1 M NaCl and 1 mM acetic acid was depressed by approximately 40% by slow HPs, but the small amplitude of the depolarizing receptor potential for 10 mM quinine-HCl (Q-HCl) and 1 M sucrose was completely depressed by slow HPs and furthermore changed to the hyperpolarizing direction. The duration of the depolarizing receptor potentials depressed by slow HPs prolonged with increasing period of PSN stimulation. As tastant-induced depolarizing receptor potentials were increased, the amplitude of PSN-induced slow HPs inhibiting the receptor potentials gradually decreased. The mean reversal potentials of the slow HPs were approximately -1 mV under NaCl and acetic acid stimulations, but approximately -14 mV under Q-HCl and sucrose stimulations. This implies that when a slow HP was evoked on the same amplitude of depolarizing receptor potentials, the depression of the NaCl and acetic acid responses in taste cells was larger than that of Q-HCl and sucrose responses. It is concluded that slow HP-induced depression of gustatory depolarizing receptor potentials derives from the interaction between gustatory receptor current and slow hyperpolarizing current in frog taste cells and that the interaction is stronger for NaCl and acetic acid stimulations than for Q-HCl and sucrose stimulations.     

Electrical responses of supporting cells in the frog taste organ to chemical stimuli

1. The mean resting potential of supporting cells in the frog taste organ was -19.1 mV. The supporting cells responded to the four basic taste stimuli with a depolarization but responded to water with a depolarization or a hyperpolarization. 2. The membrane resistances of supporting cells decreased during stimulation with sucrose, NaCl and acetic acid, but increased during stimulation with Q-HCl and water. 3. Reversal potential of the depolarizing response for 0.5 M NaCl in supporting cells was +7.6 mV. The depolarizing responses for Q-HCl and acetic acid were independent of the membrane potential level. 4. These results suggest that the characteristics of taste responses in supporting cells are similar to those in taste cells.     

Interaction Between Gustatory Depolarizing Receptor Potential and Efferent-Induced Slow Depolarizing Synaptic Potential in Frog Taste Cell

Electrical stimulation of parasympathetic nerve (PSN) efferent fibers in the glossopharyngeal nerve induced a slow depolarizing synaptic potential (DSP) in frog taste cells under hypoxia. The objective of this study is to examine the interaction between a gustatory depolarizing receptor potential (GDRP) and a slow DSP. The amplitude of slow DSP added to a tastant-induced GDRP of 10 mV was suppressed to 60% of control slow DSPs for NaCl and acetic acid stimulations, but to 20–30% for quinine–HCl (Q-HCl) and sucrose stimulations. On the other hand, when a GDRP was induced during a prolonged slow DSP, the amplitude of GDRPs induced by 1 M NaCl and 1 M sucrose was suppressed to 50% of controls, but that by 1 mM acetic acid and 10 mM Q-HCl unchanged. It is concluded that the interaction between GDRPs and efferent-induced slow DSPs in frog taste cells under hypoxia derives from the crosstalk between a gustatory receptor current across the receptive membrane and a slow depolarizing synaptic current across the proximal subsynaptic membrane of taste cells.     

Membrane resistance change of the frog taste cells in response to water and Nacl

The electrical properties of the frog taste cells during gustatory stimulations with distilled water and varying concentrations of NaCl were studied with intracellular microelectrodes. Under the Ringer adaptation of the tongue, two types of taste cells were distinguished by the gustatory stimuli. One type, termed NaCl-sensitive (NS) cells, responded to water with hyperpolarizations and responded to concentrated NaCl with depolarizations. In contrast, the other type of cells, termed water-sensitive (WS) cells, responded to water depolarizations and responded to concentrated NaCl with hyperpolarizations. The membrane resistance of both taste cell types increased during the hyperpolarizing receptor potentials and decreased during the depolarizing receptor potentials, Reversal potentials for the depolarizing and hyperpolarizing responses in each cell type were a few millivolts positive above the zero membrane potential. When the tongue was adapted with Na-free Ringer solution for 30 min, the amplitude of the depolarizing responses in the NS cells reduced to 50% of the control value under normal Ringer adaptation. On the basis of the present results, it is concluded (a) that the depolarizing responses of the NS and WS cells under the Ringer adaptation are produced by the permeability increase in some ions, mainly Na+ ions across the taste cell membranes, and (b) that the hyperpolarizing responses of both types of taste cells are produced by a decrease in the cell membrane permeability to some ions, probably Na+ ions, which is slightly enhanced during the Ringer adaptation.     

Enhancement of Gustatory Neural Responses by Parasympathetic Nerve in the Frog

The autonomic nervous system affects the gustatory responses in animals. Frog glossopharyngeal nerve (GPN) contains the parasympathetic nerve. We checked the effects of electrical stimulation (ES) of the parasympathetic nerves on the gustatory neural responses. The gustatory neural impulses of the GPNs were recorded using bipolar AgCl wires under normal blood circulation and integrated with a time constant of 1 s. Electrical stimuli were applied to the proximal side of the GPN with a pair of AgCl wires. The parasympathetic nerves of the GPN were strongly stimulated for 10 s with 6 V at 30 Hz before taste stimulation. The integrated neural responses to 0.5 M NaCl, 2.5 mM CaCl₂, water, and 1 M sucrose were enhanced to 130–140% of the controls. On the other hand, the responses for 1 mM Q-HCl and 0.3 mM acetic acid were not changed by the preceding applied ES. After hexamethonium (a blocker of nicotinic ACh receptor) was intravenously injected, ES of the parasympathetic nerve did not modulate the responses for all six taste stimuli. The mechanism for enhancement of the gustatory neural responses is discussed.     

Vasopressin increases frog gustatory neural responses elicited by NaCl and HCl.

1. The effect of arginine vasopressin (AVP) on frog gustatory responses was investigated by recording integrated responses of the whole glossopharyngeal nerve by stimulation of the tongue with tastants. 2. After AVP (100 mUnits/ml) was perfused to the basolateral side of taste cells through the lingual artery, gustatory neural responses for NaCl and hydrochloric acid (HCl) stimuli were greatly enhanced, but the responses for CaCl2, quinine hydrochloride (Q-HCl) and galactose were not affected. 3. Three hours after the onset of AVP perfusion, the responses for NaCl and HCl increased to 260% and 270% of the respective controls. 4. The NaCl response which was insensitive to amiloride during normal saline perfusion became sensitive to amiloride during AVP perfusion. 5. When membrane-permeable 8-bromo-cyclic AMP (8-Br-cAMP, 0.1 mM) was perfused to the basolateral side of taste cells, the responses for NaCl and HCl decreased to 41 and 63% of the respective controls. 6. These results suggest that AVP may regulate the gustatory responses for monovalent salts and acids by a mechanism which is not necessary to activate adenylate cyclase.     

Topographical difference in taste organ density and its sensitivity of frog tongue

Distribution density of the taste disks of the fungiform papillae in the frog tongue was larger at the proximal portion than at the apical and middle portions. The number of myelinated afferent nerve fibres and taste cells per cm2 area of the tongue increased in the order of proximal greater than middle greater than apical portion. The amplitudes of gustatory neural responses for 0.5 M NaCl, 0.5 M KCl, 0.5 M NH4Cl, 0.05 M CaCl2, 1 mM acetic acid and 1 mM quinine-HCl (Q-HCl) were significantly larger with lingual stimulation of the proximal region than with the stimulation of the apical region. With these stimuli the mean ratio of the apical response to the proximal response was 1.00:1.54. On the other hand, this ration with deionized water was 1.00:5.00. The mean magnitudes of receptor potentials in taste cells for 1 mM acetic acid and 10 mM Q-HCl were the same among the apical, middle and proximal portions of the tongue. The mean magnitudes of receptor potentials for 0.5 M NaCl were significantly larger at the apical portion than at the other portions, whereas those for deionized water tended to be the largest at the proximal portion. It is concluded that the larger magnitude of the gustatory neural responses at the proximal portion of the tongue is due to morphological and physiological properties of the taste organ.     

The mouse taste cell response to five sugar stimuli

Intracellular recordings of mouse taste cell responses were made using glass microelectrodes filled with procion yellow dye solution. Only responses recorded from taste buds with fluorescent cells, as observed in subsequent histological preparations, were used in this study. The mouse taste cell depolarizes when stimulated with sucrose and is accompanied by either a resistance increase or no change. On the other hand, a NaCl stimulus produces a depolarization, hyperpolarization or null response and is accompanied by either a membrane resistance decrease or no change. Four sugars other than sucrose (maltose, fructose, glucose and lactose) produced the depolarization or null responses and were accompanied by an increase or no change in membrane resistance. From the above observations, it is suggested that each taste cell produces its own characteristic response profiles and membrane resistance changes for the five sugars and NaCl tested.     

Monosodium glutamate but not linoleic acid differentially activates gustatory neurons in the rat geniculate ganglion

To date, only one study has examined responses to monosodium glutamate (MSG) from gustatory neurons in the rat geniculate ganglion and none to free fatty acids. Accordingly, we recorded single-cell responses from geniculate ganglion gustatory neurons in anesthetized male rats to MSG and linoleic acid (LA), as well as to sucrose, NaCl, citric acid, and quinine hydrochloride. None of the 52 neurons responded to any LA concentration. In contrast, both narrowly tuned groups of gustatory neurons (sucrose specialists and NaCl specialists) responded to MSG, as did 2 of the broadly tuned groups (NaCl generalist(I) and acid generalists). NaCl-generalist(II) neurons responded only to the highest MSG concentration and only at low rates. No neuron type responded best to MSG; rather, responses to 0.1 M MSG were significantly less than those to NaCl for Na(+) -sensitive neurons and to sucrose for sucrose specialists. Interestingly, most Na(+) -sensitive neurons responded to 0.3 M MSG at levels comparable with those to 0.1 M NaCl, whereas sucrose specialists responded to 0.1 M MSG despite being unresponsive to NaCl. These results suggest that the stimulatory effect of MSG involves activation of sweet- or salt-sensitive receptors. We propose that glutamate underlies the MSG response of sucrose specialists, whereas Na(+) -sensitive neurons respond to the sodium cation. For the latter neuron groups, the large glutamate anion may reduce the driving force for sodium through epithelial channels on taste cell membranes. The observed concentration-dependent responses are consistent with this idea, as are cross-adaptation studies using 0.1 M concentrations of MSG and NaCl in subsets of these Na(+) -sensitive neurons.     

Proline porter II is activated by a hyperosmotic shift in both whole cells and membrane vesicles of Escherichia coli K12

Proline porter II is rapidly activated when nongrowing bacteria are subjected to a hyperosmotic shift (Grothe, S., Krogsrud, R. L., McClellan, D. J., Milner, J. L., and Wood, J. M. (1986) J. Bacteriol. 166, 253-259). Proline porter II was active in membrane vesicles prepared from bacteria grown under optimal conditions, nutritional stress, or osmotic stress. That activity was: (i) dependent on the presence of the energy sources phenazine methosulphate plus ascorbate or D-lactate; (ii) observed only when a hyperosmotic shift accompanied the transport measurement; (iii) inhibited by glycine betaine in a manner analogous to that observed in whole cells; and (iv) eliminated by lesions in proP. Membrane vesicles were able to transport serine but not glutamine and serine transport was reduced by the hyperosmotic shift. In whole cells, proline porter II activity was supported by glucose and by D-lactate in a strain defective for proline porters I and III and the F1F0-ATPase. Glucose energized proline uptake was eliminated by carbonyl cyanide m-chlorophenylhydrazone and KCN as was serine uptake. These results suggested that proline porter II was respiration-dependent and probably ion-linked. Activation of proline porter II in whole cells by sucrose or NaCl was sustained over 30 min, whereas activation by glycerol was transient. Proline porter II was activated by NaCl and sucrose with a half-time of approximately 1 min in both whole cells and membrane vesicles. Thus, activation of proline porter II was reversible. It occurred at a rate comparable to that of K+ influx and much more rapid than the genetic regulatory responses that follow a hyperosmotic shift.     

Application of transretinal current stimulation for the study of bipolar-amacrine transmission

Transretinal current flowing from the receptor side to the vitreous side depolarizes the axon terminals of retinal cells and facilitates the release of transmitter. Such current elicited a depolarizing response in off-center bipolar cells and a hyperpolarizing response in on-center bipolar cells. It also elicited a response of relatively complex waveform in amacrine cells. The responses elicited in bipolar cells were suppressed in the presence of 5-10 mM glutamate in the perfusing Ringer solution, while the responses of amacrine cells persisted, although their waveform changed to a simple one that showed monotonic depolarization irrespective of the type of amacrine cell and were accompanied by a decrease in the membrane resistance. The results indicate excitatory synaptic transmission from bipolar cells to amacrine cells. Since the response elicited by current in ON-OFF cells was almost identical to those elicited in ON or OFF amacrine cells, the transient nature of their light response cannot be due to their membrane properties. ON-OFF cells responded to transretinal current flowing in the opposite direction with a small hyperpolarization accompanied by a resistance increase. The hyperpolarizing response was suppressed by the addition of GABA in glutamate Ringer solution. The results suggest an activation by the current of GABA-ergic feedback pathways from amacrine cells to bipolar cells.     

Multiple sensitivity of single taste cells of the frog tongue to four basic taste stimuli

The electrical response of the taste cells of the frog fungiform papillae to four fundamental taste solutions (NaCl, acetic acid, quinine-HCl and sucrose) was studied by using the intracellular recording technique. The average value of resting membrane potential was 22.5 mV, inside negative. Each of the four taste solutions applied to the tongue produced a slow depolarizing potential, the receptor potential, on which no spike potential was superimposed. The amplitude of the receptor potentials increased linearly as a function of the logarithm of the concentration of the stimulus. Amplitudes of depolarizations to a given taste stimulation varied from one cell to another even within a single taste bud. Most of the cells responded to more than two of the four basic taste solutions. Sensitivity patterns in terms of the number of effective solutions and the relative effectiveness of different kinds of solutions were variable among cells. Statistical analysis suggests that at the receptor membranes of the taste cells, the sensitivities for the four basic stimuli are independent and random.     

Tonic GABAergic inhibition of taste-responsive neurons in the nucleus of the solitary tract

The effects of gamma-aminobutyric acid (GABA) and the GABAA receptor antagonist bicuculline methiodide (BICM) on the activity of taste- responsive neurons in the nucleus of the solitary tract (NST) were examined electrophysiologically in urethane-anesthetized hamsters. Single neurons in the NST were recorded extracellularly and drugs (21 nl) were microinjected into the vicinity of the cell via a multibarrel pipette. The response of each cell was recorded to lingual stimulation with 0.032 M NaCl, 0.032 M sucrose, 0.0032 M citric acid and 0.032 M quinine hydrochloride (QHCl). Forty-six neurons were tested for the effects of GABA; the activity of 29 cells (63%) was inhibited by 5 mM GABA. Whether activity was elicited in these cells by repetitive anodal current stimulation (25 microA, 0.5 s, 0.1 Hz) of the tongue (n = 13 cells) or the cells were spontaneously active (n = 13 cells), GABA produced a dose-dependent (1, 2 and 5 mM) decrement in activity. Forty- seven NST neurons were tested for the effects of BICM on their responses to chemical stimulation of the tongue; the responses of 28 cells (60%) were enhanced by 10 mM BICM. The gustatory responses of 26 of these cells were tested with three concentrations (0.2, 2 and 10 mM) of BICM, which produced a dose-dependent increase in both spontaneous activity and taste-evoked responses. Nine of these neurons were sucrose- best, seven were NaCl-best, eight were acid-best and two responded best to QHCl. The responses to all four tastants were enhanced, with no difference among neuron types. For 18 cells that were tested with two or more gustatory stimuli, BICM increased their breadth of responsiveness to their two most effective stimuli. These data show that approximately 60% of the taste-responsive neurons in the rostral NST are inhibited by GABA and/or subject to a tonic inhibitory influence, which is mediated by GABAA receptors. The modulation of these cells by GABA provides a mechanism by which the breadth of tuning of the cell can be sharpened. Modulation of gustatory activity following a number of physiological changes could be mediated by such a GABAergic circuit.      

Arterial perfusion of frog tongue for intracellular recording of taste cell receptor potential

The frog tongue was perfused through its artery with a Ringer solution using a peristaltic pump, and a method was developed to record stable intracellular receptor potentials of taste cells. Perfusing at 0.05 ml/min with a Ringer solution containing 5% dextran did not cause tongue edema, but perfusing at the same rate with Ringer without dextran caused edema. After perfusion at 0.05 ml/min with 100 mM K Ringer, the membrane potential of taste cells gradually decreased and reached a constant level in about 30 min, indicating that the intercellular fluid of the tongue could be replaced within this time period. While the artery of the frog tongue was perfused at 0.05 ml/min with Ringer containing 5% dextran, intracellular receptor potentials of taste cells elicited by four basic taste stimuli (1 M NaCl, 10 mM quinine-HCl (Q-HCl), 1 mM acetic acid and 1 M galactose) were similar to those obtained from the control taste cells under normal blood flow.     

Intracellular characteristics and responses of taste bud and lingual cells of the mudpuppy

Intracellular recordings of membrane potentials of mudpuppy lingual cells were made with micropipette electrodes. Three types of cells were distinguished by their responses to chemical stimulation. Surface epithelial (SE) cells outside of taste buds responded with large membrane potential and resistance changes to a variety of stimuli representing the four taste qualities. Salts and acids evoked particularly large potential changes, and MgCl2, acids, and quinine greatly increased the membrane resistance. One type of taste bud cell (TB-1) was characterized by large depolarizations to K salts, and the other type of taste bud cell (TB-2) characteristically hyperpolarized to MgCl2, acid, and sugar solutions. Membrane resistance changes accompanying TB-1 and TB-2 cell responses were relatively small compared to those of SE cells. Electrotonic coupling was observed between pairs of SE and TB-2 cells but not for pairs of TB-1 cells nor cells of different types. After recording cell responses, dye-marking allowed verification of results in situ and histologically. From the identification of cells in section, it is hypothesized the TB-1 and TB- 2 cells correspond to light and dark cells, respectively. Responses of TB-1 cells imply a taste receptive function; wheras TB 2-cell responses suggest secretory, supportive, and (or) receptive functions. Factors affecting cellular characteristics, non-taste bud cell responsiveness, response mechanisms, and function of electrotonic coupling are discussed in relation to taste reception.     

Effect of Gap Junction Blocker β-Glycyrrhetinic Acid on Taste Disk Cells in Frog

A gap junction blocker, 18β-glycyrrhetinic acid (β-GA), increased the membrane resistance of Ia, Ib and II/III cells of frog taste disk by 50, 160, and 300 MΩ, respectively, by blocking the gap junction channels and hemichannels. The amplitudes of gustatory depolarizing potentials in the disk cells for 4 basic taste stimuli were reduced to 40–60% after intravenous injection of β-GA at 1.0 mg/kg. β-GA of 1.0 mg/kg did not affect the resting potentials and the reversal potentials for tastant-induced depolarizing potentials in any taste disk cells. The percentage of cells responding to each of 4 basic taste stimuli and varying numbers of 4 taste qualities did not differ between control and β-GA-treated taste disk cells. This implies that gustatory depolarizing response profiles for 4 basic taste stimuli were very similar in control and β-GA-treated taste disk cells. It is concluded that β-GA at 1.0 mg/kg reduced the amplitude of gustatory depolarizing potentials in taste disk cells by strongly blocking depolarizing currents flowing through the gap junction channels and hemichannels, but probably weakly affected the gustatory transduction mechanisms for 4 taste stimuli.     

Cutting edge: a single MHC class Ia is sufficient for CD8 memory T cell differentiation

Recent studies have suggested a role for MHC class Ib molecules in providing signals for memory T cell differentiation during the early phases of acute infection. To test this hypothesis, we assessed the development of effector and memory CD8 T cells in transgenic mice expressing a single chain H-2D(d)/beta2-microglobulin (beta2M) fusion protein on a beta2M-deficient background. These mice thus express a single MHC class Ia in the absence of all other beta2M-dependent class Ia and Ib molecules. Following infection with a recombinant vaccinia virus expressing a known D(d)-restricted epitope from HIV-1 gp160, the development of effector and memory cells CD8 T cells was comparable to control mice. Furthermore, these memory cells responded rapidly and robustly to antigenic restimulation. Therefore, we conclude that full CD8 memory differentiation requires only a single MHC class Ia chain, ruling out a requirement for MHC class Ib molecules in this process.     

Voltage-gated inward currents of morphologically identified cells of the frog taste disc

We used the patch clamp technique to record from taste cells in vertical slices of the bullfrog (Rana catesbeiana) taste disc. Cell types were identified by staining with Lucifer yellow in a pipette after recording their electrophysiological properties. Cells could be divided into the following three groups: type Ib (wing) cells with sheet-like apical processes, type II (rod) cells with single thick rod-like apical processes and type III (rod) cells with thin rod-like apical processes. No dye-coupling was seen either between cells of the same type or between cells of different types. We focused on the voltage-gated inward currents of the three types of cells. Type Ib and type II cells exhibited tetrodotoxin (TTX)-sensitive voltage-gated Na+ currents. Surprisingly, type III cells showed TTX-resistant voltage-gated Na+ currents and exhibited a lack of TTX-sensitive Na+ currents. TTX-resistant voltage-gated Na+ currents in taste cells are reported for the first time here. The time constant for the inactivating portion of the voltage-gated inward Na+ currents of type III cells was much larger than that of type Ib and type II cells. Therefore, slow inactivation of inward Na+ currents characterizes type III cells. Amplitudes of the maximum peak inward currents of type III cells were smaller than those of type Ib and type II cells. However, the density (pA/pF) of the maximum peak inward currents of type III cells was much higher than that of type Ib cells and close to that of type II cells. No evidence of the presence of voltage-gated Ca2+ channels in frog taste cells has been presented up to now. In this study, voltage-gated Ba2+ currents were observed in type III cells but not in type Ib and type II cells when the bath solution was a standard Ba2+ solution containing 25 mM Ba2+. Voltage-gated Ba2+ currents were blocked by addition of 2 mM CoCl2 to the standard Ba2+ solution, suggesting that type III cells possess the voltage-gated Ca2+ channels and they do classical (calcium-influx) synaptic transmission. It appears that type III cells are taste receptor cells.     

Multi-photon microscopy of cell types in the viable taste disk of the frog

The morphology of viable taste disks of the frog was explored with multi-photon microscopy. In order to identify single sensory or supporting cells within the tissue, we searched for fluorescent dyes that stained subsets of the cell population or possibly cell types. Some cell types indeed stained preferentially with certain fluorescent dyes. A subset of glia-like cells (type Ic) stained with BCECF, a H⁺-sensitive dye, and indo-1, a Ca²⁺-sensitive dye, both presented in the membrane-permeant ester form. BCECF-ester also stained the dendrites of type III receptor cells, but indo-1 ester did not. Receptor cells of type II stained with MQAE, a positively charged Cl^–-sensitive dye. A subset of type II cells accumulated amiloride, a positively charged fluorescent diuretic. Certain supporting cells, i.e., wing cells (type Ib) and glia-like cells (type Ic), were labeled by negatively charged dyes, e.g., calcium green-1 dextran. Mucus cells (type Ia) were stained with only two of the 19 dyes examined, and Merkel-like basal cells (type IV) were stained only with a membrane-labeling voltage-sensitive dye, presumably by endocytosis. No dye was found which would stain all types of cells or all receptor cells. This finding reveals a potential problem for future functional imaging aiming at population responses, as the responses of unstained cells then would remain unobserved. Specificity of dyes with respect to cell types was sufficient to identify supporting cells and receptor cells. Cell shape could then be reconstructed, using optical slicing and rendering techniques. Thus populations of dye-loaded elongated cells, especially types Ic, II and III, could for the first time be visualized in three dimensions.This work was supported by the Deutsche Forschungsgemeinschaft (SFB 530, project B2)     

Cell envelope and shape of Escherichia coli: multiple mutants missing the outer membrane lipoprotein and other major outer membrane proteins.

Starting with an Escherichia coli strain missing the outer membrane lipoprotein, multiple mutants were constructed than in addition to this defect miss the outer membrane proteins II, Ia and Ib, or Ia, Ib, and II. In contrast to all single mutants or strains missing the lipoprotein and polypeptides Ia and Ib, drastic influences on the integrity of the outer membrane and cell morphology were observed in mutants without lipoprotein and protein II. Such strains exhibited spherical morphology. They required increased concentrations of electrolytes for optimal growth, and Mg2+ or Ca2+ were the most efficient. These mutants were sensitive to hydrophobic antibiotics and detergents. Electron microscopy revealed abundant blebbing of the outer membrane, and it could clearly be seen that the murein layer was no longer associated with the outer membrane.