Spin label study of erythrocyte deformability. Ca2+-induced loss of deformability and the effects of stomatocytogenic reagents on the deformability loss in human erythrocytes in shear flow

The Ca2+-induced loss of deformability in human erythrocytes and the recovery of the lost deformability by stomatocytogenic reagents were investigated by means of a new flow electron paramagnetic resonance (EPR) spin label method, which provides information on deformation and orientation characteristics of spin labeled erythrocytes in shear flow. The Ca2+-induced loss of deformability is attributed mainly to the increase in intracellular viscosity resulting from efflux of intracellular potassium ions and water (Gardos effect). Partial recovery of the lost deformability is demonstrated in the presence of stomatocytogenic reagents, such as chlorpromazine, trifluoperazine, W-7, and calmidazolium (R24571). The recovery can not be explained solely by suppression of the Gardos effect due to the reagents. Incorporation of an optimal amount of the reagents into the membrane appears to compensate for the membrane modification due to Ca2+ ions to restore a part of the lost deformability.     

Immune Response against Hamster Erythrocytes in the Low-Responder Mouse Strains

It was confirmed by passive transfer experiments that the function of thymus-derived cells specific for hamster erythrocytes (HRBC) was deficient in the low-responder mouse strains. 1) Antibody production against HRBC was enhanced by passive transfer of thymus cells from normal SL mice (high-responder) to normal C57BL/6 mice (low-responder). 2) The enhancing effect of passive transfer of thymus cells from SL mice was abrogated by pre-sensitization of the recipients (C57BL/6) with thymus cells from SL mice. 3) In C57BL/6 mice, antibody production against HRBC was enhanced by the transfer of lymph node or spleen cells from C57BL/6 mice which had been sensitized with HRBC in Freund's complete adjuvant or hamster lymphoma cells.     

Recognition of xenogeneic erythrocytes: the GalNAc/Gal-particle receptor of rat liver macrophages mediates or participates in recognition

We studied mechanisms that mediate recognition of human erythrocytes (HRBC) and sheep erythrocytes (SRBC) by rat liver macrophages. We used an in vitro cell binding assay that allows spontaneous formation of cell contacts. Binding of HRBC to rat macrophages shows the following characteristics: inhibition studies with several monosaccharides and oligosaccharides yield complete inhibition of cell contacts with saccharides, which block the GalNAc/Gal-particle receptor on rat liver macrophages. We found the inhibition pattern: N-acetyl-D-galactosamine, lactose greater than D-galactose, D-fucose greater than L-fucose much greater than N-acetyl-D-glucosamine. Cell binding is dependent on the presence of calcium ions, but not influenced by heat-aggregated IgG or gangliosides. The inhibition pattern was the same after treatment of HRBC with neuraminidase. Therefore, binding of HRBC, as well as binding of neuraminidase-treated HRBC, is mediated by the GalNAc/Gal-particle receptor. Binding of SRBC is partly inhibited by galactose-related saccharides. Binding is also partly inhibited by heat-aggregated IgG, gangliosides, and L-fucose. Complete inhibition of cell contacts with SRBC is achieved by combination of all inhibitors. We therefore conclude that binding of SRBC is mediated by several different mechanisms, including the GalNAc/Gal-particle receptor. Binding of neuraminidase-treated SRBC, however, was found to be completely inhibited by saccharides, which block the GalNAc/Gal-particle receptor. We conclude that the GalNAc/Gal-particle receptor mediates or participates in recognition of non-self structures.     

Immune Response against Hamster Erythrocytes in the Low-Responder Mouse Strains

1) A subcutaneous injection of hamster erythrocytes (HRBC) in Freund's complete adjuvant (FCA) or an intravenous injection of hamster lymph node (HLN) cells suppressed antibody production against HRBC in the low-responder C57BL/6 and AKR mice, when HRBC in saline were given on the same day; 2) The suppressing effect of such treatments was neither detectable in the high-responder SL mice, nor in the C57BL/6 mice, which had been pre-sensitized with HRBC in FCA or hamster lymphoma cells; 3) Positive reactions of the peritoneal macrophage disappearance test and the enhanced antibody production were detected seven days after treatment with HRBC in FCA and HRBC in saline, or HLN cells and HRBC in saline; 4) The suppressing effect of such simultaneous treatments on anti-HRBC antibody production was eliminated by a transfer of normal syngeneic thymus cells to AKR mice or a transfer of thymus cells from SL to C57BL/6 mice. Suppression of the antibody production in the low-responder mice by the described simultaneous treatments may be due to a competitive involvement of HRBC-specific thymus-derived cells (T cells) in the developmental stages of delayed hypersensitivity and antibody production. High-responder SL mice appear to have enough T cells for development of the delayed hypersensitivity and as helper cells in antibody production. These results appear to support the concept that T cells for delayed hypersensitivity and antibody production to HRBC antigen are derived from the same original pool.     

Antigen-Specific Proliferative Response of Peritoneal Exudate Lymphocytes Primed with Antigen and Bacterial Lipopolysaccharide: The Roles of Ia+ Accessory Cells and IL-2

In vitro antigen-specific proliferation was investigated in a lymphocyte population that had been taken from the peritoneal exudate cells (PEC) of C3H/HeN mice (Ia^k) primed in vivo with both bacterial lipopolysaccharide (LPS) and horse red blood cells (HRBC) and had been purified by passage through a nylon fiber column (Nfc). The proliferative response of the Nfc-passed lymphocytes primed with HRBC and LPS [T(HRBC + LPS) cells] depended on the dose of antigen in the cultures, and the response was higher than that of cells prepared from mice primed with HRBC alone [T(HRBC) cells]. No response was seen in the cells prepared from the LPS-primed mice [T(LPS) cells] or normal mice [T(N) cells]. The response of the T(HRBC) cells was abolished by previous treatment of the cells with anti-Ia^k antibody and complement (C), whereas the response of the T(HRBC + LPS) cells was retained after the same treatment, indicating that the Ia^– T(HRBC + LPS) cells can proliferate in response to antigen in spite of Ia+ accessory cell-depletion. Supernatants from the cultures of Ia^– T(HRBC + LPS) cells in the presence of HRBC showed abundant IL-2 activity, while those of Ia^– T(HRBC) cells did not. The IL-2 should be produced by the L3T4 cell population in T(HRBC + LPS) cells in response to antigen, since the previous treatment of the cells with anti-L3T4 antibody and C abrogated the production. On the other hand, the Ia^– T(HRBC + LPS) cells as well as the Ia^– T(LPS) cells could respond to IL-2 dose-dependently when recombinant IL-2 was added into the cultures, but the response of Ia^– T(HRBC) cells to IL-2 was very weak. The cell population responding to IL-2 in the T(HRBC + LPS) cells as well as T(LPS) cells must be AsGM1-positive or natural killer (NK) cells, since previous treatment of the cells with anti-AsGM1 antibody and C abrogates the response. Together these results suggest that L3T4 lymphocytes capable of producing IL-2 in response to HRBC antigen without Ia⁺ accessory cells are generated in the PEC of the mice after priming with LPS and antigen together, and the IL-2 produced by the L3T4 lymphocytes induces the proliferation of the LPS-primed AsGM1+ cells.     

Immune Response against Hamster Erythrocytes in the Low-Responder Mouse Strains

C57BL/6 and AKR mice were treated with hamster erythrocytes (HRBC) in complete Freund's adjuvant (CFA) or incomplete Freund's adjuvant (IFA) and the development of delayed hypersensitivity and antibody production were examined. 1) Delayed hypersensitivity against HRBC antigen, as determined by the peritoneal macrophage disappearance test, was detected in mice sensitized with HRBC in CFA but not in those sensitized with HRBC in IFA. 2) Antibody production against HRBC or hapten TNP after a booster injection of HRBC or trinitrophenylated HRBC (TNP-HRBC) in saline was enhanced by pretreatment with HRBC in CFA or IFA. 3) Delayed hypersensitivity was not detectable after a booster sensitization with HRBC in CFA in mice which had been pretreated with HRBC in IFA 2 weeks earlier. In the mice treated with both HRBC in IFA (day ?21) and in CFA (day ?7), however, an enhanced antibody production against HRBC or TNP was detected after an intravenous injection with HRBC or TNP-HRBC in saline (day 0). These results suggest that sensitized effector lymphocytes in delayed hypersensitivity and helper cells in antibody production may be derived from the same pool of unprimed T cells. The pool of unprimed T cells with a capacity to differentiate into either type of primed T cells may be exhausted after pretreatment with the antigen in IFA, and the primed helper T cells may not be able to differentiate into sensitized lymphocytes even after sensitization with the antigen in CFA, which favors development of delayed hypersensitivity in normal controls.     

Immune Response against Hamster Erythrocytes in the Low-Responder Mouse Strains

Patterns of proliferation of antibody-forming cells after an intravenous immunization with hamster erythrocytes (HRBC) were compared in groups of mice possessing different activities of thymus-derived lymphocytes (T cells). 1) Marked differences in the numbers of hemolysin plaque-forming cells (PFC) after HRBC injection were found among the low- and high-responder normal mice and those pretreated with HRBC in complete Freund's adjuvant (CFA) or incomplete adjuvant (IFA), and they appeared to depend primarily upon the different rates of proliferation of antibody-forming cells rather than on the numbers of antigen-specific lymphocytes initiating the antibody response. 2) The numbers of hemolytic foci were slightly larger in mice with large numbers of PFC (normal SL mice, the pretreated SL and C57BL/6 mice) than in those with small numbers of PFC (normal C57BL/6 mice). The numbers of hemolytic foci increased at almost the same rate from day 2 to day 3 in both groups, while the numbers of PFC increased more efficiently in mice with large numbers of PFC than in those with small numbers of PFC from day 2 to day 3. Individual hemolytic foci appeared to contain larger numbers of PFC in mice with large total numbers of PFC than in those with small total numbers of PFC. 3) The numbers of rosette-forming cells (RFC) were increased by pretreatment with HRBC in CFA and by pretreatment with HRBC in IFA to almost the same extent. Rates of increases in PFC were, however, larger by pretreatment with HRBC in CFA than with HRBC in IFA. These results suggested that the activity of the T cell determined not only the rates of proliferation of antibody-forming cells but also the antibody-producing capacity of each cell.     

Binding of two spin-labelled derivatives of chlorpromazine to human erythrocytes.

The binding to human intact erythrocytes of two different spin-labelled derivatives of chlorpromazine has been studied. The influence of the positively charged side chain of the drug has been the focus of our attention. The positively charged amphiphilic compound (spin derivative I) is water-soluble up to 80 microM at pH values below 5.9. The apolar analogue (spin derivative II) aggregates in aqueous buffer from the lowest concentration tested. Both spin derivatives undergo a slow reduction inside the erythrocyte. The reduced nitroxides are readily reoxidized by adding a low, non-quenching, concentration of potassium ferricyanide to the intact erythrocytes. The fractions of spin label I and II bound to the erythrocyte membrane or to the erythrocyte-extracted lipids remain constant as a function of the temperature (3-42 degrees C) and as a function of the concentration of the spin label up to 150 microM. E.s.r. spectra of both spin labels show a two-component lineshape when they are bound to intact erythrocytes. Below 35 degrees C for the positively charged spin probe, and below 32 degrees C for the apolar spin probe, the simulation of the lineshape shows that more than 50% of the spectrum originates from a slow-motion component. This slow-motion component is also found in erythrocyte-extracted lipids probed by the positively charged spin label below 25 degrees C. In contrast, no slow-motion component is detected in the range 4-40 degrees C for the apolar spin label in erythrocyte-extracted lipids. In this environment the apolar probe experiences a single fast anisotropic motion with an exponential dependence on 1/temperature. Detailed lineshape simulations take into account the exchange frequency between binding sites where the probe experiences a fast motion and binding sites where it experiences a slow motion. The exchange frequency is strongly temperature-dependent. Characterization of the different motions experienced inside the different locations has been achieved and compared for whole erythrocytes and for the extracted lipids. The biochemical nature of the binding sites (membrane protein/acidic phospholipid) giving rise to the slow-motion component is discussed as a function of the polarity of the spin-labelled drug and as a function of the temperature controlling the fluidity of the lipid bulk and influencing the distribution of the drug inside the membrane.     

Spin label study of erythrocyte deformability. III. Further characterizations of electron spin resonance spectral change in shear flow

It is demonstrated that the change in the spin label ESR spectrum induced by shear flow reflects the whole cell deformation as a function of the cell surface area-to-volume ratio (s/v), the morphology and intracellular viscosity. Since the effect of the change in the membrane mechanical property on the ESR spectrum has been described previously, the spin label ESR spectrum is now shown to contain full information concerning the whole cell deformability which is determined by the major intrinsic and extrinsic properties of the red blood cells. The result of microphotographic observations shows also that the cells in the flow are elongated and aligned approximately along the flow direction to an increasing extent as the cells flow near the surface of the flat channel walls. Thus, the entire observation confirms the view that the ESR spectral difference-shear rate profile is closely related to the elongation ratio shear rate characteristics obtained by other (optical) methods.     

Evidence for a membrane surface defect in erythrocytes in Huntington's disease

Electron spin resonance studies of erythrocyte membranes from patients with Huntington's disease and normal controls have been performed. Intact erythrocytes in each case were either untreated or subjected to proteolysis with the membrane impermeable enzymes, pronase, chymotrypsin, or trypsin. Membrane ghosts were prepared from untreated and protease-treated intact cells and spin labeled with protein- or lipid-specific spin probes. Comparison of the resulting electron spin resonance spectra confirmed our previous findings that in untreated samples the relevant parameter of the protein-specific spin label was increased in Huntington's disease (P < 0.02) suggesting an altered physical state of membrane proteins in this disorder, while no difference in erythrocyte lipid fluidity could be discerned. No significant difference in the physical state of membrane proteins in Huntington's disease and control as judged as spin labeling methods could be detemined in membrane ghosts prepared from protease-treated intact cells. These results, together with the known specificity of the proteases used in this study, suggest that a molecular defect in Huntington's disease erythrocytes is manifested in an exterior part of a membrane protein and supports our hypothesis that Huntington's disease is associated with a generalized cell membrane defect.     

Electron paramagnetic resonance reveals a large-scale conformational change in the cytoplasmic domain of phospholamban upon binding to the sarcoplasmic reticulum Ca-ATPase

We used EPR spectroscopy to probe directly the interaction between phospholamban (PLB) and its regulatory target, the sarcoplasmic reticulum Ca-ATPase (SERCA). Synthetic monomeric PLB was prepared with a single cytoplasmic cysteine at residue 11, which was then spin labeled. PLB was reconstituted into membranes in the presence or absence of SERCA, and spin label mobility and accessibility were measured. The spin label was quite rotationally mobile in the absence of SERCA, but became more restricted in the presence of SERCA. SERCA also decreased the dependence of spin label mobility on PLB concentration in the membrane, indicating that SERCA reduces PLB-PLB interactions. The spin label MTSSL, attached to Cys11 on PLB by a disulfide bond, was stable at position 11 in the absence of SERCA. In the presence of SERCA, the spin label was released and a covalent bond was formed between PLB and SERCA, indicating direct interaction of one or more SERCA cysteine residues with Cys11 on PLB. The accessibility of the PLB-bound spin label IPSL to paramagnetic agents, localized in different phases of the membrane, indicates that SERCA greatly reduces the level of interaction of the spin label with the membrane surface. We propose that the cytoplasmic domain of PLB associates with the lipid surface, and that association with SERCA induces a major conformational change in PLB in which the cytoplasmic domain is drawn away from the lipid surface by SERCA.     

Antigen-specific suppression of the plaque-forming cell response induced polyclonally by lipopolysaccharides

Horse erythrocytes (HRBC) were added with LPS in mouse spleen cell cultures, and the effects of HRBC on the LPS-induced polyclonal PFC response were investigated by enumerating total IgM-secreting PFC, anti-HRBC PFC, and PFC against sheep erythrocytes (SRBC). The addition of HRBC influenced the frequencies of anti-HRBC PFC in the total IgM-secreting PFC, but did not influence those of anti-SRBC PFC. The augmentation of the frequencies of anti-HRBC PFC occurred only when an appropriate dose of HRBC was added in the cultures containing T cells. Higher doses of HRBC decreased the frequencies of anti-HRBC PFC whether T cells were present or absent. The degree of reduction of the frequencies of anti-HRBC PFC was dependent on the dose of HRBC, but independent of the dose of LPS. The addition of HRBC at 1 day after LPS stimulation also decreased the frequency of anti-HRBC PFC, though the addition of 2 or 3 days hardly suppressed it. These results suggest that the antigen-specific augmentation occurs via helper T cells, and the suppression is ascribed to the direct action of antigen on the antigen-specific B cells.     

An electron spin resonance study of temperature-induced structural changes of the spin-labeled myosin head: an evidence for two conformational states of myosin head

Heavy meromyosin labeled at the SH1 thiol group with an iodoacetamide spin label was studied by electron spin resonance spectroscopy at various temperatures in the presence and absence of nucleotides and PPi. The electron spin resonance spectra of the spin label bound to myosin head showed temperature-dependent changes indicating changes of the structure around the SH1 thiol group of the myosin head. As the temperature was elevated, the bound spin label was more mobilized in all the systems examined. The mobilization of the bound spin label by the elevation of temperature was enhanced in the presence of nucleotides or PPi. The temperature-dependent spectral changes had isosbestic points indicating that the structural changes around the SH1 thiol group took place between two states of the bound spin label, a weakly immobilized and a strongly immobilized state.     

Effects of pH on membrane fluidity of human erythrocytes

The effects of pH on the membrane fluidity of intact human erythrocytes, ghosts, and their lipid vesicles were studied by spin label techniques in the range of pH 3.0 to 9.1. Two fatty acid spin labels, 5-nitroxide stearic acid (5NS) and 12-nitroxide stearic acid (12NS), and a maleimide spin label were used for the labeling of the membrane lipids and proteins, respectively. The outer hyperfine splitting (T parallel) was measured as a parameter of membrane fluidity. In the case of 5NS, the T parallel values for intact erythrocytes and ghosts remained almost constant over the entire pH range at 22 degrees C but those for their lipid vesicles changed slightly, indicating the vertical displacement of the labels in lipid bilayers. On the other hand, the ESR spectra of 12NS incorporated into intact erythrocytes and ghosts, as compared with their lipid vesicles, showed marked pH dependence. By means of spin labeling of membrane proteins, the conformational changes of the proteins were observed in the pH range mentioned above. These results suggest a possible association between the strong pH dependence of the T parallel values and the conformation changes of membrane proteins. The pH dependence of the membrane fluidity was also investigated in cholesterol-enriched and -depleted erythrocytes. The effects of cholesterol demonstrated that the membrane fluidity was significantly mediated by cholesterol at low pH, but not at high pH.     

Immunological properties of Propionibacterium acnes. I. Potentiation and Suppression on antibody response to sheep and hamster erythrocytes in mice.

Adjuvant activity of phenol-treated cells of Propionibacterium acnes C-7 in antibody response was investigated in ICR mice. Simultaneous administration (day 0) of P. acnes (i.p.) and sheep red blood cells (SRBC) (i.v.) enhanced the formation of direct plaque-forming cells (PFC) on days 2, and the formation of indirect PFC response on day 7 and thereafter. Conversely, pretreatment from 11 to 14 days before antigen injection suppressed markedly the antibody response. The potentiation and the suppression of immune response depended on doses of antigen and of P. acnes, the timing of adjuvant injection and the time of assay. The two opposite phenomena caused by P. acnes were also confirmed in antibody response against hamster red blood cells (HRBC). Pretreatment with P. acnes 1 to 14 days before antigen injection suppressed markedly anti-HRBC antibody response, whereas P. acnes injected simultaneously with HRBC or one day after injection of the antigen induced prolongation of antibody response and the production of 2-mercaptoethanol-resistant antibody.     

Spontaneous rosette-forming lymphocytes in sheep lymph. A marker for antigen-binding cells

Type O Rh positive human red blood cells (HRBC), native or treated with one of three enzymes (papain, trypsin, or neuraminidase), were labeled with ⁵¹Cr and then sensitized with anti-Rh immune globulin. These cells served as targets in antibody-dependent cellular cytotoxicity (ADCC) for unfractionated human mononuclear cells (MC), MC depleted of monocytes by adhesion to plastic, and MC enriched for monocytes. Enzyme-treated HRBC were lysed with greater efficiency in ADCC than native HRBC. This was explained by the finding that the enzyme modified HRBC were lysed both by lymphocytes and monocytes, whereas native HRBC were lysed only by monocytes. The lysis of native HRBC was strongly inhibited by small amounts of human serum or free IgG. In contrast, the lysis of enzyme-treated HRBC was considerably more resistant to inhibition by human serum or free IgG. The enhanced lysis of enzyme-treated HRBC could not be the result of increased binding of antibody to the target cells, since augmented lysis was observed both for HRBC sensitized before neuraminidase treatment as well as for HRBC sensitized after neuraminidase treatment. These results suggest that the surface charge on target cells plays a critical role in determining which classes of leukocytic effector cells are active in ADCC systems.     

Difference in antibody production to hererologus erythrocytes in conventional, specific-pathogen-free (SPF), germfree and antigen-free mice.

The expression of antibody-producing capacities against hamster erythrocytes (HRBC), known to be weakly immunogenic in mice, was compared among conventional, SPF, germfree and antigen-free mice. ICR strain germfree and antigen-free mice showed antibody production to HRBC comparable to that in conventional or specific-pathogen-free (SPF) mice. In the NC strain, some of the conventional mice produced low titers of antibody after a single injection of HRBC, but none of the germfree mice showed such a transient antibody production. In the ICR-KIG strain, which was selected from the colony-bred ICR strain, antibodies with high titers were produced after a single injection of HRBC under both conventional and germfree conditions. The onset of conversion from 2-mercaptoethanol (2-ME) sensitive to 2-ME resistant antibody after a single injection of HRBC was not delayed in the germfree mice when compared with the conventional or SPF mice. Antibody production to sheep erythrocytes (SRBC), known to be highly immunogenic in mice, was not influenced by exogeneous stimulation from microorganisms or diet. No differences in antibody production to SRBC were detected irrespective of the maintenance conditions of the mice. Stimulation with microorganisms or diet may not be required as essential elements for the maturation of antibody-producing capacities, but such a stimulation appears to modify the antibody response. The modifying effect was more prominent in the antibody response against weakly immunogenic antigens than against highly immunogenic ones.     

The effect of osmotic pressure of aqueous PEG solutions on red blood cells

A drastic increase of the intracellular microviscosity of red blood cells in the presence of polyethylene glycol (PEG) was established by electron spin resonance using the small spin label molecule 2,2,6,6-tetramethylpiperidine-N-oxyl-4-one (TEMPONE). The effective osmotic pressure of PEG solutions stressing the cells was estimated by comparison with those cytoplasmic rotational correlation times of TEMPONE measured in NaCl or sucrose containing media of known osmotic pressure.     

A comparison of the spin labels MAL-3 and TEMPAMINE for measuring the internal microviscosity of human erythrocytes

In a recent paper, Daveloose et al. (Daveloose, D., Fabre, G., Berleur, F., Testylier, G. and Letterrier, F. (1983) Biochim. Biophys. Acta 763, 41-49) described a technique to measure the internal microviscosity of erythrocytes using the spin label MAL-3 (2,2,5,5-tetramethyl-3-malimidopyrrolidinyl-N-oxyl) to supercede the use of TEMPAMINE (2,2,6,6-tetramethylpiperidine-N-oxyl-4-amine)ferricyanide. They cite the long time (12 h) required for TEMPAMINE to enter the erythrocyte interior and the 'unphysiological' nature of the potassium ferricyanide required to isolate the TEMPAMINE signal inside the erythrocytes as reasons for developing this technique. In the present work, the penetration of TEMPAMINE into the erythrocyte interior is found to be pH-dependent and on the order of seconds, not hours. The slow penetration of TEMPAMINE described by Daveloose et al. probably is a result of their using trimethylTEMPAMINE which is membrane-impermeable. High concentrations of ferricyanide do affect red cell morphology in a reproducible fashion. The internal microviscosity as measured by TEMPAMINE reflects this. MAL-3 does not. Sample preparation time using TEMPAMINE-ferricyanide is about 1 min compared to about 40 min for the MAL-3 technique. In the presence of ferricyanide or ascorbate, MAL-3-labeled cells show slow loss of signal intensity which indicates that MAL-3 leaks out of the cells. Both the TEMPAMINE-ferricyanide and MAL-3 techniques have advantages and disadvantages.     

Neisseria gonorrhoeae membrane microenvironment studied by spin-label electron spin resonance: comparison of colony types.

Spin-label electron spin resonance was used to characterize the microenvironment around spin probes which localize (i) in membranes, (ii) at the membrane surface, or (iii) in the cytoplasm of living Neisseria gonorrhoeae. Four colony types (T1, T2, T3, and T4) of gonococci were compared on the basis of the electron spin resonance parameters 2T parallel to, S (order parameter), and tau c (microviscosity). The concentration of spin label used had little or no effect on viability. T1 and T2 gonococci were found to have a more restricted environment for molecular motion of a membrane surface spin label than did T3 and T4. The membrane fluidity, as measured by a membrane lipid spin label, of T4 (S = 0.571) was significantly greater than that of T1 or T3 (S = 0.580). This difference was detected at 37 degrees C, at 25 degrees C, in agar-grown bacteria, and in exponential-phase cells. Studies using spin labels which probe different levels of the membrane indicated the presence of a membrane flexibility gradient. Cytoplasmic spin-label studies indicated that the cytoplasm of all gonococcal colony types was three to five times more viscous than water.