Modulation of platelet shape and membrane receptor binding by Ca2+-calmodulin complex

Genetic Engineering of the Heme Pocket in Human Serum Albumin: Modulation of O2Binding of Iron Protoporphyrin IX by Variation of Distal Amino Acids

Complexing an iron protoporphyrin IX into a genetically engineered heme pocket of recombinant human serum albumin (rHSA) generates an artificial hemoprotein, which can bind O2 in much the same way as hemoglobin (Hb). We previously demonstrated a pair of mutations that are required to enable the prosthetic heme group to bind O2 reversibly: (i) Ile-142-->His, which is axially coordinated to the central Fe2+ ion of the heme, and (ii) Tyr-161-->Phe or Leu, which makes the sixth coordinate position available for ligand interactions [I142H/Y161F (HF) or I142H/Y161L (HL)]. Here we describe additional new mutations designed to manipulate the architecture of the heme pocket in rHSA-heme complexes by specifically altering distal amino acids. We show that introduction of a third mutation on the distal side of the heme (at position Leu-185, Leu-182, or Arg-186) can modulate the O2 binding equilibrium. The coordination structures and ligand (O2 and CO) binding properties of nine rHSA(triple mutant)-heme complexes have been physicochemically and kinetically characterized. Several substitutions were severely detrimental to O2 binding: for example, Gln-185, His-185, and His-182 all generated a weak six-coordinate heme, while the rHSA(HF/R186H)-heme complex possessed a typical bis-histidyl hemochrome that was immediately autoxidized by O2. In marked contrast, HSA(HL/L185N)-heme showed very high O2 binding affinity (P1/2O2 1 Torr, 22 degrees C), which is 18-fold greater than that of the original double mutant rHSA(HL)-heme and very close to the affinities exhibited by myoglobin and the high-affinity form of Hb. Introduction of Asn at position 185 enhances O2 binding primarily by reducing the O2 dissociation rate constant. Replacement of polar Arg-186 with Leu or Phe increased the hydrophobicity of the distal environment, yielded a complex with reduced O2 binding affinity (P1/2O2 9-10 Torr, 22 degrees C), which nevertheless is almost the same as that of human red blood cells and therefore better tuned to a role in O2 transport.

O2and CO Binding Properties of Artificial Hemoproteins Formed by Complexing Iron Protoporphyrin IX with Human Serum Albumin Mutants

The binding properties of O2 and CO to recombinant human serum albumin (rHSA) mutants with a prosthetic heme group have been physicochemically and kinetically characterized. Iron(III) protoporphyrin IX (hemin) is bound in subdomain IB of wild-type rHSA [rHSA(wt)] with weak axial coordination by Tyr-161. The reduced ferrous rHSA(wt)-heme under an Ar atmosphere exists in an unusual mixture of four- and five-coordinate complexes and is immediately autoxidized by O2. To confer O2 binding capability on this naturally occurring hemoprotein, a proximal histidine was introduced into position Ile-142 or Leu-185 by site-directed mutagenesis. A single mutant (I142H) and three double mutants (I142H/Y161L, I142H/Y161F, and Y161L/L185H) were prepared. Both rHSA(I142H/Y161L)-heme and rHSA(I142H/Y161F)-heme formed ferrous five-N-coordinate high-spin complexes with axial ligation of His-142 under an Ar atmosphere. These artificial hemoproteins bind O2 at room temperature. Mutation at the other side of the porphyrin, Y161L/L185H, also allowed O2 binding to the heme. In contrast, the single mutant rHSA(I142H)-heme could not bind O2, suggesting that removal of Y161 is necessary to confer reversible O2 binding. Laser flash photolysis experiments showed that the kinetics of CO recombination with the rHSA(mutant)-heme were biphasic, whereas O2 rebinding exhibited monophasic kinetics. This could be due to the two different geometries of the axial imidazole coordination arising from the two orientations of the porphyrin plane in the heme pocket. The O2 binding affinities of the rHSA(mutant)-heme were significantly lower than those of hemoglobin and myoglobin, principally due to the high O2 dissociation rates. Changing Leu-161 to Phe-161 at the distal side increased the association rates of both O2 and CO, which resulted in enhanced binding affinity.

Characterization of bullet tuna myoglobin with reference to the thermostability-structure relationship

Myoglobin (Mb) was isolated from bullet tuna (Auxis rochei) skeletal muscle and characterized from the viewpoint of the thermostability-structure relationship. Differential scanning calorimetry (DSC) measurement showed that the thermostability of bullet tuna Mb was the lowest among all the scombridae fish Mbs so far examined. The highest value (72.8 degrees C) of melting temperature (Tm) was obtained at pH 6.52. alpha-Helical content at 10 degrees C was 34.5%, clearly lower than that of horse Mb (55.3%). The amino acid sequence was then deduced by cloning cDNA which encodes bullet tuna Mb. Bullet tuna Mb consisted of 147 amino acids, and the sequence identity was very close to that of skipjack (Katsuwonus pelamis) Mb (91.8%). A few amino acid substitutions, which could be involved in stability difference of Mb, were recognized. By mass spectrometry of lysyl endoproteinase digest of Mb, the N-terminus was found to be acetylated like that of other fish Mbs.

Oxygen equilibrium properties of myoglobin locked in the liganded and unliganded conformations

A comparison of the O(2) equilibrium curves of sperm-whale myoglobin locked in the liganded (CO-bound) and unliganded (deoxy) conformations by encapsulation in a wet porous sol-gel silica reveals a marked difference between them. The CO-bound state-locked myoglobin showed a nearly monophasic (hyperbolic) O(2) equilibrium curve with a dissociation constant of 0.2 Torr, which is smaller than that of myoglobin in solution (0.5 Torr). On the other hand, the deoxy state-locked myoglobin exhibited a multiphasic O(2) equilibrium curve that can be represented by a sum of three independent components with dissociation constants of 0.19, 0.90, and 44 Torr, respectively, indicating that deoxymyoglobin exists in multiple conformations. These results show that myoglobin can be frozen into ligand-dependent conformational populations at room temperature in the wet sol-gel and suggest that the overall O(2) equilibrium properties of myoglobin in solution are generated by a redistribution of protein conformational populations in response to ligand binding.

Measurements of the photodissociation quantum yields of MbNO and MbO(2) and the vibrational relaxation of the six-coordinate heme species

The (t approximately 0) photodissociation quantum yields (Y(0)) of MbNO and MbO(2) are measured to be 50 +/- 5 and 28 +/- 6%, respectively, using MbCO (Y(0) = 100%) as a reference. When photolysis does not take place, we find that a significant portion of the photon energy contributes to heating of the residual six-coordinate heme (MbNO and MbO(2)). The time constant for vibrational relaxation of the six-coordinate ligand-bound heme is found to be close to 1 ps for both samples. The MbO(2) sample also shows a approximately 4-ps optical response that is assigned to a rapid phase (25-30% amplitude) of O(2) geminate rebinding. We observe no additional geminate recombination in the MbO(2) sample out to 120 ps. In contrast, the MbNO sample displays significant geminate recombination over the first 120 ps, which can be adequately fit with two exponentials whose amplitudes and time constants appear to depend weakly on the pump wavelength. This more complex kinetic behavior conceivably arises due to heating of the photodissociated heme and its effect on the geminate recombination as the system cools. Overall, the data are consistent with a hypothesis that distortions along the iron-ligand bending coordinate play a key role in the photodissociation process. The transient formation of an unphotolyzable FeO(2) side-on binding geometry is suggested to be responsible for the lowered quantum yield of MbO(2) relative to MbNO.

Raman spectroscopic and electrochemical characterization of myoglobin thin film: implication of the role of histidine 64 for fast heterogeneous electron transfer

Myoglobin (Mb) thin films formed on various substrates have been characterized by using Raman spectroscopy, reflectance absorbance FT-IR, UV-vis absorption spectroscopy, and electrochemical methods. Raman spectra were obtained upon excitation within the Soret band as well as alpha-beta bands. The spin state marker bands observed from the Mb film in the 1550-1630 cm(-)(1) region (excitation at 514.5 nm) are approximately 20 cm(-)(1) higher than those of aqueous metMb having the high spin state. The 1210 cm(-)(1) band (methine bridge C-H vibration) also shifts to 1240 cm(-)(1) upon the formation of the film. These results indicate that the heme iron of myoglobin in the film is the ferric low-spin state, and the iron atom is pulled to the heme plane. A comparison of the Raman spectra of the Mb film with that of an Mb-imidazole derivative leads to the conclusion that the distal histidine is responsible for the change in the spectral characteristics. The escape of water from the sixth position upon the formation of the Mb film may result in a conformational change at the heme distal pocket: the histidine residue at the E7 helical position (H64) moves toward the central iron and is coordinated with it through the N on the imidazole ring. These structural features facilitate the fast electron transfer between the thin protein film and the electrode. Distal histidine may serve as an electron-transfer pathway as it does in cytochrome c.

Imidazole is a sensitive probe of steric hindrance in the distal pockets of oxygen-binding heme proteins

The FixL heme-based sensor, despite its low affinity for oxygen, is much more reactive than myoglobin toward the large polar ligand imidazole. To determine which features of a myoglobin heme pocket favor binding of imidazole, we have measured binding of this ligand to the FixL heme domain, elephant myoglobin, wild-type sperm whale myoglobin, and sperm whale myoglobins having alanine, valine, threonine, glutamine, leucine, phenylalanine, or tryptophan substitutions of the distal (E7) histidine residue. Except for histidine, the association rate constants dropped more than 3000-fold as the volume of the E7 side chain, at position 64, was expanded from alanine (10(6) M-1 s-1) to phenylalanine (10(3) M-1 s-1). There was inhibition of imidazole binding due to displacement of coordinated water from H64 and H64Q sperm whale myoglobins, where the E7 side chain hydrogen bonds directly to the bound ligand. The imidazole dissociation rate constants varied less dramatically and less consistently with any single factor, though they were measurably decreased by hydrogen bonding to an E7 glutamine or histidine. On the whole, the results for the sperm whale myoglobin E7 substitutions show that the rate constants for imidazole binding are useful and sensitive indicators of steric hindrance and polar interactions in the distal pockets of myoglobins. The combined effects of the glutamine 64 and phenylalanine 29 in elephant myoglobin largely account for its increased imidazole association and dissociation rate constants, respectively, compared to those of sperm whale myoglobin. An unhindered distal pocket not competent to stabilize positive poles is indicated by the large imidazole association (>/=10(4) M-1 s-1) and dissociation (>/=50 s-1) rate constants, parameters that are characteristic of FixL.

A comparison of functional and structural consequences of the tyrosine B10 and glutamine E7 motifs in two invertebrate hemoglobins (Ascaris suum and Lucina pectinata)

The architecture of the distal heme pocket in hemoglobins and myoglobins can play an important role in controlling ligand binding dynamics. The size and polarity of the residues occupying the distal pocket may contribute steric and dielectric effects. In vertebrate systems, the distal pocket typically contains a "distal" histidine at position E7 and a leucine at position B10. There are several invertebrate organisms that have hemoglobins or myoglobins that display a pattern in which residues E7 and B10 are a glutamine and tyrosine, respectively. These proteins often have very high oxygen affinities stemming from very slow ligand off rates. In this study, two such hemoglobins, one from the nematode Ascaris suum and the other from the sulfide-fixing clam Lucina pectinata, are compared with respect to conformational and functional properties. Ultraviolet resonance Raman spectroscopy and visible resonance Raman spectroscopy are used to probe, respectively, the ligand-dependent hydrogen bonding pattern of the tyrosine residues and the proximal heme pocket interactions. Fourier transform infrared absorption spectroscopy is used to probe the dielectric properties of the distal heme pocket through the stretching frequency of carbon monoxide bound to the heme. Functionality is probed through the geminate rebinding of both CO and O2. The findings reveal two very different patterns indicative of two different mechanisms for achieving low oxygen off rates. In Hb Ascaris, a hydrogen bonding network that includes the E7 Gln, B10 Tyr, and oxygen bound to the heme results in a tight cage for the oxygen. Dissociation of the O2 requires a large amplitude conformational fluctuation that results both in a spontaneous dissociation of the oxygen through the loss of hydrogen bond stabilization and in an enhanced probability for ligand escape though the transient disruption and opening of the tight distal cage. In the case of the Hb from Lucina, there is no evidence for a tight cage. Instead the data support a model in which the hydrogen bonding network is far more tenuous and the equilibrium state of distal pocket is far more open and accessible than is the case in Ascaris. The results explain why Hb Ascaris has one of the highest oxygen affinities known (P50 approximately 10(-)3 Torr) while Hb Lucina II has an oxygen affinity comparable to that of Mb (P50 = 0.13 Torr) even though both of these Hbs contain the B10 Tyr and E7 Gln motif and display very low oxygen off rates. The roles of water and proximal strain are discussed.

Identification of conformational substates involved in nitric oxide binding to ferric and ferrous myoglobin through difference Fourier transform infrared spectroscopy (FTIR)

Hemeproteins play an important role in the signaling processes mediated by nitric oxide (NO). For example, the production of NO by nitric oxide synthase, the activation of guanylate cyclase by binding NO, and the scavenging of NO by hemoglobin, myoglobin, and cytochrome c oxidase all occur through unique mechanisms of interaction between NO and hemeproteins. Unlike carbon monoxide (CO) and oxygen (O2), which have been studied extensively, the reactions of NO with ferric and ferrous hemeproteins are not as well characterized. In this work, NO binding to myoglobin is studied using cryogenic optical spectroscopy and Fourier transform infrared spectroscopy (FTIR) in order to characterize the ligand-bound and photoproduct states involved in the interaction of NO with the heme iron and the distal pocket of the protein. For ferrous nitrosyl myoglobin (MbIINO), optical spectroscopy is used to show that the ligand-bound state can be converted to >95% stable photoproduct below 10 K. The Soret peak of the photoproduct is red-shifted by 4 nm relative to deoxy-myoglobin (Mb), similar to previous results for carbonmonoxy- (MbCO) and oxy-myoglobin (MbO2) (Miller et al., 1996). MbIINO completely rebinds by 35 K, indicating that the rebinding barrier for NO is lower than MbCO, consistent with room temperature picosecond kinetic measurements. For ferric nitrosyl myoglobin (MbIIINO), we find that the photoproduct yield at cryogenic temperatures is less than unity and dependent on the distal pocket residue. Native MbIIINO has a lower photoproduct yield than the mutant, MbIII(H64L)NO, where the distal histidine is replaced by leucine. The rebinding rates for the native and mutant species are similar to each other and to MbIINO. By using FTIR difference spectroscopy (photolyzed/unphotolyzed) of isotopically labeled ferrous nitrosyl myoglobin (MbIINO), the NO stretching frequencies in both the ligand-bound states and photoproduct states are determined. Two ligand-bound conformational states (1607 and 1613 cm-1) and two photoproduct conformational states (1852 and 1857 cm-1) are observed for MbIINO. This is the first direct observation of photolyzed NO in the distal pocket of myoglobin. The ligand-bound frequencies are consistent with a bent MbIINO moiety, where the unpaired pi*(NO) electron remains localized on NO, causing nu(N-O) to be approximately 300 cm-1 lower than MbIIINO. Similar to MbO2, we suggest that Nepsilon of the distal histidine is protonated, forming a hydrogen bond to the NO ligand. For native MbIIINO, a single ligand-bound conformational state with respect to nu(N-O) is observed at 1927 cm-1. This frequency decreases to 1904 cm-1 for the mutant, MbIII(H64L)NO, contrary to the increase of the carbon monoxide (CO) stretching frequency in the isoelectronic MbII(H64L)CO mutant versus native MbCO. For linear MbIIINO, we suggest that backbonding from the unpaired pi*(NO) electron to iron results in an increased positive charge on the NO ligand, Fe(delta-)-NO(delta+). This can be facilitated by tautomerism of the distal histidine, leaving Nepsilon of the imidazole ring unprotonated and able to accept positive charge from the Fe(delta-)-NO(delta+) moiety, resulting in a higher bond order (and a 23 cm-1 shift to higher frequency) for native MbIIINO versus MbIII(H64L)NO, where this interaction is absent. These different interactions between the distal histidine and the ferrous versus ferric species illustrate potential ways the protein can stabilize the bound ligand and demonstrate the versatile nature by which NO can bind to hemeproteins.

Assignment and analysis of fluorine nuclear magnetic resonance spectra of 4-fluorotryptophan myoglobins and hemoglobins

We have obtained the 470 MHz 19F NMR spectra of wild type [4-F]Trp-labeled myoglobins (MbCO, MbO2, deoxyMb, metMb, and MbCN) and hemoglobins (HbCO, HbO2, and deoxyHb), as well as those of several mutants (W7F Mb, betaW15F Hb, betaW37S Hb, and betaY130F Hb, all as the carbonmonoxy adducts), prepared via site-directed mutagenesis. The maximum observed chemical shift range induced by folding is 6.4 ppm. Using a multipole shielding polarizability-local reaction field approach, we have computed the electrostatic field contributions to the fluorine shielding. For residues which do not have F atoms in contact with neighboring groups, we find an approximately 1 ppm mean square deviation in shift from experiment, with the R2-like structure of HbCOA being in very close accord with experiment.

Interaction of Derivatized Capped Iron(II) Porphyrin Complexes with CO and O(2)

A series of porphyrins strapped and capped by benzene and amidobenzene rings have been prepared. O(2) and CO bindings to their iron(II) complexes have been examined, and the role of hydrogen-bonding in stabilizing O(2) binding has been measured. Each comparison between the benzene ring (no H-bonding) and the amidobenzene ring analogues showed a free energy gain of approximately 1 kcal/mol (at -45 degrees C) for the amidobenzene derivatives. An X-ray structure analysis was carried out for the ferric (Cl(-)) complex of the benzene-capped porphyrin strapped by two butyl side chains. The crystals were monoclinic, with a = 10.557(3) Å, b = 31.290(5) Å, c = 11.221(3) Å, beta = 104.62(2) degrees, Z = 4, and space group P2(1)/n. The structure was solved by the Patterson method and was refined by full-matrix least-squares procedures to R = 0.040 (R(w) = 0.041) for 3844 reflections with I >/= 3sigma(F(2)).

Origin of the pH-dependent spectroscopic properties of pentacoordinate metmyoglobin variants

The pH dependence of the electronic and EPR spectra of two variants of horse heart myoglobin (Mb) in which the distal His64 ligand has been replaced by either Thr or Ile has been studied. Both of these variants exhibit spectroscopic changes with pH that are indicative of a transition between two ferric high-spin forms that occurs with a pKa of 9.49 for the His64Thr variant and 9.26 for the His64Ile variant and that is distinctly different from the pH-dependent spectroscopic changes related to titration of the distal aquo ligand of wild-type Mb. The electronic and EPR spectra of both variants at all values of pH studied are consistent with the presence of a pentacoordinate heme iron center. For the His64Thr variant, a high-resolution (1.9 A) structure determination establishes the lack of the distal aquo ligand and demonstrates an out-of-plane movement of the ferric iron toward the proximal histidine together with a decrease of the Fe-His bond length. Investigation of this pH-linked equilibrium by EPR spectroscopy reveals rhombically split high-spin signals at both pH 7 and 11 with a greater degree of rhombicity exhibited by the alkaline species. We propose that the pH-linked spectroscopic transition exhibited by these distal histidine variants results from the deprotonation of the proximal His93 residue to produce imidazolate ligation at alkaline pH.

Phospholipase C from Clostridium perfringens stimulates acetyltransferase-dependent formation of platelet-activating factor in cultured intestinal epithelial cells (INT 407)

The mechanisms by which phospholipase C from Clostridium perfringens stimulates the formation of platelet-activating factor (PAF-acether) in cultured intestinal epithelial cells (INT 407) were investigated. Although stimulation with phospholipase C caused a significant formation of PAF-acether, there was no significant increase in the cellular levels of lysoPAF-acether after stimulation. Moreover, when cells prelabeled with 3H-1-O-alkyl-2-acyl-sn-glycerophosphocholine were stimulated with phospholipase C, the 3H-lysoPAF-acether content was not increased in stimulated cells as compared with unstimulated cells. When cells were preincubated with the calmodulin inhibitor trifluoperazine (TFPA), the protein kinase C inhibitor 1-(5-isoquinolinylsulfonyl)-2-methylpiperazine (H-7), or the combined phospholipase A2-inhibitor and lipoxygenase inhibitor nordihydroguaiaretic acid (NDGA) before stimulation with phospholipase C, the PAF-acether formation was significantly decreased. The phospholipase A2 inhibitor 4-bromophenacyl bromide (BPB), on the other hand, had no significant effect on the PAF-acether formation. Preincubation with NDGA also decreased the levels of lysoPAF-acether, whereas BPB, H7, or TFPA had no such effect. These findings indicate that stimulation of acetyltransferase activity with increased acetylation of lysoPAF-acether may be one way by which phospholipase C from C. perfringens stimulates formation of PAF-acether in INT 407 cells.

Spectrin and calmodulin in spreading mouse blastomeres

The role of spectrin and its association with calmodulin in spreading mouse blastomeres was investigated. Embryonic spectrin binds 125I-calmodulin in a calcium-dependent fashion in the blot overlay technique. Double-labeling experiments show coordinate redistribution of spectrin and calmodulin in blastomeres preparing to undergo active spreading movement. At this stage cortical spectrin staining is lost from the region of cell-substrate contact and spectrin and calmodulin become concentrated in two structures closely associated with the contacted region: a group of spherical bodies located on the cytoplasmic side of the cortical layer and a subcortical ring that marks the perimeter of the contacted region. The localization pattern of spectrin and calmodulin is also coordinated with that of actin and myosin. The results suggest that spectrin plays a role in the spreading of blastomeres and that this function may involve linkage of spectrin, calmodulin, and the cortical contractile apparatus.

Platelet aggregation in myotonia

Platelet function tests were carried out on 21 patients with myotonic dystrophy (MyD) and 7 patients with myotonia congenita (MC) together with 22 healthy subjects. Compared to the controls, the MyD and MC patients showed significantly greater sensitivity to the calcium ionophore A23187 but not to adenosine diphosphate, collagen, ristocetin or adrenaline. In addition, with the MyD patients, a significantly higher concentration of chlorpromazine and lignocaine was required to inhibit aggregation induced by a standard dose of adrenaline. A similar trend was noted with the MC patients but because of the small numbers tested, statistical analysis was not possible. Plasma levels of betathromboglobulin were also significantly higher in MyD patients than in controls. Platelet adenine nucleotide levels were within the normal range in both groups of patients and were not significantly different from those in controls. These preliminary results support the view that there is a defect in calcium metabolism in myotonic dystrophy and suggest that it may be possible to use the blood platelet as a model with which to carry out further studies in this disorder.

Selective effects of TPA and IL-1 on protein phosphorylation in murine thymocytes

Protein kinase C activity was demonstrated in murine thymocytes and the effects of TPA and IL-1 on this enzyme were studied. TPA, but not IL-1, could substitute for diacylglycerol in protein kinase C activation. Although TPA and IL-1 are both potent comitogens for murine thymocytes they markedly differ in their effects on protein phosphorylation and protein kinase C activation. Treatment of intact thymocytes with TPA resulted in a marked increase in the phosphorylation of an endogenous protein with Mr approximately 44,000. Enhanced phosphorylation of this protein was also observed when protein kinase C was activated in thymocyte extracts. In contrast to TPA, IL-1 neither induced phosphorylation of the 44,000-Da protein nor activated protein kinase C. The data suggests that protein kinase C does not mediate the comitogenic effect of IL-1 in murine thymocytes.

Calmodulin in human serum and the specific release of calmodulin from calmodulin-rich platelets

Calmodulin(CaM)-like activity was detected in human serum and foetal calf serum, with an activity i0 times more than that detectable in plasma. Serum CaM was largely accounted for by release from human platelets as confirmed by both radioimmunoassay and sodium-dodecyl-sulphate/polyacrylamide-gel electrophoresis of CaM partially purified from platelet releasate obtained in response to thrombin. Lactate dehydrogenase release was unaffected by thrombin. Platelet CaM content was very variable (1.3 to 11.3 pg/mg protein; n = 15). It is suggested that intact platelets are rich in CaM and that release of CaM during preparation explains the variation in CaM content reported here and in the literature.

Changes in cell shape correlate with collagenase gene expression in rabbit synovial fibroblasts

Induction of the neutral proteinase, collagenase, is a marker for a specific switch in gene expression observed in rabbit synovial fibroblasts. A variety of agents, including 12-O-tetradecanoylphorbol-13-acetate, cytochalasins B and D, trypsin, chymotrypsin, poly(2-hydroxyethylmethacrylate), and trifluoperazine induced this change in gene expression. Induction of collagenase by these agents was always correlated with a marked alteration in cell morphology, although the cells remained adherent to the culture dishes. The amount of collagenase induced was positively correlated with the degree of shape change produced by a given concentration and, to some extent, with the duration of treatment. Altered cell morphology was required only during the first few hours of treatment with inducing agents; after this time collagenase synthesis continued for up to 6 d even when agents were removed and normal flattened cell morphology was regained. All agents that altered cell morphology also produced a characteristic switch in protein secretion phenotype, characterized by the induction of procollagenase (Mr 53,000 and 57,000) and a neutral metalloproteinase (Mr 51,000), which accounted for approximately 25% and 15% of the protein secreted, respectively. Secretion of another neutral proteinase, plasminogen activator, did not correlate with increased collagenase secretion. In contrast, synthesis and secretion of a number of other polypeptides, including the extracellular matrix proteins, collagen and fibronectin, were concomitantly decreased. That changes in cell shape correlated with a program of gene expression manifested by both degradation and synthesis of extracellular macromolecules may have broad implications in development, repair, and pathologic conditions.

Control of erythrocyte shape by calmodulin

Erythrocytes are deformable cells whose shapes can be altered by treatments with a variety of drugs. The forms the erythrocyte may assume vary continuously from the spiny "echinocytes" or crenated cells at one extreme to highly folded and dented "cupped" cells at the other extreme. Examination of 39 compounds for cup-forming activity revealed a remarkable correlation between their ability to form cupped cells and their inhibitory activity against the calcium regulatory protein, calmodulin. Calmodulin is known to interact with several erythrocyte proteins including spectrin, spectrin kinase, and the Ca++ ATPase calcium pump of the membrane. These proteins regulate the form of the cytoskeleton as well as intracellular calcium and ATP levels. It is proposed that calmodulin is required to maintain normal erythrocyte morphology and that in the presence of calmodulin inhibitors, the cell assumes a cupped shape.

Alterations of the cytoplasmic organization of WIRL cells induced by trifluoperazine

The antipsychotic drug trifluoperazine (TFP) causes a reversible rounding of cells of the rat liver epithelial cell line, WIRL. We have investigated the cytoplasmic organization of these cells after TFP treatment using SEM, TEM and immunofluorescence and have observed significant differences between the control and treated cells. Mitochondria are converted to the condensed configuration with distended cristae and the endoplasmic reticulum becomes tubular with distended cisternae. Intermediate filaments, visualized with a monoclonal antibody, are aggregated to a cap on the nucleus in an arrangement different from that induced by colcemid.

Calcium regulation of the actin-mediated cytoskeletal transformation of sea urchin coelomocytes

Coelomocytes from several echinoderm species undergo an actin-mediated cytoskeletal transformation once subjected to hypotonic shock. In this study, coelomocytes from the sea urchins Lytechinus variegatus and Arbacia punctulata were induced to "transform" by treatment with greater than 5 microM of the calcium ionophore A23187 in the presence of external Ca++. The dependence of ionophore transformation on external Ca++ and the lack of chlorotetracycline staining indicates that these cells rely on external Ca++ sources. NBD-phallacidin (7-Nitrobenz-2-oxa-1,3-diazole-phallacidin) staining of lysolecithin permeabilized cells and whole-mount transmission electron microscopy (TEM) show that similar reorganizations of the actin cytoskeleton take place during hypotonic shock and ionophore transformation, although actin filament bundling is less apparent in A23187-treated cells. As has been shown with hypotonic shock transformation, the ionophore elicited shape change is inhibited by anticalmodulin drugs. Greater than 10 microM concentrations of W 13 inhibit filopod formation, while this drug's less active structural analogue, W 12, exhibits no effects. W 13 also appears to disrupt actin filament-membrane associations in the cells. Fluorescent localization of calmodulin using a photooxidized derivative of trifluoperazine indicates a general cytoplasmic distribution with some concentration in filopod core bundles. Coelomocyte transformation may be an example of a cellular shape change regulated by Ca++ through the action of calmodulin modulation of actin-membrane interactions.

Interaction of drugs with calmodulin. Biochemical, pharmacological and clinical implications

Calmodulin is a widely distributed, highly active, calcium-binding protein that influences a number of important biological events. Accordingly, agents that inhibit the activity of calmodulin should have profound pharmacological effects. Within the past few years, a number of compounds have been identified that inhibit calmodulin. The most potent of these described so far include certain antipsychotic drugs, smooth muscle relaxants, alpha-adrenergic blocking agents and neuropeptides. Studies of the physicochemical and structural properties of a variety of calmodulin inhibitors have shown that there are ionic and hydrophobic interactions between the drug and calmodulin. From the limited studies conducted so far, we conclude that, for a compound to inhibit calmodulin, it should carry a positive charge at physiological pH, presumably to interact with negative charges on the highly acidic calmodulin, and have hydrophobic groups, presumably to interact with lipophilic regions on calmodulin. But these two factors are not the only ones that are involved in inhibiting calmodulin, for many highly charged and highly hydrophobic agents have relatively little effect on calmodulin activity. The structural relationships between these ionic and hydrophobic regions and other, as yet identified, factors are also important. Many of the biochemical actions of the phenothiazine antipsychotic agents can be explained by the common mechanism of their binding to, and inhibiting, calmodulin. The question of whether these biochemical actions can explain their pharmacological and clinical effects is still unclear. The fundamental role calmodulin plays in biology suggests that this calcium binding protein may provide a new site for the pharmacological manipulation of biological activity. The calmodulin inhibitors described thus far hardly scratch the surface of this fertile area of research.