Multiple conformations of physiological membrane-bound cytochrome c.

One-tenth of cytochrome c (cyt c) remains bound to the inner mitochondrial membrane (IMM) at physiological ionic strength (I; i.e. , I approximately 150 mM), exhibiting decreased electron transport (ET) activity. We now show that this form of membrane-bound cyt c (MB-cyt c) can be obtained in vitro and that binding to membranes at low I generates an additional conformation with higher ET activity. This low I bound form of MB-cyt c (MBL-cyt c) exhibited intrinsic ET rates similar to those of electrostatically bound cyt c (EB-cyt c). The ET activity of IMM-bound MB-cyt c approached slowly that of MBL-cyt c or EB-cyt c, suggesting that MB-cyt c converts to MBL-cyt c while bound to IMM. When maintained at physiological I, both forms of MB-cyt c were released from the membrane, indicating that they convert to an EB-cyt c-like form. This process may be very dynamic in cellular mitochondria, as binding and release for both MB-cyt c forms increased considerably with temperature. I-Dependent binding of MB-cyt c does not require IMM, and it can be reproduced using large or small unilamellar vesicles (SUV). Using SUV-cyt c complexes, we characterized the secondary structure of MB-cyt c and MBL-cyt c by circular dichroism. Conformational analysis revealed that cyt c binding as MB-cyt c decreases its alpha-helical content (70-79%) and increases its beta-sheet up to 135%. The secondary structure of MBL-cyt c was similar to that of EB-cyt c and soluble cyt c, with a modest increase in beta-sheet. Taken together, our experiments suggest that physiological cyt c exists in soluble and membrane-bound conformations with similar ET activity, which may exchange very rapidly, and that soluble hydrophilic proteins can bind transiently to biomembranes.

Persistence of cytochrome c binding to membranes at physiological mitochondrial intermembrane space ionic strength.

We have shown that cytochrome c (cyt c) diffuses primarily in three dimensions in the intermembrane space (IMS) of intact mitochondria at physiological ionic strength (I). Recently, we found that a small percentage (11.2 +/- 2.1%) of endogenous cyt c remains bound to inner mitochondrial membranes (IMM) at high, physiological I (I = 150 mM), even after extensive washing with solutions at physiological I, overnight dialysis, changes in medium osmolarity, or further purification of IMM at high I using self-generating Percoll gradients. Measurements of heme c/heme a ratios, and electron transport (ET) reactions in which cyt c participates, confirmed the presence of a low amount of this I-resistant, membrane-bound form of cyt c (MB-cyt c), that had one third of the ET activity of electrostatically-bound cyt c (EB-cyt c), and which could not account for maximal ET rates. The amount of MB-cyt c was significantly increased above endogenous MB-cyt c by exposing KCl-washed IMM to increasing concentrations of exogenous cyt c. Also, subjecting large unilamellar vesicles (LUV) to successive cycles of cyt c binding/high I KCl-washes gave progressive increases in MB-cyt c. These protocols allowed in vitro characterization of MB-cyt c. The I at which binding takes place affects the affinity of cyt c for membranes, and oxidized cyt c had a greater intrinsic affinity for IMM or SUV than reduced cyt c. MB-cyt c appears to be bound partially by hydrophobic interactions since MB-cyt c was detected on negatively charged (asolectin) LUV and also on neutral, zwitterionic (phosphatidylcholine) LUV at high I. Consistent with the concentration-dependent changes in MB-cyt c, decreasing the IMS-volume of intact mitochondria (i.e., increasing th endogenous IMS-cyt c concentration) by metabolic or osmotic means increased the amount of MB-cyt c. After cyt c was delivered into the IMS by liposome-mediated low pH-induced fusion, resonance energy transfer showed a time-dependent cyt c-membrane proximity which was consistent with slow exchange of soluble IMS-entrapped cyt c molecules with a population bound to membranes at I = 150 mM. We conclude that, even though the majority of functional IMS-cyt c diffuses in three dimensions, a small portion remains firmly bound on the surface of the IMM under I conditions that are physiological for intact mitochondria. The occurrence of MB-cyt c may reflect an intrinsic conformational flexibility in cyt c, that allows a degree of membrane penetration and the formation of hydrophobic interactions which stabilize the membrane-bound form. The persistence of cyt c-membrane interactions under physiological I conditions indicates that cyt c-mediated ET in the IMS involves both fast (3D-diffusion) and slow (2D-diffusion) pathways for electron transfer.

The ionic strength of the intermembrane space of intact mitochondria is not affected by the pH or volume of the intermembrane space.

Ionic strength affects the electron transport activity of cytochrome c through its electrostatic interactions with redox partners and membrane lipids. We previously reported (Cortese, J.D., Voglino, A.L. and Hackenbrock, C.R. (1991) J. Cell Biol. 113, 1331-1340) that the ionic strength (I) of the intermembrane space (IMS-I) in isolated, intact condensed mitochondria is similar to the external, bulk I, over a wide range of bulk I. We now consider the possible effects of IMS-pH and IMS-volume, both variable parameters of mitochondrial function in situ, on IMS-I. IMS-pH and IMS-I were measured with pH- and I-sensitive fluorescent probes (highly fluorescent FITC-dextran for IMS-pH and FITC-BSA for IMS-I). These probes were delivered into the IMS of intact mitochondria via probe encapsulation into asolectin vesicles, followed by low pH-induced fusion of the vesicles with the outer membranes of intact mitochondria. IMS-pH was found to be 0.4-0.5 units lower than bulk pH over the pH range 6.0-8.5 for mitochondria with a large IMS-volume separating the two mitochondrial membranes (condensed configuration), and 0-0.2 units lower for mitochondria with a small IMS-volume and membranes closely opposed (orthodox configuration). This small pH difference between IMS-pM and bulk pH did not influence the similarity between IMS-I and bulk I. When the IMS-volume was osmotically decreased, bringing the two mitochondrial membranes in close proximity as in the orthodox configuration, IMS-I followed the bulk I above 10 mM but did not respond to changes in bulk I below 10 mM. The lack of response of the IMS-I below 10 mM indicates that the close proximity of the two mitochondrial membranes excludes ions only at low, nonphysiological I. Since the similarity of IMS-I and bulk I is unaffected by either IMS-pH or IMS-volume above a bulk I of 10 mM, at cytosolic physiological I (i.e., 100-150 mM) cytochrome c can be expected to be a free, three-dimensional diffusant in the IMS irrespective of the pH or volume of the IMS.

Ionic strength of the intermembrane space of intact mitochondria as estimated with fluorescein-BSA delivered by low pH fusion.

The electrostatic interactions of cytochrome c with its redox partners and membrane lipids, as well as other protein interactions and biochemical reactions, may be modulated by the ionic strength of the intermembrane space of the mitochondrion. FITC-BSA was used to determine the relative value of the mitochondrial intermembrane ionic strength with respect to bulk medium external to the mitochondrial outer membrane. FITC-BSA exhibited an ionic strength-dependent fluorescence change with an affinity in the mM range as opposed to its pH sensitivity in the microM range. A controlled, low pH-induced membrane fusion procedure was developed to transfer FITC-BSA encapsulated in asolectin liposomes, to the intermembrane space of intact mitochondria. The fusion procedure did not significantly affect mitochondrial ultrastructure, electron transport, or respiratory control ratios. The extent of fusion of liposomes with the mitochondrial outer membrane was monitored by fluorescence dequenching assays using a membrane fluorescent probe (octadecylrhodamine B) and the soluble FITC-BSA fluorescent probe, which report membrane and contents mixing, respectively. Assays were consistent with a rapid, low pH-induced vesicle-outer membrane fusion and delivery of FITC-BSA into the intermembrane space. Similar affinities for the ionic strength-dependent change in fluorescence were found for bulk medium, soluble (9.8 +/- 0.8 mM) and intermembrane space-entrapped FITC-BSA (10.2 +/- 0.6 mM). FITC-BSA consistently reported an ionic strength in the intermembrane space of the functionally and structurally intact mitochondria within +/- 20% of the external bulk solution. These findings reveal that the intermembrane ionic strength changes as does the external ionic strength and suggest that cytochrome c interactions, as well as other protein interactions and biochemical reactions, proceed in the intermembrane space of mitochondria in the intact cell at physiological ionic strength, i.e., 100-150 mM.

Angiokeratoma corporis diffusum (Fabry's disease) with unusual features in a female patient. Light- and electron-microscopic investigation.

A case of clinically apparent angiokeratoma corporis diffusum (Fabry's disease) in an adult female carrier is presented. The patient had biochemical evidence of the disease, and showed multiple cutaneous lesions, in the absence of other major organ involvement. Ultrastructural examination of tissue fragments obtained by skin biopsies demonstrated the presence of a few typical electron-dense lamellar structures in endothelial cells, but not in smooth muscle cells. Electron microscopy proved to be the only effective way of detecting the intracytoplasmic inclusions, since light-microscopic histochemistry failed to reveal small amounts of the stored glycolipid. The exclusive involvement of endothelial cells suggests that they are more prone to store glycolipid than are all other types of cells usually involved in the disease. The abundance of intermediate filaments in the cytoplasm of endothelial cells is related exclusively to the high blood pressure in angiomatous arteriovenous shunts.

Ultrastructural histochemistry of infantile digital fibromatosis.

A case of infantile digital fibromatosis was studied by light and electron microscopy histochemistry. Using two different acidic solutions of phosphotungstic acid at varying pHs, the round inclusions characteristic of this tumor were shown to have a high protein content with little or no carbohydrates. The histochemical reactivity of the inclusions was similar to that of the cytoplasmic microfibrils in the tumors cells and consistent with the idea that both the inclusions and the microfibrils represent actin. There is, however, no definite proof that the tumor cells are myofibroblasts. At the present time, this tumor should be viewed as a peculiar expression of deranged assembly or metabolism of filamentous proteins or both.