Thermodynamic studies of bovine lung surfactant extract mixing with cholesterol and its palmitate derivative.

Langmuir film behavior of bovine lipid extract surfactant (BLES), mixed with cholesterol (CHOL) and cholesterol palmitate (CHOLP), has been studied by surface pressure (pi)-area (A) measurements. Associative interactions, observed for both systems, were less favored at lower BLES content. The presence of unsaturated phospholipids and surfactant proteins in BLES favored the association. Miscibility of BLES was better with CHOLP than with CHOL at all compositions, indicating more compact packing of the BLES-CHOLP than of the BLES-CHOL system. The most stable mixtures were found at 30-40 mol% CHOL and at low pi and at 20-25 mol% CHOLP but at higher pi. These results suggest that BLES-CHOL miscibility is better at low pi and low CHOL concentrations, while BLES-CHOLP miscibility is better at high pi and high CHOLP concentrations.

Structural alterations of phospholipid film domain morphology induced by cholesterol.

Structures of the monolayer films of dipalmitoylphosphatidylcholine (DPPC) mixed with different amounts of cholesterol were studied at air-water interface using surface pressure-area measurements, epifluorescence microscopy and atomic force microscopy (AFM). Pure DPPC, cholesterol or DPPC-cholesterol mixtures were dissolved in organic solvents with a small amount of fluorescently labeled phospholipid probe (NBD-PC) and spread onto the air-water interface. Surface pressure-area isotherms and epifluorescence microscopy of such films at the air-water interface suggested that DPPC undergoes a gas to fluid to condensed phase transition, while cholesterol undergoes a gas to solid-like transition. A shift of the surface pressure-area curve to lower area per molecule was observed when cholesterol was mixed with DPPC. Epifluorescence microscopy showed the formation of spiral shaped domains for mixed monolayers. Increase in cholesterol content abolished domain characteristics possibly due to fluidizing property of cholesterol. AFM measurements of monolayers, transferred onto freshly cleaved mica by Langmuir-Blodgett technique, revealed the alterations caused by cholesterol on the gel and fluid domains of such films. AFM measurements re-established similar trend in domain characteristics as evidenced in epifluorescence microscopy.

The Sendai virus membrane fusion mechanism studied using image correlation spectroscopy.

The mechanism of Sendai virus membrane fusion to cultured cell membranes was studied. Viral lipids were labeled with the lipophilic dye, 4-(4-(dihexadecylamino)styryl-N-methylquinolinium iodine) (DiQ), and viral proteins were labeled using fluorescein isothiocyanate (FITC). The redistribution of these probes from the virus to cultured cells was followed using the technique of image correlation spectroscopy. This technique assayed the intensity change and the redistribution of these probes as fusion progressed from a more to less aggregated state. The lipid probe DiQ dispersed into the membrane of the target membrane at both 22 and 37 degrees C, while the FITC-labeled proteins dispersed only at 37 degrees C. Simultaneous labeling of virus with both of these probes showed that at 37 degrees C their redistribution proceeded at different rates. These data were consistent with the formation of a hemifusion intermediate during the fusion process.

Sendai virus binds to a dispersed population of NBD-GD1a.

Receptor aggregation is believed to be an important step in the attachment of membrane enveloped virus' to target cell membranes. A likely receptor for Sendai virus is the ganglioside GD1a. In this work we have studied the membrane diffusion of the fluorescent ganglioside NBD-GD1a on the surface of CV-1 cells with standard photobleaching techniques. Using confocal laser scanning microscopy (CLSM) and Image Correlation Spectroscopy (ICS) NBD-GD1a is shown to exist in at least two populations: dispersed and aggregated. By quantifying the distribution of NBD-GD1a pre- and post-incubation with Sendai virus it is shown that the virus induces a dose-dependent clustering of NBD-GD1a. Image cross-correlation spectroscopy (ICCS) is used to further quantitatively characterize this clustering by demonstrating that it occurs due to binding of virus to the dispersed population of NBD-GD1a.

Free clathrin triskelions are required for the stability of clathrin-associated adaptor protein (AP-2) coated pit nucleation sites.

In this study image correlation spectroscopy was used to demonstrate the presence of two populations of clathrin in situ, on intact cells. In the periphery of the cell approximately 35% of the clathrin triskelions are free within the cytosol while approximately 65% are in large aggregates, presumably coated pits. Although endocytosis is inhibited at low temperature, free clathrin triskelions are still present and small AP-2 aggregates (of approximately 20 proteins), or coated pit nucleation sites, are still observed. Following hypertonic treatment, or cytoplasmic acidification, free clathrin triskelions within the cytosol are depleted and all of the clathrin becomes associated with the membrane. Under these conditions coated pit associated AP-2 remains while the smaller AP-2 aggregates, or coated pit nucleation sites, dissociate. This indicates that the stabilization of AP-2 coated pit nucleation sites requires the presence of free clathrin triskelions within the cytosol. Furthermore, this indicates that free clathrin is required for the early stages of coated pit formation and presumably the continuation of the clathrin-mediated endocytic process. We also provide indirect evidence that AP-2 binding to the membrane in coated pit nucleation sites may be regulated in part by binding to internalization-competent membrane receptors.

SP-B refining of pulmonary surfactant phospholipid films.

Pulmonary surfactant stabilizes the alveoli by lining the air-fluid interface with films that reduce surface tension to near 0 mN/m (gamma(min)). Surfactant protein B (SP-B) enhances the surface activity of surfactant phospholipids. A captive bubble tensiometer (CBT) was used to study the properties of adsorbed films of dipalmitoylphosphatidylcholine (DPPC) with acidic 1-palmitoyl-2-oleoyl-sn-glycero-3-[phospho-rac-(1-glycerol)] (POPG) or neutral 1-palmitoyl-2-oleoyl-sn-glycerol-3-phosphocholine with (7:3) and without 1% dimeric SP-B. SP-B enhanced the adsorption rate of DPPC-containing neutral or acidic lipid suspensions (1 mg/ml) to a similar extent. Quasi-static cycling of these films revealed that SP-B significantly decreased the film area reduction required to reach gamma(min) for the acidic but not for the neutral system. The results obtained with DPPC-phosphatidylglycerol (PG)-SP-B were consistent with selective DPPC adsorption into the surface monolayer during film formation. Film area reduction required to reach gamma(min) with this system (with and without calcium) approached that of pure DPPC, suggesting selective DPPC insertion and PG squeeze-out. Dynamic cycling of such films showed that larger film area reductions were required to reach gamma(min) for the neutral than for acidic system, even after 20 cycles. Fluorescence microscopy of solvent-spread DPPC-POPG-SP-B planar films revealed highly condensed structures at approximately 25 mN/m, although no specific PG phase-segregated structures could be identified. The study suggests that specific interactions of SP-B with acidic phospholipids of surfactant may be involved in the generation and maintenance of DPPC-rich films in the alveoli.

An internalization-competent influenza hemagglutinin mutant causes the redistribution of AP-2 to existing coated pits and is colocalized with AP-2 in clathrin free clusters.

Image correlation spectroscopy and cross correlation spectroscopy were used to demonstrate that approximately 25% of the internalization-competent influenza virus hemagglutinin mutant, HA+8, is colocalized with clathrin and AP-2 at the plasma membrane of intact cells, while wild-type HA (which is excluded from coated pits) does not colocalize with either protein. Clathrin and AP-2 clusters were saturated when HA+8 was overexpressed, and this was accompanied by a redistribution of AP-2 into existing coated pits. However, de novo coated pit formation was not observed. In nontreated cells, the number of clusters of clathrin or AP-2 colocalized with HA+8 was always comparable. Hypertonic treatment which disperses the clathrin lattices resulted in more clusters containing AP-2 and HA+8 than clathrin and HA+8. Less colocalization of HA+8 with clathrin was also observed after cytosol acidification, which causes the formation of deeply invaginated pits, where the HA+8 may be inaccessible to extracellular labeling by antibodies, and blocks coated vesicle budding. However, cytosol acidification elevated the number of clusters containing both HA+8 and AP-2, suggesting an increase in their level of association outside of the deep invaginations. Our results imply that AP-2 and HA+8 can colocalize in clusters devoid of clathrin, at least in cells treated to alter the clathrin lattice structure. Although we cannot ascertain whether this also occurs in untreated cells, we propose that AP-2 binding to membrane proteins carrying internalization signals can occur prior to the binding of AP-2 to clathrin. While such complexes can in principle serve to recruit clathrin for the formation of new coated pits, the higher affinity of the internalization signals for clathrin-associated AP-2 [Rapoport, I., et al. (1997) EMBO J. 16, 2240-2250] makes it more likely that once the AP-2-membrane protein complexes form, they are quickly recruited into existing coated pits.

Correlated atomic force and transmission electron microscopy of nanotubular structures in pulmonary surfactant.

Pulmonary surfactant stabilizes the lung by reducing surface tension at the air-water interface of the alveoli. Surfactant is present in the lung in a number of morphological forms, including tubular myelin (TM). TM is composed of unusual 40 x 40 nm square elongated proteolipid tubes. Atomic force microscopy (AFM) was performed on polymer-embedded Lowicryl and London Resin-White (LR-White) unstained thin sections. AFM was used in imaging regions of the sections where TM was detected by transmission electron microscopy (EM) of corresponding stained sections. Tapping- and contact-mode AFM imaging of the unstained sections containing TM indicated a highly heterogeneous surface topography with height variations ranging from 10 to 100 nm. In tapping-mode AFM, tubular myelin was seen as hemispherical protrusions of 30-70 nm in diameter, with vertical dimensions of 5-8 nm. In contact-mode AFM and with phase imaging using a sharper (>10 nm nominal radius) probe, square open-ended tubes which resembled typical electron micrographs of such regions were observed. The cross-hatch structures observed inside the tubes using EM were not observed using AFM, although certain multilobe structures and topographic heterogeneity were detected inside some tubes. Other regions of multilamellar bodies and some regions where such bilayer lamella appear to fuse with the tubes were found in association with TM using AFM. EM of acetone-delipidated tubes in LR-White revealed rectangular tubular cores containing cross-hatched structures, presumably protein skeletons. AFM surface topography of these regions showed hollow depressions at positions at which the protein was anticipated instead of the protrusions seen in the lipid-containing sections. Gold-labeled antibody to surfactant protein A was found associated somewhat randomly within the regions containing the protein skeletons. The topography of the gold particles was observed as sharp peaks in contact-mode AFM. This study suggests a method for unambiguous detection of three-dimensional nanotubes present in low abundance in a biological macromolecular complex. Only limited detection of proteins and lipids in surfaces of embedded tubular myelin was possible. EM and AFM imaging of such unusual biological structures may suggest unique lipid-protein associations and arrangements in three dimensions.

Image correlation spectroscopy. II. Optimization for ultrasensitive detection of preexisting platelet-derived growth factor-beta receptor oligomers on intact cells.

Previously we introduced image correlation spectroscopy (ICS) as an imaging analog of fluorescence correlation spectroscopy (FCS). Implementation of ICS with image collection via a standard fluorescence confocal microscope and computer-based autocorrelation analysis was shown to facilitate measurements of absolute number densities and determination of changes in aggregation state for fluorescently labeled macromolecules. In the present work we illustrate how to use ICS to quantify the aggregation state of immunolabeled plasma membrane receptors in an intact cellular milieu, taking into account background fluorescence. We introduce methods that enable us to completely remove white noise contributions from autocorrelation measurements for individual images and illustrate how to perform background corrections for autofluorescence and nonspecific fluorescence on cell population means obtained via ICS. The utilization of photon counting confocal imaging with ICS analysis in combination with the background correction techniques outlined enabled us to achieve very low detection limits with standard immunolabeling methods on normal, nontransformed human fibroblasts (AG1523) expressing relatively low numbers of platelet-derived growth factor-beta (PDGF-beta) receptors. Specifically, we determined that the PDGF-beta receptors were preaggregated as tetramers on average with a mean surface density of 2.3 clusters micrometer(-2) after immunolabeling at 4 degreesC. These measurements, which show preclustering of PDGF-beta receptors on the surface of normal human fibroblasts, contradict a fundamental assumption of the ligand-induced dimerization model for signal transduction and provide support for an alternative model that posits signal transduction from within preexisting receptor aggregates.

Fusion of Sendai virus and individual host cells and inhibition of fusion by lipophosphoglycan measured with image correlation spectroscopy.

Fusion between Sendai virus (SV) and individual host cells was investigated with confocal laser scanning microscopy (CLSM) and image correlation spectroscopy (ICS). SV was labeled with the fluorescent probe 7-octadecylamino-4-nitrobenz-2-oxa-1,3-diazole (NBD-NH-C18) and was allowed to bind to host cells (HEp-2, BALB-3T3) at 4 degrees C. The effect of lipophosphoglycan (LPG), isolated from Leishmania donovani, on virus fusion was investigated by incorporation of LPG (0, 5, 10 or 20 microM) into the host cell membrane (HEp-2) before addition of SV. LPG did not affect the number of SV bound per cell. After incubation at 37 degrees C for 15 min without LPG, CLSM revealed a redistribution of NBD-NH-C18 from the SV envelope to the host cell membrane and an increase in average fluorescence intensity, indicating dequenching. ICS analysis of images obtained after incubation at 37 degrees C showed an increased mean cluster density to 260% of the value at 4 degrees C, reflecting the disappearance of labeled SV from the cell surface and diffusion of NBD-NH-C18 into the host cell membrane. Preincubation of the cells with LPG inhibited the temperature-induced redistribution and dequenching of NBD-NH-C18 in a concentration-dependent manner, with a total inhibition of fusion at 20 microM LPG. Together, the results demonstrate that CLSM combined with ICS is a powerful tool for studies of fusion of enveloped viruses with individual host cells and that LPG inhibits the fusion process at or before the hemifusion (lipid mixing) stage of SV interaction with cells.

An image correlation analysis of the distribution of clathrin associated adaptor protein (AP-2) at the plasma membrane.

Clathrin associated adaptor protein is involved in endocytosis at the plasma membrane (AP-2) and protein sorting at the Golgi membrane (AP-1). There is a great deal of information available on the structure, function and binding characteristics of AP-2, however, there is little quantitative data on the AP-2 distribution at the membrane. Image correlation spectroscopy is a technique which yields number counts from an autocorrelation analysis of intensity fluctuations within confocal microscopy images. Image correlation spectroscopy analysis of the indirect immunofluorescence from AP-2 at the plasma membrane of CV-1 cells shows that AP-2 is in a bimodal distribution consisting of large coated pit associated aggregates of approximately 60 AP-2 molecules, and smaller aggregates containing approximately 20 AP-2 molecules, which we propose are coated pit nucleation sites. Following hypertonic treatment 25% of the AP-2 molecules dissociate from the large AP-2 aggregates and form AP-2 dimers, leaving the remaining AP-2 as large aggregates with approximately 45 molecules. The smaller AP-2 aggregates completely dissociate forming AP-2 dimers. Dispersion of AP-2 with hypertonic treatment is not seen qualitatively because the number of large AP-2 aggregates is unchanged, the aggregates are just 25% smaller. Change in temperature from 37 degrees C to 4 degrees C has no affect on the number of AP-2 aggregates or the AP-2 distribution between the two populations. These data and estimates of the coated pit size suggest that coated pits cover approximately 0.9% of the cell membrane. Combination of image correlation spectroscopy analysis and measurements of the CV-1 cell surface area show that there are approximately 6x10(5) AP-2 molecules per CV-1 cell with approximately 2x10(5) AP-2 molecules within coated pit structures.

Diffusion of transferrin receptor clusters.

Two dimensional motion of membrane receptors provides a mechanism for interaction among receptors in the plane of the membrane. In some cases the lateral diffusion leads to formation of clusters which may also be mobile. We have used image cross-correlation (ICCS) spectroscopy technique to measure the translational motion of transferrin receptors in the membrane of 3T3 fibroblasts and HEp2 carcinoma cells. The technique is based on the measurement and analysis of fluctuations in the intensity observed in fluorescence confocal microscope images measured as a function of time. The fluorescence fluctuations arise from stochastic concentration fluctuations about the equilibrium concentration caused by movement of receptors. The amplitude of the fluctuations depend on the number of fluorescent molecules in the observation volume and the dynamics provide the rate of movement. The diffusion observed by this analysis is orders of magnitude slower than that measured by conventional photobleaching techniques. The slower motion corresponds to the diffusion of receptor clusters which provide the more dominant fluctuations.

Analysis of membrane protein cluster densities and sizes in situ by image correlation spectroscopy.

Communication between cells invariably involves interactions of a signalling molecule with a receptor at the surface of the cell. Typically, the receptor is imbedded in the membrane and it is hypothesized that the binding of the signalling molecule causes a change in the state of aggregation of the receptor which, in turn, initiates a biochemical signal within the cell. Subsequently, many of the occupied receptors bind to membrane-associated structures, called coated pits, which invaginate and pinch off to form coated vesicles, thereby removing the receptors from the cell surface. The state of aggregation of membrane receptors is obviously in constant flux. Any useful approach to measuring the state of aggregation must, therefore, allow for dynamic measurements in living cells. It is possible to use fluorescently labelled signalling molecules or antibodies directed at the receptor of interest to visualize the receptor on the cell surface with a fluorescence microscope. By employing a laser confocal microscope, high resolution images can be produced in which the fluorescence intensity is quantitatively imaged as a function of position across the surface of the cell. Calculations of autocorrelation functions of these images provide direct and accurate measures of the density of fluorescent particles on the surface. Combined with the average intensity in the image, which reflects the total average number of molecules, it is possible to estimate the degree of aggregation of the receptor molecules. We refer to this analysis as image correlation spectroscopy (ICS). We show how ICS can be used to measure the density of several receptors on a variety of cells and how it can be used to measure the density of coated pits and the number of molecules per coated pit. We also show how the technique can be used to monitor fusion of virus particles to cell membranes. Further, we illustrate that, by calculating cross-correlation functions between pairs of images, we can extend the analysis to measurements of the distributions as a function of time, on the second timescale, as well as to measurements of the movement of the receptor aggregates on the surface. Finally, we illustrate that, by this approach, we can measure the extent of interaction between two different receptors as a function of time. This represents the most quantitative measurement of the extent of co-localization of receptors available and is independent of the spatial resolution of the confocal microscope. The theory of ICS and image cross-correlation spectroscopy (ICCS), focussing on the interpretation of the data in terms of the biological phenomenon being probed, is discussed.

Partitioning of proteins into plasma membrane microdomains. Clustering of mutant influenza virus hemagglutinins into coated pits depends on the strength of the internalization signal.

Internalization of membrane proteins involves their recruitment into plasma membrane clathrin-coated pits, with which they are thought to interact by binding to AP-2 adaptor protein complexes. To investigate the interactions of membrane proteins with coated pits at the cell surface, we applied image correlation spectroscopy to measure directly and quantitatively the clustering of influenza hemagglutinin (HA) protein mutants carrying specific cytoplasmic internalization signals. The HA system enables direct comparison between isolated internalization signals, because HA itself is excluded from coated pits. The studies presented here provide, for the first time, a direct quantitative measure for the degree of clustering of membrane proteins in coated pits at the cell surface. The degree of clustering depended on the strength of the internalization signal and on the integrity of the clathrin lattices and correlated with the internalization rates of the mutants. The clustering of the HA mutants fully correlated with their ability to co-precipitate alpha-adaptin from whole cells, the first such demonstration for a membrane protein that is not a member of the epidermal growth factor receptor family. Furthermore, both the clustering in coated pits and the co-precipitation with alpha-adaptin were dramatically reduced in the cold, suggesting that low temperature can interfere with the sorting of proteins into coated pits. In addition to the specific results reported here, the general applicability of the image correlation spectroscopy approach to study any process involving the clustering or oligomerization of membrane receptors at the cell surface is discussed.

Effects of size of macrocyclic polyamides on their rate of diffusion in model membranes.

A series of homologous amphiphilic molecules with surface areas in the range of 0.3 nm2 to 3.0 nm2 were prepared and used to investigate the diffusion in model dimyristoylphosphatidylcholine membranes as a function of temperature. The diffusion behavior of smaller molecules can be described by the interfacial viscosity limited free area theory promoted by Vaz and his co-workers, and that of the larger molecules can best be modeled by a recent interpretation of the theoretical description proposed by Evans and Sackmann. The experimental data show that the rate of diffusion is controlled by the size of the molecules at the interface of the lipid membrane, and provide evidence for a view of the membrane as a hydrodynamic triple layer with a low-viscosity central layer encased by two more viscous, yet fluid, layers.

Aggregation of PDGF-beta receptors in human skin fibroblasts: characterization by image correlation spectroscopy (ICS).

Receptor aggregation is believed to be an important, early step when growth factors such as PDGF stimulate proliferation and differentiation of cell populations. To investigate receptor aggregation, we utilized a novel biophysical technique, image correlation spectroscopy, to study the distribution and aggregation state of PDGF-beta receptors on the surface of human dermal fibroblasts under various experimental conditions. It was found that the cell surface receptors were pre-clustered at 4 degrees C and receptor aggregation increased for samples measured at 37 degrees C. Treatment with PDGF-BB had no measurable effect on the receptor aggregation state. The results also indicate that additions of 10% serum or an inhibitor of tyrosine kinase activity, may disperse the receptors. The results of this study are consistent with organization of PDGF-beta receptors in pre-existing membrane domains.

Quantitation of membrane receptor distributions by image correlation spectroscopy: concept and application.

Measurement of receptor distributions on cell surfaces is one important aspect of understanding the mechanism whereby receptors function. In recent years, scanning fluorescence correlation spectroscopy has emerged as an excellent tool for making quantitative measurements of cluster sizes and densities. However, the measurements are slow and usually require fixed preparations. Moreover, while the precision is good, the accuracy is limited by the relatively small amount of information in each measurement, such that many are required. Here we present a novel extension of the scanning correlation spectroscopy that solves a number of the present problems. The new technique, which we call image correlation spectroscopy, is based on quantitative analysis of confocal scanning laser microscopy images. Since these can be generated in a matter of a second or so, the measurements become more rapid. The image is collected over a large cell area so that more sampling is done, improving the accuracy. The sacrifice is a lower resolution in the sampling, which leads to a lower precision. This compromise of precision in favor of speed and accuracy still provides an enormous advantage for image correlation spectroscopy over scanning correlation spectroscopy. The present work demonstrates the underlying theory, showing how the principles can be applied to measurements on standard fluorescent beads and changes in distribution of receptors for platelet-derived growth factor on human foreskin fibroblasts.

Lateral diffusion in model membranes is independent of the size of the hydrophobic region of molecules.

We have systematically investigated the probe size and shape dependence of lateral diffusion in model dimyristoyl phosphatidylcholine membranes. Linear hydrophobic polymers, which differ in length by an order of magnitude, were used to explore the effect on the lateral diffusion coefficient of hydrodynamic restrictions in the bilayer interior. The polymers employed are isoprenoid alcohols--citronellol, solanesol, and dolichol. Tracer lateral diffusion coefficients were measured by fluorescence photobleaching recovery. Despite the large difference in lengths, the nitrobenzoxadiazole labelled alcohols all diffuse at the rate of lipid self-diffusion (5.0 x 10(-12) m2 s-1, 29 degrees C) in the liquid crystal phase. Companion measurements in isotropic polymer solution, in gel phase lipid membranes and with nonpolar fluorescent polyaromatic hydrocarbons, show a marked dependence of the lateral diffusion coefficient on the probe molecule size. Our results in the liquid crystal phase are in accord with free area theory which asserts that lateral diffusion in the membrane is restricted by the surface-free area. Probe molecules which are significantly longer than the host phospholipid, seven times longer in the case of dolichol, are still restricted in their lateral motion by the surface properties of the bilayer in the liquid crystal phase. Fluorescence quenching experiments indicate that the nitrobenzoxadiazole label does not reside at the aqueous interface, although it must reside in close proximity according to the diffusion measurements.

Average density and size of microclusters of epidermal growth factor receptors on A431 cells.

For some hormone receptors, the early events of signal transduction depend on their molecular arrangement and interactions at the cell surface. An understanding of the mechanism of signal transduction in general needs a careful analysis of the receptor distribution. Here, we present the first quantitative measurement of epidermal growth factor receptor distribution on A431 cells obtained by scanning fluorescence correlation spectroscopy. Prior to epidermal growth factor binding, the A431 cell membrane presents an average surface density of 7.7-8.4 microclusters/microns 2, each containing an average of 130 receptors.

Preparation of novel cyclosporin A derivatives.

The hydroxyl group on the 2-N-methyl-(R)-((E)-2-butenyl)-4-methyl-L-threonine residue of cyclosporin A was protected by acetylation, then the double bond on the same amino acid residue was oxidatively cleaved using a periodate/permanganate reagent. The resultant derivative of cyclosporin A contained a carboxylic acid group which was subsequently reacted with the nucleophiles 5-(aminoacetamido)fluorescein and poly(L-lysine), in the presence of 1-ethyl-3-[3-(dimethylamino)propyl] carbodiimide, to furnish novel cyclosporin A conjugates.