Rapid transport of internalized P-selectin to late endosomes and the TGN: roles in regulating cell surface expression and recycling to secretory granules

Prior studies on receptor recycling through late endosomes and the TGN have suggested that such traffic may be largely limited to specialized proteins that reside in these organelles. We present evidence that efficient recycling along this pathway is functionally important for nonresident proteins. P-selectin, a transmembrane cell adhesion protein involved in inflammation, is sorted from recycling cell surface receptors (e.g., low density lipoprotein [LDL] receptor) in endosomes, and is transported from the cell surface to the TGN with a half-time of 20-25 min, six to seven times faster than LDL receptor. Native P-selectin colocalizes with LDL, which is efficiently transported to lysosomes, for 20 min after internalization, but a deletion mutant deficient in endosomal sorting activity rapidly separates from the LDL pathway. Thus, P-selectin is sorted from LDL receptor in early endosomes, driving P-selectin rapidly into late endosomes. P-selectin then recycles to the TGN as efficiently as other receptors. Thus, the primary effect of early endosomal sorting of P-selectin is its rapid delivery to the TGN, with rapid turnover in lysosomes a secondary effect of frequent passage through late endosomes. This endosomal sorting event provides a mechanism for efficiently recycling secretory granule membrane proteins and, more generally, for downregulating cell surface receptors.     

Lysosome inhibitors enhance the ability of cyclic AMP-elevating agents to induce the LDL receptor in human vascular smooth muscle cells.

In human vascular smooth muscle cells cyclic AMP elevation by forskolin increases synthesis of the LDL receptor by a mechanism which appears independent of sterol control. This increased receptor synthesis is further enhanced by chloroquine. Both forskolin and prostaglandin E1 increase the number of cell surface LDL receptors indicating that prostaglandins could exert physiological control over LDL metabolism. This effect is enhanced synergistically by chloroquine. The stimulation by forskolin of LDL receptor synthesis and expression leads to increased metabolism of apo-B and increased hydrolysis of LDL-borne cholesteryl ester. These effects of cyclic AMP on the activity of the LDL pathway are enhanced more than additively by preincubation with the reversible lysosomal inhibitor NH4Cl. Thus cyclic AMP causes up-regulation of the LDL receptor pathway resulting in increased rates of LDL metabolism but this effect can be damped or masked in cell culture by a cyclic AMP-sensitive lysosomal event, probably the acute stimulation of lysosomal cholesterol ester hydrolase.     

Kinetics of insulin receptor internalization and recycling in adipocytes. Shunting of receptors to a degradative pathway by inhibitors of recycling

The kinetics of receptor internalization and recycling was directly determined in adipocytes by measuring 125I-insulin binding to total, intracellular, and cell-surface insulin receptors. In the absence of insulin 90% of all receptors were on the cell-surface and 10% were intracellular. Insulin (100 ng/ml) rapidly altered this distribution by translocating surface receptors to the cell-interior through a temperature and energy dependent process. Surface-derived receptors were seen within cells as early as 30 s and accumulated intracellularly at the rate of approximately 20,000/min (t 1/2 = 2.7 min). After 6 min the size of the intracellular receptor pool plateaued (for up to 2 h), with 30% of surface receptors residing within the cell. This plateau was due to the attainment of an equilibrium between receptor uptake and recycling, since removal of insulin (to stop receptor uptake) was followed by both a rapid depletion of intracellular receptors and a a concomitant and stoichiometric reappearance of receptors on the cell-surface. Receptors were efficiently recycled, with little or no net loss observed even after 4 h of insulin treatment; however, recycling could be partially inhibited (approximately 10%) by several agents (e.g. chloroquine and Tris). Tris treatment of adipocytes in the presence of insulin led to 50% loss of surface and total receptors at 2 and 4 h, respectively. Since chloroquine prevented the decrease in total receptors, but not the loss of surface receptors, it appears that Tris impairs recycling by diverting a portion of incoming receptors to a chloroquine-inhibitable degradative site. From these results we conclude that: 1) insulin triggers endocytotic uptake of insulin-receptor complexes; 2) internalized receptors are then rapidly reinserted into the plasma membrane, and the receptors can traverse this recycling pathway within 6 min; 3) prolonged recycling does not normally result in measurable receptor loss, but when receptors are prevented from recycling, they become trapped intracellularly and are shunted to a chloroquine-sensitive degradative pathway; and 4) chloroquine and Tris are only partially effective inhibitors of receptor recycling.     

Modifications of LDL-receptor-mediated endocytosis rates in CEM lymphoblastic cells grown in lipoprotein-depleted fetal calf serum

The efficiency of supplying cholesterol by the LDL endocytic pathway of lymphoblastic T CEM cells was compared when incubated in the presence of either fetal calf serum (FCS) or lipoprotein-depleted fetal calf serum (LDFCS). In the presence of FCS, there were 8600 +/- 2000 LDL receptors/cell with a Kd of (2.2 +/- 0.8).10(-8) M and a receptor cycling time of about 7 min; about 90% of the internalized LDL was degraded. LDL degradation produced 98% of total cellular cholesterol and only 2% came from endogenous synthesis. The absence of LDL in the culture medium of lymphoblastic CEM cells deeply modified certain metabolic and structural characteristics of the cells. Their cholesterol content decreased; the total number of LDL receptors increased 6-fold, whereas their affinity for the ligand decreased by the same factor (Kd = (1.2 +/- 0.2).10(-7) M); the receptor cycling time increased 3-fold. Finally, LDL degraded by cholesterol-depleted CEM cells amounted to about 40% of that degraded by untreated CEM cells.     

Sinusoidal endothelial endocytosis of low density lipoprotein-gold conjugates in perfused livers of ethinyl-estradiol treated rats

We examined endocytosis of low-density lipoproteins (LDL) conjugated to colloidal gold by the sinusoidal endothelium in perfused livers of 17 alpha-ethinyl estradiol-treated rats. After 15 min of perfusion, the gold-LDL was randomly bound at the endothelial surface, but internalized only at coated pits. Uptake of the conjugate was to electron-lucid vacuoles. After 1 h of perfusion, there was a progressive accumulation of gold in organelles resembling lysosomes, with further accumulation seen at 2 h perfusion. Uptake of the conjugate was equivalent to the rate of 125I-LDL and competitively inhibited by a 20-fold excess of free LDL. These results suggest that a specific endocytotic pathway for LDL is present in the sinusoidal endothelium in the estrogen-hypolipidemic state.     

Low density lipoprotein receptor binding in aging human diploid fibroblasts in culture

High affinity cell surface receptors for low density lipoproteins (LDL) are inducible in cultured human lung fibroblasts by the removal of lipoproteins from the cell culture medium. The binding, uptake, and degradation of 125I-LDL by fibroblasts decrease with increasing number of population doublings. The affinity of LDL receptor binding, however, remained unchanged at different population doublings levels. Hence, the difference in LDL binding activity in the aging fibroblasts can be attributed to a reduction in the number of receptor sites on the cell membrane. Cellular uptake of [4-14C]cholesterol and 2-deoxy-D-[1-14C]glucose mediated through mechanisms independent of the LDL receptor pathway revealed no significant difference in early and late passage fibroblasts. This suggests that the alteration in the LDL receptor binding in serially passaged fibroblasts is an "age-related" phenomenon. The late population doubling fibroblasts require more LDL in the culture medium for feedback inhibition of LDL receptor synthesis. Thus, aging fibroblasts are both progressively less inducible and less suppressible in the regulation of their cell membrane LDL receptors. Similar results were also obtained with respect to the regulation of DL-3-hydroxy-3-methyl-glutaryl coenzyme A reductase in the aging fibroblasts in culture; the enzyme has become less inducible and less supressible as the fibroblasts approach the limit of their in vitro lifespan. These age-related alterations in the cellular metabolism of LDL and cholesterol might contribute to our understanding of the increased risk of athlerosclerosis in our aging population.     

Endocytosis of cationized ferritin in human resting peripheral blood by T-lymphocytes

Summary We have examined the binding and internalization of cationized ferritin in T-lymphocytes of human peripheral blood, as a model for resting cells. After 30 min of incubation only 8% of endocytotic vesicles contain cationized ferritin. T-cells internalie the equivalent of their entire surface area in approximately 54 h, a longer time than is required by non-resting cells such as PHA-stimulated human lymphocytes. These tracer experiments suggest that the endocytosis of cationized ferritin by T-lymphocytes follows a lysosome pathway similar to that described for other cell types.     

Heterozygous familial hypercholesterolemia: failure of normal allele to compensate for mutant allele at a regulated genetic locus.

In normal human fibroblasts, the synthesis of a cell surface receptor for plasma low density lipoprotein (LDL) is regulated by a sensitive system of feedback suppression. The number of functional LDL receptors declines by more than 20 fold when cellular stores of esterified cholesterol are increased by incubation of cells with an exogenous source of cholesterol. Fibroblasts from patients with the heterozygous form of familial hypercholesterolemia (FH) possess one functional allele and one nonfunctional allele at the LDL receptor locus. In the current studies, we have examined the effect that this deficiency produces upon the pattern of regulation of the single functional allele at the LDL receptor locus. Under growth conditions that induced a maximal rate of LDL receptor synthesis (that is, growth in the absence of an exogenous source of cholesterol), the FH heterozygote cells produced about one half as many functional LDL receptors as did the normal cells. More importantly, when grown in the presence of increasing amounts of exogenous cholesterol, the FH heterozygote and normal cells suppressed their respective LDL receptor activities in parallel. Over a wide range of LDL receptor activities, at each level of cellular esterified cholesterol, the FH heterozygote cells expressed about one half as many receptors as did the normal cells. These data indicate that in the FH heterozygote cells, the receptor regulatory mechanism dictates that the normal allele produce only the amount of gene product that it would normally produce at a given level of cellular esterified cholesterol. The failure of the regulatory mechanism to stimulate the normal allele at the LDL receptor locus to produce twice its normal amount of gene product leaves the FH heterozygote cells with a persistent 50% deficiency in LDL receptors under all conditions of cell growth.     

Intracellular pathway followed by the insulin receptor covalently coupled to 125I-photoreactive insulin during internalization and recycling

After it interacts with a specific receptor on the cell surface, insulin is internalized in its target cell by an adsorptive endocytotic process and eventually degraded in lysosomes. It was also recently shown that the initial surface interaction between the hormone and its receptor is followed by an internalization of the receptor, which later is recycled back to the cell surface. In the present study the insulin receptor was tagged with a 125I-photoreactive insulin analogue that can be covalently coupled to the insulin receptor by ultraviolet irradiation. Using this tool we could trace by quantitative electron microscope autoradiography the intracellular pathway followed by this labeled receptor. The quantitative analysis of the intracellular distribution of the labeled material as a function of incubation time at 37 degrees C supports the following sequence of events: association first with clear vesicles, second with multivesicular bodies, third with dense bodies, and fourth, a return to the cell surface via clear vesicles. This insulin receptor recycling process is inhibited by monensin but unaffected by cycloheximide.     

Oncostatin M up-regulates low density lipoprotein receptors in HepG2 cells by a novel mechanism.

Oncostatin M is a growth regulatory protein secreted by macrophages and activated T lymphocytes. In a hepatoma cell line (HepG2) the polypeptide very potently increased low density lipoprotein (LDL) uptake with an EC50 of 0.1-0.2 nM. The stimulation of LDL uptake was detectable by 2 h, was maximal by 8 h, and remained elevated through 20 h of oncostatin M incubation. In a similar fashion, oncostatin M also increased the number of cell surface LDL receptors by a mechanism that was inhibited by cycloheximide or the protein kinase C inhibitor H-7. Oncostatin M stimulation of LDL uptake and receptor protein occurred regardless of the state of cholesterol-dependent regulation of HepG2 LDL receptor (i.e. cells incubated in medium containing lipoproteins responded to the same extent as did cells incubated in the absence of lipoproteins). No significant effects were observed on sterol synthesis over 8 h or on DNA synthesis over 24 h. Oncostatin M induced rapid alterations in HepG2 phospholipid metabolism. Within 5-15 min there was a 20-50% increase in incorporation of 32P into several classes of phospholipids, including the phosphoinositides. Radiolabeled diacylglycerol levels were elevated 20% by 2 min and nearly 50% by 15 min. In addition, the polypeptide induced rapid increased (within 1 min) in phosphorylation of HepG proteins on tyrosine residues. Stimulation of both phosphotyrosine and LDL receptor up-regulation by oncostatin M was decreased by the tyrosine kinase inhibitor genistein. We propose that oncostatin M up-regulates HepG2 LDL receptor expression by a mechanism that includes stimulation of a tyrosine kinase followed by generation of phospholipid-related second messengers.     

Phorbol esters inhibit the binding of low-density lipoproteins (LDL) to U-937 monocytelike cells

The present study demonstrates that U-937 monocytelike human cells possess specific LDL receptors. 125I-LDL binds at 4 degrees C on the cell surface. The bound molecules are releasable by heparin. The reaction requires Ca2+ and the binding sites are sensitive to proteolysis. Unlabeled LDL compete with 125I-LDL, whereas HDL are ineffective. At 37 degrees C, LDL are internalized and degraded by a chloroquine-sensitive pathway. Tumor-promoting phorbol esters inhibit the binding of 125I-LDL to its receptor on U-937 cells. This inhibition exhibits temperature, time, and concentration dependence. At 37 degrees C, inhibition is 50% at 5 X 10(-9) M of TPA. After removal of phorbol esters, treated cells recover their 125I-LDL-binding activity in 60 min. The inhibitory activities of various phorbol esters are proportional to their tumor-promoting activities. Inhibition appears to be due to a reduction in the number of available LDL receptors rather than a decrease in receptor affinity.     

LDL receptor binds newly synthesized apoE in macrophages. A precursor pool for apoe secretion.

There appear to be multiple post-translational sites for regulation of macrophage apolipoprotein (apo)E secretion, including the presence of a distinct cell surface pool of apoE. Cell surface proteoglycans have been shown to be involved in forming this pool. The current studies were designed to investigate the role of an additional cell surface site, i.e., the low density lipoprotein (LDL) receptor. Antiserum to the LDL receptor displaced apoE from the macrophage cell surface and into the medium during a 4 degrees C incubation from apoE-expressing J774 cells, from proteoglycan-depleted apoE-expressing J774 cells, and from human monocyte-derived macrophages. Similar results were obtained when purified monoclonal antibody to the LDL receptor was added to human monocyte-derived macrophages. J774 cells transfected to express an LDL receptor binding-defective mutant of apoE did not show a similar response to addition of LDL receptor antibody. Studies were conducted in which cells were pulse labeled for 30 min, followed by various periods of chase at 4 degrees C or 37 degrees C in the presence or absence of LDL receptor antibody. The results of these studies indicated that nascent macrophage-derived apoE binds to the LDL receptor, and that this apoE served as a precursor pool for apoE released into the medium.These studies establish a role for the LDL receptor in forming the cell surface pool of apoE and, along with data regarding the importance of proteoglycans, indicate that cell surface binding sites for nascent macrophage-derived apoE are heterogeneous. The heterogeneity of such sites could have implication for the size and turnover of this cell surface pool.     

Targeting of tumor cells by low density lipoproteins: principle and use of ellipticin derivatives]

Cells acquire cholesterol via de novo synthesis and high affinity receptor-mediated uptake of low-density lipoprotein (LDL). Some tumor tissues display increased receptor-mediated uptake of LDL as compared with the corresponding normal tissues. This increased LDL receptor activity is unexplained: a high cholesterol demand for cell growth or a mechanism directly linked to cell transformation. LDL has therefore been proposed as a potential carrier for chemotherapeutic agents. Various methods have been used to incorporate antineoplastic lipophilic drugs into LDL. The resultant drug-LDL complexes have been shown to be cytotoxic towards tumor cells in vitro, via the LDL receptor dependent pathway. However little is now on the in vivo fate of this complex. We described the incorporation of lipophilic derivatives of ellipticine into LDL by a fusion or facilitated transfer technique between drug containing microemulsions and LDL. The drug-LDL complex expressed similar metabolic activity, in vitro and in vivo, than native LDL. Initial experiments with melanoma B16 tumor-bearing mice suggest that LDL may be a potential drug carrier in the treatment of malignant diseases. The knowledge of the molecular mechanism of the expression of the LDL receptor in tumor cells and the ability to downregulate the LDL receptor in the normal tissues, will define the application field of this targeting approach.     

The apolipoprotein E-dependent low density lipoprotein cholesteryl ester selective uptake pathway in murine adrenocortical cells involves chondroitin sulfate proteoglycans and an alpha 2-macroglobulin receptor

Cells acquire lipoprotein cholesterol by receptor-mediated endocytosis and selective uptake pathways. In the latter case, lipoprotein cholesteryl ester (CE) is transferred to the plasma membrane without endocytosis and degradation of the lipoprotein particle. Previous studies with Y1/E/tet/2/3 murine adrenocortical cells that were engineered to express apolipoprotein (apo) E demonstrated that apoE expression enhances low density lipoprotein (LDL) CE uptake by both selective and endocytic pathways. The present experiments test the hypothesis that apoE-dependent LDL CE selective uptake is mediated by scavenger receptor, class B, type I (SR-BI). Surprisingly, SR-BI expression was not detected in the Y1/E/tet/2/3 clone of Y1 adrenocortical cells, indicating the presence of a distinct apoE-dependent pathway for LDL CE selective uptake. ApoE-dependent LDL CE selective uptake in Y1/E/tet/2/3 cells was inhibited by receptor-associated protein and by activated alpha(2)-macroglobulin (alpha(2)M), suggesting the participation of the LDL receptor-related protein/alpha(2)M receptor. Reagents that inhibited proteoglycan synthesis or removed cell surface chondroitin sulfate proteoglycan completely blocked apoE-dependent LDL CE selective uptake. None of these reagents inhibited SR-BI-mediated LDL CE selective uptake in the Y1-BS1 clone of Y1 cells in which LDL CE selective uptake is mediated by SR-BI. We conclude that LDL CE selective uptake in adrenocortical cells occurs via SR-BI-independent and SR-BI-dependent pathways. The SR-BI-independent pathway is an apoE-dependent process that involves both chondroitin sulfate proteoglycans and an alpha(2)M receptor.     

Effect of preferential insertion of LDL receptors near coated pits

Recent experiments suggest that low density lipoprotein (LDL) receptors on human fibroblasts are not inserted into the plasma membrane uniformly, as earlier experiments indicated, but are inserted into specialized regions, called plaques, where coated pits form. If the consequent reduction in the time required for LDL receptors to diffuse to coated pits were significant, this could alter conclusions drawn from previous calculations based on the assumption that LDL receptors are inserted uniformly. In particular, the conclusion could be wrong that diffusion of LDL receptors to coated pits is the rate limiting step in the interaction of cell surface LDL receptors with coated pits. Here we calculate the extent of the reduction in mean travel time of an LDL receptor to a coated pit, as a function of the plaque radius. We find that only if LDL receptor insertion is limited to a very small portion of the plasma membrane near coated pit sites is there a substantial decrease in the average time it would take an LDL receptor to diffuse to a coated pit. In order for preferential insertion of LDL receptors into plaques to cut the mean receptor travel time in half, plaques would have to take up no more than 10% of the cell surface area; to reduce the travel time by a factor of 10 plaques would have to cover only 2% of the cell surface, approximately twice the area covered by coated pits at 37°C.     

Regulation of LDL receptor, apoB, and apoE protein and mRNA in Hep G2 cells.

The regulation of low-density lipoprotein (LDL) receptor activity, protein synthesis, and cellular mRNA content was evaluated in the human hepatoma cell line Hep G2. Incubation of the cells with LDL led to a complete downregulation of LDL receptor mRNA and LDL receptor protein synthesis. This LDL regulation of the LDL receptor and its mRNA was both time- and concentration-dependent. In contrast to protein synthesis and cellular mRNA concentrations of the LDL receptor, which were reduced to undetectable levels by prolonged incubation in the presence of LDL, LDL receptor activity was reduced to only 44% of preincubation levels. These findings support the presence of a second metabolic pathway for LDL uptake in human hepatocytic cells. The effect of LDL on cellular LDL receptor expression was specific for LDL because incubation in the presence of HDL did not affect any of these study end points. The potential coordinate regulation of the expression of the LDL receptor with its principal ligands, apolipoproteins (apo) B and E, was also investigated. In contrast to the LDL receptor mRNA downregulation with LDL incubation, cellular apoB and apoE mRNA concentrations were not affected by either LDL or HDL. Secretion of apoB, however, was significantly increased by incubating Hep G2 cells with LDL. These findings indicate that, in contrast to LDL receptor which is regulated at the mRNA level, the ligands for the LDL receptor are regulated either co- or post-translationally.     

Cholesterol metabolism in pigeon aortic smooth muscle cells lacking a functional low density lipoprotein receptor pathway

Cholesterol metabolism was examined in aortic smooth muscle cells from atherosclerosis-susceptible White Carneau pigeons that have been shown to lack a functional LDL receptor pathway. In cells incubated in the presence of whole serum or low density lipoprotein (LDL) the rate of cholesterol synthesis from [1-14C]acetate or of HMG-CoA reductase activity was 20-100 times greater than for mammalian cells incubated under the same conditions. Unexpectedly, cholesterol synthesis decreased by nearly 50% after preincubation for 24 hr with lipoprotein-deficient serum (LPDS). This occurred without a change in cellular cholesterol content. Neither the high rate of cholesterol synthesis nor the effect of LPDS could be accounted for by differences in cell turnover or state of growth. Cholesterol added in ethanol was ineffective in altering cellular cholesterol synthesis or esterification even though a near doubling in cellular free cholesterol content occurred. Cholesterol synthesis and esterification were, however, able to be regulated with 25-OH cholesterol and mevalonolactone, as indicated by their ability to suppress cholesterol synthesis and to stimulate cholesterol esterification. In spite of the high rate of endogenous cholesterol synthesis, cellular cholesterol content was maintained at a constant level by the efficient efflux of the newly synthesized cholesterol from the cell. Unlike mammalian cells that require a cholesterol acceptor in the medium for efflux to occur, cholesterol efflux from pigeon cells occurred in the absence of a cholesterol acceptor. This suggests either that pigeon cells utilize a different mechanism for cholesterol efflux or that they produce their own cholesterol acceptor. As a result of a lack of a functional LDL receptor pathway, pigeon smooth muscle cells do not maintain cholesterol homeostasis through the controlled uptake of exogenous LDL cholesterol, as do mammalian cells. Rather, pigeon smooth muscle cells would appear to regulate cholesterol concentrations at the level of either cholesterol synthesis or efflux.     

Monensin interrupts the recycling of low density lipoprotein receptors in human fibroblasts

In cultured human fibroblasts, each LDL receptor mediates the internalization of approximately 100 particles of LDL every 20 hr. We provide evidence that this reutilization of LDL receptors involves the recycling of receptors into and out of the cell and that the carboxylic ionophore monensin blocks the return of the receptors to the surface. In the presence of monensin and LDL, 75% of the receptors disappeared from the cell surface within 15 min and more than 90% disappeared within 60 min. The receptors that left the surface were trapped intracellularly within perinuclear vacuoles, as visualized by indirect immunofluorescence with the use of LDL, monensin caused about 50% of the receptors to be trapped intracellularly within 15 min. The receptors that remained on the surface after monensin treatment could be trapped within the cell if LDL was added subsequently in the continued presence of monensin. Monensin did not decrease surface LDL receptors in fibroblasts from a patient (J.D.) with the internalization-defective form of familial hypercholesterolemia. In these mutant cells, LDL receptors are not localized to coated pits. The current data are interpreted to indicate that: in normal fibroblasts about 50% of surface LDL receptors absence of LDL; the remaining 50% of surface receptors can be induced to recycle by the presence of LDL; and monensin interrupts this recycling by preventing the receptor from returning to the surface, thereby causing the receptors to accumulate within the cell.     

Immunoprecipitation of the low-density-lipoprotein (LDL) receptor and its precursor from human monocyte-derived macrophages.

Synthesis of the low-density-lipoprotein (LDL) receptor protein by cultured human monocyte-derived macrophages was demonstrated by immunoprecipitation of [35S]methionine-labelled cell extracts with a monoclonal antibody to the bovine adrenal LDL receptor. Although the antibody does not bind to or inhibit binding of 125I-LDL to the LDL receptor on intact fibroblasts, it specifically binds to a protein in extracts of human skin fibroblasts, of Mr approx. 130,000 under non-reducing conditions, that is able to bind LDL. In monocyte-derived macrophages, as in fibroblasts, the receptor is synthesized as a low-Mr precursor that is converted into the mature protein. The half-life of the precursor in human macrophages is approx. 44 min. In cells from two homozygous familial-hypercholesterolaemic subjects, only the precursor form of the receptor is synthesized. Detection of abnormalities of LDL-receptor synthesis in human mononuclear cells may be a useful aid in diagnosis of familial hypercholesterolaemia that is simpler and quicker than methods requiring growth of cultured skin fibroblasts.