VEGF causes pulmonary hemorrhage, hemosiderosis, and air space enlargement in neonatal mice.

To determine whether increased levels of VEGF disrupt postnatal lung formation or function, conditional transgenic mice in which VEGF 164 expression was enhanced in respiratory epithelial cells were produced. VEGF expression was induced in the lungs of VEGF transgenic pups with doxycycline from postnatal day 1 through 2 and 6 wk of age. VEGF levels were higher in bronchoalveolar lavage fluid (BALF) and lung homogenates of VEGF transgenic mice compared with endogenous VEGF levels in controls. Neonatal mortality was increased by 50% in VEGF transgenic mice. Total protein content in BALF was elevated in VEGF transgenic mice. Surfactant protein B protein expression was unaltered in VEGF transgenic mice. Although postnatal alveolar and vascular development were not disrupted by VEGF expression, VEGF transgenic mice developed pulmonary hemorrhage, alveolar remodeling, and macrophage accumulation as early as 2 wk of age. Electron microscopy demonstrated abnormal alveolar capillary endothelium in the VEGF transgenic mice. In many locations, the endothelium was discontinuous with segments of attenuated endothelial cells. Large numbers of hemosiderin-laden macrophages and varying degrees of emphysema were observed in adult VEGF transgenic mice. Overexpression of VEGF in the neonatal lung increased infant mortality and caused pulmonary hemorrhage, hemosiderosis, alveolar remodeling, and inflammation.

In vitro model for developmental progression from vasculogenesis to angiogenesis with a murine endothelial precursor cell line, MFLM-4.

In the embryo, vascular networks are developed through both vasculogenesis, the assembly of vessels from endothelial progenitor cells or hemangioblasts, and angiogenesis, the sprouting of vessels from preexisting capillaries. Cell culture models using endothelial cells (EC) and various extracellular matrix components have been useful in understanding the cellular and molecular factors involved in angiogenesis. However, there are few models of vasculogenesis. Using a murine endothelial precursor cell line, MFLM-4, derived from e14.5 lung mesenchyme, we have developed a culture system that not only recapitulates many of the characteristics of vasculogenesis but also progresses into angiogenesis. By 8 h, MFLM-4 cultured on the basement membrane preparation Matrigel invade the matrix, coalesce, and assemble into large clusters of cells resembling blood islands. During vascular development, blood islands are the focal areas for coalescence of endothelial precursors. For MFLM-4, this phase of in vitro vasculogenic clustering does not require proliferation. If proliferation is not blocked, MFLM-4 progresses into an angiogenic phase with the clusters forming multicell angiogenic sprouts. Through 3 days of culture, lumens form within the clusters, adjacent clusters are connected with tube-like structures, and eventually an extensive network or plexus of clusters connected by capillary-like tubes is formed. MFLM-4 cultured on Matrigel provides an in vitro system for analysis of the multistage, concurrent processes of vasculogenesis and angiogenesis.

Embryonic vasculogenesis by endothelial precursor cells derived from lung mesenchyme.

During development, the lung mesenchyme has a dynamic relationship with the branching airway. Embryonic lung mesenchyme is loosely packed and composed of indistinguishable cells, yet it is the source of vascular progenitors that will become endothelial cells, smooth muscle cells and fibroblasts. In the lung, vessel development in the periphery proceeds first through vasculogenesis, the migration and assembly of cells into a primitive network, and subsequently, through angiogenesis, the sprouting of vessels from this network. As a way to assess the cellular and molecular mechanisms of lung vascularization, we have isolated and cloned cell lines from mouse fetal lung mesenchyme (MFLM). Two of these MFLM cell lines, MFLM-4 and MFLM-91U, display characteristics of an endothelial lineage. RNA analysis demonstrates transcripts for the vascular endothelial growth factor receptors R1 and R2, the receptor tyrosine kinases, Tie-1 and Tie-2, as well as the Tie-2 ligands, Ang-1 and -2. The MFLM cell lines form extensive networks of capillary-like structures with lumens when cultured on a reconstituted basement membrane. In vivo, following blastocyst injection, the MFLM cells chimerize endothelium of the lung and areas of the heart vasculature. The results from these studies suggest that MFLM-4 and MFLM-91U, derived from embryonic lung mesenchyme, can function in vitro and in vivo as endothelial precursors and as models of cardiopulmonary vascularization. Dev Dyn 2000;217:11-23.

AF12198, a novel low molecular weight antagonist, selectively binds the human type I interleukin (IL)-1 receptor and blocks in vivo responses to IL-1.

Interleukin-1 (IL-1) -alpha and -beta are potent regulators of inflammatory responses. The naturally occurring interleukin-1 receptor antagonist (IL-1ra) is effective in vitro and in vivo in modulating biological responses to IL-1. We have previously reported the discovery of IL-1 antagonist peptides from the search of phage display libraries. Further characterization of this group of peptides has led to a 15-mer, AF12198, Ac-FEWTPGWYQJYALPL-NH2 (J represents the unnatural amino acid, 2-azetidine-1-carboxylic acid), with both in vitro and in vivo IL-1 antagonist activity. AF12198 selectively binds the human type I IL-1 receptor but not the human type II receptor or the murine type I receptor. In vitro, AF12198 inhibits IL-1-induced IL-8 production by human dermal fibroblasts with a half-maximal inhibition concentration or IC50 of 25 nM and IL-1-induced intercellular adhesion molecule-1 (ICAM-1) expression by endothelial cells with an IC50 of 9 nM. When given as an intravenous infusion to cynomolgus monkeys, AF12198 blocks ex vivo IL-1 induction of IL-6 and down modulates in vivo induction of IL-6. This is the first small molecule to show IL-1 receptor antagonist activity in vivo.

Suppressive effect of antioxidants on intercellular adhesion molecule-1 (ICAM-1) expression in human epidermal keratinocytes.

Intercellular adhesion molecule-1 (ICAM-1) is strongly expressed by human epidermal keratinocytes during the course of inflammatory skin diseases. To test the possibility that reactive oxygen species produced in the skin during an inflammatory response affect ICAM-1 expression, cultured human epidermal keratinocytes were treated with H2O2 at concentrations that did not damage the cells, and cell-surface ICAM-1 expression was analyzed. Expression of ICAM-1 was induced on keratinocytes by treatment with 300 microM H2O2 for 1 h. The antioxidant N-acetyl-L-cysteine strongly inhibited H2O2-induced ICAM-1 expression, whereas the antioxidants pyrrolidine dithiocarbamate and alpha-tocopherol were less inhibitory. N-acetyl-L-cysteine also suppressed keratinocyte surface expression of ICAM-1 induced by the cytokines interferon-gamma (IFN-gamma) or tumor necrosis factor-alpha (TNF-alpha), whereas pyrrolidine dithiocarbamate and alpha-tocopherol suppressed IFN-gamma-induced surface expression but not TNF-alpha-induced expression. We found that N-acetyl-L-cysteine treatment reduced ICAM-1 mRNA levels when keratinocytes were stimulated with either IFN-gamma or TNF-alpha; however, pyrrolidine dithiocarbamate and alpha-tocopherol had no effect on either IFN-gamma- or TNF-alpha-induced ICAM-1 mRNA levels. Our results indicate that reactive oxygen species may be involved in the skin inflammatory process by increasing epidermal ICAM-1 expression and that some antioxidants may be effective in suppressing the epidermal ICAM-1 expression induced by reactive oxygen species and cytokines in inflammatory skin diseases.

A fluorometric assay for the quantitation of cell adherence to endothelial cells.

A rapid quantitative fluorometric assay was developed for analysis of leukocyte adherence to endothelial cells. In this method adherent monocyte and T cell lines are labeled with the fluorescent dye Calcein AM without affecting cell function. Following coincubation with endothelial cells and gentle washing to remove nonadhering cells, the relative fluorescence intensity of the adhering cells is determined with a fluorescence microtiter plate reader. Relative fluorescence intensity increases linearly with cell number over a wide range of concentrations. By comparison of fluorescence levels of adhering cells to a dilution series of labeled cells alone, the number of adhering cells can be determined. We compared this adherence assay with the 51Cr-labeling assay and found comparable adherence. However, we found the fluorescence assay to be more rapid as the use and special handling of radioactive material is eliminated. To monitor the reliability and reproducibility of this method, we followed the adherence of Calcein AM-labeled THP-1 cells, a human monocytic cell line, to human endothelial cells treated with interleukin-1.

Identification of a specific receptor for interleukin-1 in vascular smooth muscle cells: regulation by interleukin-1 and interleukin-6.

A fluorescently labeled ligand was utilized to establish the existence of an interleukin-1 (IL-1) receptor in vascular smooth muscle. The binding of the phycoerythrin-labeled IL-1 beta to the murine T cell line, EL-4, was examined as a positive control. The phycoerythrin-labeled IL-1 beta identified a specific IL-1 receptor in the EL-4 cells. Vascular smooth muscle cells were also positively stained by the fluorescent ligand. The binding of phycoerythrin-labeled IL-1 beta to these cells was saturable and reversed by 100-fold excess unlabeled IL-1 beta. Incubation of the vascular smooth muscle cells with IL-1 beta (25 ng/ml) or IL-6 (250 ng/ml) for 18 h increased and decreased, respectively, the percentage of cells positively stained by phycoerythrin-labeled IL-1 beta which suggests these cytokines regulate IL-1 receptor expression in these cells. These data indicate a specific receptor for IL-1 exists in vascular smooth muscle cells.

Human aortic endothelial cells express the type I but not the type II receptor for interleukin-1 (IL-1).

Endothelial cells (EC) are very responsive to the proinflammatory cytokine interleukin-1 (IL-1). EC are induced by IL-1 to secrete chemotactic factors and to increase expression of cell surface adhesion molecules leading to increased leukocyte adhesion. Activated EC further contribute to the inflammatory response by secreting additional cytokines. IL-1 interacts with EC through high-affinity cell-surface receptors. However, the low number of receptors present on EC has made characterization difficult. Further, recent evidence has suggested diversity in the responses of EC from different regions of the vascular system. Interested in the effect of IL-1 on early atherosclerotic lesion formation, we have characterized the IL-1 receptors on human aortic endothelial cells (HAEC). Using a direct binding assay, we found that HAEC have 1,000-3,000 IL-1 receptors per cell and bind IL-1 alpha with a Kd of 3.5 x 10(-10) M. We found that a monoclonal antibody specific for the type I receptor completely blocks IL-1 alpha binding. The blocking antibody also completely inhibits the IL-1 induced increase in intracellular adhesion molecule 1 (ICAM-1) expression by HAEC. Using solution hybridization and ribonuclease protection with an antisense probe, a sensitive method for detection of low abundance mRNA species we found that HAEC as well as human umbilical vein EC (HUVEC) have significant levels of mRNA for the type I IL-1 receptor. To test whether HAEC might also contain transcripts for the type II IL-1 receptor, we compared levels of mRNAs by polymerase chain reaction (PCR) amplification of cDNAs reverse-transcribed from total RNA. We found only transcripts for the type I receptor and not the type II receptor in HAEC. Based on this data, we conclude that aortic endothelial cells respond to IL-1 through the type I receptor.

Induction of interleukin 1 beta expression from human peripheral blood monocyte-derived macrophages by 9-hydroxyoctadecadienoic acid.

Oxidatively modified low density lipoproteins (LDL) have recently been proposed to play a role in atherogenesis by promoting foam cell formation and endothelial cell toxicity. The purpose of the present study was to determine whether modified LDL could also induce macrophage release of interleukin 1 beta (IL-1 beta), a cytokine which enhances vascular smooth muscle cell proliferation, another feature of the atherosclerotic process. LDL were oxidatively modified by incubation with either Cu2+ (Cu(2+)-LDL) or human peripheral blood monocyte-derived macrophages (M-LDL). Incubation of these modified LDL with macrophages (6 x 10(6) cells/culture) resulted in a dose-dependent induction of IL-1 beta release. At 300 micrograms protein/ml, Cu(2+)-LDL and M-LDL induced 422 and 333 pg of IL-1 beta/culture, respectively. Saponified Cu(2+)-LDL and M-LDL were shown to contain 9- and 13-hydroxyoctadecadienoic acid (HODE), lipid oxidation products of linoleate. When tested for activity in macrophage culture (3 x 10(6) cells/culture), it was found that 9-HODE and 13-HODE (final concentration 33 microM) induced the release of 122 and 43 pg of IL-1 beta/culture, respectively, whereas untreated cells released only 4 pg of IL-1 beta/culture. Incubation of macrophages with cholesteryl-9-HODE also induced IL-1 beta release; however, the degree of induction of IL-1 beta release by 9-HODE or its cholesteryl ester relative to modified LDL suggests that other components in oxidized LDL may also contribute to IL-1 beta induction. 9-HODE was rapidly taken up by macrophages, and the kinetics were similar to IL-1 beta release. A 1.5- to 6-fold increase in the level of IL-1 beta mRNA was detected as little as 3-h post-9-HODE treatment. The induction of IL-1 beta release from human monocyte-derived macrophages by 9-HODE and cholesteryl-9-HODE suggests a role for modified LDL, and its associated linoleate oxidation products, in vascular smooth muscle cell proliferation.

Suppression of interleukin-1 beta and LDL scavenger receptor expression in macrophages by a selective protein kinase C inhibitor.

A human monocytic cell line, THP-1, stimulated with 40 nM phorbol myristate acetate (PMA), differentiated to macrophage-like cells, and exhibited increased expression and release of interleukin-1 beta and expression of acetylated low density lipoprotein (ac-LDL) receptors. A selective inhibitor, MDL 29,152 (4-propyl-5-(4-quinolinyl)-2(3H)-oxazolone) was used to show that this induction required activation of protein kinase C. MDL 29,152 acts in the catalytic domain of protein kinase C and is at least 200-fold selective for protein kinase C over cAMP-dependent protein kinase in THP-1 cells. MDL 29,152 (50 microM) reduced levels of interleukin-1 beta mRNA in PMA-stimulated cells by 76% and eliminated detectable interleukin-1 beta in the media. Flow cytometric analysis showed that 48 h after THP-1 activation, approximately 50% of the cells expressed ac-LDL receptors, while in the presence of 100 microM MDL 29,152, less than 5% of the cells expressed receptors. The relationship between THP-1 differentiation and protein kinase C activation was determined by following the expression of the cell surface antigen MO-1. Expression of MO-1 antigen increases as monocytes differentiate to macrophages. After 48 h of phorbol activation, 90% of the THP-1 population was MO-1-positive; less than 16% of the population was MO-1-positive when 100 microM MDL 29,152 was present. By dual analysis, it was found that within the differentiated, MO-1-positive population, only approximately 50% of the cells also expressed ac-LDL receptors. Based on these findings, we conclude that protein kinase C promotes processes important in THP-1 activation and differentiation to macrophage-like cells including interleukin-1 beta expression and secretion, ac-LDL receptor and MO-1 expression.

Inhibition of IL-1 beta expression in THP-1 cells by probucol and tocopherol.

The cytokine interleukin-1, IL-1, likely plays an important role in the early stages of atherogenesis. The possible action of probucol and tocopherol on the expression and secretion of IL-1 beta was investigated using the human monocytic leukemia cell line, THP-1. Both probucol and D-alpha-tocopherol inhibit the phorbol ester-induced release of IL-1 beta without altering differentiation. Analysis of IL-1 beta mRNA levels revealed that probucol and tocopherol had an inhibitory effect on the activation of expression of the IL-1 beta gene. The data suggest that the beneficial effects of probucol may be related to inhibition of IL-1 at an early phase of atherosclerotic plaque formation.

Accessory cells induce a polyphosphatidylinositol response when cultured with mitogen-activated T lymphocytes.

Monocytes (MO) influenced phosphoinositide metabolism when human T lymphocytes, isolated from peripheral blood, were activated by polyclonal mitogens. In the 3 hr immediately following mitogenic challenge, the synthesis of phosphatidylinositol (PI) was augmented and the synthesis of PI-4-phosphate (PIP) and PI-4,5-bisphosphate (PIP2) was induced in cultures of T lymphocytes and MO. In addition, MO induced a rapid and transient degradation of PIP and PIP2 in T cells prelabeled with [32P]PL and subsequently activated by mitogen. Induction of a PIP/PIP2 response correlated well with induction of DNA replication by MO when T cells were activated by phytohemagglutinin or by neuraminidase plus galactose oxidase. MO did not influence polyphosphoinositide metabolism when T cells were stimulated by the nonmitogenic lectin wheat germ agglutinin. Interleukin 1 could not substitute for monocytes in inducing a polyphosphoinositide response. By causing a rapid and transient release of the second messengers diacylglycerol and inositol phosphates and by subsequently increasing their cellular precursors, MO may induce the interleukin 2 responsive state in T lymphocytes.

Normal and mutant human adenosine deaminase genes.

Adenosine deaminase (ADA) deficiency in humans is one cause of severe combined immunodeficiency. When ADA fails to catalyze the deamination of adenosine and deoxyadenosine, the levels of deoxyadenosine that accumulate are toxic to lymphoid cells. Patients with complete ADA deficiency (e.g., with less than 5% normal ADA catalytic activity) lack both B- and T-lymphocyte function. B-lymphoblast cell lines derived from patients with ADA deficiency have been analyzed at multiple levels. Blot hybridization and S1 nuclease analysis of ADA messenger RNA (mRNA) indicates that the majority of ADA-deficient cell lines have ADA mRNA in the same abundance and size as in normal cell lines. Sequence analysis of ADA cDNAs derived from these mRNAs shows that the majority of mutations are single base changes that alter the amino acid sequence. Expression analysis proves that these point mutations lead to deficiency of ADA catalytic activity. Several cell lines have mutations that alter mRNA transcription or processing. These include a point mutation in one allele of an ADA-deficient cell line that leads to deletion of exon 4 during mRNA splicing. In addition, two cell lines are homozygous for large deletions of the gene that are the result of homologous recombination. Subjects with partial ADA deficiency have undetectable ADA activity in their erythrocytes, variable activity in their lymphoid cells, and normal immunological function. Analysis of the ADA catalytic activity of partially deficient cell lines indicates that the mutations involved affect protein stability. However, the mutations causing partial ADA deficiency are as yet undefined.

Mutant human adenosine deaminase alleles and their expression by transfection into fibroblasts.

Adenosine deaminase (ADA) deficiency in humans is one cause of severe combined immunodeficiency disease. Single base mutations affecting the ADA protein have been identified for both alleles of the ADA-deficient cell line GM2606 and for one allele of the ADA-deficient cell line GM2825A. One allele of GM2606 has a mutation altering amino acid 101 from Arg to Trp, and the other allele has a mutation altering amino acid 211 from Arg to His. As previously reported, one ADA allele of GM2825A has a single base mutation changing Ala-329 to Val-329, and the other allele has a mutation which eliminates exon 4 from the mature mRNA. Sequence analysis of polymerase chain reaction-amplified GM2825A DNA showed a single base change of A to G within the invariant bases of the 3' splice site of intron 3 that can account for the mis-splicing of exon 4. To test the effect on ADA catalytic activity of these mutations and the mutations previously found in the ADA-deficient line GM2756, expression vectors containing normal and mutant ADA-coding sequences under transcriptional regulation of the Rous sarcoma virus long terminal repeat were constructed and transfected into human fibroblasts. All transfected cells had levels of ADA mRNA 15-25 times higher than the endogenous ADA message. Yet, cells transfected with the normal ADA-coding sequences had ADA enzymatic levels 40 times higher than cells transfected with any of the mutant ADA sequences. This analysis demonstrates that while the mutant ADA-coding sequences are transcribed, they do not encode a functional ADA protein.

Mutations in the human adenosine deaminase gene that affect protein structure and RNA splicing.

Adenosine deaminase (ADA; adenosine aminohydrolase, EC 3.5.4.4) deficiency is one cause of the genetic disease severe combined immunodeficiency. To identify mutations responsible for ADA deficiency, we synthesized cDNAs to ADA mRNAs from two cell lines, GM2756 and GM2825A, derived from ADA-deficient immunodeficient patients. Sequence analysis of GM2756 cDNA clones revealed a different point mutation in each allele that causes amino acid changes of alanine to valine and arginine to histidine. One allele of GM2825A also has a point mutation that causes an alanine to valine substitution. The other allele of GM2825A was found to produce an mRNA in which exon 4 had been spliced out but had no other detrimental mutations. S1 nuclease mapping of GM2825A mRNAs showed equal abundance of the full-length ADA mRNA and the ADA mRNA that was missing exon 4. Several of the ADA cDNA clones extended 5' of the major initiation start site, indicating multiple start sites for ADA transcription. The point mutations in GM2756 and GM2825A and the absence of exon 4 in GM2825A appear to be directly responsible for the ADA deficiency. Comparison of a number of normal and mutant ADA cDNA sequences showed a number of changes in the third base of codons. These changes do not affect the amino acid sequence. Analyses of ADA cDNAs from different cell lines detected aberrant RNA species that either included intron 7 or excluded exon 7. Their presence is a result of aberrant splicing of pre-mRNAs and is not related to mutations that cause ADA deficiency.

The selective activation of T8+ cells by neuraminidase and galactose oxidase is mediated by activated T4+ cells.

The response of highly enriched populations of human T8+ lymphocytes to the oxidative mitogenic enzymes neuraminidase (NA) and galactose oxidase (GO) was enhanced by NAGO-primed T4+ lymphocytes. No similar enhancement occurred when the cells were primed with phytohemagglutinin (PHA). In the absence of subclass contamination (1%), the T8+ and T4+ cells responded equally to NAGO by the criterion of DNA replication. The addition of a small number, 2-10%, of NAGO-T4+ cells to the NAGO-T8+ cells enhanced DNA synthesis by as much as 8.5-fold. Augmentation of the cellular response did not occur unless the T4+ cells were activated by NAGO. The converse situation, 2-10% of NAGO-T8+ cells in a primarily NAGO-T4+ cell population, did not increase the DNA synthetic response of the NAGO-T4+ cells. The NAGO-T4+ cells did not augment the early event of increased phosphatidylinositol metabolism or the midcycle event of induction of receptors for interleukin 2 (IL2) and transferrin. The NAGO-T4+ cells therefore increased the probability that fully activated T8+ lymphocytes crossed the G1/S boundary. The basis for this effect was not an enhanced responsiveness of the NAGO-T8+ cells to IL2 or to other soluble growth mediators in medium conditioned by NAGO-activated lymphocytes. The results of this investigation thus implicate a control point in the NAGO-T8+ lymphocyte cell cycle that is positively modulated by the NAGO-T4+ cells themselves or by a product of their activation.

Complete sequence and structure of the gene for human adenosine deaminase.

The nucleotide sequence of the human adenosine deaminase gene was determined. The gene was isolated in a series of overlapping lambda phage clones containing human germ line DNA. A total of 36,741 base pairs were sequenced, including 32,040 base pairs from the transcription initiation site to the polyadenylation site, 3935 base pairs of 5'-flanking DNA, and 766 base pairs of 3'-flanking DNA. The gene contains 12 exons separated by 11 introns. The exons range in size from 62 to 325 base pairs while the introns are 76-15 166 base pairs in size. The area sequenced contains 23 copies of Alu repetitive DNA and a single copy of an "O" family repeat. All but one of these repeat sequences are located in the first three introns or the 5'-flanking region. The apparent promoter region of the gene lacks the "TATA" and "CAAT" sequences often found in eucaryotic promoters and is extremely G/C rich. Contained within this region are areas homologous to other G/C-rich promoters, including six decanucleotide sequences that are highly homologous to sequences identified as functional binding sites for transcription factor Sp1.

The phosphatidylinositol response and proliferation of oxidative enzyme-activated human T lymphocytes: suppression by plasma lipoproteins.

The phosphatidylinositol (PI) response and DNA synthesis of neuraminidase and galactose oxidase (NAGO)-stimulated human T lymphocytes are suppressed by low density lipoproteins (LDL). To understand the mechanism of lymphocyte activation more fully, the PI response and DNA synthesis and suppression of these events by LDL in NAGO-stimulated T lymphocytes were characterized. Between 30 min and 6 hr after NAGO stimulation, there was an increase of 32Pi incorporation into PI without increased incorporation into the phosphorylated forms of PI or into other phospholipids. DNA synthesis as determined by [3H]thymidine incorporation depended on the lymphocyte-accessory monocyte ratio and total cell density. Optimal stimulation of the PI response and DNA synthesis occurred at the same concentration of neuraminidase and galactose oxidase. While the PI response was only partially suppressed by LDL with optimal suppression at 10 to 20 micrograms of protein/ml, DNA synthesis was completely suppressed although at much higher LDL concentrations, greater than 100 micrograms protein/ml. As monocyte numbers are increased, LDL suppression of DNA synthesis is decreased. The ability of NAGO to stimulate the PI response and DNA synthesis in a similar way, and the suppression of both events by LDL, suggests the PI response is important for lymphocyte activation and proliferation. Stimulation of human T lymphocytes by oxidative mitogens, neuraminidase, and galactose oxidase caused increased phosphatidylinositol metabolism and increased DNA synthesis. Both responses were suppressed by low density lipoproteins.