High-throughput evaluation of cryoprotective agents for mixture effects that reduce toxicity.

Vitrification is a promising approach for cryopreserving complex biological structures such as organs. However, to prevent ice formation, high concentrations of cell-permeable cryoprotective agents (CPAs) are required, which can be highly toxic. The current reliance on a small number of CPAs limits optimization of low-toxicity compositions for vitrification. To address this, there is growing interest in uncovering novel chemicals with analogous protective qualities. This may not only enhance vitrification efficacy but also mitigate toxic effects. In the current study, we employed a high throughput method to assess the toxicity of 21 compounds at room temperature, both individually and in binary combinations. Our analysis revealed that toxicity increases with both exposure duration and concentration, and that several CPA combinations result in reduced overall toxicity. Notably, among all tested mixtures, four binary combinations—formamide/glycerol, dimethyl sulfoxide/1,3-propanediol, 1,2-propanediol/diethylene glycol, and 1,3-propanediol/diethylene glycol—produced a statistically significant decrease in toxicity, resulting in significantly higher viability for the 6 mol/kg mixture than both corresponding 6 mol/kg single CPA solutions. The high-throughput approach presented here will aid in building a comprehensive CPA toxicity database, which will improve our understanding of toxicity mechanisms and support the development of predictive models for identifying novel CPA mixtures with low toxicity.

Screening for cryoprotective agent toxicity and toxicity reduction in mixtures at subambient temperatures.

Cartilage-derived anti-tumor factor (CATF) inhibits the proliferation and DNA synthesis of bovine pulmonary artery endothelial (BPAE) cells in culture (Takigawa, M.et al. Cell. Biol. Inn. Rep., 9, 619–625, 1985). In the present study, we partially purified CATF by monitoring inhibition of DNA synthesis in BPAE cells and tested the effects of the purified materials on the growth of solid tumors and tumor-induced angiogenesis. Crude CATF (CATF20–300k), the fraction of 20 k to 300 k daltons, separated by ultrafiltration was further separated into three fractions by ultrafiltration. The fraction of 100 k to 300 k daltons (CATF100–300k) caused slightly more inhibition than CATF20–300k of DNA synthesis in BPAE cells. On the other hand, the fraction of 20 k to 50 k daltons had only a slight effect, and the fraction of 50 k to 100 k had even less effect on DNA synthesis in BPAE cells. CATF100–300k caused slightly more inhibition than CATF20–300k of the growth of solid tumors of B16 melanoma, while the fraction of 20 k to 100 k daltons did not inhibit the growth of tumors at all. CATF100–300k also inhibited B16 melanoma-induced angiogenesis in chick embryo chorioallantoic membranes (CAM), whereas the fraction of 20 k to 100 k daltons had little effect on the angiogenesis. CATF100–300k was further purified by DEAE-Sepharose CL-6B chromatography. The main peak with activity on DNA synthesis in BPAE cells was eluted with 0.3 to 0.35 M NaCl at pH 8.0. The activity of this peak on DNA synthesis in BPAE cells was about 70 fold that of CATF100–300k. The purified CATF also inhibited the growth of B16 melanoma and B16 melanoma-induced angiogenesis in CAM. On the other hand, the inactive fraction on DNA synthesis in BPAE cells obtained by DEAE-Sepharose chromatography was also inactive in inhibiting the growth of B16 melanoma and B16 melanoma-induced angiogenesis in CAM. These findings strongly suggest that CATF is an anionic macromolecule(s) and has anti-angiogenic activity, thereby inhibiting the growth of solid tumors.

Endothelial cells used in transendothelial migration assays have been derived from different sources including human umbilical vein and bovine pulmonary artery. As plasma-induced activation of endothelial cells for enhanced transmigration of leukocytes may be pathophysiologically relevant, the role of species differences between plasma and endothelial cells was studied. The effects of human plasma on human umbilical vein and bovine pulmonary artery endothelial cells with regard to transmigration of neutrophils were compared. Transendothelial migration of neutrophils was tested employing endothelial cell-covered Transwell cell culture chamber inserts with micropore filters of 5 microns pore size. Results showed that human plasma induces neutrophil transmigration of both human umbilical vein and bovine pulmonary artery endothelial cell monolayers. Plasma samples from 50 to 69 year old men with and without carotid arteriosclerosis (N = 152) stimulated endothelial cells of human origin significantly stronger for transmigration of neutrophils than endothelial cells of bovine origin. Analyses of limits of agreement and of correlation of the two methods indicated that bovine pulmonary artery and human umbilical vein endothelial cells cannot replace each other in assays of plasma-induced activation of transmigration. Plasma levels of soluble adhesion molecule P-selectin, which are elevated in patients with arteriosclerotic diseases, are inversely related to plasma-induced activation of neutrophil transmigration of endothelial monolayers of human origin but not of bovine origin (N = 65). As rates of transmigration that were induced by plasma from subjects without carotid arteriosclerosis (N = 73) were significantly related between human and bovine endothelium, elevated plasma levels of soluble P-selectin may thus affect results of assays for plasma-induced activation of neutrophil transmigration of endothelial monolayers in a species-specific manner.

Mechanisms by which extracellular ATP and UTP stimulate the release of prostacyclin from bovine pulmonary artery endothelial cells

Extracellular ATP causes an increase in the concentration of cytoplasmic free calcium ([Ca2+]i) in bovine pulmonary-artery endothelial (BPAE) cells that results in the synthesis and release of prostacyclin (PGI2), a potent vasodilator and inhibitor of platelet aggregation. We show here that PGI2 release in BPAE cells correlates with the concentration of the fully ionized form of extracellular ATP (ATP4-) and not with the concentration of other ionic forms of ATP. Concentrations as low as 10 nM-ATP4- elicited an increase in PGI2 release [EC50 (concn. giving half-maximal stimulation) 3 microM] in BPAE cells incubated in an iso-osmotic medium, pH 7.4, lacking Ca2+ and Mg2+. When the pH or the Mg2+ concentration of the medium was varied so as to maintain a constant level of ATP4-, while varying the concentration of proton-ATP (HATP3-) or MgATP2- respectively, PGI2 release remained constant. An inhibitory effect of extracellular Mg2+ on PGI2 release could be attributed solely to a decrease in the concentration of ATP4-. In contrast with Mg2+, extracellular Ca2+ stimulated PGI2 release induced by ATP. Several results suggest that extracellular Ca2+ modulates PGI2 release by increasing Ca2+ uptake through an ATP(4-)-activated plasma-membrane channel. In BPAE cells incubated in Ca(2+)-free medium, ATP elicited a transient increase in [Ca2+]i that declined to the basal level within 60 s. In cells incubated in Ca(2+)-containing medium, ATP caused an increase in [Ca2+]i that had two components: a transient peak in [Ca2+]i (0-60 s) and a sustained increase in [Ca2+]i that was maintained for several minutes after ATP addition. Increasing the concentration of extracellular calcium from 0.25 mM to 10 mM had no effect on the transient rise in [Ca2+]i induced by ATP, but significantly enhanced the magnitude of the sustained increase in [Ca2+]i. Alterations in the magnitude of the sustained increase in [Ca2+]i would likely modulate PGI2 release, which was not complete until 2 min after ATP addition. Extracellular Ca2+ also stimulated PGI2 release induced by bradykinin. Bradykinin caused a sustained increase in [Ca2+]i in BPAE cells in the presence of extracellular Ca2+. Finally, the magnitude of PGI2 release induced by UTP, a more potent agonist than ATP, correlated with the concentration of extracellular fully ionized UTP (UTP4-). These findings support the hypothesis that nucleotide receptors in BPAE cells recognize the fully ionized form of ATP and UTP and are coupled to signal-transduction pathways involving the mobilization of intracellular Ca2+, the influx of extracellular Ca2+ and the subsequent release of PGI2.

Increased permeability is an early and universal response of the vasculature to radiation injury, yet the biological basis of this reaction is poorly understood. The present study determined the time course and the dose-response relationship of radiation-induced hyperpermeability in cultured bovine pulmonary artery endothelial (BPAE) cells. BPAE cells were grown to a confluent monolayer on microcarrier beads, and column chromatography methods were used to evaluate permeability to two low molecular weight compounds: sodium fluorescein (NaFlsc, mol. wt. = 342) and cyanocobalamin (B12, mol. wt. = 1355). This is a novel in vitro model to study mechanisms and modifiers of radiation-induced permeability of endothelial cells under flow conditions using nonradioactive tracers. Cell-covered beads were exposed to a single dose of 10 Gy Of 137Cs gamma rays and placed in the column, and permeability was measured every 30 min for 3 h. There was a time-dependent increase in permeability to both tracers, reaching significance by 2 h. Increased permeability was accompanied by perturbations in F-actin distribution in the BPAE cells as determined by rhodamine-phalloidin fluorescence microscopy. Neither catalase nor captopril ameliorated this hyperpermeability, but dibutyryl cAMP partially prevented it. At 3 h after 0, 1, 2, 5 and 10 Gy irradiation, permeability values of 11.8 +/- 2.1, 13.9 +/- 2.2, 20.9 +/- 3.6, 24.8 +/- 2.8 and 27.2 +/- 3.3 (10(-5) cm/s, +/- SEM), respectively, were observed using NaFlsc. The increase was significant (P < 0.05) at 2 Gy or higher. Permeability to B12 was significantly elevated after 5 or 10 Gy. These results suggest that permeability of endothelial cells to low molecular weight solutes increases within 3 h after therapeutic doses of radiation, and that cAMP ameliorates this response.

Endothelial cell migration is critical for proper blood vessel development. Signals from growth factors and matrix proteins are integrated through focal adhesion proteins to alter cell migration. Hydrogen peroxide-inducible clone 5 (Hic-5), a paxillin family member, is enriched in the focal adhesions in bovine pulmonary artery endothelial (BPAE) cells, which migrate to lysophosphatidic acid (LPA) on denatured collagen. In this study, we investigate the role of Hic-5 in LPA-stimulated endothelial cell migration. LPA recruits Hic-5 to the focal adhesions and to the pseudopodia in BPAE cells plated on collagen, suggesting that recruitment of Hic-5 to focal adhesions is associated with endothelial cell migration. Knockdown of endogenous Hic-5 significantly decreases migration toward LPA, confirming involvement of Hic-5 in migration. To address the role of Hic-5 in endothelial cell migration, we exogenously expressed wild-type (WT) Hic-5 and green fluorescent protein Hic-5 C369A/C372A (LIM3 mutant) constructs in BPAE cells. WT Hic-5 expression increases chemotaxis of BPAE cells to LPA, whereas migration toward LPA of the green fluorescent protein Hic-5 C369A/C372A-expressing cells is similar to that shown in vector control cells. Additionally, ERK phosphorylation is enhanced in the presence of LPA in WT Hic-5 cells. A pharmacological inhibitor of MEK activity inhibits LPA-stimulated WT Hic-5 cell migration and ERK phosphorylation, suggesting Hic-5 enhances migration via MEK activation of ERK. Together, these studies indicate that Hic-5, a focal adhesion protein in endothelial cells, is recruited to the pseudopodia in the presence of LPA and enhances migration.

We examined the mechanism of endothelin (ET)-1 regulation by cigarette smoke extract (CSE) and the effect of platelets on CSE-induced stimulation of ET-1 gene expression in human and bovine pulmonary artery endothelial cells (PAECs). Our data show that CSE (1%) induces ET-1 gene expression (after 1 h) and ET-1 peptide synthesis (after 4 h) in bovine PAECs. The induction of preproET-1 mRNA level was due to de novo transcription, and new protein synthesis was not required for this induction. The protein kinase C inhibitors staurosporine (10(-8) mol/l) and calphostin C (10(-7) mol/l) abolished the induction of ET-1 gene expression by CSE in bovine and human PAECs. Although a lower concentration of platelets (10(6) cells/ml in bovine PAECs; 10(7) cells/ml in human PAECs) did not significantly alter ET-1 gene expression in PAECs, incubation of platelets with CSE (1%) and PAECs produced a significant increase in preproET-1 mRNA and ET-1 peptide compared with the values in the presence of CSE (1%) alone. CSE (1%) induced platelet aggregation and increased the expression of platelet membrane glycoproteins ex vivo. Thus our data suggest that CSE stimulates ET-1 gene expression via PKC in PAECs. CSE and platelets showed a synergistic effect on ET-1 gene expression, possibly through the activation of platelets by CSE.

REGULATION OF EXTRACELLULAR MATRIX PROTEINS BY TRANSFORMING GROWTH FACTOR β1IN CULTURED PULMONARY ENDOTHELIAL CELLS

Heat shock proteins (HSPs) have been remarkably conserved throughout evolution. It has been assumed that induction of HSPs remains a stereotypic response to injury, important for survival of eukaryotic cells during euthermic injury. However, there are few studies of this phenomenon in endothelial cells, and none in pulmonary endothelial cells. We studied the induction of synthesis of 70-kD proteins in bovine pulmonary artery endothelial cells (BPAECs) in response to heat shock and to euthermic injury induced by bacterial endotoxin. First, in response to heat, BPAECs showed rapid and reversible heat-induced synthesis of 70-kD proteins, readily detectable by one-dimensional SDS-PAGE of [35S]methionine-labeled BPAECs. Heat shock at 42 degrees C for 3 h or 43 degrees C for 2 h suppressed total protein synthesis by 30% (P less than 0.001) but an increased rate of synthesis of 70-kD protein continued, representing an increasing fraction of total protein synthesis. Heat-induced synthesis of 70-kD protein returned to baseline levels 8 h after heat shock. Northern analysis showed that mRNA for a protein homologous to a conserved amino acid sequence in the family of species-homologous 70-kD heat shock proteins (HSP 70) was induced by a 15-min incubation at 42 degrees C and remained detectably increased for 6 h. We next assessed whether euthermic injury by bacterial endotoxin (LPS) generated a similar response. LPS was cytotoxic by BPAECs as assessed morphologically, by release of 51Cr from prelabeled cells, and by a significant suppression of total protein synthesis (range, 35 to 70%; P less than 0.001). Despite cytotoxicity, LPS did not induce 70-kD protein at a level that could be detected by SDS-PAGE, and no increase in mRNA for HSP 70 was detected by Northern analysis. LPS-injured BPAECs remained "competent" to induce both 70-kD proteins and mRNA for HSP 70 in response to heat shock. We conclude that at least quantitatively, induction of HSP 70 by BPAECs is not a stereotypic response to injury but rather is at least relatively injury-specific. However, competence to induce HSP 70 appears to be extremely resilient: it is retained in dysfunctional BPAECs in the face of profound inhibition of global protein synthesis, suggesting an important homeostatic role.

1. Barrier function and cytosolic free calcium content [Ca2+]i was measured in monolayers of bovine pulmonary artery endothelial cells (BPAEC) and bovine aortic endothelial cells (BAEC). 2. Thrombin (1 u ml-1) increased albumin transfer across monolayers of BPAEC but not BAEC, yet induced biphasic increases in [Ca2+]i in both endothelial cell types, consisting of a rapid, initial phasic component which decayed to a lower, more sustained plateau phase. 3. 4 beta-Phorbol 12-myristate 13-acetate (PMA; 0.3-3000 nM) increased albumin transfer across monolayers of BPAEC and BAEC, but had no effect on basal levels of [Ca2+]i in either endothelial cell type. 4. Treatment of BPAEC and BAEC with forskolin (30 microM), an activator of adenylate cyclase, had no effect on resting transfer of albumin, but inhibited that stimulated by PMA (600 nM). It also inhibited the thrombin (1 u ml-1)-induced increase in albumin transfer across monolayers of BPAEC, but enhanced the plateau phase of the associated increase in [Ca2+]i. 5. Treatment of BPAEC and BAEC with either atriopeptin II (100 nM), an activator of particulate guanylate cyclase, or 8 bromo cyclic GMP (30 microM) had no effect on resting or PMA (600 nM)-stimulated transfer of albumin. Both agents did, however, inhibit the thrombin (1 u ml-1)-induced increase in albumin transfer across monolayers of BPAEC, but had no effect on the associated increase in [Ca2+]i. 6. These data suggest a dissociation between the ability of agents that increase or decrease albumin transfer and their effects on [Ca2+]i. Consequently, activation of protein kinase C may be the major stimulus for trans-endothelial transfer of macromolecular solutes. Endothelial barrier function is enhanced by elevation of either cyclic AMP or cyclic GMP content. Cyclic AMP appears to act by inhibiting the actions of protein kinase C, while cyclic GMP may act to inhibit a key step proximal to activation of this enzyme.

L-Arginine (L-Arg) is metabolized to nitric oxide (NO) by NO synthase (NOS) or to urea by arginase (AR). L-Arg is transported into bovine pulmonary arterial endothelial cells (BPAECs) by cationic amino acid transporter-2 (CAT-2). We hypothesized that cytokine treatment would increase L-Arg metabolism and increase CAT-2 mRNA expression. BPAECs were incubated for 24 h in medium (control) or medium with lipopolysaccharide and tumor necrosis factor-alpha (L-T). L-T increased nitrite production (3.1 +/- 0.4 nmol/24 h vs. 1.8 +/- 0.1 nmol/24 h for control; P < 0.01) and urea production (83.5 +/- 29.5 nmol/24 h vs. 17.8 +/- 8.6 nmol/24 h for control; P < 0.05). L-T-treated BPAECs had greater endothelial and inducible NOS mRNA expression compared with control cells. Increasing the medium L-Arg concentration resulted in increased nitrite and urea production in both the control and the L-T-treated BPAECs. L-T treatment resulted in measurable CAT-2 mRNA. L-T increased L-[(3)H]Arg uptake (5.78 +/- 0.41 pmol vs. 4.45 +/- 0.10 pmol for control; P < 0.05). In summary, L-T treatment increased L-Arg metabolism to both NO and urea in BPAECs and resulted in increased levels of CAT-2 mRNA. This suggests that induction of NOS and/or AR is linked to induction of CAT-2 in BPAECs and may represent a mechanism for maintaining L-Arg availability to NOS and/or AR.

Biosynthesis and modulation of endothelin from bovine pulmonary arterial endothelial cells

1. We combined patch clamp and fura-2 fluorescence methods to characterize human TRP3 (hTRP3) channels heterologously expressed in cultured bovine pulmonary artery endothelial (CPAE) cells, which do not express the bovine trp3 isoform (btrp3) but express btrp1 and btrp4. 2. ATP, bradykinin and intracellular InsP3 activated a non-selective cation current (IhTRP3) in htrp3-transfected CPAE cells but not in non-transfected wild-type cells. During agonist stimulation, the sustained rise in [Ca2+]i was significantly higher in htrp3-transfected cells than in control CPAE cells. 3. The permeability for monovalent cations was PNa > PCs approximately PK >> PNMDG and the ratio PCa/PNa was 1.62 +/- 0.27 (n = 11). Removal of extracellular Ca2+ enhanced the amplitude of the agonist-activated IhTRP3 as well as that of the basal current The trivalent cations La3+ and Gd3+ were potent blockers of IhTRP3 (the IC50 for La3+ was 24.4 +/- 0.7 microM). 4. The single-channel conductance of the channels activated by ATP, assessed by noise analysis, was 23 pS. 5. Thapsigargin and 2,5-di-tert-butyl-1, 4-benzohydroquinone (BHQ), inhibitors of the organellar Ca2+-ATPase, failed to activate IhTRP3. U-73122, a phospholipase C blocker, inhibited IhTRP3 that had been activated by ATP and bradykinin. Thimerosal, an InsP3 receptor-sensitizing compound, enhanced IhTRP3, but calmidazolium, a calmodulin antagonist, did not affect IhTRP3. 6. It is concluded that hTRP3 forms non-selective plasmalemmal cation channels that function as a pathway for agonist-induced Ca2+ influx.