Assignment of SV40-Immortalized Cells to More Than One Complementation Group for Immortalization

Human cell lines have been assigned to four complementation groups for immortalization [O.M. Pereira-Smith and J. R. Smith, Proc. Natl. Acad. Sci. USA 85, 6042-6046, 1988]. Three SV40-immortalized epithelial cell lines were fused to cell lines representative of each of these four complementation groups. All three formed senescent hybrids with an SV40-immortalized cell line representative of group A, indicating that SV40 genes do not always cause immortalization via the same genetic mechanism. In contrast to the results of studies with other human cell lines, each of these three cell lines was assigned to more than one complementation group for immortalization. Thus these cell lines may have lost the function of two or more putative senescence genes.     

Alterations in p53 and p16INK4 expression and telomere length during spontaneous immortalization of Li-Fraumeni syndrome fibroblasts.

Normal cells have a strictly limited growth potential and senesce after a defined number of population doublings (PDs). In contrast, tumor cells often exhibit an apparently unlimited proliferative potential and are termed immortalized. Although spontaneous immortalization of normal human cells in vitro is an extremely rare event, we observed this in fibroblasts from an affected member of a Li-Fraumeni syndrome kindred. The fibroblasts were heterozygous for a p53 mutation and underwent senescence as expected at PD 40. In four separate senescent cultures (A to D), there were cells that eventually recommenced proliferation. This was associated with aneuploidy in all four cultures and either loss (cultures A, C, and D) or mutation (culture B) of the wild-type (wt) p53 allele. Loss of wt p53 function was insufficient for immortalization, since cultures A, B, and D subsequently entered crisis from which they did not escape. Culture C has continued proliferating beyond 400 PDs and thus appears to be immortalized. In contrast to the other cultures, the immortalized cells have no detectable p16INK4 protein. A culture that had a limited extension of proliferative potential exhibited a progressive decrease in telomere length with increasing PD. In the culture that subsequently became immortalized, the same trend occurred until PD 73, after which there was a significant increase in the amount of telomeric DNA, despite the absence of telomerase activity. Immortalization of these cells thus appears to be associated with loss of wt p53 and p16INK4 expression and a novel mechanism for the elongation of telomeres.     

Tat-mediated protein delivery can facilitate MHC class I presentation of antigens

We have previously shown that the tat protein of HIV-1 can be used as a carrier to promote the intracellular delivery of heterologous proteins. Here we have tested if the tat-delivery technology can be used to direct MHC class I presentation of native protein, using ovalbumin (OVA) as a model system. We show that a tat-ovalbumin conjugate (tatOVA) can be delivered into cells and that subsequent processing and presentation occurs, resulting in effective and specific killing of these target cells by an OVA specific cytotoxic T-lymphocyte (CTL) line. Comparison with the E.G7 line that expresses the OVA gene indicates that tat-mediated delivery is as efficient as endogenous expression in this system. Tat-mediated antigenic protein delivery may be useful both as a research technique and, potentially, as a therapeutic or prophylactic vaccine.     

Effects of nitric oxide on platelet-activating factor- and alpha-adrenergic-stimulated vasoconstriction and glycogenolysis in the perfused rat liver.

Effects of nitric oxide (NO) on hemodynamic and glycogenolytic responses to platelet-activating factor (PAF) and phenylephrine were investigated in perfused livers derived from fed rats. Infusion of NO (34 microM) into perfused livers inhibited PAF (0.22 nM)-induced increases in hepatic glucose output and portal pressure approximately 90 and 85%, respectively, and abolished effects of PAF on hepatic oxygen consumption. NO attenuated PAF-stimulated increases in glucose output and portal pressure, the latter indicative of hepatic vasoconstriction, with a similar dose dependence with an IC50 of approximately 8 microM. In contrast to its effects on PAF-induced responses in the perfused liver, NO inhibited increases in hepatic portal pressure in response to phenylephrine (10 microM) approximately 75% without altering phenylephrine-stimulated glucose output and oxygen consumption. Similarly, infusion of NO into perfused livers significantly inhibited increases in hepatic portal pressure but not in glucose output in response to a submaximal concentration of phenylephrine (0.4 microM). Like NO, sodium nitroprusside (83 microM) significantly inhibited hemodynamic but not glycogenolytic responses to phenylephrine in perfused livers. However, PAF (0.22 nM)-stimulated alterations in hepatic portal pressure, glucose output, and oxygen consumption were unaffected by infusion of sodium nitroprusside (83 microM) into perfused livers. These results provide the first evidence for regulatory effects of NO in the perfused liver and support the contention that PAF, unlike phenylephrine, stimulates glycogenolysis by mechanisms secondary to hepatic vasoconstriction. These observations raise the intriguing possibility that NO may act in liver to regulate hemodynamic responses to vasoactive mediators.     

Nonequilibrium kinetics of a cyclic GMP-binding protein in dictyostelium discoideum

Chemoattractants added to cells of the cellular slime mold dictyostelium discoideum induce a transient elevation of cyclic GMP levels, with a maximum at 10 s and a recovery of basal levels at approximately 25 s after stimulation. We analyzed the kinetics of an intracellular cGMP binding protein in vitro and in vivo. The cyclic GMP binding protein in vitro at 0 degrees C can be described by its kinetic constants K(1)=2.5 x 10(6) M(- 1)s(-1), k(-1)=3.5 x 10(-3)s(-1), K(d)=1.4 x 10(-9) M, and 3,000 binding sites/cell. In computer simulation experiments the occupancy of the cGMP binding protein was calculated under nonequilibrium conditions by making use of the kinetic constants of the binding protein and of the shape of the cGMP accumulations. These experiments show that under nonequilibrium conditions by making use of the kinetic constants of the binding protein and the shape of the cGMP accumulations. These experiments show that under nonequilibrium conditions the affinity of the binding protein for cGMP is determined by the rate constant of association (k(1)) and not by the dissociation constant (k(d)). Experiments in vivo were performed by stimulation of aggregative cells with the chemoattractant cAMP, which results in a transient cGMP accumulation. At different times after stimulation with various cAMP concentrations, the cells were homogenized and immediately thereafter the number of binding proteins which were not occupied with native cGMP were determined. The results of these experiments in vivo are in good agreement with the results of the computer experiments. This may indicate that: (a) The cGMP binding protein in vivo at 22 degrees C can be described by its kinetic constants: K(1)=4x10(6)M(-1)s(-1) and K(-1)=6x10(-3)s(-1). (b) Binding the cGMP to its binding protein is transient with a maximum at about 20-30 s after chemotactic stimulation, followed by a decay to basal levels, with a half-life of approximately 2 min. (c) The cGMP to its binding proteins get half maximally occupied at a cGMP accumulation of δ[cGMP](10)=2x10(-8) M, which corresponds to an extracellular stimulation of aggregative cells by 10(-10) M cAMP. (d) Since the mean basal cGMP concentration is approximately 2x10(-7) M, the small increase of cGMP cannot be detected accurately. Therefore the absence of a measurable cGMP accumulation does not argue against a cGMP function. (e) There may exist two compartments of cGMP: one contains almost all the cGMP of unstimulated cells, and the other contains cGMP binding proteins and the cGMP which accumulates after chemotactic stimulation. (f) From the kinetics of binding, the cellular responses to the chemoattractant can be divided into two classes: responses which can be mediated by this binding protein (such as light scattering, proton extrusion, PDE induction, and chemotaxis) and responses which cannot be (solely) mediated by this binding protein such as rlay, refractoriness, phospholipids methylation, and protein methylation.     

Structure/function studies on vascular cell adhesion molecule-1.

Vascular cell adhesion molecule-1 (VCAM1) is a member of the immunoglobulin (Ig) superfamily which interacts with the integrin very late antigen-4 (VLA4). The VCAM1/VLA4 interaction mediates both adhesion and signal transduction and is thought to play an important role in inflammatory and immune responses in vivo. The major form of human VCAM1 contains seven extracellular Ig-like domains, with domain 1 designated as the most N-terminal. We have examined the relationship between human VCAM1 structure and function using a combination of domain truncation mutants and proteolytic fragmentation of recombinant soluble VCAM1. We have characterized two regions of VCAM1, localized to domains 4 and 5, which are highly sensitive to proteolytic cleavage, localized the epitope of the blocking monoclonal antibody 4B9 to domain 1, and found that domains 1-3 are sufficient for both its adhesive function and its ability to initiate T cell activation.     

Cloning of murine and rat vascular cell adhesion molecule-1.

Vascular cell adhesion molecule-1 (VCAM1) is a member of the immunoglobulin (Ig) superfamily which interacts with the integrin very late antigen 4 (VLA4). We have cloned the cDNAs for both murine and rat VCAM1 from endotoxin-treated lung libraries. Both sequences encode proteins with seven extracellular Ig-like domains, which show 75.9% and 76.9% identity, respectively, with human VCAM1. Both murine and human cell lines show VLA4-dependent binding to COS cells transiently expressing murine and rat VCAM1. Two mAbs, M-K/1 and M-K/2, which recognize an antigen on murine bone marrow stromal cell lines, bind to murine VCAM1 expressed in COS cells and block VCAM1-dependent adhesion, confirming that these mAbs recognize murine VCAM1.     

Poly(adenosine diphosphate-ribosylation) of nuclear matrix proteins in alkylating agent resistant and sensitive cell lines

Using Walker 256 breast carcinoma cell lines either with or without acquired resistance to alkylating agents, the structural framework proteins of the nucleus, the nuclear matrix proteins, were found to be effective acceptors for poly(ADP-ribose). Incubation of isolated nuclei with nicotinamide adenine [32P] dinucleotide ([32P] NAD), followed by the isolation of the nuclear matrix, demonstrated that two polypeptides of approximate molecular weight (Mr) 105 000 and 116 000 were extensively poly(ADP-ribosylated). By an in vitro [32P] NAD assay, the nuclear matrix fraction was found to maintain approx. 15% of the total nuclear matrix activity of poly(ADP-ribose) polymerase. Confirmation that the trichloroacetic acid (TCA) precipitable material represented ADP-ribose units was achieved by enzymatic digestion of the nuclear matrix preparation with snake venom phosphodiesterase (SVP). Within 15 min, greater than 85% of the 32P label was digested by SVP and the final digestion products were found to be phosphoribosyl-AMP (PR-AMP) and adenosine 5'-monophosphate (5'-AMP) by thin layer chromatographic analysis. The average polymer chain length was estimated to be 6-7 ADP-ribose units. Because poly(ADP-ribose) polymerase has a putative role in DNA repair, a comparison of the nuclear matrix fractions from Walker resistant and sensitive tumor cell lines was made. In both cell lines, the quantitative and qualitative patterns of the nuclear matrix associated poly(ADP-ribosylation) were similar.     

Site-selected fluorescence spectra of porphyrin derivatives of heme proteins

The emission spectra of the porphyrin in metal-free and Zn cytochrome c and in metal-free mesoporphyrin derivatives of horseradish peroxidases A and C, leghemoglobin, and myoglobin were examined as a function of temperature and excitation wavelength. At room temperature, the emission spectra were unresolved and were independent of excitation wavelength. At low temperature (4.2 K), the spectra depended upon excitation wavelength: using narrow-band excitation into the high-energy side of the 0-1 and 0-0 bands gave unresolved emission spectra whereas excitation into the low-energy side produced quasi-line spectra. The resolved spectra were different for the five proteins and further varied with pH, indicating chromophore-protein interactions. The spectra are interpreted in terms of site selection and phonon interactions.