Endogenous tumor necrosis factor alpha mediates enhanced apoptosis of cultured villous trophoblasts from intrauterine growth-restricted placentae.

Tumor necrosis factor alpha (TNFalpha) has been implicated in the abnormally high levels of trophoblast apoptosis seen in placentae from pregnancies complicated by small births. We examined the hypothesis that at physiological (35-50 mmHg) oxygen tensions, the production of TNFalpha stimulates the apoptosis of placental trophoblasts associated with infants that are intrauterine growth-restricted (IUGR). Highly purified cytotrophoblasts (CT) from IUGR-complicated pregnancies spontaneously underwent a higher rate of apoptosis after 24 h of culture at a normoxic (for villous CT) tension of 38 mmHg than did CT from normal placentae. Real-time PCR analysis of TNFalpha mRNA revealed approximately threefold higher levels in IUGR trophoblasts after culturing at a pO2 of 38 mmHg. A higher level of TNFalpha receptor p55 (which mediates apoptosis) was found in IUGR CT by western blot analysis at pO2 of <10, 38, and 140 mmHg. Neutralizing antibody to TNFalpha significantly inhibited the apoptosis of IUGR trophoblasts cultured at 38 mmHg and addition of TNFalpha significantly elevated apoptosis of normal and IUGR trophoblasts but less in IUGR cells cultured at <10 mmHg. We conclude that at physiological oxygen tensions (38 mmHg), villous CT from IUGR pregnancies, when compared with uncomplicated pregnancies, undergo more TNFalpha-induced apoptosis both because of elevated expression of TNFalpha and TNF receptor p55.

Effect of oxygen levels in villous trophoblast apoptosis.

In pregnancies complicated by intrauterine growth restriction (IUGR), the villous trophoblast shows increased apoptosis and immature cytotrophoblasts (CT) may be exposed to both higher or lower oxygen levels than normal placentae. We propose that villous CT undergo higher frequencies of apoptosis at extreme oxygen tensions. The apoptosis of CT isolated from normal term placentae was examined before culture and after 24 h of culture at different oxygen tensions with or without TNFalpha. The apoptosis frequencies of cells cultured for 24 h at O2 levels of approximately 15 mm and approximately 38 mm Hg were similar to the frequency before culture. Both constitutive and TNFalpha-induced apoptosis and cell loss were highest at low (<10 mm Hg) and high ( approximately 140 mm Hg) oxygen tensions. Further, the ratios of induced to constitutive apoptosis, constant from approximately 15 mm to approximately 140 mm Hg, indicate induced apoptosis to be rather insensitive to changes in oxygen levels. These results show that primary villous trophoblasts from normal placentae undergo minimal apoptosis unless subjected to extreme oxygen tensions <15 mm or 140 mm Hg. The results indicate that normal villous trophoblasts are remarkably resistant to hypoxia-induced apoptosis.

The effect of oxygen tension on intracellular survival signalling in primary villous trophoblasts.

Villous trophoblasts undergo increased apoptosis and experience a wider gradient of oxygen tensions (pO(2)) in pregnancies complicated by intrauterine growth restriction. We hypothesize that pO(2)affects trophoblast apoptosis by altering survival signalling through the phosphatidylinositol-3 (PI-3)-kinase and mitogen activated protein kinase (MAPK) pathways. Cytotrophoblasts were cultured at pO(2)from <10 to approximately 140 mmHg with Epidermal Growth Factor (EGF), Vascular Endothelial Growth Factor (VEGF) and basic Fibroblast Growth Factor (bFGF) at concentrations of 0.1 to 10 ng/ml for 1 to 12 h, then assessed for apoptosis (TUNEL) and specific protein expressions (Western blot analysis). Spontaneous apoptosis was highest at <10 mmHg and lowest approximately 15 mmHg. Only EGF activated either signalling pathway at any pO(2). Inhibition of both pathways was required to inhibit EGF-stimulated survival. Maximal EGF activation of either pathway was insensitive to pO(2). At lower oxygen tensions, MAPK phosphorylation was maximal at 1 ng/ml EGF compared with 10 ng/ml for the PI-3-kinase path. The EGF receptor was spontaneously phosphorylated with increasing culture times at lower oxygen levels, an effect reflected down-stream by PI-3-kinase and Akt phosphorylation. We conclude that strong survival signalling in trophoblasts requires both PI-3- and MAP-kinase pathways, is rather insensitive to pO(2)changes and is spontaneously activated with increasing hypoxic exposure.

Selective cytotoxicity of gemcitabine in bladder cancer cell lines.

We have examined the cytotoxic effect of gemcitabine in intravesical therapy using an in vitro co-cultured spheroid model composed of transitional cell carcinoma (TCC) and fibroblasts from both human and rat species. Immunohistochemistry analysis of the co-cultured spheroids, using cytokeratin-13 and vimentin antibodies against TCC and fibroblasts, respectively, showed the central location of fibroblasts within the spheroid, whereas TCC formed the peripheral layers. Spheroids composed of human TCC and fibroblasts (MGH-U3/CRL-1120 or RT-112/CRL-1120) as well as rat TCC and their corresponding fibroblasts (AY-27/RF-Ed1) displayed the same drug tolerance profile after an exposure of 0, 1, 3, 5, 7 and 14 days. As confirmed by time-lapse photography, MTT essay and vital dye staining, gemcitabine selectively killed the human and rat bladder cancer cell lines, but did not affect un-transformed human and rat fibroblast lines.

The spectrum of pathogenic bacteria in positive blood cultures.

There is compelling evidence to suggest that the profiles of pathogenic bacteria which cause septicaemia shock vary from one region to another due to differences in the source of contamination. Blood cultures were prepared from 3,481 patients with symptoms of systemic bacterial contamination. The blood cultures of 558 (16.02%) patients showed at least one kind of bacterial infection. This rate was markedly higher than that reported in Germany (12.8%) and Japan (12.3%). Systemic bacterial infection was significantly higher in males than in females (82% versus 18%). Most of the patients surveyed (62%) were adults and the rest were either infants (19%) or neonates (19%). When blood samples of these patients were cultured, and isolated bacteria were characterized by a variety of diagnostic tests, over twenty different strains of bacteria were identified and characterized. More than 29% of positive cultures were Enterobacter spp. while Staphylococcus aureus (20%) and Brucella spp. (8%) ranked second and third highest among the infections. The results suggest that agents which cause infections vary with respect to region and that knowledge of local risk factors may aid in patient diagnosis and treatment.

Liposome associated interferon-alpha-2b functions as an anti-fibrogenic factor in dermal wounds in the guinea pig.

We have previously reported that interferon-alpha-2b (IFN-alpha-2b) can be encapsulated in liposomes without compromising its anti-fibrogenic effects on dermal fibroblasts in vitro. This study was conducted to determine whether this preparation applied topically to guinea pig wounds can affect their healing. The rationale for this approach is that systemic administration of IFN-alpha-2b by injection for treatment of dermal fibrosis is uncomfortable, requires a large quantity of the cytokine and cannot be easily used in children. Liposomes are potentially useful vehicles for the topical delivery of drugs. Empty sonicated liposome vesicles were mixed with various concentrations of IFN-alpha-2b and then dried and rehydrated. An enzyme-linked immunosorbent assay (ELISA) was used to determine the efficiency of encapsulation and the stability of the preparation under experimental conditions. A total of 36 full thickness skin wounds (6/animal, 3 on each side) were made with an 8 mm disposable punch. Each wound on the right side received cream (100 mg/wound) containing 3000 units of liposome-encapsulated IFN-alpha-2b, while wounds on the left side received cream containing empty liposomes. There was a significant reduction in rate of contraction of wounds treated with IFN-alpha-2b as early as 5 days after wounding. This reduction remained significant up to 10 days. Northern analysis, used to evaluate the expression of mRNAs for type I and type III collagens in response to IFN-alpha-2b showed a marked reduction in abundance of the transcripts for the pro-alpha1(I) chain of type 1 collagen on days 11 and 14 after wounding. Similarly, the level of mRNA for type III procollagen was markedly reduced as early as day 7 and remained depressed up to day 14. These findings were consistent with results obtained for the total collagen content in tissue samples. Cellularity of the IFN-alpha-2b-treated wounds, assessed by vimentin content, was also markedly reduced at day 7 and remained depressed up to day 14. Liposome associated IFN-alpha-2b applied 5 days after completion of epithelialization reduced mRNA for the pro-alpha1(I) chain of type 1 collagen, confirming its transepidermal penetration and effectiveness. The activity of liposome-associated IFN-alpha-2b in vivo supports the concept of the topical use of this anti-fibrogenic agent for treatment of fibroproliferative disorders.

Mannose-6-phosphate/IGF-II receptors mediate the effects of IGF-1-induced latent transforming growth factor beta 1 on expression of type I collagen and collagenase in dermal fibroblasts.

We have previously shown that insulin-like growth factor-1 (IGF-1) induces the expression of latent transforming growth factor beta 1 (LTGF-beta 1) through activation of c-fos and c-jun oncogenes. In this study we investigated whether IGF-1 induced latent TGF-beta 1 has autocrine effects on dermal fibroblasts and described a possible mechanism. Human dermal fibroblasts were treated with either vehicle, IGF-1 alone, or IGF-1 with either anti-TGF-beta 1 neutralizing antibody or mannose-6-phosphate (M6P) and levels of mRNAs for TGF-beta 1, collagenase and the pro alpha 1(I) chain of type I collagen were then evaluated by Northern analysis. Conditioned medium was also collected from treated and untreated cells and assayed for TGF-beta 1 protein by enzyme-linked immunosorbent assay. The results of the Northern analysis revealed a differential effect on the expression of the pro alpha 1(I) chain of type I collagen and collagenase in dermal fibroblasts: mRNA for the former being significantly increased in response to IGF-1 treatment while that for collagenase was markedly suppressed. These effects of IGF-1 were blocked to a significant extent by TGF-beta 1 neutralizing antibody at a concentration of 0.5-2.0 micrograms/ml. As the TGF-beta 1 induced by IGF-1 is inactive in the traditionally used mink lung epithelial cell growth inhibition assay, we explored the possible role of IGF-II/M6P receptors in facilitating these autocrine effects. The results showed that the greater than two-fold increase (201.9 +/- 38 vs 81.8 +/- 13, p < 0.05) in mRNA for the pro alpha 1(I) chain of type I collagen induced by IGF-1 was at least 60% inhibited by M6P in a time-dependent fashion. A direct correlation between the expression of TGF-beta 1 and the pro alpha 1(I) chain of type I collagen was found in response to either IGF-1 alone or IGF-1 with M6P. Treatment of cell cultures with TGF-beta 1 neutralizing antibody mimicked the effect of M6P. In contrast to the effects on expression of type I collagen, the level of collagenase mRNA was markedly reduced by IGF-1 alone and was restored by the administration of M6P. The levels of TGF-beta 1 in conditioned medium from treated and untreated cells showed a similar pattern to that of the mRNA detected by Northern analysis. These findings suggest that IGF-1 induces latent TGF-beta 1 and that the matrix-modulating autocrine effects of LTGF-beta 1 on dermal fibroblasts are facilitated by M6P/IGF-II receptors on these cells.

The antifibrogenic effects of liposome-encapsulated IFN-alpha2b cream on skin wounds.

Interferons (IFN), including IFN-alpha2b, have been used as antifibrogenic factors to modulate the excessive production of extracellular matrix (ECM) associated with dermal fibroproliferative disorders. This study was conducted to examine the ability of a dermal cream containing liposome-encapsulated IFN-alpha2b (LIPO+IFN) to affect the synthesis of ECM in open and reepithelialized wounds. Full-thickness skin wounds in 32 female Hartley guinea pigs (6 wounds per animal, 3 on each side) were made with an 8-mm biopsy punch. Each wound on the right side received 3,000 U LIPO+IFN, whereas wounds on the left side received cream containing empty liposomes. Histologic examination revealed a significant reduction in scar formation in LIPO+IFN-treated but not in vehicle-treated wounds. Northern analysis showed reductions in type I procollagen mRNA in healed wounds treated with LIPO+IFN (day 4 groups: 1596.9 +/- 207 vs. 3710.2 +/- 493 densitometry units, p < 0.01, n = 8). This was consistent with a reduction in the concentration of collagen in the tissue, assayed as 4-hydroxyproline (day 4 group: 38.5 +/- 3.8 vs. 54.5 +/- 3.9 microg per tissue, p < 0.01, n = 8). Even when applied to reepithelialized wounds, LIPO+IFN caused a marked reduction in type I collagen mRNA (1938.5 +/- 579 vs. 4085.7 +/- 1271 densitometry units, p < 0.01, n = 8). These findings support the concept of the early topical use of this antifibrogenic agent for treatment of dermal fibroproliferative disorders, such as hypertrophic scars.

Placental trophoblasts resist infection by multiple human immunodeficiency virus (HIV) type 1 variants even with cytomegalovirus coinfection but support HIV replication after provirus transfection.

Whether cell-free human immunodeficiency virus type 1 (HIV-1) can productively infect placental trophoblasts (which in turn could transmit the virus into the fetal circulation) is controversial but essential to know for the evaluation of alternative routes (such as cell-mediated infection or trophoblast damage). We have addressed infection factors such as cell purity, source, culture methods, and activation states as well as virus variant and detection methods to conclusively determine the outcome of trophoblast challenge by free virus. Pure (> 99.98%) populations of trophoblasts from 11 different placentas were challenged at a multiplicity of infection (MOI) as high as 6 with five different HIV-1 variants, three of which are non-syncytium-forming, macrophage-tropic isolates from infected infants, with and without coinfection with cytomegalovirus; these preparations were monitored for productive infection for up to 3 weeks after challenge by five different criteria, the most sensitive of which were cocultivation with target cells that can detect virus at an MOI of 10(-7) and HIV DNA PCR that detects 30 virus copies per 10(5) cells. Infection was never detected. However, molecularly cloned T-cell (pNL4-3)- and macrophage (pNLAD8)-tropic provirus plasmids, when transfected into primary trophoblasts, yielded productive infections, indicating that trophoblasts do not suppress late-stage virus replication and assembly. Because of the purity of the trophoblast preparations, the extended length of the infection culture period, the number of trophoblast preparations and virus types examined, the sensitivity of the bioassays and molecular detection assays, and the observations that trophoblasts can support virus replication from provirus, the results of this study strongly argue that free virus cannot infect primary villous trophoblasts.

Geographical variation in 18S rRNA gene sequence of Cryptosporidium parvum.

Using 18S rRNA as a probe, an EcoR1 fragment containing 1507 nt of 18S rRNA from C. parvum was identified, cloned and sequenced. Comparison of this sequence with the partial sequence of the small subunit rRNA of Cryptosporidium published by Johnson, Fielke, Lumb & Baverstock (International Journal for Parasitology 20: 141-147, 1990) and 1516 bp of the 18S rRNA nucleotide sequence of C. parvum published by Cai, Collins, McDonald & Thompson (Biochimica et Biophysica Acta, 1131: 317-320, 1992) revealed 97% and 91.6% sequence homology, respectively. These data suggest that differences exist among the same species of Cryptosporidium from different geographical areas.

Identification of low-molecular-mass coproantigens of Cryptosporidium parvum.

Cryptosporidium parvum coproantigens (CCAg) of 18 and 20 kDa were identified in the stool eluates of calves and humans infected with Cryptosporidium species. Monospecific antibodies raised against the 20-kDa antigen recognized both 18- and 20-kDa CCAg in all positive but no negative control samples. These antibodies reacted with C. parvum sporozoites in an immunofluorescence assay. Human immune sera recognized the 20-kDa antigen in infected calf stool eluates. Both 18- and 20-kDa CCAg remained intact in commonly used preservatives and at various temperatures. These CCAg may be useful in designing sensitive, reliable methods for diagnosing cryptosporidiosis.

Characterization of an immunodominant Giardia lamblia protein antigen related to alpha giardin.

The trophozoites of Giardia lamblia possess several protein antigens, predominant among them a protein of approximately 32,000 Da. In the present study, we used monospecific antibodies that recognize this protein to demonstrate its presence on a variety of G. lamblia isolates from human and animal sources. Immune electron microscopy was used to localize 32-kDa antigen on the trophozoite membrane and disk. Immunofluorescent assays employing monospecific antibodies confirmed the presence of 32-kDa antigen on the membrane and disk and its absence on flagella or nuclei. The N-terminal 17 amino acids of the 32-kDa antigen are identical to alpha-1-giardin, a protein component of microribbons on the ventral disk. These results suggest that the 32-kDa immunodominant trophozoite antigen is alpha-1-giardin.

Purification of Cryptosporidium oocysts and sporozoites by cesium chloride and Percoll gradients.

The lack of an adequate system for the in vitro cultivation of Cryptosporidium spp. has forced researchers to work on infected feces or tissues. Molecular and immunological analyses of Cryptosporidium stages must be preceded by complex preparatory steps involving the concentration, storage, purification, excystation of oocysts, and purification of sporozoites. This paper describes two new procedures for the purification of Cryptosporidium. The first, consisting of pretreatment of oocysts with sodium hypochlorite followed by concentration using a Percoll gradient, is suitable for nucleic acid analyses. The second, a concentration of untreated oocysts using a Cesium chloride gradient, is suitable for biochemical and immunological studies, but requires "fresh" oocysts.