Search

Type I interferons (IFN alpha and IFN beta) are presently used in the adjuvant treatment of several human cancers. However, these cytokines have demonstrated only modest success in breast cancer therapy, and research efforts have focused on improving their efficacy. Recent progress in understanding the molecular mechanisms underlying the antiproliferative effects of IFNs has identified the cytoplasmic transcription factor Stat1 as a critical mediator. It is, therefore, possible that IFN-induced growth inhibition of mammary epithelial cells is counteracted by other cytokines that also use Stat1. One such candidate IFN-antagonist with particular relevance to breast cancer is the mammotropic hormone prolactin (PRL). The main goal of this study was to examine whether PRL would interfere with type I IFN (IFN alpha/beta) signal transduction by competing for limited cytoplasmic Stat factors. A second aim was to test whether pretreatment of mammary tumor cell lines with IFN gamma could enhance the effect of IFN alpha/beta. By analyzing the effect of PRL on IFN alpha/beta-induced tyrosine phosphorylation of Stat proteins and their binding to IFN-regulated genes, we now report that costimulation of PRL receptors did not interfere with IFN alpha/beta signals in several human breast cancer cell lines, including T47D, MCF-7, and BT-20. Specifically, PRL did not affect IFN alpha/beta-induced tyrosine phosphorylation or heterodimerization of Stat1 and Stat2 in any cell line. Instead, IFN alpha/beta- and PRL-induced tyrosine phosphorylation of Stat1 was additive and occurred without evidence of competition for limited concentrations of cytoplasmic Stat1. A similar additive relationship was observed on IFN alpha/beta- and PRL-induced Stat3 tyrosine phosphorylation. Furthermore, electrophoretic mobility shift assays showed that type I IFNs induced predominantly Stat1-Stat2 or Stat1-Stat3 heteromeric complexes with various IFN-response elements of IFN-stimulated genes, whereas PRL induced Stat1 homodimers. Despite significant mutual use of Stats by IFNs and PRL, these results indicated a high degree of signaling specificity in the two receptor systems, and that cytoplasmic levels of Stat proteins were not limiting. Similarly, PRL did not interfere with the growth-inhibitory effect of IFN beta. On the other hand, the study indicated that pretreatment of human breast cancer cell lines with IFN gamma enhanced the growth-inhibitory action of type I IFNs, suggesting a possible avenue for improving the effect of type I IFNs in the treatment of breast cancer patients.

Exposure of human lymphoblastoid cell cultures to leukocyte interferon initiated the formation of membranous tubuloreticular inclusions (TRI) within the endoplasmic reticulum or perinuclear envelope. In four different cell lines originating from patients with lymphocytic cancers (Daudi, Raji, H-SB2, and SB), this unique ultrastructural effect displayed a log-linear relationship to increasing doses of interferon-alpha and was dose and time dependent. TRI morphogenesis began within 12 hr in Daudi or Raji cells exposed continuously to 500 IU of leukocyte interferon/ml, but only at 20 hr after 2- to 4-hr pulse exposures to 50 to 100 IU/ml. The TRI accumulation, determined by thin-section counts, reached maximal levels of up to 16% of cell sections within 48 to 72 hr. Experiments with Raji cells indicated a decrease in TRI formation during successive cell divisions; detection of TRI after a pulse of interferon was enhanced when DNA replication was arrested. TRI morphogenesis appeared to be independent of several other known biological actions of interferons. It manifested later than the induction of 2'-5'-oligoadenylate synthetase activity which has been correlated with establishment of the antiviral state and occurred in cell lines, including DaudiKIFNR, which resist the growth-inhibitory effects of leukocyte interferon. Participation of new polypeptide synthesis was indicated by experiments with inhibitors of transcription (actinomycin D) or translation (cycloheximide): TRI morphogenesis was blocked when actinomycin D was added 4 hr after interferon and was reduced when cycloheximide was added for the interval of 13 to 23 hr after interferon.

A surface-adherent cell population developed spontaneously from Friend erythroleukemia cells (FLC745) which previously had grown continuously in suspension cultures. The adherent cells have been transferred through more than 60 passages and apparently represent a stable variant, herein designated FLC745-Ad. The cell size, chromosome complement, and tumorigenicity in DBA/2 mice were similar to the parental FLC745 line, although the FLC745-Ad cultures grew at a slightly slower rate and synthesized less hemoglobin in response to dimethyl sulfoxide. Production of C-type particles was abundant in both suspension and adherent cells, but the latter appeared to contain more intracisternal A-type particles. Scanning electron microscopy disclosed that most adherent cells were globoid and nonspreading. They attached to the substrate by a circumferential lamellar skirt of unusual breadth. Adhesion of the FLC745-Ad cells was not influenced by serum concentration, trypsinization, 5-bromodeoxyuridine, or N6, O2'-dibutyryl adenosine 3':5'-monophosphate. Treatment with cytochalasin B increased the cell size and the area of cell attachment.