The somatic mutation frequency of the transforming growth factor beta receptor type II gene varies widely among different cancers with microsatellite instability

Disruption of the DNA mismatch repair system, characterized by microsatellite instability (MSI), plays an important role in the course of human carcinogenesis. Frequent somatic mutations in a polyadenine (poly(A)) tract and two GT repeats within the coding region of the transforming growth factor beta (TGFbeta) receptor II (RII) gene were reported in colorectal cancers with MSI. We examined mutations of RII in cancers of various organs with MSI and found deletions at the poly(A) tract in eight of nine (89%) gastric cancers and four of five (80%) colorectal cancers. In contrast, no mutations were found in cancers of the pancreas, endometrium, or lungs. These results suggest that TGFbeta-mediated growth control plays a very important role in the stomach and colorectum.

Modulation of thermal induction of hsp70 expression by Ku autoantigen or its individual subunits

Previously, we proposed a dual control mechanism for the regulation of the heat shock response in mammalian cells: a positive control mediated by the heat shock transcription factor HSF1 and a negative control mediated by the constitutive heat shock element-binding factor (CHBF). To study the physiological role of CHBF in the regulation of heat shock response, we purified CHBF to apparent homogeneity and showed it to be identical to the Ku autoantigen, a heterodimer consisting of 70-kDa (Ku-70) and 86-kDa (Ku-80) polypeptides. To study further the functional significance of Ku/CHBF in the cellular response to heat shock, we established rodent cell lines that stably and constitutively overexpressed one or both subunits of the human Ku protein, and examined the thermal induction of hsp70 and other heat shock proteins in these Ku-overexpressing ing cells. We show that expression of the human Ku-70 and Ku-80 subunits jointly or of the Ku-70 subunit alone specifically inhibits heat-induced hsp70 expression. Conversely, expression of human Ku-80 alone does not have this effect. Thermal induction of other heat shock proteins in all of the Ku-overexpressing cell lines appears not to be significantly affected, nor is the state of phosphorylation or the DNA-binding ability of HSF1 affected. These findings support a model in which hsp70 expression is controlled by a second regulatory factor in addition to the positive activation of HSF1. The Ku protein, specifically the Ku-70 subunit, is involved in the regulation of hsp70 gene expression.

Preoperative segmental localization of focal hepatic lesion on MRI

In order to evaluate the accuracy of preoperative segmental localization of hepatic mass lesions, 68 cases were prospectively studied and evaluated by the correlation of magnetic resonance imaging (MRI) and surgical findings. Right, middle and left hepatic veins were seen in 100%, 97% and 94% of the subjects respectively on MRI. The right and left portal veins were seen in 100% and 95.6% respectively. The accuracy in determining the segmental location of hepatic lesions was 89.7%. The accuracy could be increased to 91.3% if the lesion was located at single segment whereas it was only 86.4% when the lesion extended across several segments. It is possible to say that MRI is an important modality in the preoperative segmental localization of masses.

A constitutive heat shock element-binding factor is immunologically identical to the Ku autoantigen

Analysis of the heat shock element (HSE)-binding proteins in extracts of rodent cells, during heat shock and their post-heat shock recovery, indicates that the regulation of heat shock response involves a constitutive HSE-binding factor (CHBF), in addition to the heat-inducible heat shock factor HSF1. We purified the CHBF to apparent homogeneity from HeLa cells using column chromatographic techniques including an HSE oligonucleotide affinity column. The purified CHBF consists of two polypeptides with apparent molecular masses of 70 and 86 kDa. Immunoblot and gel mobility shift analysis verify that CHBF is identical or closely related to the Ku autoantigen. The DNA binding characteristics of CHBF to double-stranded or single-stranded DNA are similar to that of Ku autoantigen. In gel mobility shift analysis using purified CHBF and recombinant human HSF1, CHBF competes with HSF1 for the binding of DNA sequences containing HSEs in vitro. Furthermore, when Rat-1 cells were co-transfected with human Ku expression vectors and the hsp70-promoter-driven luciferase reporter gene, thermal induction of luciferase is significantly suppressed relative to cells transfected with only the hsp70-luciferase construct. These data suggest a role of CHBF (or Ku protein) in the regulation of heat response in vivo.

Suppression of heat-induced hsp70 expression by the 70-kDa subunit of the human Ku autoantigen

Expression of the 70-kDa polypeptide of human Ku autoantigen in rat cells is shown to suppress specifically the induction of hsp70 upon heat shock. Thermal induction of other heat shock proteins is not significantly affected, nor is the state of phosphorylation or the DNA-binding ability of the heat shock transcription factor HSF1. These findings support a model in which hsp70 gene expression is controlled by a second regulatory factor in addition to the positive activator HSF1. The Ku autoantigen, or a protein closely related to it, is likely to be involved in the regulation of hsp70 expression.

Heat shock protein hsp70 accelerates the recovery of heat-shocked mammalian cells through its modulation of heat shock transcription factor HSF1

The role of mammalian 70-kDa heat shock protein (hsp70) in regulating cellular response to heat shock was examined by using three closely related rat cells: control Rat-1 cells, thermotolerant Rat-1 (TT Rat-1) cells, and heat-resistant M21 cells, a derivative of Rat-1 cells that constitutively overexpress human hsp70. In all these cells, after a prescribed heat shock, the level of the phosphorylated form of heat shock transcription factor HSF1 and that of HSF1 capable of binding to its cognitive DNA sequence heat shock element (HSE) exhibit similar time dependence. The amount of a constitutive HSE-binding activity (CHBA), on the other hand, inversely correlates with those of the two aforementioned forms of HSF1. The recovery kinetics from heat shock are different for the three cell lines, with the thermal-resistant TT Rat-1 and M21 cells showing faster recovery in terms of the state of phosphorylation of HSF1 and its ability to bind HSE or in terms of the reappearance of CHBA. Treatment with okadaic acid, a serine/threonine phosphatase inhibitor, delays the recovery kinetics of Rat-1 cells but not that of thermal-resistant M21 cells. These results are interpreted in terms of a role for hsp70 in the recovery of heat-shocked mammalian cells.

Correlation of heat resistance and HSP-70A mRNA levels in human tumor cells measured by competitive quantitative polymerase chain reaction

PURPOSE

Several HSP-70 genes have been cloned and sequenced in human cells. Among these genes, the HSP-70A mRNA and protein levels correlate best with the development and decay of thermotolerance and intrinsic thermal sensitivity. Leukemic and nonleukemic human tumor cells express low levels of the normally heat inducible HSP-70A mRNA in control nonheated cells. Using a competitive quantitative polymerase chain reaction, we have measured the levels of mRNA for this gene and have correlated it with both transient and intrinsic thermal sensitivity of tumor cells. Such studies were also extended to tumor samples obtained from patients.

METHODS AND MATERIALS

In these studies, the plasmid phHSP-70 which contains the entire human HSP-70A gene was modified by the insertion of the T7 promoter at the 5'-end untranslated region as well as the insertion of a 23 bp synthetic linker at the BamH1 site in the promoter region of the HSP-70A gene. The PCR primers were located such that the amplified fragment contained the linker. Using the T7 polymerase, the HSP-70A mRNA was transcribed from this plasmid (phHSP-70L) in vitro. A known amount of HSP-70A mRNA was then added to the total RNA prepared from the cell samples or from the tumor tissues obtained from patients. Using the components of the PCR reaction plus known amounts of HSP-70A mRNA synthesized from phHSP-70L and unknown amounts of total cellular RNA, the samples were amplified and analysed on a denaturing acrylamide gel. The PCR products obtained from phHSP-70L were 23 bp larger than the PCR products obtained from the cell samples due to the addition of the synthetic linker to the HSP-70A gene in phHSP-70L and therefore, the two products could be easily distinguished from each other and quantitated. The alpha-32P-dCTP incorporated in each sample was quantitated by AMBIS Scanner. When the 32P-counts were equal in the known and the unknown samples, the amount of the HSP-70A mRNA was taken to be equal in the known and the unknown sample.

RESULTS

The results show that HSP-70A mRNA levels can be used to predict the survival levels during the development and decay of thermotolerance. In nonleukemic human tumor cell lines, there are as much as 40-50-fold induction of HSP-70A mRNA levels during the peak of thermotolerance. In leukemic cell lines, however, HSP-70A mRNA levels are induced only by three-fold during the same time period. These differences between the levels of HSP-70A mRNA positively correlate with the amount of tolerance development in leukemic and nonleukemic tumor cells. HSP-70A mRNA levels also vary in different tumor cells under nonheated conditions and there is a positive correlation between HSP-70A mRNA levels in nonleukemic human tumor cells and the level of their intrinsic thermal resistance.

CONCLUSION

HSP-70A mRNA levels can be used to predict the intrinsic thermal sensitivity of nonleukemic human tumor cells.

Selective expression of heat shock genes during differentiation of human myeloid leukemic cells

Several studies have indicated a role for heat shock proteins during development and differentiation. In these studies we have examined the patterns of activation of the HSP-70A, HSP-70B, HSP-70B' and HSP-28 mRNAs and proteins during the differentiation of immature human leukemic cells to more mature progenitors by several differentiation-inducing agents. K562 cells activate the mRNA for HSP-70A, HSP-70B' and HSP-28 genes in the presence of hemin or sodium butyrate as cells differentiate into late erythroblasts. K562 cells become progressively more resistant to killing by heat shock during their differentiation to late erythroblasts. Further, selective inhibition of HSP-70A by antisense oligonucleotides to reduce HSP-70 kDa accumulation results in consistent reduction of hemoglobin production by 25-30% in K562 cells exposed to hemin, HL-60 cells differentiate into mature macrophages within 3 days following addition of PMA. HSP-70A mRNA levels increase within the first 2 h of PMA treatment and remain elevated for up to 3 days during the cells' gradual differentiation into mature macrophages. PMA and sodium butyrate treatment also cause elevated levels of HSP-28 mRNA expression; this increase is barely detectable at 24 h but is considerable at 72 h when about 90% of HL-60 cells are differentiated into mature macrophages or monocytes. These studies show that HSP-70A, HSP-70B' and HSP-28 may have specific roles during the differentiation of blood cell progenitors into erythrocytes or macrophages. Further, differentiation alters the thermal sensitivity of leukemic cells.

Lower heat shock factor activation and binding and faster rate of HSP-70A messenger RNA turnover in heat sensitive human leukemias

Normal bone marrow progenitors and some leukemic cells develop only a limited amount of thermotolerance. Further, once developed, thermotolerance decays at a faster rate than that normally observed in cells of nonhemopoietic origin. Thermotolerance induction and maintenance correlates with reduced levels of expression of various M(r) 70,000 heat shock proteins (HSP-70) mRNAs after heat shock. We have now compared the accumulation of HSP-70 proteins in heat-shocked human leukemic cells KG-1, HL-60, and K562 to that in Ht1080, a colon carcinoma cell line. We have found reduced accumulation of HSP-70 proteins in all leukemic cells. The rate of decay of HSP-70A mRNA, measured following heat shock by using actinomycin D treatment to inhibit further RNA synthesis, was more rapid in KG-1 and HL-60 cells compared to Ht1080 cells. The half-life of HSP-70A mRNA was 2 h in KG-1 and HL-60 cells while in Ht1080 cells it was > 7 h. HSP-70A mRNA is known to decay with a half-life of 2 h in unheated cells; this is increased to > 7 h following heat shock. We therefore postulate that leukemic cells lack the mechanism to stabilize HSP-70A mRNA after heat shock. One postulated mechanism for HSP-70 mRNA decay rate is known to be due to the nucleotide sequences at the 3'-untranslated region. We examined the 3'-untranslated region in leukemic cells. No sequence variations, however, were observed at either the genomic or the complementary DNA levels between leukemic or nonleukemic tumor cells. Heat shock factor activation and binding by gel retardation assays showed that KG-1 and HL-60 cells had a reduced heat shock factor binding to the heat shock element when compared to K562 and Ht1080 cells. Furthermore, HSF-1 mRNA was found to be expressed at relatively lower levels in HL-60 cells when compared to Ht1080 or KG-1 cells. In conclusion, reduced HSP synthesis and accumulation of leukemic cells after heat shock correlates with the reduction in heat shock factor-heat shock element binding and a faster HSP-70A mRNA decay rate that is observed in these cells.