Leukemia Research Vol. 16, No. 6/7, pp. 597-605, 1992. Printed in Great Britain.
0145-2126/92 $5.00 + .00 Pergamon Press Ltd
H I G H CONSTITUTIVE EXPRESSION OF H E A T SHOCK PROTEIN 90o: IN HUMAN ACUTE LEUKEMIA CELLS YuJI YUFU, JUNJI NISHIMURA and HAJIME NAWATA Third Department of Internal Medicine, Faculty of Medicine, Kyushu University, Fukuoka, Japan (Received 5 July 1991. Revision accepted 30 November 1991) Abstract--The constitutive expression of the genes for four heat shock proteins (hsps) was studied in leukemia cell lines, cells obtained from patients with acute leukemia, and normal blood cells by means of Northern-blot analysis. Western-blot analysis with hsp90 antibody showed that the leukemia cells contained larger amounts of hsp90 than the normal peripheral mononuclear cells. The expression of the hsp90~ gene was enhanced in the leukemia cell lines and the acute leukemia cells from patients as compared with the normal blood cells. In contrast, the expression of the hsp90fl gene could hardly be recognized in either the acute leukemia cells or the normal blood cells. An increased expression of hsp70 gene was observed in only one patient. The expression of the hsp27 gene was enhanced in one-half the patients with common acute lymphoblastic leukemia. Thus, exclusively the hsp90o~gene was expressed highly in the leukemia cells, indicating its association with cellular proliferation. Key words: Leukemia cells, heat shock protein, hsp90, hsp70, hsp27, cellular proliferation.
INTRODUCTION
in leukemia cells. A recent report demonstrated a specific increase of hsp27 synthesis in common A L L [10]. Furthermore, the synthesis of hsp70 was enhanced in mitogen-activated T lymphocytes [11] and in tumor promoter-stimulated monocytes [12] in humans. Thus, there have been several findings which indicate a correlation between the constitutive synthesis of hsps and the growth or transformation of hematopoietic cells. However, few reports have studied the expression of hsp genes in leukemia cells obtained from patients as compared with normal blood cells. It is also of interest to examine whether the increased synthesis of hsps may be ascribed to an increased m R N A or to selective translation. In this study, we investigated the constitutive expression of hsp genes in normal hematopoietic cells, leukemia cell lines, and cells from patients with acute leukemia to detect any possible differences.
THE exposure of cells to elevated temperatures rapidly induces a group of specific polypeptides known as heat shock proteins (hsps). The most prominent hsps of mammalian cells have molecular weights of approximately 90000 (hsp90), 70000 (hsp70) and 27 000 (hsp27) [1-3]. The increased synthesis of hsps following heat shock is correlated in time with the development of heat resistance, indicating that a function of the hsps may be to protect the cells against heat-induced damage [4, 5]. Recent reports suggest that hsp cognates, which are constitutively synthesized and slightly heat-inducible, may play a role in cell growth and differentiation [6]. We reported previously that several human leukemia cell lines and a few leukemia cells obtained from patients with acute leukemia showed an elevated synthesis of hsp90 even without heat shock, which was markedly suppressed after the induction of differentiation [7, 8]. Hickey et al. isolated two distinct cDNAs encoding hsp90cr and hsp90fl [9]. It is unclear, however, which form of hsp90 is increased
MATERIALS AND METHODS Cells Mononuclear cells from the peripheral blood or bone marrow were obtained from 10 patients with acute leukemia and from 4 healthy donors by centrifugation on a Ficoll-Hypaque gradient. The majority of patient samples contained more than 70% leukemic cells. Clinical characteristics of the patients appear in Table 1. There were 5 men and 5 women aged 29-78 years. Leukemia was classified according to the recommendations of the French-American-British (FAB) Cooperative Group [13]. Five human
Abbreviations: hsp(s), heat shock protein(s); SDSPAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; CML, chronic myelogenous leukemia; AML, acute myeloid leukemia; A L L , acute lymphoblastic leukemia. Correspondence to: Yuji Yufu, M.D., Nakabaru Hospital, Internal Medicine, Mitarai 6, Shime-machi, Kasuyagun, Fukuoka 811-22, Japan. 597
Y. YUFU et al.
598
TABLE 1. CLINICAL DATA ON 10 LEUKEMIC PATIENTS
Patient No.
Age/sex
Diagnosis (FAB*)
1 2 3 4 5 6 7 8 9 10
78/M 42/F 37/M 53/F 45/M 57/F 44/M 29/F 34/M 53/F
AML (M4) AML (M2) AML (M2) AML (M7) AML (M5) AML (M2) ALL (L2) ALL (L2) ALL (L2) ALL (L1)
Immunophenotype
Common Common Common Common
ALL ALL ALL ALL
Leukemic cell (%) (Source) 82% 88% 88% 98% 75% 92% 76% 52% 96% 93%
(BMt) (BM) (BM) (PB~) (BM) (BM) (PB) (PB) (BM) (PB)
Disease state Initial Initial Refractory Refractory Relapse Refractory Refractory Relapse Initial Relapse
* FAB, French-American-British classification. t BM, bone marrow. $ PB, peripheral blood.
leukemia cell lines [14, 15], K562 (myeloid/erythroid blast crisis of CML), HL-60 (AML with promyelocytic differentiation), KG-1 (AML), U937 (histiocytic lymphoma) and MOLT-4 (thymic T-cell leukemia) were all provided by the Japanese Cancer Research Resources Bank (Tokyo, Japan). Cells were maintained in RPMI 1640 medium supplemented with 10% fetal calf serum. In some experiments, heat shock was given to K562 cells at 42°C for 120 min prior to the preparation of total RNA.
Western-blot analysis Whole cell extracts (50 ~tg of proteins) were separated by 10% SDS-PAGE according to the method of Laemmli [16] and transferred to a 0.45 Ixm nitrocellulose filter (Schleicher & Schuell Inc., Dassel, Germany). The filter was preincubated overnight at room temperature with blocking buffer (3% bovine serum albumin, 50 mM TrisHC1, pH 7.5,150 mM NaC1) and then overlaid with rabbit antiserum against mouse hsp90 (1 : 100 dilution), which was kindly provided by Dr S. Koyasu (The Tokyo Metropolitan Institute of Medical Science) [17], for 4 h at room temperature. Since major hsps are highly conserved proteins in evolution, rabbit antibody against mouse hsp90 can cross-react with human hsp90 [18]. After extensive washing with 0.2% NP-40 in 50 mM Tris-HC1, pH 7.5, 150 mM NaCl, [125I]protein A (30mCi/mg; Amersham International, Buckinghamshire, U.K.) at 1 ~tCi/ml in blocking buffer was added for 1 h. The filter was washed and exposed to Kodak X-Omat AR film (Eastman Kodak Co., Rochester, NY) with an intensifying screen at -70°C for autoradiography. Molecular weights of the relevant bands were determined by use of the molecular weight standards (Bio-Rad Laboratories, Richmond, CA). Northern-blot analysis Total cellular RNA was isolated by a guanidinium isothiocyanate procedure and cesium chloride centrifugation [19]. Following denaturation with 1 M glyoxal and 50% dimethyl sulfoxide, equal amounts of total RNA (10 ~tg) were fractionated on 1% agarose gel, transferred to nylon filters (Biodyne B, Pall Ultrafine Filtration Co., East Hills, NY), prehybridized and hybridized [19] with 32p-labeled plasmid DNA probes, pHS801, pHS811, pHS709 and pHS208, which contain cDNAs coding for human hsp90o;,
hsp90#, hsp70 and hsp27, respectively [9]. These plasmids were a gift of Drs E. Hickey and L. Weber (University of South Florida). After hybridization was carried out as previously described [7], the filters were processed for autoradiography. Quantitative measurements of the hybridization bands were made by scanning autoradiograms with a Jookoo PAN-802 densitometer (Tokyo, Japan). Data were analyzed by Student's t-test.
RESULTS We first examined the amount of constitutive hsp90 in leukemia cells and normal blood cells by Westernblot analysis with hsp90 antiserum. It was evident that all the leukemia cells examined including HL60 cells and the samples obtained from 5 patients with acute leukemia contained larger amounts of hsp90 than the mononuclear cells from the peripheral blood of normal subjects (Fig. 1). However, it was not known whether the increased hsp90 was hsp90c~, hsp90/3 or both, since the hsp90 antiserum used as a probe would react with both forms of hsp90. The expression of hsp90ol m R N A , which is markedly induced by heat shock, was rather intense under unstressed conditions in all leukemia cell lines (Fig. 2). The expression of hsp90# m R N A , which is only slightly heat-inducible, resembled that of hsp90tr m R N A although it was pronounced in unstressed K562 ceils (Fig. 2). Next, the expression of mRNAs for hsp90c~ and hsp90# was examined in mononuclear cells sampled from the normal peripheral blood and bone marrow (Fig. 3). The normal hematopoietic cells expressed much smaller amounts of mRNAs for both of hsp90o~ and hsp90/3 than K562 cells. When observed in more detail, it appeared that whereas the expression of hsp90tr m R N A was present to some extent in normal hematopoietic cells, that of hsp90/3 m R N A could
U
FIG. 1. Western-blot analysis of constitutive hsp90 in leukemia cells and normal blood cells probed with hsp90 antiserum. Equal amounts of proteins (50 gg) were applied to each lane. Cells received no heat treatment. PBMNC refers to peripheral blood mononuclear cells. Numbers at the top correspond to the patient numbers shown in Table 1.
599
FIG. 2. Northern-blot analysis of mRNAs for hsp90o: and hsp90/3 in various leukemia cell lines. K562(C) and K562(HS) are control K562 cells and heat-shocked K562 cells, respectively. No cell lines received treatment except for K562(HS).
600
FIG. 3. Northern-blot analysis of mRNAs for hsp90o~ and hsp90/3 in normal blood cells. Untreated K562 cells [K562(C)] were added for comparison. All blood cells were untreated. PBMNC and BMMNC refer to peripheral blood mononuclear cells and bone marrow mononuclear cells, respectively.
601
FIG. 4. Northern-blot analysis of mRNAs for various hsps in acute leukemia cells from the patients. (A) Levels of mRNAs for hsp90ol and hsp90/3. The lowest panel shows the bands of 28S ribosomal RNA (28SrRNA) to ensure the equality of the amounts of total RNA loaded. (B) Levels of hsp70 mRNA. (C) Levels of hsp27 mRNA. Numbers at the top correspond to the patient numbers shown in Table 1. Control [K562(C)] and heat-shocked [K562(HS)] K562 cells, and peripheral blood mononuclear cells (PBMNC) were added for comparison. No cells received treatment except for K562(HS) cells.
602
High hsp90o~expressionin leukemiacells hardly be recognized. Differing results were observed in the cells obtained from patients with acute leukemia (Fig. 4). As compared with mononuclear cells from normal peripheral blood, the expression of hsp90a~was easily detected in all acute leukemia cells. In particular, 7 of the 10 samples of leukemia cells (patients Nos. 1, 2, 3, 5, 6, 7 and 9) showed an increased expression of hsp90tr mRNA at levels comparable to those of K562 cells (Fig. 4A). Densitometric quantitation of the hybridization bands for the hsp90cr mRNA was made on the autoradiograms in Figs 3 and 4. The values were expressed as relative densities to that of control K562 cells. The acute leukemia cells (n = 10) expressed a significantly (p < 0.01) larger amount of hsp90tr mRNA than the normal mononuclear cells (n = 4). In contrast, the expression of hsp90fl mRNA, which was also strong in the K562 cells, was not increased in the acute leukemia cells (Fig. 4A). It was now evident that the expression of hsp90 genes differed between normal blood cells and acute leukemia cells, with a preferential expression of hsp90tr mRNA in the acute leukemia cells. We next investigated the constitutive expression of two other hsp genes, hsp70 and hsp27, in acute leukemia cells. The expression of hsp70 mRNA was detected in only one of the 10 samples, that for patient No. 1 (Fig. 4B). Two of 4 samples obtained from patients with common ALL (patient Nos. 7 and 9) and one of 6 samples of AML cells (patient No. 6) expressed a high level of hsp27 mRNA comparable to that of K562 cells (Fig. 4C). DISCUSSION In this study, we showed that acute leukemia cells synthesized larger amounts of constitutive hsp90 than did normal blood cells, and that the increased hsp90 was most likely hsp90a~, not hsp90fl. The hsps are induced by a variety of factors other than heat stress, among them anticancer agents [2,3,20]. Such exogenous factors generally induce hsp90 and hsp70 in parallel or hspT0 almost exclusively, but the increase in hsp90c~ in the acute leukemia cells was not accompanied by a corresponding increase of hsp70. In addition, the increased hsp90a~ was observed not only in refractory patients with leukemia but also in those who had relapsed or who were newly diagnosed and not yet exposed to any anticancer drug prior to sampling. These findings preclude the possibility that the increased expression of hsp90a~ gene may be an artifact. The enhanced expression of hsp90o: mRNA has also been reported in murine embryonal carcinoma cell lines [21] and in tumor cells from patients with
603
ovarian cancer [22] although that of hsp90fl was not examined. There are a few reports on some differences between the expression of the genes for hsp90o: and hsp90fl. In HeLa cells, the adenovirus E I A gene product and serum stimulate the transcription of hsp90a~ mRNA but not hsp90/3 mRNA [23, 24]. The synthesis of hsp90a~ is reported to decrease following the induction of differentiation whereas that of hsp90fl did not change in murine embryonal carcinoma cells [25]. These findings suggest that while the transcription of the hsp90tx gene may be regulated by cellular proliferation, that of the hsp90fl gene may be independent. However, other investigators have indicated that the hsp90fl gene exhibits a greater mitogen-stimulated induction than the hspg0a~ gene of human lymphocytes [26]. In this study, we noted that leukemia cell lines showed an increased expression of both the hsp90a~ and hspg0fl genes as compared with normal blood cells, whereas acute leukemia cells obtained from the patients actively expressed only the hsp90a~ gene. It is not known with which property of the acute leukemia cells the increased expression of hsp90a~ gene is associated. It has been demonstrated that hsp90 binds some oncogene products having tyrosine-specific kinase activity [27, 28] and also steroid hormone receptors [28, 29], possibly regulating the activity of its iigands. In addition, hsp90 is reported to interact with actin [17]. These findings suggest that hsp90 may function as a molecular shuttle for such important cellular proteins as hspT0 [30, 31]. It is known that hspg0c~ mRNA is expressed in normal blood cells and is not specific to leukemia cells, although the amount is small. Therefore, it is unlikely that hsp90~ is involved primarily in leukemogenesis. An increased amount of hsp90ol may be required for the active and indefinite proliferation of leukemia cells by the efficient and active transportation of important proteins related to growth regulation. The mechanism(s) by which the expression of the hsp90a, gene is increased in leukemia cells is unclear. The hsp90o: mRNA has been reported to be more stable than the mRNA encoding hspT0 in HeLa cells [24]. The hspg0a~ mRNA is transcribed more actively in murine embryonal carcinoma cells than in fibroblasts [21]. The increased expression of the hsp90a~ gene in leukemia cells may also be attributable to stabilization, active transcription or both. A short, highly conserved sequence, the heat shock element (HSE), is required for the transcriptional activation of heat shock genes [2, 32]. HSEs are binding sites for a specific transcription factor called the heat shock factor (HSF) [2, 33]. The HSE-binding activity of HSF increases dramatically in response to heat shock. Interestingly, murine embryonal carcinoma cells con-
604
Y. YUFUet al.
tain significant levels of HSE-binding activity under unstressed conditions [34]. It would be intriguing to determine whether the increased expression of hsp90cr gene in leukemia cells may be explained by the same mechanism. As with hsp90, the synthesis of hsp70 is increased in murine embryonal carcinoma cells [35], serumstimulated H e L a cells [36] and mitogen-stimulated human lymphocytes [11]. In addition, h e p a t o m a cells contain larger amounts of hsp70 than do the hepatocytes of normal rats [37]. H o w e v e r , except in one case in our study no e n h a n c e m e n t of hsp70 gene expression was observed in acute leukemia cells. hsp70 forms a multigene family consisting of at least 5 genes [38]. Differing results could be obtained depending on which c D N A for the hsp70 family is used as probe. In fact, conflicting results have been reported on changes in the expression of hsp70 gene in mitogen-stimulated h u m a n lymphocytes [11, 39]. Different probes should be used to obtain additional information on the expression of the hsp70 gene, A recent report indicates that an increased amount of hsp27 could be a m a r k e r for c o m m o n A L L [10]. We also recognized the increased expression of the hsp27 gene in c o m m o n A L L . A minority of A M L cells, however, contained significant levels of hsp27 gene. Additional cases should be examined to determine whether the enhanced expression of the hsp27 gene is confined to a specific type of acute leukemia.
Acknowledgements--We would like to thank Drs M. Katsuno and K. Shibata for providing clinical specimens. This work was supported in part by a Grant-in-Aid for Cancer Research (02151033) from the Ministry of Education, Science and Culture in Japan and a grant from the Fukuoka Cancer Society.
REFERENCES 1. Schlesinger M. J. (1986) Heat shock proteins: the search for functions. J. Cell Biol. 103, 321. 2. Burdon R. H. (1986) Heat shock and the heat shock proteins. Biochem. J. 240, 313. 3. Subjeck J. R. & Shyy T.-T. (1986) Stress protein systems of mammalian cells. Am. J. Physiol. 250 (Cell Physiol. 19), CI. 4. Li G. C. & Werb Z. (1982) Correlation between synthesis of heat shock proteins and development of thermotolerance in Chinese hamster fibroblasts. Proc. natn. Acad. Sci. U.S.A. 79, 3218. 5. Johnston R. N. & Kucey B. L. (1988) Competitive inhibition of hsp70 gene expression causes thermosensitivity. Science 242, 1551. 6. Carper S. W., Duffy J. J. & Gerner E. W. (1987) Heat shock proteins in thermotolerance and other cellular processes. Cancer Res. 47, 5249. 7. Yufu Y., Nishimura J., Takahira H., Ideguchi H. & Nawata H. (1989) Down-regulation of a Mr 90,000
heat shock cognate protein during granulocytic differentiation in HL-60 human leukemia cells. Cancer Res. 49, 2405. 8. Yufu Y., Nishimura J., Ideguchi H. & Nawata H. (1990) Heterogeneity of the synthesis of heat shock proteins in human leukaemic cells. Br. J. Cancer 62, 65. 9. Hickey E., Brandon S. E., Sadis S., Smale G. & Weber L. A. (1986) Molecular cloning of sequences encoding the human heat-shock proteins and their expression during hyperthermia. Gene 43, 147. 10. Strahler J. R., Kuick R., Eckerskorn C., Lottspeich F., Richardson B. C., Fox D. A., Stoolman L. M., Hanson C. A., Nichols D., Tueche H. J. & Hanash S. M. (1990) Identification of two related markers for common acute lymphoblastic leukaemia as heat shock proteins. J. clin. Invest. 85, 200. 11. Ferris D. K., Harel-Bellan A., Morimoto R. I., Welch W. J. & Farrar W. L. (1988) Mitogen and lymphokine stimulation of heat shock proteins in T lymphocytes. Proc. natn. Acad. Sci. U.S.A. 85, 3850. 12. Fincato G., Polentarutti N., Sica A., Mantovani A. & Colotta F. (1991) Expression of a heat-inducible gene of the HSP70 family in human myelomonocytic cells: regulation by bacterial products and cytokines. Blood 77, 579. 13. Bennet J. M., Catovsky D., Daniel M. T., Flandrin G., Galton D. A. G., Galnick H. R. & Sultan C. (1985) Proposed revised criteria for the classification of acute myeloid leukemia: a report of the French-AmericanBritish Cooperative Group. Ann. Int. Med. 103, 620. 14. Koeffler H. P. (1986) Human acute myeloid leukemia lines: models of leukemogenesis. Semin. Hemat. 23, 223. 15. Minowada J., Ohnuma T. & Moore G. E. (1972) Rosette-forming human lymphoid cell lines. I. Establishment and evidence for origin of thymus-derived lymphocytes. JNCI 49, 891. 16. Laemmli U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, Lond. 227, 680. 17. Koyasu S., Nishida E., Kadowaki T., Matsuzaki F., Iida K., Harada F., Kasuga M., Sakai H. & Yahara I. (1986) Two mammalian heat shock proteins, HSP90 and HSP100, are actin-binding proteins. Proc. natn. Acad. Sci. U.S.A, 83, 8054. 18. Kelley P. M. & Schlesinger M. J. (1982) Antibodies to two major chicken heat shock proteins cross-react with similar proteins in widely divergent species. Mol. Cell Biol. 2, 267. 19. Maniatis T., Fritsch E. F. & Sambrook J. J. (1982) In Molecular Cloning, p. 187. Cold Spring Harbor Laboratories, New York. 20. Schaefer E. L., Morimoto R. I., Theodorakis N. G. & Seidenfeld J. (1988) Chemical specificity for induction of stress response genes by DNA-damaging drugs in human adenocarcinoma cells. Carcinogenesis 9, 1733. 21. Legagneux V., Mezger V., Qu61ard C., Barnier J. V., Bensaude O. & Morange M. (1989) High constitutive transcription of HSP86 gene in murine embryonal carcinoma cells. Differentiation 41, 42. 22. Mileo A. M., Fanuele M., Battaglia F., Scambia G., Benedetti-Panici P., Mancuso S. & Ferrini U. (1990) Selective over-expression of mRNA coding for 90 kDa stress-protein in human ovarian cancer. Anticancer Res. 510, 903.
High hsp90tr expression in leukemia cells 23. Simon M. C., Kitchener K., Kao H.-T., Hickey E., Weber L., Voellmy R., Heintz N. & Nevins J. R. (1987) Selective induction of human heat shock gene transcription by the adenovirus E1A gene products, including the 12S E1A product. Mol. Cell Biol. 7, 2884. 24. Hickey E., Brandon S. E., Smale G., Lloyd D. & Weber L. A. (1989) Sequence and regulation of a gene encoding a human 89-kilodalton heat shock protein. Mol. Cell Biol. 9, 2615. 25. Barnier J. V., Bensaude O., Morange M. & Babinet C. (1987) Mouse 89 kD heat shock protein: two polypeptides with distinct developmental regulation. Expl Cell Res. 170, 186. 26. Hansen L. K., Houchins J. P. & O'Leary J. J. (1991) Differential regulation of HSC70, HSP70, HSP90oL, and HSP90/3 mRNA expression by mitogen activation and heat shock in human lymphocytes. Expl Cell Res. 192, 587. 27. Oppermann H., Levinson W. & Bishop J. M. (1981) A cellular protein that associates with the transforming protein of Rous sarcoma virus is also a heat-shock protein. Proc. natn. Acad. Sci. U.S.A. 78, 1067. 28. Ziemiecki A., Catelli M.-G., Joab I. & Moncharmont B. (1986) Association of the heat shock protein HSP90 with steroid hormone receptors and tyrosine kinase oncogene products. Biochem. biophys. Res. Commun. 138, 1298. 29. Schuh S., Yonemoto W., Brugge J., Bauer V. J., Riehl R. M., Sullivan W. P. & Toft D, O. (1985) A 90,000dalton binding protein common to both steroid receptors and the Rous sarcoma virus transforming protein, pp 60v-src. J. biol. Chem. 260, 14292. 30. Deshaies R. J., Koch B. D., Werner-Washburne M., Craig E. A. & Schekman R. (1988) A subfamily of
31. 32. 33.
34. 35.
36. 37.
38.
39.
605
stress proteins facilitates translocation of secretory and mitochondrial precursor polypeptides. Nature, Lond. 332, 800. Chirico W. J., Waters M. G. & Blobel G. (1988) 70K heat shock related proteins stimulate protein translocation into microsomes. Nature, Lond. 332, 805. Pelham H. R. B. (1982) A regulatory upstream promoter element in the Drosophila hsp70 heat-shock gene. Cell 30, 517. Wu C., Wilson S., Walker B., Dawid I., Paisley T., Zimarino V. & Ueda H. (1987) Purification and properties of Drosophila heat shock activator protein. Science 238, 1247. Mezger V., Bensaude O. & Morange M. (1989) Unusual levels of heat-shock element-binding activity in embryonal carcinoma cells. Mol, Cell Biol. 9, 3888. Bensaude O. & Morange M. (1983) Spontaneous high expression of heat-shock proteins in mouse embryonal carcinoma ceils and ectoderm from day 8 mouse embryo. EMBO J. 2, 173. Wu B. J. & Morimoto R. 1. (1985) Transcription of the human hsp70 gene is induced by serum stimulation. Proc. hath. Acad. Sci. U.S.A. 82, 6070. Cairo G., Schiaffonati L., Rappocciolo E., Tacchini L. & Bernelli-Zazzera A. (1989) Expression of different members of heat shock protein 70 gene family in liver and hepatomas. Hepatology 9, 740. Minota S., Cameron B., Welch W. J. & Winfield J. B. (1988) Autoantibodies to the constitutive 73-kD member of the hsp70 family of heat shock proteins in systemic lupus erythematosus. J. expl Med. 168, 1475. Kaczmarek L., Calabretta B., Kao H.-T., Heintz N., Nevins J. & Baserga R. (1987) Control of hsp70 RNA levels in human lymphocytes. J, Cell Biol. 104, 183.
0145-2126/92 $5.00 + .00 Pergamon Press Ltd
H I G H CONSTITUTIVE EXPRESSION OF H E A T SHOCK PROTEIN 90o: IN HUMAN ACUTE LEUKEMIA CELLS YuJI YUFU, JUNJI NISHIMURA and HAJIME NAWATA Third Department of Internal Medicine, Faculty of Medicine, Kyushu University, Fukuoka, Japan (Received 5 July 1991. Revision accepted 30 November 1991) Abstract--The constitutive expression of the genes for four heat shock proteins (hsps) was studied in leukemia cell lines, cells obtained from patients with acute leukemia, and normal blood cells by means of Northern-blot analysis. Western-blot analysis with hsp90 antibody showed that the leukemia cells contained larger amounts of hsp90 than the normal peripheral mononuclear cells. The expression of the hsp90~ gene was enhanced in the leukemia cell lines and the acute leukemia cells from patients as compared with the normal blood cells. In contrast, the expression of the hsp90fl gene could hardly be recognized in either the acute leukemia cells or the normal blood cells. An increased expression of hsp70 gene was observed in only one patient. The expression of the hsp27 gene was enhanced in one-half the patients with common acute lymphoblastic leukemia. Thus, exclusively the hsp90o~gene was expressed highly in the leukemia cells, indicating its association with cellular proliferation. Key words: Leukemia cells, heat shock protein, hsp90, hsp70, hsp27, cellular proliferation.
INTRODUCTION
in leukemia cells. A recent report demonstrated a specific increase of hsp27 synthesis in common A L L [10]. Furthermore, the synthesis of hsp70 was enhanced in mitogen-activated T lymphocytes [11] and in tumor promoter-stimulated monocytes [12] in humans. Thus, there have been several findings which indicate a correlation between the constitutive synthesis of hsps and the growth or transformation of hematopoietic cells. However, few reports have studied the expression of hsp genes in leukemia cells obtained from patients as compared with normal blood cells. It is also of interest to examine whether the increased synthesis of hsps may be ascribed to an increased m R N A or to selective translation. In this study, we investigated the constitutive expression of hsp genes in normal hematopoietic cells, leukemia cell lines, and cells from patients with acute leukemia to detect any possible differences.
THE exposure of cells to elevated temperatures rapidly induces a group of specific polypeptides known as heat shock proteins (hsps). The most prominent hsps of mammalian cells have molecular weights of approximately 90000 (hsp90), 70000 (hsp70) and 27 000 (hsp27) [1-3]. The increased synthesis of hsps following heat shock is correlated in time with the development of heat resistance, indicating that a function of the hsps may be to protect the cells against heat-induced damage [4, 5]. Recent reports suggest that hsp cognates, which are constitutively synthesized and slightly heat-inducible, may play a role in cell growth and differentiation [6]. We reported previously that several human leukemia cell lines and a few leukemia cells obtained from patients with acute leukemia showed an elevated synthesis of hsp90 even without heat shock, which was markedly suppressed after the induction of differentiation [7, 8]. Hickey et al. isolated two distinct cDNAs encoding hsp90cr and hsp90fl [9]. It is unclear, however, which form of hsp90 is increased
MATERIALS AND METHODS Cells Mononuclear cells from the peripheral blood or bone marrow were obtained from 10 patients with acute leukemia and from 4 healthy donors by centrifugation on a Ficoll-Hypaque gradient. The majority of patient samples contained more than 70% leukemic cells. Clinical characteristics of the patients appear in Table 1. There were 5 men and 5 women aged 29-78 years. Leukemia was classified according to the recommendations of the French-American-British (FAB) Cooperative Group [13]. Five human
Abbreviations: hsp(s), heat shock protein(s); SDSPAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; CML, chronic myelogenous leukemia; AML, acute myeloid leukemia; A L L , acute lymphoblastic leukemia. Correspondence to: Yuji Yufu, M.D., Nakabaru Hospital, Internal Medicine, Mitarai 6, Shime-machi, Kasuyagun, Fukuoka 811-22, Japan. 597
Y. YUFU et al.
598
TABLE 1. CLINICAL DATA ON 10 LEUKEMIC PATIENTS
Patient No.
Age/sex
Diagnosis (FAB*)
1 2 3 4 5 6 7 8 9 10
78/M 42/F 37/M 53/F 45/M 57/F 44/M 29/F 34/M 53/F
AML (M4) AML (M2) AML (M2) AML (M7) AML (M5) AML (M2) ALL (L2) ALL (L2) ALL (L2) ALL (L1)
Immunophenotype
Common Common Common Common
ALL ALL ALL ALL
Leukemic cell (%) (Source) 82% 88% 88% 98% 75% 92% 76% 52% 96% 93%
(BMt) (BM) (BM) (PB~) (BM) (BM) (PB) (PB) (BM) (PB)
Disease state Initial Initial Refractory Refractory Relapse Refractory Refractory Relapse Initial Relapse
* FAB, French-American-British classification. t BM, bone marrow. $ PB, peripheral blood.
leukemia cell lines [14, 15], K562 (myeloid/erythroid blast crisis of CML), HL-60 (AML with promyelocytic differentiation), KG-1 (AML), U937 (histiocytic lymphoma) and MOLT-4 (thymic T-cell leukemia) were all provided by the Japanese Cancer Research Resources Bank (Tokyo, Japan). Cells were maintained in RPMI 1640 medium supplemented with 10% fetal calf serum. In some experiments, heat shock was given to K562 cells at 42°C for 120 min prior to the preparation of total RNA.
Western-blot analysis Whole cell extracts (50 ~tg of proteins) were separated by 10% SDS-PAGE according to the method of Laemmli [16] and transferred to a 0.45 Ixm nitrocellulose filter (Schleicher & Schuell Inc., Dassel, Germany). The filter was preincubated overnight at room temperature with blocking buffer (3% bovine serum albumin, 50 mM TrisHC1, pH 7.5,150 mM NaC1) and then overlaid with rabbit antiserum against mouse hsp90 (1 : 100 dilution), which was kindly provided by Dr S. Koyasu (The Tokyo Metropolitan Institute of Medical Science) [17], for 4 h at room temperature. Since major hsps are highly conserved proteins in evolution, rabbit antibody against mouse hsp90 can cross-react with human hsp90 [18]. After extensive washing with 0.2% NP-40 in 50 mM Tris-HC1, pH 7.5, 150 mM NaCl, [125I]protein A (30mCi/mg; Amersham International, Buckinghamshire, U.K.) at 1 ~tCi/ml in blocking buffer was added for 1 h. The filter was washed and exposed to Kodak X-Omat AR film (Eastman Kodak Co., Rochester, NY) with an intensifying screen at -70°C for autoradiography. Molecular weights of the relevant bands were determined by use of the molecular weight standards (Bio-Rad Laboratories, Richmond, CA). Northern-blot analysis Total cellular RNA was isolated by a guanidinium isothiocyanate procedure and cesium chloride centrifugation [19]. Following denaturation with 1 M glyoxal and 50% dimethyl sulfoxide, equal amounts of total RNA (10 ~tg) were fractionated on 1% agarose gel, transferred to nylon filters (Biodyne B, Pall Ultrafine Filtration Co., East Hills, NY), prehybridized and hybridized [19] with 32p-labeled plasmid DNA probes, pHS801, pHS811, pHS709 and pHS208, which contain cDNAs coding for human hsp90o;,
hsp90#, hsp70 and hsp27, respectively [9]. These plasmids were a gift of Drs E. Hickey and L. Weber (University of South Florida). After hybridization was carried out as previously described [7], the filters were processed for autoradiography. Quantitative measurements of the hybridization bands were made by scanning autoradiograms with a Jookoo PAN-802 densitometer (Tokyo, Japan). Data were analyzed by Student's t-test.
RESULTS We first examined the amount of constitutive hsp90 in leukemia cells and normal blood cells by Westernblot analysis with hsp90 antiserum. It was evident that all the leukemia cells examined including HL60 cells and the samples obtained from 5 patients with acute leukemia contained larger amounts of hsp90 than the mononuclear cells from the peripheral blood of normal subjects (Fig. 1). However, it was not known whether the increased hsp90 was hsp90c~, hsp90/3 or both, since the hsp90 antiserum used as a probe would react with both forms of hsp90. The expression of hsp90ol m R N A , which is markedly induced by heat shock, was rather intense under unstressed conditions in all leukemia cell lines (Fig. 2). The expression of hsp90# m R N A , which is only slightly heat-inducible, resembled that of hsp90tr m R N A although it was pronounced in unstressed K562 ceils (Fig. 2). Next, the expression of mRNAs for hsp90c~ and hsp90# was examined in mononuclear cells sampled from the normal peripheral blood and bone marrow (Fig. 3). The normal hematopoietic cells expressed much smaller amounts of mRNAs for both of hsp90o~ and hsp90/3 than K562 cells. When observed in more detail, it appeared that whereas the expression of hsp90tr m R N A was present to some extent in normal hematopoietic cells, that of hsp90/3 m R N A could
U
FIG. 1. Western-blot analysis of constitutive hsp90 in leukemia cells and normal blood cells probed with hsp90 antiserum. Equal amounts of proteins (50 gg) were applied to each lane. Cells received no heat treatment. PBMNC refers to peripheral blood mononuclear cells. Numbers at the top correspond to the patient numbers shown in Table 1.
599
FIG. 2. Northern-blot analysis of mRNAs for hsp90o: and hsp90/3 in various leukemia cell lines. K562(C) and K562(HS) are control K562 cells and heat-shocked K562 cells, respectively. No cell lines received treatment except for K562(HS).
600
FIG. 3. Northern-blot analysis of mRNAs for hsp90o~ and hsp90/3 in normal blood cells. Untreated K562 cells [K562(C)] were added for comparison. All blood cells were untreated. PBMNC and BMMNC refer to peripheral blood mononuclear cells and bone marrow mononuclear cells, respectively.
601
FIG. 4. Northern-blot analysis of mRNAs for various hsps in acute leukemia cells from the patients. (A) Levels of mRNAs for hsp90ol and hsp90/3. The lowest panel shows the bands of 28S ribosomal RNA (28SrRNA) to ensure the equality of the amounts of total RNA loaded. (B) Levels of hsp70 mRNA. (C) Levels of hsp27 mRNA. Numbers at the top correspond to the patient numbers shown in Table 1. Control [K562(C)] and heat-shocked [K562(HS)] K562 cells, and peripheral blood mononuclear cells (PBMNC) were added for comparison. No cells received treatment except for K562(HS) cells.
602
High hsp90o~expressionin leukemiacells hardly be recognized. Differing results were observed in the cells obtained from patients with acute leukemia (Fig. 4). As compared with mononuclear cells from normal peripheral blood, the expression of hsp90a~was easily detected in all acute leukemia cells. In particular, 7 of the 10 samples of leukemia cells (patients Nos. 1, 2, 3, 5, 6, 7 and 9) showed an increased expression of hsp90tr mRNA at levels comparable to those of K562 cells (Fig. 4A). Densitometric quantitation of the hybridization bands for the hsp90cr mRNA was made on the autoradiograms in Figs 3 and 4. The values were expressed as relative densities to that of control K562 cells. The acute leukemia cells (n = 10) expressed a significantly (p < 0.01) larger amount of hsp90tr mRNA than the normal mononuclear cells (n = 4). In contrast, the expression of hsp90fl mRNA, which was also strong in the K562 cells, was not increased in the acute leukemia cells (Fig. 4A). It was now evident that the expression of hsp90 genes differed between normal blood cells and acute leukemia cells, with a preferential expression of hsp90tr mRNA in the acute leukemia cells. We next investigated the constitutive expression of two other hsp genes, hsp70 and hsp27, in acute leukemia cells. The expression of hsp70 mRNA was detected in only one of the 10 samples, that for patient No. 1 (Fig. 4B). Two of 4 samples obtained from patients with common ALL (patient Nos. 7 and 9) and one of 6 samples of AML cells (patient No. 6) expressed a high level of hsp27 mRNA comparable to that of K562 cells (Fig. 4C). DISCUSSION In this study, we showed that acute leukemia cells synthesized larger amounts of constitutive hsp90 than did normal blood cells, and that the increased hsp90 was most likely hsp90a~, not hsp90fl. The hsps are induced by a variety of factors other than heat stress, among them anticancer agents [2,3,20]. Such exogenous factors generally induce hsp90 and hsp70 in parallel or hspT0 almost exclusively, but the increase in hsp90c~ in the acute leukemia cells was not accompanied by a corresponding increase of hsp70. In addition, the increased hsp90a~ was observed not only in refractory patients with leukemia but also in those who had relapsed or who were newly diagnosed and not yet exposed to any anticancer drug prior to sampling. These findings preclude the possibility that the increased expression of hsp90a~ gene may be an artifact. The enhanced expression of hsp90o: mRNA has also been reported in murine embryonal carcinoma cell lines [21] and in tumor cells from patients with
603
ovarian cancer [22] although that of hsp90fl was not examined. There are a few reports on some differences between the expression of the genes for hsp90o: and hsp90fl. In HeLa cells, the adenovirus E I A gene product and serum stimulate the transcription of hsp90a~ mRNA but not hsp90/3 mRNA [23, 24]. The synthesis of hsp90a~ is reported to decrease following the induction of differentiation whereas that of hsp90fl did not change in murine embryonal carcinoma cells [25]. These findings suggest that while the transcription of the hsp90tx gene may be regulated by cellular proliferation, that of the hsp90fl gene may be independent. However, other investigators have indicated that the hsp90fl gene exhibits a greater mitogen-stimulated induction than the hspg0a~ gene of human lymphocytes [26]. In this study, we noted that leukemia cell lines showed an increased expression of both the hsp90a~ and hspg0fl genes as compared with normal blood cells, whereas acute leukemia cells obtained from the patients actively expressed only the hsp90a~ gene. It is not known with which property of the acute leukemia cells the increased expression of hsp90a~ gene is associated. It has been demonstrated that hsp90 binds some oncogene products having tyrosine-specific kinase activity [27, 28] and also steroid hormone receptors [28, 29], possibly regulating the activity of its iigands. In addition, hsp90 is reported to interact with actin [17]. These findings suggest that hsp90 may function as a molecular shuttle for such important cellular proteins as hspT0 [30, 31]. It is known that hspg0c~ mRNA is expressed in normal blood cells and is not specific to leukemia cells, although the amount is small. Therefore, it is unlikely that hsp90~ is involved primarily in leukemogenesis. An increased amount of hsp90ol may be required for the active and indefinite proliferation of leukemia cells by the efficient and active transportation of important proteins related to growth regulation. The mechanism(s) by which the expression of the hsp90a, gene is increased in leukemia cells is unclear. The hsp90o: mRNA has been reported to be more stable than the mRNA encoding hspT0 in HeLa cells [24]. The hspg0a~ mRNA is transcribed more actively in murine embryonal carcinoma cells than in fibroblasts [21]. The increased expression of the hsp90a~ gene in leukemia cells may also be attributable to stabilization, active transcription or both. A short, highly conserved sequence, the heat shock element (HSE), is required for the transcriptional activation of heat shock genes [2, 32]. HSEs are binding sites for a specific transcription factor called the heat shock factor (HSF) [2, 33]. The HSE-binding activity of HSF increases dramatically in response to heat shock. Interestingly, murine embryonal carcinoma cells con-
604
Y. YUFUet al.
tain significant levels of HSE-binding activity under unstressed conditions [34]. It would be intriguing to determine whether the increased expression of hsp90cr gene in leukemia cells may be explained by the same mechanism. As with hsp90, the synthesis of hsp70 is increased in murine embryonal carcinoma cells [35], serumstimulated H e L a cells [36] and mitogen-stimulated human lymphocytes [11]. In addition, h e p a t o m a cells contain larger amounts of hsp70 than do the hepatocytes of normal rats [37]. H o w e v e r , except in one case in our study no e n h a n c e m e n t of hsp70 gene expression was observed in acute leukemia cells. hsp70 forms a multigene family consisting of at least 5 genes [38]. Differing results could be obtained depending on which c D N A for the hsp70 family is used as probe. In fact, conflicting results have been reported on changes in the expression of hsp70 gene in mitogen-stimulated h u m a n lymphocytes [11, 39]. Different probes should be used to obtain additional information on the expression of the hsp70 gene, A recent report indicates that an increased amount of hsp27 could be a m a r k e r for c o m m o n A L L [10]. We also recognized the increased expression of the hsp27 gene in c o m m o n A L L . A minority of A M L cells, however, contained significant levels of hsp27 gene. Additional cases should be examined to determine whether the enhanced expression of the hsp27 gene is confined to a specific type of acute leukemia.
Acknowledgements--We would like to thank Drs M. Katsuno and K. Shibata for providing clinical specimens. This work was supported in part by a Grant-in-Aid for Cancer Research (02151033) from the Ministry of Education, Science and Culture in Japan and a grant from the Fukuoka Cancer Society.
REFERENCES 1. Schlesinger M. J. (1986) Heat shock proteins: the search for functions. J. Cell Biol. 103, 321. 2. Burdon R. H. (1986) Heat shock and the heat shock proteins. Biochem. J. 240, 313. 3. Subjeck J. R. & Shyy T.-T. (1986) Stress protein systems of mammalian cells. Am. J. Physiol. 250 (Cell Physiol. 19), CI. 4. Li G. C. & Werb Z. (1982) Correlation between synthesis of heat shock proteins and development of thermotolerance in Chinese hamster fibroblasts. Proc. natn. Acad. Sci. U.S.A. 79, 3218. 5. Johnston R. N. & Kucey B. L. (1988) Competitive inhibition of hsp70 gene expression causes thermosensitivity. Science 242, 1551. 6. Carper S. W., Duffy J. J. & Gerner E. W. (1987) Heat shock proteins in thermotolerance and other cellular processes. Cancer Res. 47, 5249. 7. Yufu Y., Nishimura J., Takahira H., Ideguchi H. & Nawata H. (1989) Down-regulation of a Mr 90,000
heat shock cognate protein during granulocytic differentiation in HL-60 human leukemia cells. Cancer Res. 49, 2405. 8. Yufu Y., Nishimura J., Ideguchi H. & Nawata H. (1990) Heterogeneity of the synthesis of heat shock proteins in human leukaemic cells. Br. J. Cancer 62, 65. 9. Hickey E., Brandon S. E., Sadis S., Smale G. & Weber L. A. (1986) Molecular cloning of sequences encoding the human heat-shock proteins and their expression during hyperthermia. Gene 43, 147. 10. Strahler J. R., Kuick R., Eckerskorn C., Lottspeich F., Richardson B. C., Fox D. A., Stoolman L. M., Hanson C. A., Nichols D., Tueche H. J. & Hanash S. M. (1990) Identification of two related markers for common acute lymphoblastic leukaemia as heat shock proteins. J. clin. Invest. 85, 200. 11. Ferris D. K., Harel-Bellan A., Morimoto R. I., Welch W. J. & Farrar W. L. (1988) Mitogen and lymphokine stimulation of heat shock proteins in T lymphocytes. Proc. natn. Acad. Sci. U.S.A. 85, 3850. 12. Fincato G., Polentarutti N., Sica A., Mantovani A. & Colotta F. (1991) Expression of a heat-inducible gene of the HSP70 family in human myelomonocytic cells: regulation by bacterial products and cytokines. Blood 77, 579. 13. Bennet J. M., Catovsky D., Daniel M. T., Flandrin G., Galton D. A. G., Galnick H. R. & Sultan C. (1985) Proposed revised criteria for the classification of acute myeloid leukemia: a report of the French-AmericanBritish Cooperative Group. Ann. Int. Med. 103, 620. 14. Koeffler H. P. (1986) Human acute myeloid leukemia lines: models of leukemogenesis. Semin. Hemat. 23, 223. 15. Minowada J., Ohnuma T. & Moore G. E. (1972) Rosette-forming human lymphoid cell lines. I. Establishment and evidence for origin of thymus-derived lymphocytes. JNCI 49, 891. 16. Laemmli U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, Lond. 227, 680. 17. Koyasu S., Nishida E., Kadowaki T., Matsuzaki F., Iida K., Harada F., Kasuga M., Sakai H. & Yahara I. (1986) Two mammalian heat shock proteins, HSP90 and HSP100, are actin-binding proteins. Proc. natn. Acad. Sci. U.S.A, 83, 8054. 18. Kelley P. M. & Schlesinger M. J. (1982) Antibodies to two major chicken heat shock proteins cross-react with similar proteins in widely divergent species. Mol. Cell Biol. 2, 267. 19. Maniatis T., Fritsch E. F. & Sambrook J. J. (1982) In Molecular Cloning, p. 187. Cold Spring Harbor Laboratories, New York. 20. Schaefer E. L., Morimoto R. I., Theodorakis N. G. & Seidenfeld J. (1988) Chemical specificity for induction of stress response genes by DNA-damaging drugs in human adenocarcinoma cells. Carcinogenesis 9, 1733. 21. Legagneux V., Mezger V., Qu61ard C., Barnier J. V., Bensaude O. & Morange M. (1989) High constitutive transcription of HSP86 gene in murine embryonal carcinoma cells. Differentiation 41, 42. 22. Mileo A. M., Fanuele M., Battaglia F., Scambia G., Benedetti-Panici P., Mancuso S. & Ferrini U. (1990) Selective over-expression of mRNA coding for 90 kDa stress-protein in human ovarian cancer. Anticancer Res. 510, 903.
High hsp90tr expression in leukemia cells 23. Simon M. C., Kitchener K., Kao H.-T., Hickey E., Weber L., Voellmy R., Heintz N. & Nevins J. R. (1987) Selective induction of human heat shock gene transcription by the adenovirus E1A gene products, including the 12S E1A product. Mol. Cell Biol. 7, 2884. 24. Hickey E., Brandon S. E., Smale G., Lloyd D. & Weber L. A. (1989) Sequence and regulation of a gene encoding a human 89-kilodalton heat shock protein. Mol. Cell Biol. 9, 2615. 25. Barnier J. V., Bensaude O., Morange M. & Babinet C. (1987) Mouse 89 kD heat shock protein: two polypeptides with distinct developmental regulation. Expl Cell Res. 170, 186. 26. Hansen L. K., Houchins J. P. & O'Leary J. J. (1991) Differential regulation of HSC70, HSP70, HSP90oL, and HSP90/3 mRNA expression by mitogen activation and heat shock in human lymphocytes. Expl Cell Res. 192, 587. 27. Oppermann H., Levinson W. & Bishop J. M. (1981) A cellular protein that associates with the transforming protein of Rous sarcoma virus is also a heat-shock protein. Proc. natn. Acad. Sci. U.S.A. 78, 1067. 28. Ziemiecki A., Catelli M.-G., Joab I. & Moncharmont B. (1986) Association of the heat shock protein HSP90 with steroid hormone receptors and tyrosine kinase oncogene products. Biochem. biophys. Res. Commun. 138, 1298. 29. Schuh S., Yonemoto W., Brugge J., Bauer V. J., Riehl R. M., Sullivan W. P. & Toft D, O. (1985) A 90,000dalton binding protein common to both steroid receptors and the Rous sarcoma virus transforming protein, pp 60v-src. J. biol. Chem. 260, 14292. 30. Deshaies R. J., Koch B. D., Werner-Washburne M., Craig E. A. & Schekman R. (1988) A subfamily of
31. 32. 33.
34. 35.
36. 37.
38.
39.
605
stress proteins facilitates translocation of secretory and mitochondrial precursor polypeptides. Nature, Lond. 332, 800. Chirico W. J., Waters M. G. & Blobel G. (1988) 70K heat shock related proteins stimulate protein translocation into microsomes. Nature, Lond. 332, 805. Pelham H. R. B. (1982) A regulatory upstream promoter element in the Drosophila hsp70 heat-shock gene. Cell 30, 517. Wu C., Wilson S., Walker B., Dawid I., Paisley T., Zimarino V. & Ueda H. (1987) Purification and properties of Drosophila heat shock activator protein. Science 238, 1247. Mezger V., Bensaude O. & Morange M. (1989) Unusual levels of heat-shock element-binding activity in embryonal carcinoma cells. Mol, Cell Biol. 9, 3888. Bensaude O. & Morange M. (1983) Spontaneous high expression of heat-shock proteins in mouse embryonal carcinoma ceils and ectoderm from day 8 mouse embryo. EMBO J. 2, 173. Wu B. J. & Morimoto R. 1. (1985) Transcription of the human hsp70 gene is induced by serum stimulation. Proc. hath. Acad. Sci. U.S.A. 82, 6070. Cairo G., Schiaffonati L., Rappocciolo E., Tacchini L. & Bernelli-Zazzera A. (1989) Expression of different members of heat shock protein 70 gene family in liver and hepatomas. Hepatology 9, 740. Minota S., Cameron B., Welch W. J. & Winfield J. B. (1988) Autoantibodies to the constitutive 73-kD member of the hsp70 family of heat shock proteins in systemic lupus erythematosus. J. expl Med. 168, 1475. Kaczmarek L., Calabretta B., Kao H.-T., Heintz N., Nevins J. & Baserga R. (1987) Control of hsp70 RNA levels in human lymphocytes. J, Cell Biol. 104, 183.
