differential expression of heat shock proteins in pancreatic carcinoma

[CANCER RESEARCH 54, 547-551. January 15, 1994]

Differential Expression of Heat Shock Proteins in Pancreatic Carcinoma

Thomas M. Gress, l Friederike Miiller-Piilasch, Christoph Weber, Markus M. Lerch, Helmut Friess, Markus Biichler, Hans G. Beger, and Guido Adler

Department of Internal Medicine I, University of Ulm, 89081 Ulm, Germany [T. M. G., F. M-P., C. W., M. M. L., G. A.], and Department of Surgery, University of Ulm, 89075 UIm, Germany [H. F., M. B., H. G. B.]

ABSTRACT M A T E R I A L S AND M E T H O D S

In the present study we sought to determine by Northern blot analysis and mRNA in situ hybridization whether gene expression of heat shock proteins (HSPs) (HSP 89a, HSP 891~, HSP 70, and ubiquitin) is altered in pancreatic carcinoma, compared to control tissues (normal pancreas and chronic pancreatitis tissue). HSP 89a was selectively overexpressed in pancreatic carcinoma, and tumor cells were shown to contain the largest amount of HSP 89a mRNA. Steady state levels of HSP 70 mRNA were increased in pancreatic carcinoma (tumor and connective tissue cells) and in chronic pancreatitis (connective tissue cells and residues of exocrine acinar cells). HSP 891B and ubiquitin B were constitutively expressed at high levels in pancreatic tissue from all three groups; HSP 89/3 mRNA was found in cells of parenchymai and stromal origin. A strong correlation was found between the expression of HSP 70 and the expression of transforming growth factor 61. The finding that HSPs are differentially expressed in pancreatic cancer, compared to normal pancreas and chronic pancreatitis tissue, and the cancer specifity of HSP 89a suggest that HSPs play a specific role in the pathogenesis of pancreatic cancer, e.g., by participating in regulatory processes or in tumor immunity, as proposed previously.

I N T R O D U C T I O N

HSPs 2 are produced by eukaryotic and prokaryotic cells in response to a variety of cellular injuries (1, 2). Most HSPs have been grouped into families of different molecular masses. Members of different families are characterized not only by their size but also by a number of common physiological functions. These are mostly based on the ability of HSPs to form complexes with other proteins, thus altering

their functional status (2). In a variety of animal models, HSPs of the 70-, 90-, and 100-kDa classes elicit tumor-specific immunity to tumor

challenges (see Ref. 3 for a review). Enhanced expression of HSP genes of the 90- and 70-kDa families has been observed after oncogene transformation of cell lines (4-8). Experiments reported so far primarily focused on in vi tro or animal studies, and little is known about HSP expression in human malignant diseases. The aim of our study was to investigate whether gene expression of HSPs in pancreatic carcinoma is altered, compared to control tissues (normal pancreas and chronic pancreatitis). For this purpose we measured the gene

expression of several HSPs at the transcript level. We chose HSPs of

the 90- and 70-kDa classes because these members of the multigene family have been shown to be relevant for regulatory processes in transformed cells (4--8) and for tumor rejection mechanisms (see Ref.

3 for a review). Ubiquitin was chosen as a control HSP, because it appears to be ubiquitously expressed at a high level in all eukaryotic cells (2). Tissue from patients with chronic pancreatitis was included to study the cancer specifictiy of alterations in HSP gene expression. Transcript levels for TGF-/31 and for several oncogenes known to be overexpressed in pancreatic carcinoma (see Ref. 9 for a summary)

were measured to identify possible correlations between the gene expression of the latter and gene expression of HSPs.

Received 8/23/93; accepted 11/5/93. The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom requests for reprints should be addressed, at Medizinische Klinik, Abt. I, Universit~it Ulm, Robert Koch Str. 8, 89081 Ulm, Germany.

2 The abbreviations used are: HSP, heat shock protein; TGF-/31, transforming growth factor/31; UbB, ubiquitin B; UbC, ubiquitin C.

Materials

Human pancreatic tissue from patients with adenocarcinoma of the pancreas (n = 5; 3 men and 2 women) or chronic pancreatitis (n = 5; 4 men and 1 woman) and from organ donors (n = 5; 3 women and 2 men) was provided by the Department of Surgery of the University of Ulm, with written consent from each patient and after approval by the local Ethics Committee. The histological diagnosis was confirmed for each individual tissue block by standard light microscopic evaluation of sections stained with hematoxylin and eosin. There were two grade 2 and three grade 3 duct cell adenocarcinomas among the cancer tissues (10). Only chronic pancreatitis tissue samples showing similar degrees of inflammation and fibrosis were chosen for this study. Surgical procedures consisted either of a partial duodenopancreatectomy for cancer patients or a resection of the head of the pancreas, preserving the duodenum, for chronic pancreatitis patients. Cancer patients were not subjected to any sort of chemotherapy or radiation therapy prior to resection. For RNA extraction tissue was flash frozen in liquid nitrogen immediately after surgical removal. For in situ hybridization tissue was fixed overnight in 10% neutral buffered formalin and embedded in paraffin.

Cloned human DNA probes were purchased from the American Type Cul- ture Collection (Rockville, MD) [HSP 70 (American Type Culture Collection 57494), Ki-ras-2 (American Type Culture Collection 41027), c-myc (American Type Culture Collection 41010), c-raf-1 (American Type Culture Collection 41050), and TGF-/31 (American Type Culture Collection 59955)] and from Stressgen (Victoria, British Columbia, Canada) [HSP 89c~ (SPD-930) and HSP 89/3 (SPD-940)]. Human ubiquitin and 18S rRNA were previously cloned in our laboratory (11).

Methods

Unless stated otherwise, standard protocols for experimental procedures and solution preparation were as described by Sambrook et aL (12).

Extraction of Total RNA. RNA from shock-frozen human pancreatic tissue was extracted by use of a standard guanidinium thiocyanate extraction, followed by centrifugation in a cesium cloride gradient (Pharmacia LKB, Freiburg, Germany).

Northern Blot Analysis and Hybridizations. As determined by A260 measurement and ethidium bromide staining, 30/zg of total RNA derived from pancreatic tissues were size fractionated on 1% agarose-8% formaldehyde denaturing gels and were transfered to Hybond N membranes (Amersham). Cloned DNA probes were prepared by use of Qiagen columns (Qiagen, DiJs- seldorf, Germany) according to the manufacturer's instructions. Inserts were labeled by random hexamer priming using [3ZP]dCTP as radioactive label. Prehybridizations and hybridizations were carried out in 6• standard saline citrate (1 • SSC: 150 mM NaCI, 1.5 mM sosium citrate), 5• Denhardt's solution, 0.5% sodium dodecyl sulfate, 100 /xg/ml yeast tRNA, 50 ixg/ml sonicated total human placental DNA (Sigma), 10 /xg/ml polyuridine homo- polymer (Pharmacia), 50% formamide, at 42~ Filters were washed at high stringency. X-ray films (Kodak XAR) were exposed for 1-7 days at -70~ using intensifying screens. Absorbance values for hybridization signals were measured with an image analysis system comprising a charged coupled device video camera, a flame grabber, and a computer (developed in collaboration with Prof. H. Frey, Fachhochschule Ulm, Ulm, Germany). To exclude differences due to unequal loading of gels, the absorbance value for each individual hybridization signal was divided by the absorbance value for the hybridization signal obtained for each particular lane with an 18S rRNA probe used as internal control. Statistical differences between the mean absorbance values for each tissue class were determined by use of the analysis of variance test. Correlations were calculated using standard linear regression.

547

Research. on July 3, 2015. © 1994 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

HEAT SHOCK PROTEIN EXPRESSION IN PANCREATIC CARCINOMA

Fig. I. A-D, upper, Northern blot hybridizations with cloned DNA probes for HSP 89a (A), HSP 89/3 (B), HSP 70A (C), and ubiquitin (UBI); the UbB band is shown (D). A-D, lower, control hybridizations with a probe for 18S rRNA used for normalization of hybridization results. Thirty/xg of total RNA from distinct human tissue samples were applied to each lane. CA, pancreatic carcinoma; CP, chronic pancreatitis; CO, control pancreas. Different Northern blots were used for each hybridization. However, the same samples were run in the same order in all blots.

A B

C D

In Situ Hybridization. mRNA in situ hybridization was basically done as described by Pardue (13), on sections from three different samples of each tissue group (pancreatic carcinoma, chronic pancreatitis, and control pancreas). Inserts of the HSP 89a, 89/3, and 70 clones were subcloned into the vector Bluescript II KS+ (Stratagene). Complementary RNA probes for each DNA strand were generated by run-off transcriptions using either T3 or T7 RNA polymerase (GIBCO BRL), ribonucleotides (Pharmacia), and [35S]ribocytidine triphosphate as radioactive label. Depending on the cloning orientation and the primer used (T3 or T7), one probe detected the coding strand (antisense probe) and the second detected the noncoding strand (sense probe). Sense probes detecting the noncoding strand were used as negative controls. As an additional negative control, one slide for each experimental approach was treated with ribonuclease A prior to hybridization, to deplete the sample of mRNA. Both controls produced no specific hybridization signal. Prehybridizations and hybridizations were carried out using stringent conditions in 50% formamide, 5 • Denhardt's solution, 10 mM dithiothreitol, 10% dextran sulfate, 250 /xg/ml sonicated human placental DNA, 250 /xg/ml yeast tRNA, 0.2% sodium dodecyl sulfate, 0.75 M NaCI, 25 mM EDTA, 25 mM piperazine-N,N'-bis(2- ethanesulfonic acid), pH 6.8 at 42-50~ After hybridization slides were in- cubated with RNase A at 37~ followed by stringent washing procedures (0.5 x standard saline citrate at 50~ After being covered with photoemulsion (Ilford K2), all slides were exposed at 10~ for 3-4 weeks. After development, sections were stained with hematoxylin and eosin and evaluated on a Zeiss Axioplan microscope.

R E S U L T S

Northern Blot Analysis. Representa t ive autoradiographs of the

relevant hybridizat ions are shown in Fig. 1. For a quanti tat ive evalu-

ation of dif ferences in gene express ion we de te rmined the absorbance

values of hybridizat ion signals (Table 1). HSP 89ot express ion was

enhanced 4 -5 - fo ld in pancreat ic carc inoma, compared to chronic pan-

creatitis and controls (Fig. 1A; signals in chronic pancreati t is and

control samples are barely detectable). For HSP 89/3 no significant

differences be tween the levels of express ion in pancreat ic carc inoma

and normal controls could be detected. Express ion of HSP 89/3 was found to be reduced in chronic pancreati t is , compared to pancreat ic

carc inoma, a l though differences were small (Fig. 1B). HSP 70 was the

only gene expressed at a h igher level in chronic pancreati t is than in

Table 1 Absorbance of hybridization signals Absorbance values of hybridization signals in controls were arbitrarily set to 1. The

expression levels in pancreatic carcinoma and chronic pancreatitis are shown as a multiple of the expression in controls. Mean --- SD of the normalized absorbance values are shown.

Absorbance

Chronic Probe Carcinoma pancreatitis Control

HSP 89a 3 • 1 a 0.8 - 0.3 1 _-• 0.2 HSP 89/3 1.1 • 0.1 0.7 • 0.2 a l + 0.3 HSP 70 4.1 • 1.4 a 2.7 -4- 1.8 a 1 • 0.3 UbB 0.9 - 0.2 1 ___ 0.1 1 • 0.1

'~ Significant difference (P < 0.05), as determined by analysis of variance.

controls , a l though the differences were not statistically signif icant due

to s trong variat ions be tween different samples. The absorbance values

for hybridizat ion signals of HSP 70 were roughly 3 t imes higher in

chronic pancreati t is and 4 t imes higher in pancreat ic carc inoma, com-

pared to normal controls (Fig. 1C). Mammal i an cells contain three

size classes of ubiqui t in-specif ic m R N A (ubiquit in A, B, and C), of

approximate ly 50, 1100, and 2500 nucleot ides . The relative amoun t s

of the three size classes of R N A vary be tween different species and

may show smaller variat ions be tween different t issues of the same species (14). We used a h u m a n complemen ta ry D N A probe for ubiq-

uitin that has been shown to identify all three size classes o f R N A in

a series of mouse and rat t issues (11). UbB was highly expressed in all

examined tissues, whereas U b C express ion was m u c h lower. The

hybridizat ion signal of UbB was used for quanti tat ive evaluat ion of

hybridizat ion signals. After normal iza t ion to the 18S r R N A hybrid-

ization signal, the level of express ion of UbB and U b C m R N A did not

differ in the t issues examined (Fig. 1D; Table 1).

To s tudy the degree of correlat ion be tween the pat terns o f expres-

sion of different HSP famil ies and of TGF-/31 and the oncogenes

Ki-ras, c-myc, and c-raf, we analyzed the transcript levels of the latter by Nor thern blot hybridizat ion. Correlat ions were de te rmined by cal-

culat ing the linear regression. Data obta ined for all pancreat ic t issues

used in this s tudy (carcinoma, pancreati t is , and control) were inc luded

in the linear regression analyses. A s t rong l ink was found be tween the

express ion pat terns observed for HSP 70 and TGF-/31 (Fig. 2). Only

a weak correlat ion was found be tween the express ion of HSP 89ot and that o f TGF-/31 ( r 2 = 0.52). No l ink was observed be tween the

express ion of HSP 89/3 or ubiquit in and that of TGF-/31 (data not

shown) . Transcript levels of the oncogenes c-raf, c-myc, and Ki-ras and of HSPs showed no signif icant correlat ion (data not shown).

In Situ H y b r i d i z a t i o n . Representa t ive in situ hybridizat ions are

shown in Fig. 3. In pancreat ic carc inoma tumor cells conta ined the

largest amount of HSP 8 9 a m R N A (Fig. 3A). Only background la-

1.0

O 13_ ~ 1 7 6 -I- " /

J i . . .D

r squared=0.86 0 .0 �9 '

0.0 0.5 1.0 1.5 2.0 OD TGFB

Fig. 2. Linear regression analysis, showing the degree of correlation observed in all examined pancreatic tissues for the expression of HSP 70 and TGF-/31. The absorbance values of hybridization signals are shown on the x- and y-axes. Absorbance values obtained for pancreatic carcinoma (ff]), chronic pancreatitis (11), and normal pancreas (A) are included. The correlation was highly significant (P < 0.005). OD, absorbance.

548




Fig. 3. mRNA in situ hybridizations with 35S-labeled RNA probes for HSP 89ct (A and B), HSP 89/3 (C and D), or HSP 70 (E and F). Panels illustrate representative findings observed for hybridizations with the antisense probe in three or more specimens in each tissue group. Hybridizations with the sense probe generated background activity only (data not shown). Calibration bars, 100 ixm. A, pancreatic carcinoma; HSP 89a hybridization. Tumor cells (arrows) strongly express HSP 89ct mRNA, whereas the neighboring interstitium shows only background label. Cells on the right side represent tumor cells and cellular debris charcteristically found in the lumen of neoplastic ductules in pancreatic carcinoma. B, chronic pancreatitis; HSP 89ct hybridization. Neither acinar cells (*), connective tissue (open arrow), nor vascular structures (closed arrow) are labeled with any specific silver grain label above background activity. C, Pancreatic carcinoma; HSP 89/3 hybridization. The neoplastic ductules formed by the tumor and the connective tissue cells (arrowheads) between the tumor tissue are strongly labeled. D, chronic pancreatitis; HSP 89/3 hybridization. Normal ducts (*) express HSP 89/3 at a high level and are not different from neoplastic cells in C. High levels of HSP 89/3 mRNA can also be detected over tubular complexes ( 'k) and the connective tissue. E, pancreatic carcinoma; hybridization with HSP 70. Tumor cells forming neoplastic ductules and malignant cells in the lumen of these ductules express HSP 70. This label is not tumor specific and is also found over interstitial connective tissue cells (arrowheads). F, chronic pancreatitis; HSP 70 hybridization. Acinar structures (*) as well as small ductules (open arrow) and surrounding connective tissue cells express HSP 70.

beling was detected in stromal cells and in the surrounding connective tissue. In chronic pancreatitis (Fig. 3B) and control tissue (data not shown), only background activity was observed in parenchymal and stromal cells. The in situ hybridization for HSP 89/3 revealed that transcription was not confined to one particular cell type but could be found in parenchymal and stromal cells of all examined specimens from tumor tissue and chronic pancreatitis samples (Fig. 3, C and D). HSP 70 mRNA was found in tumor cells as well as in stromal cells from pancreatic carcinoma (Fig. 3E). In chronic pancreatitis HSP 70 mRNA was detected in residues of exocrine acinar cells and in the interstitial connective tissue (Fig. 3F). Only background labeling was

seen over pancreatic acinar cells and stromal cells in control pancreata (data not shown).

DISCUSSION

In the present study we could show that HSPs are differentially expressed in pancreatic carcinoma, chronic pancreatitis, and normal pancreas. HSP 89ot and HSP 70 gene expression was enhanced in pancreatic carcinoma. HSP 89ct overexpression was specific for malignant tissue, with tumor cells containing the largest amount of HSP 89ct mRNA. HSP 70 was the only member of this multigene family

549




showing altered transcription patterns in chronic pancreatitis, compared to controls. Mammalian HSPs of the 85-90-kDa class comprise two protein species encoded by separate genes (15). In humans these mRNAs are referred to as HSP 89ol (2.95 kilobases) and HSP 89/3 (2.7 kilobases) (15). Whereas HSP 89/3 is constitutively expressed in a series of cultured human cells (5, 15), HSP 89a expression appears to be inducible by different stimuli. Adenovirus E1A gene products (8) and transformation with the H-ras oncogene (5) induce a selective overexpression of HSP 89ot. The reported in vitro expression patterns are identical to those observed in the present study for the two 90-kDa HSPs in human pancreatic carcinoma. We could show that HSP 89/3 is constitutively expressed at high levels in tumor, stromal, and non- tumorous pancreatic (exocrine and endocrine) cells. Overexpression of HSP 89a was detected exclusively in pancreatic carcinoma, and the tumor cells were shown to be the predominant site of transcription of HSP 89a mRNA.

The HSP 70 family is composed of several different proteins that play a major role in the folding, unfolding, and translocation of polypetides, as well as in the assembly and disassembly of protein complexes (1, 2). In the same way as HSP 89c~, the 70-kDa protein is inducible by E1A gene products (8) and a variety of other cellular injuries (1, 2). HSP 70-like proteins are overexpressed in cultured cells after transformation with H-ras (4) or c-myc (6) and after co- transformation with H-ras/p53 (7). Thus, involvement of the 70-kDa HSP in regulatory processes in malignant cells can be assumed, which could be the reason for the finding that the gene coding for HSP 70 was expressed at a roughly 4 times higher rate in pancreatic carcinoma, compared to control pancreas. However, the observation that gene expression of HSP 70 was increased in chronic pancreatitis shows that trancript levels of HSP 70 in the human pancreas are influenced by a variety of injuries, such as acute and chronic inflammation. Inflammatory cells synthetize high levels of HSPs (HSP 70 in particular), probably representing autoprotective mechanisms (16). Stromal cells may be the source of the increased amount of HSP 70 mRNA in pancreatic carcinoma and chronic pancreatitis and were shown to express HSP 70 mRNA by in situ hybridization. The link found between HSP 70 and TGF-/31 expression may be due to the 1. inflammatory and fibroblastic reaction in chronic pancreatitis and 2. pancreatic carcinoma. TGF-/31 serves as a chemoattractant for inflammatory cells, which are able to secrete large amounts of TGF-/31 3. protein (17). Besides promoting inflammatory reactions, TGF-/31 is 4. one of the most potent natural immunosuppressants (18). Tumor cells produce large amounts of TGF-/31, and it has been speculated that this may contribute to mechanisms by which tumor cells avoid immune 5. recognition (19). Torre-Amione et al. (20) recently reported that an immunogenic tumor engineered to secrete TGF-/31 escaped immune surveillance and grew progressively in transiently immunosuppressed 6. mice. Increased expression of the immunosuppressant TGF-/31 may 7. be an essential precondition for the survival of cells expressing the immunogenic HSPs in tumors and in chronic inflammation. However, Takenaka and Hightower (21) recently reported that members of the 8. HSP 70 and HSP 90 families were inducible in cultured chicken embryo cells upon stimulation with exogenously administered TGF- /31. It was hypothesized that cells rapidly increase their chaperoning 9. capacity for newly synthesized polypeptides in preparation for an increase in the rate of synthesis of proteins up-regulated by TGF-/31. 10.

Ubiquitin is a highly conserved small HSP which, as implied by its 11. name, is ubiquitously expressed at a high level and is inducible by heat shock (2, 14). Ubiquitin proteins are found in cells either free or

12. bound to a variety of cellular proteins; this binding appears to be essential for selective degradation of intracellular proteins (2). Three 13. different size classes of mRNA are known for ubiquitin, with relative amounts varying between species (14). UbB mRNA appears to be the 14.

550

most abundant class of ubiquitin-specific mRNA in human pancreas. No significant differences concerning the expression of mRNAs coding for UbB and UbC could be detected between pancreatic tissues used in this study.

This represents the first report implicating HSPs in pancreatic cancer. Overexpression of HSPs in human malignant diseases has been described for HSP 27 [e.g., breast cancer (see Ref. 22) and malignant fibrous histiocytoma (see Ref. 23)] and for HSP 70 (leukemias) (24). The following models are being discussed at present to explain the role of HSPs in malignant cells. (a) HSPs of the 90- and 70-kDa classes elicit tumor-specific immunity in a number of experimental animal models and may be essential for tumor rejection mechanisms (see Ref. 3 for a review). (b) An increased amount of HSPs may be needed for regulatory and stabilizing processes in rapidly reproducing malignant cells, by making use of their chaperoning capacity (2, 4-8). (c) HSP 70 protects tumor cells from tumor necrosis factor cytotoxicity (25). HSPs may thus be essential factors in autoprotective mechanisms in tumor cells. (d) The presence of mutant or abnormal proteins in cells stimulates HSP synthesis (26). Therefore, overexpression of HSPs may be a nonspecific reaction of cells to any sort of acute or chronic cellular injury.

Our data show that HSP genes are differentially expressed in pancreatic carcinoma, chronic pancreatitis, and control pancreas and that one member of the 90-kDa class is selectively overexpressed in pancreatic carcinoma. These findings suggest that increased gene expression of HSP genes is not solely a nonspecific reaction of pancreatic cells to acute or chronic cellular injury and that HSPs may play a role in the pathogenesis of pancreatic cancer, e.g., by participating in tumor immunity or in regulatory processes, as described above.

ACKNOWLEDGMENTS

We thank Prof. S. Kaufmann for valuable advice and discussions and U. Lacher and E. Preiss for excellent technical assistance.

REFERENCES

Kaufmann, S. H. E. Heat shock proteins and the immune response. Immunol. Today, 11: 129-136, 1990. Lindquist, S., and Craig, E. A. The heat-shock proteins. Annu. Rev. Genet., 22: 631-677, 1988. Shrivastava, P. K., and Maki, R. G. Stress-induced proteins in immune response to cancer. Curr. Top. Microbiol. Immunol., 167: 109-123, 1991. Konno, A., Sato, N., Yagihashi, A., Torigoe, T., Cho, J., Torimoto, K., Hara., I., Wada, Y., Okubo, M., Takahasi, N., and Kikuchi, K. Heat- or stress-inducible transformation-associated cell surface antigen on the H-ras oncogene-transfected rat fibroblasts. Cancer Res., 49: 6578-6582, 1989. Lebeau, J., Le Chalony, C., Prosperi, M-T., and Goubin, G. Constitutive overexpression of a 89 kDa heat shock protein gene in the HBL100 human mammary cell line converted to a tumorigenic phenotype by the EJ/T24 Harvey-ras oncogene. Onco- gene, 6: 1125-1132, 1991. Kingston, R. E., Baldwin, A. S., and Sharp, P. A. Regulation of heat shock protein 70 gene expression by c-myc. Nature (Lond.), 312: 280-282, 1984. Pinashi-Kimhi, O., Michalowitz, D., Ben-Zeev, A., and Oren, M. Specific interaction between the p53 cellular tumour antigen and major heat shock proteins. Nature (Lond.), 320: 182-185, 1986. Simon, M. C., Kitchener, K., Kao, H. T., Hickey, E., Weber, L., Voellmy, R., Heintz, N., and Nevins, J. R. Selective induction of human heat shock gene transcription by the adenovirus E1A gene products, including the E1A product. Mol. Cell. Biol., 7: 2884-2890, 1987. Gillespie, J., and Poston, G. J. The molecular biology of pancreatic adenocarcinoma. In: C. D. Johnson and C. W. Imrie (eds.), Pancreatic Disease, Progress and Prospects, pp. 47-64. London: Springer, 1991. Hermanek, K. D., Scheibe, O., Spiessi, B., and Wagner, G. TNM Klassifikation Maligner Tumoren. New York: Springer, 1987. Gress, T. G., Hoheisel, J. D., Lennon, G. G., Zehetner, G., and Lehrach, H. Hybrid- ization fingerprinting of high-density cDNA-library arrays with cDNA pools derived from whole tissues. Mammalian Genome, 3: 609-619, 1992. Sambrook, J., Fritsch, E. E, and Maniatis, T. Molecular Cloning, a Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 1989. Pardue, M. In situ hybridization. In: B. D. Hames and S. J. Higgins (eds.), Nucleic Acid Hybridization, a Practical Approach, pp. 179-202. Oxford, UK: IRL Press, 1988. Wiborg, O., Pedersen, M. S., Wind, A., Berglund, L. E., Marcker, K., and Vuust, J.



HEAT SHOCK PROTEIN EXPRESSION 1N PANCREATIC CARCINOMA

The human ubiquitin multigene family: some genes contain multiple directly repeated ubiquitin coding sequences. EMBO J., 4: 755-759, 1985.

15. Hickey, E., Brandon, S. E., Smale, G., Lloyd, D., and Weber, L. A. Sequence and regulation of a gene encoding a human 89-kilodalton heat shock protein. Mol. Cell. Biol., 9: 2615-2626, 1989.

16. Kantengwa, S., Donati, Y. R. A., Clerget, M., Maridonneau-Parini, I., Sinclair, F., Mariethoz, E., Perin, M., Rees, A. D. M., Slosman, D. O., and PoUa, B. S. Heat shock proteins: an autoprotective mechanism for inflammatory cells?. Semin. Immunol., 3: 49-56, 1991.

17. Wahl, S. M., Hunt, D. A., Wakefield, L. M., McCartney-Francis, N., Wahl, L. M., Roberts, A. B., and Spore, M. B. Transforming growth factor-beta (TGF-beta) induces monocyte chemotaxis and growth factor production. Proc. Natl. Acad. Sci. USA, 84: 5788-5792, 1987.

18. Wahl, S. M., Hunt, D. A., Wong, H. L., Dougherty, S., McCartney-Francis, N., Wahl, L., EUingworth, L., Schmidt, J., Hall, G., Roberts, A. B., and Sporn, M. B. Trans- forming growth factor-/3 is a potent immunosuppressive agent that inhibits IL-1- dependent lymphocyte proliferation. J. Immunol., 140: 3026-3032, 1988.

19. Lucas, C., Wallick, S., Fendly, B. M., Figari, I., and Palladino, M. A. TGF/3: a possible autocrine immune regulator. CIBA Found. Symp., 157: 98-114, 1991.

20. Torre-Amione, G., Beauchamp, R. D., Koeppen, H., Park, B. H., Schreiber, H., Moses, H. L., and Rowley, D. A. A highly immunogenic tumour transfected with a

murine transforming growth factor type/31 cDNA escapes immune surveillance. Proc. Natl. Acad. Sci. USA, 87: 1486-1490, 1990.

21. Takenaka, I. M., and Hightower, L. E. Transforming growth factor-/31 rapidly induces Hsp70 and Hsp90 molecular chaperones in cultured chicken embryo cells. J. Cell. Physiol., 152: 568-577, 1992.

22. Thor, A., Benz, C., Moore, D., Goldman, E., Edgerton, S., Landry, J., Schwartz, L., Mayall, B., Hickey, E., and Weber, L. A. Stress response protein (SRP-27) determi- nation in primary human breast carcinomas: clinical, histologic, and prognostic correlations. J. Natl. Cancer Inst., 83: 170-178, 1991.

23. Tetu, B., Lacasse, B., Bouchard, H. L., Lagace, R., Huot, J., and Landry, J. Prognostic influence of HSP-27 expression in malignant fibrous histiocytoma: a clinicopatho- logical and immunohistochemical study. Cancer Res., 52: 2325-2328, 1992.

24. Michevi, N. E, Monson, J. M., and Hahn, G. M. Expression of HSP-28 and three HSP-70 genes during the development and decay of thermotolerance in leukemic and nonleukemic human tumors. Cancer Res., 51: 6608--6614, 1991.

25. J~ittela, M., Wissin, D., Bauer, P. A., and Li, G. C. Major heat shock protein HSP70 protects tumor ceils from tumor necrosis factor cytotoxicity. EMBO J., 11: 3507- 3512, 1992.

26. Ananthan, J., Goldberg, A. L., and Voellmy, R. Abnormal proteins serve as eukaryotic stress signals and trigger the activation of heat shock genes. Science (Washington DC), 232: 522-524, 1986.

551



1994;54:547-551. Cancer Res Thomas M. Gress, Friederike Müller-Pillasch, Christoph Weber, et al. CarcinomaDifferential Expression of Heat Shock Proteins in Pancreatic

Updated version

http://cancerres.aacrjournals.org/content/54/2/547

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

[email protected] at

To request permission to re-use all or part of this article, contact the AACR Publications


http://cancerres.aacrjournals.org/content/54/2/547

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

mailto:[email protected]


differential expression of heat shock proteins in pancreatic carcinoma

Documents