prevention of carcinogenesis by protease inhibitors1€¦ · prevention of carcinogenesis by...

(CANCER RESEARCH (SUPPL.) 54. l9<Ns-:(K)5s. April 1. 1W4]

Prevention of Carcinogenesis by Protease Inhibitors1

Ann R. Kennedy2

Depurimeli! of Rinliiiiitm Oncology. University of Pennsylvania School of Medicine. Philadelphia, Pennsylvania IV104

Abstract

Protease inhibitors are very effective in their ability to suppress carci-

iKiurncsis in many different in v/ro and in vitro assay systems. Oneparticularly effective protease inhibitor, the soybean-derived Bowman-

Kirk inhibitor, has been extensively studied in our laboratory. Our resultshave indicated that Bowman-Birk inhibitor suppresses carcinogenesis I )

induced by several different types of carcinogens. 2) in three differentspecies (mice, rats, and hamsters), 3) in several different tissues/organs[colon, liver, lung, esophagus, and cheek pouch (oral epitheliumÂ». 4) whenadministered to animals by several different routes (including the diet), 5)involving several different types of tumors (squamous cell carcinomas,adenocarcinomas, angiosarcomas, etc.), and 6) in different cell types[epithelial cells (in the colon, liver, lung, esophagus, and cheek pouch) asÂ»ellas connective tissue cells (fibrohlasts, both in vitro and those in theliver which give rise to angiosarcomas)|. Thus, the remarkable ability ofBowman-Birk inhibitor to serve as an anticarcinogenic agent has been

demonstrated in a variety of different carcinogenesis assay systems.Although the mechanism of action of protease inhibitors as anticarci

nogenic agents is unknown, many hypotheses have been presented. Ourresults suggest that anticarcinogenic protease inhibitors are capable ofreversing the initiating event in carcinogenesis, presumably by stoppingan ongoing process begun by carcinogen exposure. We have observedseveral effects of protease inhibitors which are thought to be related totheir anticarcinogenic activity: these include I) the ability to affect theexpression of certain oncogenes (f.#., c-myc and c-fox) and 2) the ability toaffect the levels of certain types of proteolytic activities ie.g., N-l-butuxy-carbonyl-Val-Pro-Arg-7-amino-4-nifthylcoumarin-hydrolyzing activity)which are elevated in carcinogen-exposed tissues.

We have also observed other effects of anticarcinogenic protease inhibitors which may be related to carcinogenesis. For example, we havereported that the inhibitors can reduce the carcinogen-induced, elevated

levels of gene amplification to nearly normal levels. While all of theseeffects of protease inhibitors may contribute to the prevention of carcinogenesis, the mechanism by which protease inhibitors prevent cancercannot be determined with certainty until the mechanisms involved incancer induction are known. While the mechanism remains unclear, it isclear that protease inhibitors can reverse a number of carcinogen-induced

cellular changes which may play important roles in carcinogenesis.

Introduction

The ability of protease inhibitors to serve as cancer-chemopreven-

tive agents in both in vitro (1) and in vivo (2) systems has recentlybeen reviewed. Although the mechanisms for the anticarcinogenicactivity of protease inhibitors are unknown, many hypotheses havebeen discussed (1-3). Our own studies have suggested that protease

inhibitors suppress both the initiation and promotion stages of carcinogenesis by stopping an ongoing process involved in carcinogenesis,as described below.

Materials, Methods, and Results

In Vitro Model Systems. Protease inhibitors have been studied ascancer-chemopreventive agents in a number of different in vitro

model systems. Most of our in vitro transformation studies have beenperformed in C3H lOT'/z cells, an in vitro transformation assay system

originally developed by Reznikoff et al. (4, 5). In this cell line,carcinogen treatment of cells leads to the formation of transformedfoci within a background monolayer of normal-appearing cells, as

shown in Figs. 1 and 2. The advantages and disadvantages of thedifferent in vitro transformation systems commonly used in carcinogenesis research are described elsewhere (6, 7).

The ability of protease inhibitors to suppress radiation-inducedtransformation in vitro was first observed in C3H10T'/2 cells (8). It is

now known that many different protease inhibitors are capable ofsuppressing radiation- and chemical carcinogen-induced malignant

transformation in a variety of in vitro transformation systems, asrecently reviewed (1). A wide variety of transformation-inducing

agents have been utilized in these studies, including chemical carcinogens that do or do not need to be metabolically activated, bothionizing and nonioni/ing radiation, and steroid hormones. The factthat protease inhibitors suppress transformation induced by agentsproducing very different kinds of cellular damage suggests that carcinogenesis produced by the different agents may involve a commonpathway affected by the anticarcinogenic protease inhibitors.

Not all protease inhibitors are capable of suppressing transformation in vitro; the inhibitory profiles of those that affect transformationhave been discussed elsewhere (1). For the suppression of transformation, the most effective of the protease inhibitors we have studiedis chymostatin, a quite specific and potent inhibitor of chymotrypsin,as described in detail elsewhere (9). With only picomolar concentrations in the medium, chymostatin has the ability to suppress radiation-

induced transformation in vitro (9). Other inhibitors of chymotrypsin,such as tosyl-phenylalanine chloromethyl ketone, arc also very effec

tive in their ability to inhibit malignant transformation (9). Inhibitorsof the Bowman-Birk inhibitor family of protease inhibitors are effec

tive at suppressing transformation in the nanomolar concentrationrange (10). The structure of the soybean-derived BBI1 is shown in Fig.

3. The two well characterized protease inhibitory sites in BB1 arcshown in Fig. 3 [the structure of BBI has been described by Odani andIkenaka (II)): only the chymotrypsin inhibitory site is involved in thesuppression of transformation by BBI (10).

The fact that BBI was so effective at suppressing transformationinduced by a variety of carcinogens led us to develop this proteaseinhibitor into a form that could be utilized in large-scale animal and

human cancer prevention trials. Because PBBI would be prohibitivelycostly for use in such large-scale studies, we have developed an

extract of soybeans enriched in BBI, termed BBIC. which is beingused for large-scale studies (2, 12). For the suppression of HI vitro

transformation, BBIC works as well as PBBI (10. 12).A discussion of the potential mechanisms of action of the anticar

cinogenic protease inhibitors in the suppression of transformationrequires a discussion about the mechanisms thought to be involved incarcinogen-induced transformation HI vitro. Results from our HI vitrostudies have suggested that a high-frequency initiating event is involved in transformation (13-1^). The frequency of initiation is suf

ficiently high that it is not likely to be a mututional event. It ishypothesized that the high-frequency initiating event in transforma

tion reflects a change in the pattern of gene expression that, like the

1 Presented at the 4th International Conference on Anticarcinogenesis Â£ RadiationProtection. April IH-23, lu"3, Baltimore. Ml). This research has heen supported by NIHGrants CA22704 and CA4f>4Â«6.

2 To whom requests for reprints should he addressed.

1The abbreviations used arc: BBI, Bowman-Kirk inhibitor; PBBI. pure Bowman-Birk

inhibitor; BBIC, Bowman-Birk inhibitor concentrate: IMI-., intermediate marker end-points; Hoc. jV-/-butoxycarbonyl; MCA, 7-;)min(Ã¯-4-niethylcouniariii.

1999s

on June 29, 2020. © 1994 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

PROTEASE INHIBITOR PREVENTION OF CARCINOGENESIS

Fig. 1. One focus of transformed cells; the transformed focus can be easily distinguished from the background monolayer of nontransformcd C3H10T1/; cells.

SOS system in bacteria, then determines the frequency of rare geneticevents (13-19). Initiation is clearly more stable than the SOS response, however. Radiation-induced recombination in yeast, as re

ported by Fabre and Roman (20), is a system that remains activatedfor a longer period of time and represents a type of model systemwhich could account for our observations on the kinetics of thetransformation process. It is of interest that such a system can beturned off by protease inhibitors that inhibit chymotrypsin. as shownby Wintersberger (21). Although theoretically appealing for our "ongoing process" hypothesis, a system comparable to radiation-induced

recombination in yeast has not yet been found in mammalian cells.There is now much evidence that a high-frequency event is also

involved in the initiating event that leads to cancer in animals. Severalinvestigators from several different laboratories have been able tocount the number of initiated cells which give rise to cancers inanimals in their in vitro/in vivo systems; results from some of thesestudies using radiation as the carcinogenic agent are summarized inTable 1 (data from Refs. 22-24). As can be observed in the results

shown in Table 1, frequencies of initiation and/or tumor formationfrom initiated cells are approximately 1-4% for several different

systems involving cancers of breast, thyroid, and lung tissue. Similarly, high frequencies of initiation by chemical carcinogens havebeen discussed elsewhere (18). Such frequencies are far too high torepresent single gene mutations as the initiating events in thesetissues.

Thus, initiation of carcinogenesis in both in vitro and in vivosystems is likely to be brought about by a high-frequency event, as has

been reviewed (18). Our data suggest that carcinogen exposure induces an ongoing cellular process (13-18); protease inhibitors appear

to be able to stop this process at any point before malignant cells arise(9, 10, 19). We have observed that protease inhibitors are capable ofsuppressing transformation even when applied to cultures long aftercarcinogen exposure, provided that cells are able to proliferate at thetime of exposure to protease inhibitors (6, 9, 10, 19, 25). Our resultsshow that carcinogen-treated cells are still maximally affected by the

anticarcinogenic action of protease inhibitors at 10 days and 13 celldivisions after exposure to a carcinogen (9), as demonstrated in Fig. 4.Other transformation studies of ours have suggested that proteaseinhibitors are capable of reversing the initiated state of carcinogen-

treated cells (9). Protease inhibitors are effective even when administered to cultures for only a 1-day treatment after carcinogen expo

sure (8, 19, 25).Many protease inhibitors are strongly antagonistic to the enhancing

effects of promoting agents in the induction of transformation in vitro(25, 26). Protease inhibitors have been thought to be highly "antipro-motional" in many of the in vivo two-stage carcinogenesis experi

ments which have been performed (27, 28). It has recently beenproposed that Bloom's syndrome is a human disease representing a

defect in the area of promotion (discussed in Ref. 29). We haveobserved that protease inhibitors, including BBI and BBIC, greatlyreduce the levels of spontaneously occurring chromosome aberrationsand sister chromatid exchanges in cells of patients with Bloom'ssyndrome (29). Bloom's syndrome is an autosomal recessive genetic

disease in which high levels of chromosomal abnormalities arethought to be related to the high risk of cancer in people with thedisease (30). We have hypothesized that radiation and chemical carcinogens may induce in normal cells a state or ongoing processresembling that which always occurs in the Bloom's syndrome cells,

with protease inhibitors being capable of stopping that process (29).

Fig. 2. Higher power view of the edge of thefocus of transformed cells shown in Fig. 1. As can beobserved, the transformed cells exhibit a criss-crosspattern of orientation and are highly polar and baso-

philic compared to the nontransformed cells in themonolayer.



PROTEASE INHIBITOR PREVENTION OK CARC1NOOENESIS

Fig. 3. Covalent structure of the Bowman-Birk

protease inhibitor. Two protease inhibitor (reactive)sites are shown as black circles. Modified fromRef. 11.

TrypsiuInhibitorySite

ChymotrypsinInhibitorySite

Recently, evidence has been presented that such an ongoing processcan be induced by carcinogens; it has been reported that radiation canproduce chromosomal aberrations via an indirect process operatinglong after exposure (31). Our hypothesized carcinogen-induced pro

cess could be characterized by a state of genomic instability. It is awell known phenomenon that a state of genomic instability exists inmalignant cells (32, 33). The presence of a variety of chromosomalabnormalities in preneoplastic cells (33) suggests that a state ofgenomic instability precedes malignancy. Many premalignant cellshave also acquired the ability to amplify genes, another measure ofgenomic instability (34). From the data cited above, it seems reasonable to suggest that the state of genomic instability can occur veryearly in carcinogenesis and that this state represents an ongoingprocess which could serve as a target for cancer-chemopreventive

agents such as protease inhibitors.With the availability of relatively new molecular biology tech

niques, an overwhelming number of specific mutational changes ingenes have been reported. However, what may be far more importantto the field of cancer chemoprevention than the generation of anyspecific mutational change is the process driving initiated/premalig-

nant cells to continuously produce the specific mutational changes. Ifthe system can be turned off before a critical mutation is produced,cancer development could be prevented.

Protease inhibitors have been shown to affect a number of differentcndpoints which could be related to their ability to suppress transformation in vitro. The major lines of investigation regarding the mechanism of the protease inhibitor suppression of transformation relate tothe ability of anticarcinogenic protease inhibitors to affect 1) thelevels of certain types of proteolytic activities, 2) the expression ofcertain oncogcnes, and 3) gene amplification. The most direct approach to determining the mechanism of action of the protease inhibitors is to identify and characterize the protease or proteases withwhich they are interacting; such proteases have been identified bysubstrate hydrolysis and affinity chromatography (35-43). Utilizing

substrate hydrolysis, we have examined the ability of cell homog-

enates to cleave specific substrates and have then determinedthe ability of various protease inhibitors to affect that hydrolyzingactivity. With affinity chromatography, specific proteases directlyinteracting with the anticarcinogenic protease inhibitors have beenidentified. Utilizing in vitro systems, the Boc-Val-Pro-Arg-MCA-hydrolyzing activity and the succinyl-Ala-Ala-Pro-Phe-7-amino-4-methylcoumarin-hydrolyzing activity were identified by substrate

hydrolysis (44), and a M, 43,000 protease has been identified byaffinity chromatography (38, 40). Although the functions of theseproteases are not known, the Boc-Val-Pro-Arg-MCA-hydrolyzing

activity has characteristics similar to those of specific proteasesknown to be involved in growth factor processing, which has led tospeculation that it has a similar function (36).

Our studies on the effects of protease inhibitors on the expressionof oncogenes have focused mainly on c-myc, c-fos, c-crb-B, andc-H-ras (44-51). Our initial studies in this area of research were withC3H10T'/2 cells. In these studies, total RNA was extracted from

logarithmically growing irradiated and control cells maintained withor without protease inhibitors. In these studies, we observed thatc-myc transcripts are reduced in irradiated C3H10T'/2 cells treated

with protease inhibitors, as discussed in detail elsewhere (44). Underconditions in which protease inhibitor treatment results in a reductionin c-myc expression, there is no effect on proliferation rates, as shown

in Fig. 5. Further studies in this area of research demonstrated thatc-fos expression was similarly affected by the anticarcinogenic protease inhibitors (47). The effect of protease inhibitors on c-myc

expression is thought to be related to the ability to suppress transformation in vitro, for the following reasons. 1) The protease inhibitorsuppressing effect on transformation is correlated with the ability todown-regulate c-myc expression (49). 2) Protease inhibitors candown-regulate c-myc expression in normal, but not transformed,C3H10T'/2 cells (49). This phenomenon parallels the effect of pro

tease inhibitors on transformation; protease inhibitors suppress the

Table 1 Frequency of initiation by radiation in in vivo and in vitro experiment*

CellsystemIn

vivo rat tracheas exposedtocarcinogen;cellsuspensionsobtained

from individualtracheasandcultured ///vitroin

vitro carcinogen exposure; invivogrowthof tumors(thyroidclonogenic

celltransplantationsystem)In

vivo carcinogen exposureofmammaryepithelialcells;mammaryoutgrowths derivedbyinjecting

cells into mammaryfatpadsof miceNo.

ofcells(surviving)exposed

tocarcinogen=

150 colony-formingcells/dish3.3

surviving"clonogens'Vanimalgraft

site=3

surviving stem cells/fat padType

of"radiation

(dose)X-rays

(600cGy)Neutrons(40cGy)X-rays

(500cGy)â€¢y-rays

(100 cGy)frequency

ofinitiationFrequencyof altered cells"(numberof

altered cells/number ofcolony-formingcells): X-rays, 2 X 10~";neutrons,

2.8 X 10~2Incidenceof gross tumors:42/10"clonogens,

4.2 X 10~2/clonogen100%

of carcinogen-exposedanimalshad

initiated cells;initiationfrequency,1.5 X 10~2/>Authors

(reference)Terzaghi-Howe

(24)Clifton

et al(22)Ethier

and Ullrich (23)

a 50% of the altered cells maintained in culture developed tumorigenic potential (M. Tcrzaghi-Howe, personal communication).' R. Ullrich, personal communication.

2001s



PROTEASE INHIBITOR PREVENTION OF CARCINOGENESIS

Fig. 4. In the experiments shown, cultures ofC3H10T1/:: cells were irradiated and allowed to pro

liferate until they were nearly confluent, at whichtime they were suhcultured to a lower density andseeded into dishes so that there would he approximately 3(X) viable cells/dish. At that point, proteaseinhibitors were added to cultures, which were thenmaintained as for a routine transformation assay. Theresults showed that protease inhibitors are capable ofsuppressing radiation-induced transformation invitro even when applied to cultures as long as 10days and 13 cellular generations after cells wereexposed to the carcinogenic agent. Data are fromRef. 9.

process involved in the induction of transformation but have no effecton cells that are already transformed (25). 3) Protease inhibitors onlyaffect transformation when applied to proliferating cells; similarly,protease inhibitors only affect c-myc expression when applied to

proliferating cells (45, 46, 49, 51). Another promising area for mechanistic studies involving protease inhibitors is gene amplification,because that is one type of genetic event which can be induced in awidespread fashion in a population of mammalian cells by a numberof different carcinogens (52, 53). We have observed that radiation-

induced gene amplification can be affected by the anticarcinogenicprotease inhibitors in a manner that corresponds to their ability tosuppress radiation-induced transformation (54), suggesting that the

protease inhibitor suppression of gene amplification may be related tothe anticarcinogenic activity of the inhibitors.

10

(TLUQ.

<J IO

IO*

Normal Cellsw/wo Protease Inhibitor

Irradiated Cellsw/wo Protease Inhibitor

3 4 5 6 7 8 9 10 II 12 13 14

TIME (Days After Plating)Fig. 5. Growth curves for irradiated and normal C3HK)T'/2 cells which have or have

not been maintained in the presence of anticarcinogenic protease inhibitors; at concentrations of protease inhibitors in the cellular medium which suppress transformation (andc-myc expression), there is no effect on cell growth. Data are from Ref. 45.

In Vivo Studies. Many different types of in vivo carcinogenesisstudies have been performed with protease inhibitors, as recentlyreviewed (2). It is clear from the results of these studies that certainprotease inhibitors are powerful anticarcinogenic agents, with theability to completely prevent cancer in many different animal modelsystems. Our own studies have focused on BBI administered toanimals as PBBI or BBIC (reviewed in Ref. 2). In our studies, BBI hasbeen shown to suppress carcinogenesis 1) in three different species(mice, rats, and hamsters), 2) in several organ systems/tissue types[colon, liver, lung, esophagus, and cheek pouch (oral epithelium)], 3)in cells of both epithelial and connective tissue origin, 4) when it wasgiven to animals by several different routes of administration (including i.p. or i.v. injection, topical administration, and dietary administration), 5) leading to different types of cancer (e.g., squamous cellcarcinomas, adenocarcinomas, angiosarcomas, etc.), and 6) inducedby a wide variety of chemical and physical carcinogens. We originallyidentified BBI as an anticarcinogenic agent in the C3H10T'/2 in vitro

transformation assay system (55), as described above. It was thenshown to work in several different animal model carcinogenesissystems (reviewed in Ref. 2). The anticarcinogenic effects of BBI invivo are similar to those observed in our in vitro studies, as summarized here. 1) Like the irreversible effect on in vitro transformation,BBI has an irreversible suppressive effect on carcinogenesis in vivo,as shown in the hamster oral carcinogenesis assay system (demonstrated in Fig. 6; data are from Ref. 56). 2) Like the effect of proteaseinhibitors on in vitro transformation, BBI treatment can be delayed fora long time after carcinogen treatment and still have a suppressiveeffect on in vivo carcinogenesis, as shown in Fig. 6. 3) The dose-

response relationship for the BBIC and PBBI suppression of oral (56)and colon (57) carcinogenesis is like that observed in vitro (9, 10), inthat the same suppressive effect is observed for differences in dose/concentration of BBI which vary over orders of magnitude. 4) Likethe in vitro studies showing that protease inhibitors suppress transformation without toxicity (8, 25), BBI is able to suppress carcinogenesis in animals without toxic side effects (2). In fact, anticarcinogenic levels of BBIC have a significant life-lengthening effect in mice

(12), as illustrated in Fig. 7. BBI has recently advanced to the humantrial stage. It is planned that the IME to be studied in people will bethose that were originally identified in in vitro systems and that also

2002s



PROTEASE INHIBITOR I'RKVENTION OF C'ARCINOGENESIS

Fig. ft. Suppression of dimethylben/anthracenc (/M//M )-induced oral carcinogcnesisin hamsters by purified UBI preparations. PUBI, and BBIC. As can he observed. I'HBI isas effective as BBIC in the suppression of oral carcinogenesis. In addition. BBIC' does not

have to be continuously present during the carcinogenesis assay period to have asuppressive effect on carcinogenesis. The results showing a suppressive effect on carcinogenesis when BBIC' was present for the first third of the assay period demonstrate that

there is an irreversible suppressive effect on carcinogenesis. and the results of studies onthe effects of BBIC' when treatments began during the later periods of the transformationassay indicate that BBIC' treatment can be delayed for a relatively long time after

carcinogen exposure and still have a suppressive effect on carcinogenesis. Data are fromRef. 56.

appear to play a role in in vivo carcinogenesis systems. As shown inFig. 8, both the expression of specific oncogenes (e.g., c-mvc) andspecific proteolytic activities ((â€¢.#.,Boc-Val-Pro-Arg-MCA-hydrolyz-ing activity) are elevated above normal levels in carcinogen-treated

epithelial tissue in vivo. For both of these IME, PBBI/BBIC treatmentresults in normal levels of expression in the in vivo tissues studied,and this effect on the IME is correlated with a reduction in tumoryields. Our hypothesis is that human premalignant tissue or tissue ata higher than normal risk of developing into cancer will be analogousto carcinogen-treated tissue in the animal model carcinogenesis assay

systems we have studied. Thus, it is expected that we will observeelevated levels of these IME in the human tissues at elevated risk forcancer development. As discussed above, in our previous animalstudies we have observed that BBI is capable of bringing suchelevated levels of IME back to normal levels. In the in vivo studiesperformed. BBI treatment did not affect normal levels of either c-myc

expression (48) or proteolytic activity (40) but was able to reduce theelevated levels of each of these IME to normal levels in the carcinogen-treated tissues.

Discussion

While the mechanism of action of protease inhibitors in the prevention of cancer is not yet clear, it is clear that the inhibitors arepowerful anticarcinogenic agents. Those protease inhibitors with theability to suppress transformation are capable of reversing severalcarcinogen-induced changes to a normal state. While it is possible thatwe have not yet studied the critical carcinogen-induced change that is

involved in the conversion of a cell to a malignant state, it is highlylikely that the anticarcinogenic protease inhibitors are able to reverseother important carcinogen-induced changes related to cancer devel

opment in the same manner that we have already observed for theIME discussed above. The mechanism(s) involved in cancer causation

2003s

must he known with certainty before we can feel confident about themechanism by which protease inhibitors prevent cancer.

Protease inhibitors are unlike most of the other potential classes ofcancer-chemopreventive agents that have been studied, in a number of

different ways. The most dramatic difference we have discovered istheir ability to affect the carcinogenic process in an irreversiblemanner. When protease inhibitor administration is stopped in either invivo (56, 58) or in vitro (9. 10) experiments, malignant cells or tumorsdo not arise in the assay systems we have utilized. Other cancer-preventive compounds we have studied, including vitamin li. dimcth-ylsulfoxide. ÃŸ-carotene.retinoids, etc., have a reversible effect on invitro transformation (t'.#.. see Ret. 59). When most cancer-chemopre

ventive agents are removed from carcinogen-treated cell cultures,

transformed cells do arise (59). While there are other compoundswhich appear to have the ability to reverse the initiated state of cellsin a manner similar to that which we have observed for the proteaseinhibitors, they are not commonly used as cancer-chemopreventive

agents, because the agents which have been described are potentcarcinogens in other organ systems. For example, sulfur mustardappears to inactivate initiated cells in the skin (60) but serves as acarcinogen in the respiratory tract (61). Similarly, aflatoxin B, andactinomycin D are thought to inactivate initiated cells in the skin butare potent carcinogens for other tissues (60).

Protease inhibitors are also unlike many of the other agents we haveused as cancer-chemopreventive agents, in that they arc effective at

extremely low levels while most of the other agents we have studied

a.sa

Mice on BBI Containing Diet

Rge (Weeks)

Fig. 7. Effects of high levels of dietary BBIC on the lifespan of mice. BBIC has asignificant life-lengthening effect in mice, as described in detail elsewhere (12).

10-fold increasein protease levels

DMBA BB) treatment

Boc-Val-Pro-Arg-MCA"Activity in Oral Epithelium

c-myc Expression _in Colonie Epithelium

V-)BBI-returntonormal level of

\ proteaseactivity\+BBI-return

to\normal level of

^ c-myc expression

Radiation6-lotd increase

in c-myc RNA levels

Fig. X. IME shown to be affected hy BBI in animal studies. BBIC" treatment results in

normal levels of c-mvc expression (4S) and Boc-Val-Pro-Arg-MCA-hydroly/ing activity(35) in animals treated with carcinogens and having elevated levels of these IME.



PROTEASE INHIBITOR PREVENTION OF CARCÃ•NOGENESIS

are effective only at very high levels. For example, vitamin E iseffective only at levels near the toxic level in our in vitro studies (62).Protease inhibitors such as BBI are effective in the nanomolar range(10) and have shown no toxicity at any concentration evaluated in ouri/i vitro studies (reviewed in Ref. l). The dose-response curve for the

BBI suppression of carcinogenesis in both in vivo (56, 57) and in vitro(9, 10) systems is unusual, in that concentrations/dose levels of BBIvarying over orders of magnitude have the same suppressive effect.This phenomenon may be due to the characteristics of the targetprotease involved in carcinogenesis. Studies on the M, 44.000 protease described above have shown that this protease must be activatedby trypsin to be active (40, 41). It is thought that proteases, such asthis, requiring trypsin activation are carefully controlled by cells andonly very small amounts of these enzymes are active at a given time;thus, very low concentrations of protease inhibitors would be effectiveat inhibiting such a protease (40).

Perhaps the most important difference between the anticarcinogcnicprotease inhibitors and other cancer-preventive agents is their ability

to affect so many different kinds of carcinogencsis. As mentionedabove, their anticarcinogenic effects are not restricted to specificorgans or carcinogens. The effects of most cancer-chemopreventive

agents are far more limited than are the effects of the anticarcinogenicprotease inhibitors. Protease inhibitors can almost be considered "universal" anticarcinogenic agents in their wide-ranging ability to affect

the carcinogenic process.The one type of carcinogenesis that the protease inhibitors have not

been expected to be able to affect is gastric carcinogenesis. Becausethe inhibitory activity of protease inhibitors is pH dependent, theywould not be expected to have activity at the low pH of the stomachenvironment or to function as cancer-chemopreventive agents there.

There are epidemiolÃ³gica! data, however, demonstrating that highlevels of soybean in the diet are associated with low gastric cancerrates (63). While this effect could be due to other anticarcinogenicagents in soybeans (reviewed in Ref. 64), it could also be due tosoybean protease inhibitors such as BBI. [In our studies, BBI was theonly compound in soybeans with the ability to suppress in vitrotransformation (10)]. Protease inhibitors could serve as cancer-che

mopreventive agents for gastric carcinogenesis by being internalizedby the cells lining the stomach. Once internalized, protease inhibitorssuch as BBI would have activity at the normal pH of the stomachepithelial cells.

High levels of protease inhibitors in the diet are associated with lowincidence rates for breast, colon, prostate, oral, and pharyngeal cancers, as recently reviewed (2). The particular kind of protease inhibitoractivity most closely associated with the ability to prevent cancer ischymotrypsin inhibitory activity (reviewed in Ref. 12). Soybeans havea particularly high content of chymotrypsin inhibitory activity (65).Both the Japanese (66) and members of the Seventh Day Adventists(67, 68) have high dietary levels of soybean products and extremelylow cancer rates for the common Western cancers. The Japanese dohave high rates of stomach cancer (69, 70); however, there is noevidence to suggest that high levels of soybean products in the dietcontribute to the high stomach cancer rates in Japan. If there weresuch an association, one would expect stomach cancer rates to behigher in the vegetarian Seventh Day Adventist population too. andthey are not. Further evidence against such an association includes thedata cited above, which indicate that diets high in soybean content areassociated with a decreased stomach cancer risk (63).

While there are no doubt many factors contributing to the cancerincidence and mortality data utilized for epidemiological studies, thedata showing the association between high dietary levels of soybeanproducts and low cancer mortality rates suggest that soybeans maycontain compounds capable of serving as anticarcinogenic agents in

human populations. As discussed here, at least one compound insoybeans, BBI, has clear anticarcinogenic activity in both in vivo andi/i vitro carcinogenesis assay systems.

References

14.

Kennedy, A. R. In vitro studies of anticarcinogcnic protease inhibitors. In: W. Trolland A. R. Kennedy (eds.). Protease Inhibitors as Cancer Chemopreventive Agents, pp.65-91. New York: Plenum Publishing Corp., 1993.

Kennedy, A. R. Overview: anticarcinogenic activity of protease inhibitors. In: W.Troll and A. R. Kennedy (cds.). Protease Inhibitors as Cancer ChemopreventiveAgents, pp. 9-64. New York: Plenum Publishing Corp., 1993.

Troll. W.. Frenkel. K.. and Wiesner. R. Protease inhibitors as anticarcinogens. 3. Nail.Cancer Inst.. 73: 1245-1250. 1984.Reznikoff. C. A., Brankow. D. W.. and Heidelberger. C. Establishment of a clonedline of C3H mouse embryo cells sensitive lo postcontlucnce inhibition of celldivision. Cancer Res..33: 3231-3238, 1973.

Rcznikoff. C. A.. Bertram. J. S.. Brankow. D. W.. and Heidelberger. C. Quantitativeand qualitative studies of chemical transformation of cloned C3H mouse embryo cellssensitive to poslconflucnce inhibition of cell division. Cancer Res., 33: 3239-3249,1973.Kennedy. A. R. Promotion and other interactions between agents in the induction oftransformation in vitro in fibroblasts. In: T. J. Slaga (ed.). Mechanisms of TumorPromotion. Vol. III. Tumor Promotion and Carcinogenesis In Vitro, pp. 13-55. BocaRaton, FL: CRC' Press, Inc.. 1984.

Kennedy. A. R. Use of cell and organ cultures in the identification and characterization of agents that modify carcinogcnesis. Cancer Res. (Suppl.), 5.?: 2446s-2448s,1993.Kennedy. A. R.. and Liltle, J. B. Protease inhibitors suppress radiation inducedmalignant transformation m vitro. Nature (Lond.), 276: 825-826, 1978.Kennedy. A. R. The conditions for the modification of radiation transformation invitro by a tumor promoter and protease inhibitors. Carcinogenesis (Lond.), 6: 1441-

1446, 1985.Yavelow. J.. Collins. M.. Birk. Y., Troll, W., and Kennedy, A. R. Nanomolarconcentrations of Bowman-Birk soybean protease inhibitor suppress X-ray inducedtransformation in vitro. Proc. Nail. Acad. Sci. USA, Â»2:5395-5399, 1985.

Odani, S.. and Ikenaka. T. Studies on soybean trypsin inhibitors. VIII. Disulfidebridges in soybean Bowman-Birk proteinase inhibitor. J. Biochcm. (Tokyo), 74:697-715, 1973.Kennedy. A. R.. Szuhaj. B. F., Ncwberne. P. M.. and Billings, P. C'. Prcparalion and

production of a cancer Chemopreventive agent. Bowman-Birk Inhibitor Concentrate.Nutr. Cancer, IV: 281-302, 1993.Kennedy, A. R., Fox, M.. Murphy, (i.. and Little. J. B. Relationship between X-rayexposure and malignant transformation in C3H KIT1/: cells. Proc. Nati. Acad. Sci.USA, 77: 7262-7266, 1980.

Kennedy. A. R.. and Little. J. B. Investigation of the mechanism for the enhancementof radiation transformation in vitro by TPA. Carcinogenesis (Lond.). /.â€¢1039-1047.

1980.Kennedy. A. R.. and Little. J. B. High efficiency, kinetics and numerology oftransformation by radiation in vitro. In: 1. H. Burchenal and J. F. Oeltgen (eds.).Cancer: Achievements, Challenges and Prospects for the 1980's, Vol. 1, pp. 49I-5IX).

New York: GruÃ±eand Stralton Inc.. 1981.Kennedy. A. R.. Cairns. J.. and Little, J. B. Timing of the steps in transformation ofC3H10T1 : cells by X-irradiation. Nature (Lond.). 307: 85-86, 1984.Kennedy. A. R.. and Little, J. B. Evidence that a second event in X-ray inducedoncogenic transformation /'/( vitnt occurs during cellular proliferation. Radial. Res.,

99: 228-248. 1984.Kennedy. A. R. Evidence that the tirsi step leading to carcinogen-induced malignanttransformation is a high frequency, common event. In: ÃŒ.C. Barrett and R. W.Tennant (eds.). Carcinogencsis: A Comprehensive Survey, Vol. 9, Mammalian CellTransformation: Mechanisms of Carcinogenesis and Assays for Carcinogens, pp.355-364. New York: Raven Press, 1985.

Kennedy. A. R. Antipain, but not cycloheximide. suppresses radiation transformationwhen present for only one day al five days post-irradiation. Carcinogenesis (Lond.).3: 1093-1(195. 1982.

Fahre, F.. and Roman. H. Genetic evidence for inducibility of recombination competence in yeast. Prix:. Nail. Acad. Sci. USA, 74: 1667-1671, 1977.Wintersbcrgcr. U. The selective advantage of cancer cells: a consequence of genomemobilization in the course of the induction of DNA repair processes? (model studieson yeast.) Adv. Enzyme Regul.. 22.- 311-323, 1984.

Clifton. K. H., Domann, F. E.. and Groch, K. M. On the cells of origin of radiogenicthyroid cancer: new studies based on an old idea. J. Radial. Res. (Suppl.), 2: 143-155,

1991.Ethier. S. P.. and Ullrich. R. L. Detection of ductal dysplasia in mammary outgrowthsderived from carcinogen-treated virgin female BALB/c mice. Cancer Res., 42:1753-1760, 1982.Tcrzaghi-Howe. M. Induction of preneoplastic alterations by X-rays and neutrons inexposed rat tracheas and isolated trachÃ©alepithelial cells. RadiÃ¢t.Res.. 120: 352-363,1989.Kennedy. A. R., and Little. J. B. Effects of protease inhibitors on radiation transformation in vitro. Cancer Res.. 41: 2103-2108. 1981.Kennedy. A. R. Prevention of radiation-induced transformation in vitro. In: K. N.Prasad (ed.). Vitamins, Nutrition and Cancer, pp. 166-179. Basel: S. Karger, 1984.

Troll, W.. Klassen. A., and Janoff, A. Tumorigcnesis in mouse skin: inhibition bysynthetic inhibitors of proteases. Science (Washington DC), 16V: 1211-1213, 1970.

2004s



I'KOIIAS! INIUmroR I'RI-VLNTION <>l (AK( IMMUNI SIS

28. Hozumi, M., Ogawa, M., Sugimura, T., Takeuchi, T., and Umegawa, H. Inhibition oflumorigencsis in mouse skin by leupeptin, a protease inhibitor from actinomyccles.Cancer Res., 32: 1725-1728, 1972.

29. Kennedy, A. R., Radner, B., and Nagasawa, H. Protease inhibitors reduce thefrequency of spontaneous chromosome abnormalities in cells from patients withBloom syndrome. Proc. Nail. Acad. Sci. USA, SI: 1827-1830, 1984.

30. German. J. Bloom's Syndrome X. The cancer proneness points to chromosome

mutation as the crucial event in human neoplasia. In: Chromosome Mutation andNeoplasia, pp. 347-357. New York: Alan R. Liss, Inc., 1983.

31. Kadhim, M. A., Macdonald, D. A., Goodhead, D. T., Lorimore, S. A.. Marsden, S. J.,and Wright, E. G. Transmission of chromosomal instability after plutonium a-particleirradiation. Nature (Lond.), 355: 738-740, 1992.

32. Nowell, P. C. The clonal evolution of tumor cell populations. Science (WashingtonDC), 194: 24-28, 1976.

33. Rowley. J. D.. and Ultmann, J. E. Chromosomes and Cancer. New York: AcademicPress, 1983.

34. TIsty, T. D. Normal diploid human and rodent cells lack a detectable frequency ofgene amplification. Proc. Nati. Acad. Sci. USA, 87: 3131-3136, 1990.

35. Messadi, D. V., Billings, P., Shklar, G., and Kennedy, A. R. Inhibition of oralcarcinogenesis by a protease inhibitor. J. Nail. Cancer Inst., 76: 447-452, 1986.

36. Billings, P. C., Carew, J. A., Keller-McGandy, C. E., Goldberg, A., and Kennedy,A. R. A serine protease activity in C3H/10T'/2 cells that is inhibited by anticarcino-genic protease inhibitors. Proc. Nati. Acad. Sci. USA, 84: 4801-4805, 1987.

37. Kennedy. A. R.. and Billings, P. C. Anticarcinogcnic actions of protease inhibitors.In: P. A. Cerutti, O. F. Nygaard. and M. G. Simic (eds.), Anticarcinogcnesis andRadiation Protection, pp. 285-295. New York: Plenum Press, 1987.

38. Billings, P. C.. St. Clair, W., Owen, A. J., and Kennedy, A. R. Potential intracellulartarget proteins of the anticarcinogenic Bowman-Birk protease inhibitor identified byaffinity chromatography. Cancer Res., 48: 1798-1802, 1988.

39. Carew, J. A., and Kennedy, A. R. Identification of a proteolytic activity whichresponds to anticarcinogenic protease inhibitors in C3H10T1/2 cells. Cancer Lett., 49:153-163, 1990.

40. Billings, P. C., Habrcs, J. M., Liao, D. C., and Tuttle, S. W. A protease activity inhuman fibroblasts which is inhibited by the anticarcinogenic Bowman-Birk proteaseinhibitor. Cancer Res., 5;.- 5539-5543, 1991.

41. Billings, P. C.. and Habres, J. M. A growth regulated protease activity which isinhibited by the anticarcinogenic Bowman-Birk protease inhibitor. Proc. Nati. Acad.Sci. USA. Â«9:3120-3124, 1992.

42. Billings. P. C. Approaches lo studying the large! enzymes of anticarcinogenic protease inhibitors. In: W. Troll and A. R. Kennedy (eds.). Protease Inhibitors as CancerChemoprcvcntive Agents, pp. 191-198. New York: Plenum Publishing Corp., 1993.

43. Habres, J. M.. and Billings, P. C. Intestinal epithelial cells contain a high molecularweight protease subject to inhibition by anticarcinogenic protease inhibitors. CancerLett.,Â«: 135-142. 1992.

44. Kennedy, A. R. Potential mechanisms of antitumorigenesis by protease inhibitors, in:G. Bronzetli, A. Hayatsu, S. DeFlora, M. D. Waters, and D. M. Shankel (eds.).Antimutagcnesis and Anticarcinogenesis: Mechanisms: HI, pp. 301-308. New York:

Plenum Publishing Corp., 1993.45. Chang, J. D., Billings, P., and Kennedy, A. R. c-mvc expression is reduced in

antipain-treated proliferating C3H10T1/: cells. Biochem. Biophys. Res. Commun..133: 830-835, 1985.

46. Chang. J. D.. and Kennedy, A. R. Cell cycle progression of C3HIOT'/2 and 3T3 cellsin the absence of a transient increase in c-myc RNA levels. Carcinogenesis (Lond.).9: 17-20, 1988.

47. Caggana, M., and Kennedy, A. R. e-fos mRNA levels are reduced in the presence ofantipain and the Bowman-Birk inhibitor. Carcinogenesis (Lond.), 10: 2145-2148.1989.

48. SI. Clair, W. H., Billings, P. C., and Kennedy. A. R. The effects of the Bowman-Birkprotease inhibitor on c-myc expression and cell proliferation in the unirradiatcd and

irradiated mouse colon. Cancer Lett., 52: 145-152, 1990.49. Chang, J. D., Li, J-H., Billings, P. C., and Kennedy, A. R. Effects of protease

inhibitors on c-myc expression in normal and transformed C3H10T1/: cells. Mol.Carcinog., 3: 226-232, 1990.

50. Li. J-H., Billings, P. C., and Kennedy, A. R. Induction of oncogene expression bysodium arsenite in C3H/10T1/: cells: inhibition of c-myc expression by proteaseinhibitors. Cancer J., 5: 354-358, 1992.

51. Chang, J. D.. and Kennedy. A. R. Suppression of c-myc by anticarcinogenic protease

inhibitors. In: W. Troll and A. R. Kennedy (eds.). Protease Inhibitors as CancerChemopreventive Agents, pp. 265-280. New York: Plenum Publishing Corp., 1993.

52. Lavi, S. Carcinogen-mediated amplification of viral DNA sequences in simian virus40-transformed Chinese hamster embryo cells. Proc. Nati. Acad. Sci. USA, 78:6144-6148, 1981.

53. Lavi, S. Carcinogen-mediated amplification of specific DNA sequences. J. Cell.Biochem., 18: 149-156, 1986.

54. Flick, M. B., and Kennedy. A. R. Effect of protease inhibitors on DNA amplificationin SV40-transformed Chinese hamsler embryo cells. Cancer Lett.. 56: 102-108. 1991.

55. Yavelow, J., Finlay, T. H., Kennedy, A. R., and Troll, W. Bowman-Birk soybeanprotease inhibitor as an anticarcinogcn. Cancer Res., 43: 2454-2459, 1983.

56. Kennedy, A. R., Billings, P. C., Maki, P. A., and Newberne, P. Effects of variouspreparations of dietary protease inhibitors on oral carcinogenesis in hamsters inducedby DMBA. Nutr. Cancer, 19: 191-200. 1993.

57. St. Clair, W., Billings, P., Carew, J., Keller-McGandy, C., Newberne, P., andKennedy. A. R. Suppression of DMH-induced carcinogenesis in mice by dietaryaddition of the Bowman-Birk protease inhibitor. Cancer Res., 50: 580-586, 1990.

58. Witschi, H.. and Kennedy, A. R. Modulation of lung tumor development in mice withthe soybean-derived Bowman-Birk protease inhibitor. Carcinogenesis (Lond.), 10:2275-2277, 1989.

59. Kennedy, A. R. Implications for mechanisms of tumor promotion and its inhibition byvarious agents from studies of in vitro Iransformation. In: R. Langenbach, J. C.Barrett, and E. Elmore (eds.). Tumor Promoters. Biological Approaches for Mechanistic Studies and Assay Systems, pp. 201-212. New York: Raven Press, 1988.

60. Deyoung, L., Mufson, R. A., and Boutwell, R. K. An apparent inactivation of initiatedcells by the potent inhibitor of two-stage mouse skin tumorigenesis, bis(2-chloro-ethyljsulfide. Cancer Res., 37: 4590-4594, 1977.

61. Wada, S., Miyanishi, M., Nishimoto, Y., Kambe, S., and Miller. R. W. Mustard gasas a cause of respiratory neoplasia in man. Lancet. /: 1161-1163. 1968.

62. Radner. B. S., and Kennedy. A. R. Suppression of X-ray induced transformation byvitamin E in mouse C3H/10T'/2 cells. Cancer Lett., 32: 25-32, 1986.

63. Hirayama. T. Relationship of soybean paste soup intake to gastric cancer risk. Nutr.Cancer, 3: 223-233, 1982.

64. Messina, M., and Barnes, S. The role of soy products in reducing risk of cancer. J.Nati. Cancer Inst., 83: 541-546, 1991.

65. Birk, Y. Proteinase inhibitors from plant sources. Methods Enzymol.. 45: 695-751,

1975.66. Doell, B. H., Ebden, C. J., and Smith, C. A. Trypsin inhibitor activity of conventional

foods which are part of the British diet and some soya products. Qua!. Plant. PlantFoods Hum. Nutr., 31: 139-150, 1981.

67. Mills, P. K., Beeson, W. L., Abbey, D. E., Fraser. G. E., and Phillips, R. L. Dietaryhabits and past medical history as related to fatal pancreas cancer risk amongAdventists. Cancer (Phila.), 61: 2578-2585, 1988.

68. Phillips, R. L. Role of life-style and dietary habits and risk of cancer amongSevenlh-Day Adventists. Cancer Res., 35: 3513-3522, 1975.

69. Armstrong, B., and Doll, R. Environmental factors and cancer incidence and mortalityin different countries, with special reference to dietary factors. Int. J. Cancer, /5:617-631, 1975.

70. Doll, R., and Peto, R. The causes of cancer: quantitative estimates of avoidable risksof cancer in the United States today. J. Nati. Cancer Inst., 66: 1193-1308, 1981.

2005s



1994;54:1999s-2005s. Cancer Res Ann R. Kennedy Prevention of Carcinogenesis by Protease Inhibitors

Updated version

http://cancerres.aacrjournals.org/content/54/7_Supplement/1999s

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

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/54/7_Supplement/1999sTo request permission to re-use all or part of this article, use this link



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

mailto:[email protected]



prevention of carcinogenesis by protease inhibitors1€¦ · prevention of carcinogenesis by...

Documents