characterization of protein tyrosine kinases from human breast … · exchange chromatography using...

[CANCER RESEARCH 52. 4773-4778, September I. 1992]

Characterization of Protein Tyrosine Kinases from Human Breast Cancer:Involvement of the c-src Oncogene Product1

A. E. Ottenhoff-Kalff, G. Rijksen, E. A. C. M. van Beurden, A. Hennipman, A. A. Michels, and G. E. J. Staal2

Department of Hematology, Laboratory- of Medical Enzymology Â¡A.E. O-K., (j. R., E. A. C. M. v. B., A. A. M., G. E. J. S.J, and Department of Surgery [A. H.],University Hospital Utrecht, 3508 GA Utrecht, The Netherlands

ABSTRACT

Tyrosine phosphorylation is an important regulatory mechanism inresponse to the action of growth factors and oncogenes. Since manyoncogenes code for i\rosine kinases, increased or altered oncogene expression may be reflected in increased tyrosine kinase activity. In arecent study (Hennipman et al.. Cancer Res., 49: 516-521, 1989), wefound that the tyrosine kinase activity of the cytosolic and membranefractions of malignant human breast tissue was significantly highercompared to the benign or the normal breast tissue. Moreover, theincrease in the cytosolic fractions was found to be of prognostic value. Inthe present study we determined the protein tyrosine kinase (PTK)activity of another 72 breast cancer specimens, and it could be shownagain that the PTK activity in all 72 of these tumors was elevatedcompared to normal controls. We characterized these cytosolic PTKs byaniÃ³nexchange chromatography using fast protein liquid chromatogra-phy, and it could be shown that at least two different forms of PTK exist.Using antibodies against a number of known oncogene products, wecould determine that at least 70% of the PTK activity in the cytosoloriginated from the presence of the f-src oncogene product. Both of thePTK activity peaks seen in the fast protein liquid chromatography patterns could be precipitated with the anti-Src antibody. Furthermore,using the MCF-7 breast cancer cell line, it could be shown that theantibody against c-src also precipitated a part of the cytosolic PTKactivity. In normal human peripheral lymphocytes, no precipitation ofthe cytosolic and membrane PTK activity could be achieved using theanti-Src antibody. Inasmuch as the cytosolic PTK activity parallels themalignancy in breast tumors (Hennipman el al.. Cancer Res., 49: 516-521, 1989), and the majority of this activity is precipitated by anti-Srcantibodies, the c-src protooncogene may play a key role in the manifestation of breast cancer.

INTRODUCTION

Many groups have been investigating the expression of oncogenes in breast cancer to obtain clues as to the genetic background predisposing to breast cancer. For example, Slamon etal. (l, 2) have found amplification of the c-erbB-2/neu/HER-2gene in approximately 30% of primary breast carcinomas andcorrelated the amplification with a decreased time to relapseand overall patient survival. However, other investigators (3-5)have failed to find such a correlation. On the other hand, reports on amplification of the c-myc oncogene (6, 7) and theint-2 oncogene (8) have been published. Also, the presence ofepidermal growth factor receptors in human breast cancers hasbeen demonstrated and may be associated with metastatic potential (9, 10). All of these reports have one thing in common:amplification never occurs in more than 10-30% of the investigated tumors.

Tyrosine phosphorylation is now recognized as an importantregulatory mechanism in response to a number of processesincluding the action of growth factors and oncogenes (11-14).

Received 3/16/92; accepted 6/23/92.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.

1Supported by the Netherlands Cancer Foundation (UUKC Grant 89-01).2 To Â»horn requests for reprints should be addressed, at Department of Hema

tology. Laboratory of Medical Enzymology, University Hospital Utrecht. P. O. Box85500, 3508 GA Utrecht, The Netherlands

Since many oncogenes code for tyrosine kinases, increased oraltered expression of oncogenes may be reflected in increasedtyrosine kinase activity. In a recent study (15), we have determined the PTK3 activity in the cytosolic and solubili/ed mem

brane fractions from normal, benign, and malignant humanbreast tissue to determine whether there was a correlation between the PTK activity and the degree of malignancy in thesetissues. We found (15) that the tyrosine kinase activity of thecytosolic and membrane fractions of malignant breast tissuewas significantly higher compared to the benign or the normalbreast tissue. The benign tissues took an intermediate placebetween the normal and malignant tissues. Surprisingly, thePTK activity was increased the most in the cytosolic fraction(up to 25 times that of the normal tissue). It appeared that PTKfrom this fraction also had a prognostic value: patients withPTK values below the average had a significantly better disease-free survival than did patients with a PTK activity above theaverage. It should be noted that, in 100% of the breast cancerspecimens we tested, we found an elevated PTK activity in thecytosolic as well as in the solubilized membrane fraction.

In the present study we determined the PTK activity of another 72 breast cancer specimens. It could be shown again thatthe PTK activity in all 72 of these tumors was elevated compared to normal breast tissue obtained after a reduction mam-moplasty. Furthermore, we undertook an effort to characterizeand identify the proteins responsible for the high PTK activityin the cytosolic fraction of these tumors. We performed aniÃ³nexchange chromatography using fast protein liquid chromatography and found that multiple forms of the PTKs exist. Also,we tried to immunoprecipitate the PTK activity from the cytosolic fractions of breast cancer tissues with antibodies to knownoncogene products and growth factor receptors. We show herethat at least 70% of the cytosolic protein tyrosine kinase activityfrom the breast tumors originates from the presence of a c-src-like oncogene product. In the breast cancer cell line MCF-7,anti-c-src could also precipitate a part of the PTK activity in thecytosol, while no precipitation of PTK activity could beachieved using cytosol and solubilized membranes preparedfrom normal human peripheral lymphocytes.

MATERIALS AND METHODS

Patients and Materials. Specimens of breast tumors from patientswith breast cancer were obtained after surgery' and immediately fro/en

in liquid nitrogen. The specimens were stored until further use at-70"C.

The specimens of breast tissue from patients with juvenile hypertrophy were obtained when a reduction mammoplasty was performed.These specimens did not show any abnormality on histolÃ³gica!examination and were therefore accepted to represent the enzyme activitiesof normal breast tissue.

' The abbreviations used are: PTK, protein tyrosine kinase: BSA. bovine serumalbumin; IGF. insulin-like growth factor; PMSF, phenylmelhylsulfonyl fluoride;TIU, trypsin inhibitor unit; PBS. phosphate-buffered saline; RIPA buffer, radio-immunoprecipitation buffer; FPLC, fast protein liquid chromatography; ELISA,enzyme-linked Â¡mmunosorbemassay: PBST. PBS-0.1% Tween-20.

4773

Association for Cancer Research. by guest on August 29, 2020. Copyright 1992 Americanhttps://bloodcancerdiscov.aacrjournals.orgDownloaded from

https://bloodcancerdiscov.aacrjournals.org

INVOLVEMENT OF C-SRC IN HUMAN BREAST CANCER

All chemicals used in buffers, poly(glutamic acid:tyrosine, 4:1), andmouse immunoglobulins (used as a negative control) were from Sigma(St. Louis, MO). Anti-phosphotyrosine IG2, fish gelatin, and gold-labeled goat anti-mouse IgG plus IgM were from Amersham (Amer-sham, Buckinghamshire, England), silver lactate was from Fluka(Buchs, Switzerland), hydroquinone was from Janssen Biochimica(Beerse, Belgium). Rabbit anti-mouse peroxidase was from Dakopatts(Denmark), and protein A-Sepharose was from Pharmacia (Uppsala,Sweden). BSA used for the immunoblotting was essentially fatty acidfree (Sigma). Modified 3,3'-5,5'-tetramethylbenzidine substrate was

from Eurodiagnostics (Apeldoorn, the Netherlands). The antibodiesanti-neu, anti-insulin-receptor, anti-IGFl receptor, anti-fes, and anti-abl were all from Oncogene Science (Manhasset, NY). Anti-Src 327was kindly donated by Dr. J. S. Brugge (Philadelphia, PA). The antibodies to neu, insulin receptor, and IGF1 receptor are all directedagainst an epitope located in the tyrosine kinase domain of the respective proteins. The antibodies anti-neu and anti-abl were polyclonal rab

bit antibodies, and the other antibodies used were all mouse monoclonalantibodies. The MCF-7 cell line was used at passages 83-103. The cellswere cultured in Dulbecco's modified Eagle's medium supplemented

with 10% heat-inactivated fetal calf serum, 2 mivi L-glutamine, 100units/ml penicillin, and 100 Â¿ig/mlstreptomycin, in a 5% humidifiedCO2 atmosphere at 37'C.

Sample Preparation. Before enzymological assay the specimenswere cleared of fat and debris. All specimens had a wet weight of20-180 mg. The extraction buffer contained 10 IHMTris-HCl, pH 7.4;0.25 Msucrose; 1 HIMMgCI2; 1 m\t EDTA; 1 min dithiothreitol; 1 mMPMSF, and 0.055 TIU/ml aprotinin. Samples were homogenized in 4volumes of extraction buffer using an Omni 1000 homogenizer (Om-nilabo). All debris and nuclei were removed by centrifugation at 800 xg for 10 min (4Â°C),and the supernatant was centrifuged at 48,000 x gfor 60 min (4Â°C).The 48,000 x g supernatant fraction was used as the

cytosolic fraction. The remaining pellet was resuspended in solubiliza-tion buffer which contained the following: 50 miviTris-HCl, pH 7.5; 20

mM magnesium acetate; 5 m\i NaF; 0.2 HIMEDTA; 0.8 mm EGTA; 1min dithiothreitol; and 0.5% Nonidet P-40. Thirty IM Na3VO4 wereadded just before use. The membrane fraction was kept on ice for 1 h,in which it was sonicated twice for 10s (4Â°C).The solubilized mem

brane fraction was obtained by clarification at 48,000 x g for 60 min(4Â°C).When using the breast cancer cell line MCF-7, the cells were

washed with ice-cold PBS three times and scraped in a small volume(usually 3 ml for 40 x IO6 cells) of extraction buffer. The cells werehomogenized by 20 strokes in an Elvehjem-Potter tube with a tight-fitting pestle, followed by 5 s of sonication. Subsequently, the obtainedhomogenized cell fraction was treated as the homogenized samplesdescribed above. Lymphocytes were obtained by elutriation (16). Whenusing lymphocytes, the cells were suspended in RIPA buffer (describedbelow) including proteolytic inhibitors and vigorously shaken, kept onice for 30 min, sonicated twice for 10 s, and left on ice for another 30min. Subsequently, the homogenate was centrifuged at 48,000 g for 60min. The supernatant was used as a source of cytosol and solubilizedmembranes.

Determination of Tyrosine Kinase Activity. For determining the specific activity, and when testing FPLC fractions, the nonradioactive dotblot assay described in (17) was used. Tyrosine kinase activity is expressed as pmol phosphotyrosine/min-mg of protein. Using this assay,

higher values were obtained for the specific activities than reportedbefore (15), when we used the conventional radioactive assay. Theprotein content was determined according to the method of Lowry et al.(18).

When determining the amount of precipitation by the antibodies, thekinase assay was carried out according to an ELISA, as described in(19), with some modifications. Briefly, 96-weIl plastic plates werecoated with 125 iJ\ 0.1 mg/ml poly(glutamic acid:tyrosine, 4:1) in PBSovernight at 37Â°C.After coating, the poly(glutamic acid:tyrosine) was

removed by emptying the wells, and the wells were washed with 200 n\PBST. For assaying the PTK activity, 50 n\ of enzyme solution with aprotein concentration varying between 0.0002 and 0.02 mg/ml wereadded to the wells. The reaction was started by the addition of 50 n\ of1 HIMATP containing 60 UMNa3VO4 in PBS and allowed to continue

at 37Â°Cfor 60 min. The reaction was stopped by emptying the plate,

and the plate was washed four times with 200 M!PBST. Incubation atroom temperature for 60 min with 100 M!of a 1:1000 dilution ofanti-phosphot> rosine and 1:100 dilution of normal rabbit serum inPBST was followed by emptying and washing the wells four times.Next, 100 M!of a 1:1000 dilution of rabbit anti-mouse peroxidase inPBST with a 1:20 dilution offish gelatin was incubated in the wells for60 min at room temperature. After emptying the wells and washing fourtimes, 100 M!modified 3,3'-5,5'-tetramethylbenzidine was added at a

1:20 dilution in water, and after 20 min the reaction was stopped with100 M!2 MH2SO4. The extinction in the wells was measured at 450 nm.

Fast Protein Liquid Chromatograph}1. An aniÃ³n exchange column

(Mono Q HR 5/5; Pharmacia) was used for the FPLC analyses. Thecolumn was equilibrated in buffer A, containing 10 mM Tris-HCl (pH7.4), 1 mM EDTA, 1 mM dithiothreitol, and 1 HIMMgCl2. Cytosolicproteins were extracted as described above, except that buffer A wasused, containing 1 HIMPMSF and 0.055 TIU/ml aprotinin. Proteinswere loaded onto the column and eluted with a discontinuous gradientof 0-1 MNaCl in buffer A as indicated in "Results."

Immunoprecipitation Assays. For immunoprecipitation assays,RIPA buffer was used, which contained the following: 20 mMTris-HCl,pH 8.0; 150 HIMNaCl; 10 IHMNaH2PO4; 5 IHMEDTA; 30 MMNa3VO4;10% glycerol; 1% (v/v) Nonidet P-40; 1% (w/v) sodium desoxycholate;0.1% sodium dodecyl sulfate; 1 mM PMSF; 0.055 TIU/ml aprotinin.

Cytosolic fractions were extracted, as described above, in incubationbuffer (50 mMTris-HCl, pH 7.5; 20 HIMmagnesium acetate; 5 HIMNaF;0.2 HIMEDTA; 0.8 mM EGTA; 10 HIMdithiothreitol; 10% glycerol).Thirty MMNa3VO4, 1 mw PMSF, and 0.055 TIU/ml aprotinin wereadded prior to extraction. Cytosolic proteins (0.1 to 0.5 mg; determinedby the method of Bradford, Ref. 20) were adjusted to 1% Nonidet P-40and diluted in RIPA buffer to obtain a final volume of 200 n\. Antibodies were added, together with preswollen and washed protein ASepharose (3 mg/precipitation). When mouse antibodies were used, 10Mg of goat anti-mouse immunoglobulins were included to facilitatebinding to protein A. The following amounts of antibodies were used:for anti-neu, anti-insulin-receptor, anti-IGFl-receptor, anti-fes, and anti-abl, 1 Mg/precipitation; for anti-Src, 5 M'/precipitation. The amount ofcontrol antibody used (usually anti-mouse or anti-rabbit immunoglobulins) was always equal (in amount of protein) to the amount of specificantibody. Immunoprecipitation was carried out overnight in a rotatingdevice at 4Â°C.Supernatant and pellet were separated by centrifugation,

and the pellet was washed three times with RIPA. Supernatant andpellet were tested in the ELISA tyrosine kinase assay.

Immunoblotting. Proteins were loaded onto a 7.5% sodium dodecylsulfate-polyacrylamide gel and electrophoresed according to themethod of Laemmli (21). After electrophoresis, the proteins were transferred electrophoretically to a polyvinylidene difluoride filter andblocked with 5% BSA (essentially fatty acid free) in PBS. After washingthree times with PBS/0.1% BSA, the membranes were subsequentlyincubated overnight at room temperature with anti-Src 327 (1:30) inPBS containing 1:20 normal goat serum and 0.1% BSA. After washingthree times with PBS/0.1% BSA, the membrane was incubated for 4 hat room temperature with a 1:100 dilution of gold-labeled goat anti-mouse IgG plus IgM in PBS/0.1% BSA containing 1:20 fish gelatin.Subsequently, the membrane was washed three times with PBS/0.1%BSA, followed by two washes of 1 min each with water. The proteinbands were made visible with the gold-silver staining procedure, asdescribed previously (17).

RESULTS

Determination of Protein Tyrosine Kinase Activity. ThePTK activity was determined in 72 tissues from breast cancerbiopsies and in 4 tissues which were obtained after a reductionmammoplastic operation. The cytosolic and solubilized membrane fractions were used as a source of PTK activity. Theresults are shown in Table 1. The PTK activity of the cytosolfrom normal breast tissue was consistently low, ranging from 5to 46 pmol phosphotyrosine/min â€¢¿�mg protein. In contrast, the

PTK activities in the cytosol obtained from the breast cancer4774




Table 1 Protein tyrosine kinase activities (pmol phosphotyrosine/min â€¢¿�mg) ofmalignant and normal breast tissues

Cytosol Solubilized membrane

Average Â±SD Range Average Â±SD Range

Normal tissueTumor tissue24

Â±16Â°329 Â±173e5-46 106-894204

Â±180*871 Â±615''52-438160-2760

"n = 67.

specimens were much higher, ranging from 106 to 894 pmolphosphotyrosine/min-mg protein. For the solubilized mem

brane fraction, the increase in PTK activity was not as great asfor the cytosol; a 4-fold difference between normal and malignant cytosolic fractions could be measured (Table 1).

The difference in PTK activity was greater in the cytosolicfraction; moreover, it was found to be of prognostic significancein earlier work. Therefore, our attention was focused on thesecytosolic PTKs. We undertook further experiments to elucidatethe identity and nature of these proteins.

Fast Protein Liquid Chromatography of Cytosolic PTK fromBreast Cancer. To determine whether the cytosolic PTK activity consisted of multiple forms of PTK, we tried to separatethem on an aniÃ³n exchange chromatography column usingFPLC. In total, 13 tumor and 6 normal cytosols were tested. Atypical elution profile of a cytosolic fraction from a breast cancer specimen is shown in Fig. \A. At least two peaks of PTKactivity are seen, the first eluting at 110 Â±50 (SD) mM NaCl,the second eluting at 240 Â±40 m\i NaCl. A small peak of PTKactivity was also seen at approximately 400 ITIMNaCl. In addition, a small amount of PTK activity was sometimes seen in theflowthrough fraction, although this was not a consistent observation. In Fig. IB, the elution profile of the cytosolic fractionfrom normal breast tissue is shown. As can be seen, the sametwo peaks of PTK activity are present, albeit with a much loweractivity than that of the breast cancer cytosolic fraction. It appears that the higher PTK activity seen in the breast cancercytosolic fractions is not due to the appearance of extra formsof PTK.

Immunoprecipitation of Cytosolic PTK Activity. To determine the identity of the protein(s) with elevated cytosolic PTKactivity, immunoprecipitation experiments with antibodies toknown oncogene products and growth factor receptors wereundertaken. The following antibodies were tested: anti-c-erbB-2, anti-insulin-receptor, anti-IGFl-receptor, anti-fes, anti-c-abl, and anti-c-src. Although these antibodies are directedagainst enzymes in or attached to the plasma membrane, weused them because the enzymes may end up in the cytoplasm byproteolytical processes or down-regulation. Of all of the above-mentioned antibodies, only one was able to precipitate the PTKactivity from the cytosolic fractions used. Remarkably, anti-Src(monoclonal antibody 327; Ref. 22) was able to precipitate over60% of the PTK activity present in the cytosolic fractions ofbreast cancer specimen. When the supernatant from the anti-Src precipitate was used for a second precipitation assay, another 10% of the cytosolic PTK activity could be precipitated,resulting in a total PTK precipitation of 70% (Fig. 2). No activity was detected in the pellets of the aspecific immunopre-cipitations with the control antibodies, whereas the specificimmunoprecipitations exhibited a very high activity in theELISA (data not shown). The antibody did not influence thePTK activity when added directly to cytosol tested in theELISA, discarding the possibility of an inhibiting effect of theantibody directly on the enzymes (Fig. 2).

For normal breast tissue, the results were somewhat controversial, probably due to the very low activity of the samples. Forfive cytosolic fractions tested, the mean precipitation of PTKactivity by the specific antibody was 24.0 Â±20.4%. In threecytosolic fractions, immunoprecipitation of approximately 40%of the PTK activity from the cytosolic fraction was achieved. Intwo other cases, no difference was observed between the PTKactivity of control and specific supernatants. In all cases, thespecific pellets, but not the control pellets, exhibited moderateactivity in the ELISA.

Immunoblotting of Immunoprecipitates with Anti-c-src. Toconfirm that the protein we precipitated was indeed c-src, weperformed immunoblots with anti-c-src antibodies as describedin "Materials and Methods." In Fig. 3 (Lane 2), the precipitated

Src from the tumor cytosol can be seen. As a control, we usedthe src protein isolated from human platelets. Fig. 3 also shows(Lane 4) the precipitated c-src protein from human platelets. Inboth the platelet control and the tumor material, the src proteinhas a molecular weight of approximately 56,000. Additionalbands with a lower molecular weight can also be seen (fig. 3,Lanes 2 and 4). Presumably, these represent degration forms ofthe src protein. In a recent publication by Huang et al. (23),degradation products with a molecular weight lower than

1O 2O 30

fraction number

a

o00Cxj

Fraction number

Fig. 1. Fast protein liquid chromatography of the cytosolic fraction of normaland malignant breast tissue. The cytosolic fraction of a malignant breast tissue (. i ior normal breast tissue (B) was loaded onto an aniÃ³nexchange column (Mono Q,Pharmacia) and eluted with a nonlinear gradient of 0-1 M NaCl. PTK activity isexpressed in arbitrary units (A). , OD 280.

4775




rei. precipitation1ZU100

80

60

40

20

0TTTTumorsTumors Lymphocytes MCF7control

Fig. 2. Immunoprecipitation of PTK activity using monoclonal antibody 327to the c-src protein. The activity is expressed as a percentage of the activity presentin the supernatant after immunoprecipitation with aspecific antibodies (mousetotal IgG). Tumor control: antibodies were added immediately prior to measuringPTK activity. All experiments were performed in duplicate. Values shown arefrom nine different tumors. For MCF-7 cells and lymphocytes, values are, respectively, the mean of two or three independent experiments.

60,000 can also be seen in an immunoblot with anti-Src (monoclonal antibody 327). A very faint band can be detected at A/r50,000 (Fig. 3, Lanes 1 and 3), which is aspecific and representsthe IgG heavy chain. A band with a higher molecular weight ofapproximately 110,000 can also be seen (Fig. 3, Lane 2); theidentity of this protein is unknown.

FPLC after Immunoprecipitation with Anti-Src. To see if wecould identify one or both of the PTK activity peaks in theFPLC patterns of the cytosol, we performed FPLC analysis ofa cytosolic fraction after immunoprecipitation with anti-Src ora control antibody. In Fig. 4, the results of such an experimentare shown. It is clear that both PTK activity peaks in the elutionpattern contain c-src immunoreactive material. The PTK activity of the whole elution profile is lowered, with the exception ofthe flowthrough fraction. Indeed, the c-src kinase was detectedby Western blotting with the antibody to c-Src in both peaks

(data not shown).Immunoprecipitation of c-src from Lymphocytes and from the

MCF-7 Breast Cancer Cell Line. To eliminate the possibilitythat we were dealing with the activity of c-src-like proteinsoriginating from infiltrating lymphocytes in the tumor specimens, we undertook immunoprecipitation experiments usingthe combined cytosolic and membrane extract from lymphocytes obtained by elutriation (16). It was made certain that thecontaminating platelets present in the lymphocyte fraction werenegligible. Using the anti-Src antibody, no precipitation of PTKactivity could be obtained in the supernatant of the immuno-precipitates compared to the mouse immunoglobulin control(Fig. 2). Accordingly, in Western blots, no c-src protein bandwas present in anti-Src pellets of lymphocytes (Fig. 3, Lane 6).The faint bands (Fig. 3, Lanes 5 and 6) represent the IgG heavychain and are not specific.

We also undertook some immunoprecipitation experimentswith the cytosolic fractions of the breast cancer cell line MCF-7.In the cytosol from the MCF-7 cells, PTK activity was alsoprecipitated by the anti-Src antibody 327 (Fig. 2). Precipitationof 40% of the activity was obtained. The specific immunopre-cipitate showed activity as well in the ELISA, whereas no activity could be detected in the aspecific immunoprecipitate. InWestern blots of the anti-Src pellets from the MCF-7 cytosol, a

distinct protein band comparable to that in the tumors was seenat Mr -56,000 (Fig. 3, Lane 8). Also, a faint band with a

molecular weight of approximately 110,000, comparable withthat in the anti-Src precipitates from the tumor, was detected.

DISCUSSION

In a previous study (15), we showed that the PTK activity inthe cytosolic and membrane fractions of malignant breast tumors was higher than that in benign tumor, whereas that of thebenign tumors was in turn higher than that of normal tissue. Inthe present study, we determined the protein tyrosine kinaseactivity in a larger group of patients with breast cancer compared to normal breast tissue. It could be shown again that in alltumors tested, the breast cancer specimens showed a higheractivity compared to normal breast tissue. Because of the significance which could be attributed to this cytosolic PTK activity in our earlier study (15), we tried to characterize the PTKproteins responsible for the enhancement in PTK activity in thecytosol.

To elucidate the nature of the cytosolic PTKs, we performedseveral experiments. Characterization of the PTK activity in thecytosolic fraction of the breast tumors on an aniÃ³n exchangecolumn showed that at least two forms of PTK activity werepresent. These two forms were present in both the normal andthe malignant cytosol, indicating that the increase of PTK activity in tumors probably does not result from the appearance of

1 2345678

200-

116â€”97â€”

46-

Fig. 3. Immunoblot of immunoprecipitates with anti-Src 327. Immunoprecip-itates were loaded onto a 7.5% sodium dodecyl sulfate-polyacrylamide gel. transferred electrophoretically onto a polyvinylidene difluoride membrane, and probedwith anti-Src 327, followed by a gold-labeled goat anti-mouse second antibody.Immunocomplexes were detected by the immunÂ»gold-silver staining method.Lanes I, 3, 5, and 7, control immunoprecipitates (mouse immunoglobulins):Lanes 2, 4, 6, and 8, anti-Src 327 immunoprecipitates; Lanes I and 2, malignantmamma tumor: Lanes 3 and 4, human peripheral platelets: Lanes 5 and 6, humanperipheral lymphocytes: Lanes 7 and 8, MCF-7 breast cancer cell line.

90

75

....

fraction rxmber

Fig. 4. FPLC pattern of tyrosine kinase activity after immunoprecipitation.After immunoprecipitation with the anti-Src antibody or the control mouse Ig.cytosolic proteins were loaded onto a Mono Q column and eluted with a nonlinear0-1 M NaCl gradient. The PTK activity is expressed in arbitrary units. +, untreated cytosol; O, cytosol after precipitation with control mouse IgG: A. cytosolafter anti-Src 327 precipitation.

4776




a distinct tyrosine kinase but rather from an increased expression or activation of kinases already present in the tissue oforigin. Apparently, the increase in PTK activity is representedin all of the PTK forms present in breast tissue. The two mainforms of PTK could be due to conformational changes of oneparticular enzyme or perhaps to a different phosphorylationstate, or they could just be different enzymes.

It is to be noted that most of the tyrosine kinase activityresides in receptors that are membrane bound. Yet, in breastcancer tissues most of the PTK activity is found to reside in thecytosol. It was remarkable, therefore, that 70% of the cytosolicPTK activity could be precipitated with antibody 327 to c-src(22).

The fact that there was no immunoprecipitation of PTK activity from human peripheral lymphocytic extracts points totwo things. First of all, it means that the antibody against thec-src protein is specific for this Src family member and does notreact with lek. Second, it can be concluded that the PTK activitywe precipitate with the anti-Src antibody is not a result ofinfiltrating lymphocytes in the tumor tissue. This is also supported by the fact that we are able to precipitate part of the PTKactivity in cytosolic extracts from the MCF-7 breast cancer cellline.

The c-src protooncogene product is usually classified as acytosolic tyrosine kinase, but it is attached to the membrane bymyristic acid (24-26). Myristic acid is bound to the c-src protein at the NH2-terminal glycine residue at position 2, immediately after or even during protein synthesis (27, 28). Thefinding that at least 70% of the cytosolic PTK activity in thebreast tumor specimens is associated with the presence of c-srctherefore suggests that this protein is somehow detached fromthe membrane. It is possible that the protein is disrupted fromthe membrane during extraction of cytosolic proteins, althoughwe do not think this is likely, using our protocol for isolation ofthe cytosol. It is well documented that the c-src protein is sensitive to proteolytic degradation by cytosolic proteases, leadingto degradation products of MT54,000-45,000 (23, 29). It is alsoknown that tumor tissues possess high amounts of proteolyticenzymes, perhaps leading to a considerable amount of Src inthe cytoplasm. The finding in the immunoblots of proteinbands of molecular weight lower than 60,000 indeed points tothe existence of these degraded forms. Another possibilitywhich cannot be excluded is that the src protein is not alwaysattached to the membrane but also resides in the cytosol, perhaps becoming attached after recruitment to a receptor in themembrane, as is the case for p56lck, also a member of the Src

family (30). In a recent article by Srivastava et al. (31), it wasshown that the PTK activity of mammary carcinomas from ratswas equally distributed between the particulate and cytosolicfractions. The authors also pointed out that it was unlikely thatthis cytosolic activity was derived from the particulate fraction.

It is interesting to note that in the breast tumor specimen,precipitation of c-src also led to the precipitation of another Mr~ 110,000 protein. The nature of this protein is unknown.However, the involvement of a dimer of the Mr 56,000 proteinwould seem unlikely since the immunoprecipitates are boiledunder reducing conditions.

The involvement of c-src in breast cancer has been reportedonly once before in the literature. Rosen and coworkers (32)showed that the autophosphorylating activity as well as thephosphorylation of casein by Src-kinase was higher in a specimen of breast tumor than in normal tissue from the same patient. However, they also showed that the amount of src proteinin the samples was the same for both normal and malignant

tissues. The Src-kinase activity of most, but not all, breast cancer cell lines tested was also considerably higher than that ofnormal human fibroblasts (32).

In a recent study by Cartwright et al. (33), activation of thepp60c crf protein kinase was found to be an early event in co-

Ionic carcinogenesis. Previous studies (34, 35) had alreadyshown that the in vitro protein tyrosine kinase activity ofpp60c srf from many colon carcinomas is significantly higher

than that from normal mucosa adjacent to the tumor. It wasfound that, with increasing progression of the colonie tumorstoward malignant phenotypes, the activity of the c-src A/r60,000 protein tyrosine kinase increases also (33). It is interesting to speculate on the mechanism leading to the elevatedactivity of the c-src protein. In the study by Rosene/ al. (32), theelevated activity was shown to be a result of the higher specificactivity of the Src protein, since the amount of protein was thesame in normal and malignant tissue. In a recent study by Wanget al. (36), the higher pp60c >rcactivity detected mainly in colon

carcinomas did not originate from mutations in codons whichwould be expected to confer higher activity on the enzyme. Theauthors were in fact unable to detect activating mutations of thec-src gene.

The regulation of the tyrosine kinase activity of Src is quitecomplex, and most of the studies addressing regulation havebeen carried out with the chicken c-src protein. Not only are theCOOH-terminal tyr527 and tyr416 involved in the regulation(37-39), recent evidence has shown that several NH2-terminaltyrosine residues are involved in regulation as well (40-42). Iftyr527 is dephosphorylated, the enzyme becomes active and

transforming. The transforming activity of the viral counterpartof c-src, v-src, is in part due to the absence of the 19 COOH-terminal residues, which contain tyr527 and are replaced by 12

other residues (43), conferring a very high tyrosine kinase activity on the protein (44). The presence of tyr416, however, isalso important; double mutants of c-src containing a tyrosine-to-phenylalanine mutation at positions 527 and 416 lose theirtransforming activity (38). Apparently, the phosphorylation oftyr416 is also necessary for transformation and increases thetyrosine kinase activity of c-src (38). For p56kk, it is shown that,upon activation of CD4 or CDS, p56lck binds to these molecules

and is dephosphorylated by CD45, which is a protein tyrosinephosphatase (45). Subsequently, p56k"k becomes activated as a

result of this dephosphorylation. It is tempting to speculate thatthe balance between the protein tyrosine phosphatase and protein tyrosine kinase activity in the breast cancer tissue is disturbed compared to normal tissue, perhaps resulting in dephosphorylation of the human c-src protein at tyrosine residue 530,which is therefore constitutively active. This could explain theelevated PTK activities measured in the cytosolic fractions ofthe breast tumors. However, it is not excluded that the amountof src protein in the breast tumors is higher than in the normaltissue, resulting in elevated activity. Overexpression of c-src hasbeen shown to result in an increase in Src activity (37). Preliminary results point to a combination of these two possibilities;the protein tyrosine phosphatase activity in malignant tumorsseems to be elevated compared to normal mammary tissue, andthe amount of c-src protein in tumor tissue, as measured byWestern blotting, is about 7.5 times greater than in normaltissue.4

Although it is proposed that the antibody 327 is specific forc-src, further investigation using other (specific) antibodies isneeded to confirm that we are dealing with c-src in the breast

4 A. E. Ottenhoff-KalffVr al., manuscript in preparation.

4777




tumor specimens. Future experiments will focus on furtheridentification and characterization of the src protein and itsregulation in breast cancer tissue.

ACKNOWLEDGMENTS

The authors are grateful to G. P. M. Schipper-Kester and S. S.Adriaansen-Slot for performing the tyrosine kinase assays on the breastcancer tissue series and to L. Houben (Department of Pulmonary Diseases) for performing the elutriation to obtain the lymphocytes. Theauthors also wish to thank Dr. Joan S. Brugge for very generouslyproviding anti-Src 327 antibodies.

REFERENCES1. Slamon. D. J., Clark, G. M., Wong, S. G.. Levin, W. J., Ullrich, A., and

McGuire, W. L. Human breast cancer: correlation of relapse and survivalwith amplification of the HER-2/neu oncogene. Science (Washington DC),235: 177-182, 1987.

2. Slamon, D. J.. Godolphin. W., Jones. L. A.. Holt, J. A.. Wong, S. G., Keith,D. E., Levin, W. J.. Stuart. S. G., Udove, J., Ullrich, A., and Press, M. F.Studies of the HER-2/neu proto-oncogene in human breast and ovariancancer. Science (Washington DC), 244: 707-712, 1989.

i. Ali, I. U., Campbell, G., Lidereau. R.. and Callahan. R. Lack of evidence forthe prognostic significance of c-erbB-2 amplification in human breast carcinoma. Oncol. Res., 3: 139-146. 1988.

4. Ali, I. U., Campbell, G.. Lidereau, R., and Callahan, R. Amplification ofc-erhB-2 and aggressive human breast tumors? Science (Washington DC).240: 1795-1796, 1988.

5. Van de Vijver, M. J., Van de Bersselaar, R., Devilee, P., Cornelisse, C.,Pcterse, J., and NÃ¼sse,R. Amplification of the neu (c-erbB-2) oncogene inhuman mammary tumors is relatively frequent and is often accompanied byamplification of the linked c-erbA oncogene. Mol. Cell. Biol., 7: 2019-2023,1987.

6. Escot, C.. Theillet, C.. Lidereau, R.. Spyratos. F., ChÃ¡mpeme. M-H.. Gest,J.. and Callahan. R. Genetic alteration of the c-myc protooncogene (MYC) inhuman primary breast carcinomas. Proc. Nati. Acad. Sci. USA, S3: 4834-4838. 1986.

7. Guerin, M.. Barrois, M., Terrier, M. J.. Spielmann, M., and Riou. G. Over-expression of either c-myc or c-erbB-2/neu proto-oncogenes in human breastcarcinomas: correlation with poor prognosis. Oncol. Res., 3: 21-31, 1988.

8. Lidereau. R., Callahan, R.. Dickson. C., Peters. G.. Escot, C. and Ali, I. U.Amplification of the inl-2 gene in primary human breast tumors. Oncol. Res.,2: 285-291, 1988.

9. Singletary. S. E., Baker, F. L.. Spitzer. G.. Tucker, S. L.. Tomasovic. B.,Brock, W. A., Ajani. J. A., and Kelly. A. M. Biological effect of epidermalgrowth factor on the in vitro growth of human tumors. Cancer Res., 47:403-406, 1987.

10. Sainsbury. J. R. C., Sherbet, G. V., Farndon, J. R., and Harris. A. L. Epidermal growth factor receptors and oestrogen receptors in human breastcancer. Lancet, 2: 364-366, 1985.

11. Hunter. T., and Cooper, J. A. Protein-tyrosine kinases. Annu. Rev. Bio-chem., 54:897-930. 1985.

12. Ullrich, A., and Schlessinger, J. Signal transduction by receptors with tyrosine kinase activity. Cell, 61: 203-212. 1990.

13. Druker, B. J., MamÃ³n, H. J., and Roberts. T. M. Oncogenes, growth factors,and signal transduction. N. Engl. J. Med., 321: 1383-1391, 1989.

14. Heldin, C-H., and Westermark. B. Growth factors as transforming proteins.Eur. J. Biochem.. 184: 487-496. 1989.

15. Hennipman. A., van Oirschot, B. A., Smits, J.. Rijksen, G.. and Staal, G. E.J. Tyrosine kinase activity in breast cancer, benign breast disease, and normalbreast tissue. Cancer Res., 49: 516-521, 1989.

16. De Boer, M., and Roos, D. Metabolic comparison between basophils andother leukocytes from human blood. J. Immunol., 136: 3447-3454, 1986.

17. Rijksen, G., van Oirschot, B. A., and Staal. G. E. J. Nonradioactive assays ofprotein-tyrosine kinase activity using anti-phosphotyrosine antibodies. Methods Enzymol.. 200: 98-107, 1991.

18. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall. R. J. Proteinmeasurement with the Folin phenol reagent. J. Biol. Chem.. 193: 265-275.1951.

19. LÃ¡zaro.I., Gonzalez, M.. Roy, G.. Villar, L. M., and Gonzalez-Porque. P.Description of an enzyme-linked immunosorbent assay for the detection ofprotein tyrosine kinase. Anal. Biochem., 192: 257-261. 1991.

20. Bradford, M. M. A rapid and sensitive method for the quantitation of mi-crogram quantities of protein utilizing the principle of protein-dye binding.

Anal. Biochem., 72: 248-254, 1976.21. Laemmli, U. K. Cleavage of structural proteins during the assembly of the

head of bacteriophage T4. Nature (Lond.), 227: 680-685, 1970.22. Lipsich, L. A., Lewis, A. J.. and Brugge, J. S. Isolation of Monoclonal

antibodies that recognize the transforming proteins of avian sarcoma viruses.J. Virol.. Â¥Â«.-352-360.1983.

23. Huang, M-M.. Bolen. J. B.. Barnwell. J. W.. Shattil, S. J.. and Brugge. J. S.Membrane glycoprotein IV (CD36) is physically associated with the Fyn.Lyn. and Yes protein-tyrosine kinases in human platelets. Proc. Nati. Acad.Sci. USA, 88: 7844-7848, 1991.

24. Courtneidge, S. A., Levinson. A. D.. and Bishop. J. M. The protein encodedby the transforming gene of avian sarcoma virus (pp60src) and a homologousprotein in normal cells (pp60p'"to's"') are associated with the plasma mem

brane. Proc. Nati. Acad. Sci. USA, 77: 3783-3787. 1980.25. Krueger. J. G., Wang. E.. and Goldberg, A. R. Evidence that the src gene

product of Rous sarcoma virus is membrane associated. Virology, 101: 25-40. 1980.

26. Pellman, D.. Garber, E. A., Cross, F. R.. and Hanafusa, H. Fine structuralmapping of a critical NH2-terminal region of p60*"'. Proc. Nati. Acad. Sci.USA, 82: 1623-1627. 1985.

27. Buss, J. E., Kamps, M. P., and Sefton, B. M. Myristic acid is attached to thetransforming protein of Rous sarcoma virus during or immediately aftersynthesis and is present in both soluble and membrane-bound forms of theprotein. Mol. Cell. Biol.. 4: 2697-2704. 1984.

28. Wilcox, C., Hu, J-S., and Olson, E. N. Acylation of proteins with myristicacid occurs cotranslationally. Science (Washington DC), 238: 1275-1278,1987.

29. Morgan. D. O., Kaplan, J. M., Bishop, J. M., and Varmus, H. E. Mitosis-specific phosphorylation of pftW "" by p34cdc2-associaled protein kinase.Cell. 57: 775-786, 1989.

30. Sefton. B. M. The lek tyrosine protein kinase. Oncogene, 6: 683-686. 1991.31. Srivastava, A. K., Chiasson, J-C., Chiasson, J-L., Lacroix, A., and Windsich.

L. Biochemical characteristics of cytosolic and paniculate forms of proteintyrosine kinases from A'-methyl-A'-nitrosurea (MNU)-induced rat mammarycarcinoma. Mol. Cell. Biochem., 106: 87-97, 1991.

32. Rosen, N.. Bolen, J. B., Schwartz, A. M., Cohen. P.. DeSeau, V., and Israel.M. A. Analysis of pp60c'"r protein kinase activity in human tumor cell linesand tissues. J. Biol. Chem.. 261: 13754-13759, 1986.

33. Cartw right, C. A., Meisler, A. I., and Eckhart. W. Activation of the ppou*'1"

protein kinase is an early event in colonie carcinogenesis. Proc. Nati. Acad.Sci. USA, 87: 558-562, 1990.

34. Bolen. J. B.. Veillette, A., Schwartz, A. M., DeSeau, V., and Rosen, N.Activation of pp60csrc protein kinase activity in human colon carcinoma.Proc. Nati. Acad. Sci. USA, 84: 2251-2255, 1987.

35. Cartwright, C. A.. Kamps, M. P.. Meisler. A. !.. Pipas, J. M., and Eckhart,W. pp60c'"'<'activation in human colon carcinoma. J. Clin. Invest.. 83:2025-

2033. 1989.36. Wang. P., Fromowitz. F.. Koslow, M.. Hagag, N., Johnson. B., and Viola. M.

c-src structure in human cancers with elevated ppoO0srcactivity. Br. J. Cancer, 64: 531-533, 1991.

37. Kmiecik, T. E.. Johnson. P. J., and Shalloway, D. Regulation by the auto-phosphorylation site in overexpressed ppÃ³CK'5â„¢.Mol. Cell. Biol.. 8: 4541-

4546. 1988.38. Harvey, R., Hehir. K. M., Smith, A. E.. and Cheng, S. H. pp60c-"f Variants

containing lesions that affect phosphorylation at tyrosines 416 and 527. Mol.Cell. Biol.. 9: 3647-3656. 1989.

39. Abraham, N., and Veillette, A. Activation of p56/rA through mutation of aregulatory carboxy-terminal tyrosine residue requires intact sites of auto-phosphorylation and myristylation. Mol. Cell. Biol.. 10: 5197-5206. 1990.

40. Espino. P. C., Harvey. R.. Schweickhardt. R. L.. White. G. A.. Smith, A. E.,and Cheng. S. H. The amino-terminal region of ppoO0'"* has a modulator)

role and contains multiple sites of tyrosine phosphorylation. Oncogene. 5:283-293. 1990.

41. Hirai. H., and Varmus, H. E. Site-directed mutagenesis of the SH2- andSH3-coding domains of c-src produces varied phenotypes. including onco-genic activation of p60c'"-. Mol. Cell. Biol.. 10: 1307-1318. 1990.

42. O'Brien, M. C., Fukui, Y.. and Hanafusa. H. Activation of the proto-oncogene pou*'*" by point mutations in the SH2 domain. Mol. Cell. Biol., 10:2855-2862. 1990.

43. Takeya, T., and Hanafusa, H. Structure and sequence of the cellular genehomologous to the RSV src gene and the mechanism for generating thetransforming virus. Cell. 32: 881-890. 1983.

44. Jove. R.. and Hanafusa. H. Cell transformation by the viral src oncogene.Annu. Rev. Cell Biol.. 3: 31-56, 1987.

45. Veillette, A.. Bookman. M. A., Horak. E. M.. Samelson, L. E., and Bolen, J.B. Signal transduction through the CD4 receptor involves the activation ofthe internal membrane tyrosine-protein kinase p56lck. Nature (Lond.). 338:257-259, 1989.

4778



characterization of protein tyrosine kinases from human breast … · exchange chromatography using...

Documents