- Email: [email protected]
Capsaicin-Responsive NADH Oxidase Activities from Urine of Cancer Patients
ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS
Vol. 358, No. 2, October 15, pp. 336 –342, 1998 Article No. BB980877
Capsaicin-Responsive NADH Oxidase Activities from Urine of Cancer Patients Ferda Yantiri,* D. James Morre´,*,1 Kader Yagiz,* Silvia Barogi,* Sui Wang,† Pin-Ju Chueh,† NaMi Cho,* Dagmar Sedlak,* and Dorothy M. Morre´† *Department of Medicinal Chemistry and Molecular Pharmacology and †Department of Foods and Nutrition, Purdue University, West Lafayette, Indiana 47907
Received March 30, 1998, and in revised form July 22, 1998
NADH oxidases of low specific activities from urine of cancer patients were found to be inhibited or stimulated by the vanilloid capsaicin (8-methyl-N-vanillyl-6-noneamide). Similar activities, inhibited or stimulated by capsaicin, were reported previously for sera of cancer patients but not for sera of normal volunteers or for patients with disorders other than cancer. Like those from sera, the activities from urine were resistant to heat and to digestion with proteinase K. Two different fractions with capsaicin-responsive NADH oxidase activities were obtained by FPLC. One fraction in which the 33-kDa band was the major component exhibited NADH oxidase activity stimulated by capsaicin. Another fraction in which 66-kDa and 45-kDa bands were major components exhibited NADH oxidase activities inhibited by capsaicin. A monoclonal antibody generated to a ca 34-kDa form of the NADH oxidase from sera reacted with a urine protein of a ca 33-kDa band in the capsaicin-stimulated fraction. The 33-kDa protein was of low abundance and was estimated to be present in amounts between 5 and 100 mg/L, depending on the particular patient. © 1998 Academic Press
In a series of studies from our laboratory, we have described an NADH oxidase activity from the external surface of plasma membranes of HeLa cells that was inhibited by the vanilloid, capsaicin (1). A shed form of the activity was found in culture media conditioned by the growth of HeLa cells (2). A 33.5-kDa protein with capsaicin-inhibited NADH oxidase activity has been purified from this source (3) and from sera of cancer 1 To whom correspondence should be addressed at Department of Medicinal Chemistry and Molecular Pharmacology, 1333 HANS Life Science Research Building, Purdue University, West Lafayette, IN 47907-1333. (Fax) 765-494-4007.
336
patients (4). The serum form of the activity was widely distributed among cancer patients encompassing a range of tumor types, including leukemias, lymphomas, and solid tumors (breast, prostate, lung, colon, ovarian, and others) (5). When sera of cancer patients were assayed under oxidizing conditions in the presence of 0.03% hydrogen peroxide, most were inhibited by capsaicin but in about 10% of the sera, the activity was stimulated by capsaicin. The NADH activity from sera of healthy volunteers was unaffected by capsaicin. With pooled sera stored frozen, two fractions with antisera cross-reactive forms of the activity, resistant to proteinase K and heat, and with molecular mass corresponding to 26 –29 kDa were resolved by ion exchange and hydrophobic interaction chromatography (S. Barogi, Purdue University, unpublished results). The 26- to 29-kDa form in one fraction was inhibited by capsaicin while the 26- to 29-kDa form in the other fraction was stimulated by capsaicin. In this report, an activity similar to that of sera was found in urine of cancer patients but not in urine of healthy volunteers. The capsaicin responses were more nearly equally distributed among inhibition and stimulation even under oxidizing conditions. When fractionated by FPLC, two fractions were obtained. In one fraction, the activity was inhibited by capsaicin. In the other fraction, the activity was stimulated by capsaicin. Capsaicin inhibition was associated with a larger complex including 66- and 45-kDa proteins. The fraction with capsaicin-stimulated activity contained only the single 33-kDa band. The findings point to the possibility of noninvasive cancer detection using the capsaicin-responsive NADH oxidase from urine as the test parameter. Additionally, the findings indicate two forms of the activity. As for sera, one was stimulated by capsaicin and the other was inhibited by capsaicin. 0003-9861/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.
CAPSAICIN-RESPONSIVE NADH OXIDASES FROM HUMAN URINE
MATERIALS AND METHODS
Urine Samples Urine samples were from the patient population of Plymouth Community Hospital, Plymouth, Indiana, were patient samples contributed by Dr. Lawrence Hellson, Chappaqua, New York, or were from healthy laboratory volunteers collected at Purdue University. Urine was analyzed fresh or frozen following collection and stored frozen at 220°C. Informed consent was obtained and confidentiality of medical records was assured by assigning a number to each urine sample. Patients were confirmed as having been diagnosed with active, advanced disease (metastatic for breast and colon) at the time of urine collection and cancers were identified only as to organ site. All patients had received standard radiation and/or chemotherapy appropriate to their disease conditions.
NADH Oxidation NADH oxidase activity was determined as the disappearance of NADH measured at 340 nm with 430 nm as reference using an SLM Aminco DW-2000 spectrophotometer in the dual-wavelength mode of operation or a Hitachi U3210 spectrophotometer at 340 nm with continuous recording over 5- or 10-min intervals once steady-state rates were obtained. The reaction mixture contained 50 mM Tris– Mes buffer (pH 7.0), 2 mM KCN to inhibit any potential mitochondrial oxidase activity, and 150 mM NADH in a total volume of 2.5 ml. Assay was at 37°C with constant stirring. With purified fractions, 1 mM reduced glutathione was added to reduce the protein in the presence of substrate. After 10 min, 0.03% hydrogen peroxide (or 10 mM oxidized glutathione) was added to reoxidize the protein under renaturing conditions and in the presence of substrate to start the reaction. A millimolar extinction coefficient of 6.22 was used to determine NADH disappearance. To determine capsaicin inhibition, capsaicin in DMSO2 was added to a final concentration of 1 mM (0.1% DMSO) and the rate of NADH oxidation determined over 10 min. Capsaicin at a final concentration of 100 mM (0.2% DMSO) was then added and the assay was continued for an additional 10 min.
Ammonium Sulfate Precipitation To between 100 and 1000 ml of urine, sufficient solid ammonium sulfate was added and dissolved to achieve 30% of saturation. After equilibration on ice with stirring for 30 min, the precipitated protein was collected by centrifugation at 10,000g for 10 min. To the 30% of saturation ammonium sulfate supernatant was added solid ammonium sulfate to 70% of saturation and the precipitation and collection steps were repeated. In some experiments, other ammonium sulfate fractions were collected in a similar manner.
Proteinase K Digestion The 30 to 70% of saturation ammonium sulfate pellet or the 50 to 75% of saturation ammonium sulfate pellet was resuspended in deionized water in a ratio of 4 ml per 100 ml starting urine (25-fold concentration). To each milliliter of resuspended pellet was added 200 mg proteinase K (Sigma) and the mixture was incubated at 37°C for 2 h. The supernatant following centrifugation at 10,000g for 30 min was used in subsequent steps.
FPLC The proteinase K digest (12 ml) was applied to a Pharmacia HiLoad 26/60 Superdex 200 pg FPLC Gel Filtration column. The
2
Abbreviations used: DMSO, dimethyl sulfoxide; PVDF, polyvinylidene fluoride.
337
eluant was 50 mM NaCl plus 20 mM Tris–HCI, pH 7. The flow rate was 2.65 ml/min.
Analytical SDS–PAGE Electrophoresis Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS– PAGE) was with the buffer system of Laemmli (6) on acrylamide slab gels. Proteins were denatured in sample buffer by boiling for 3 min and analyzed by SDS–PAGE (12% acrylamide). The standard marker proteins (Sigma markers, low molecular weight) were a mixture of bovine serum albumin (66 kDa), chicken egg ovalbumin (45 kDa), glyceraldehyde-3-phosphate dehydrogenase (36 kDa), bovine erythrocyte carbonic anhydrase (29 kDa), bovine pancreas trypsinogen (24 kDa), soybean trypsin inhibitor (20 kDa) and a-lactalbumin from bovine milk (14.2 kDa) or prestained molecular weight markers (low molecular weight range, Sigma) for protein staining or immunoblotting, respectively. The gels were stained for protein using 0.1% Coomassie brilliant blue R-250 or silver (7).
Preparative SDS–PAGE Electrophoresis The gels were either stained with Coomassie blue to visualize proteins and bands excised directly from the gel or transferred to ProBlott PVDF membrane (8) and then stained with Coomassie blue. The preparative gels were 12% acrylamide, 1.5 mm thick. The gels were preelectrophoresed with 0.1 mM thioglycolate to block free radicals. Proteins were denatured in sample buffer by boiling for 5 min and cooled on ice for 10 min. A Hoefer (High-Resolution Standard 16 cm SLAB Gel System) dual-slab unit preparative system with cooling was used. Separation was at 200 V for 4 h and 45 min. The buffer system of Laemmli (6) was used.
Western Blot Analysis with Detection by Electrochemiluminescence (ECL) Proteins were transferred onto nitrocellulose (0.2 mm) and blocked in 10% nonfat milk in TBS-T for 1 h. The nitrocellulose was washed twice with TBS-T and incubated with primary antibody 1:25,000 overnight at 4°C. The nitrocellulose was incubated with horseradish peroxidase-conjugated second antibody for 1 h at room temperature after washing with TBS-T four times for 15 min each. The immunoreactive bands were visualized by incubation for 1 min in detection agent purchased from Amersham (Arlington Heights, IL). The detection agent contains luminol which is oxidized in alkaline conditions by horseradish peroxidase/hydrogen peroxide. Phenols were present as enhancers.
Western Blot Analysis with Detection by Color Development Using Alkaline PhosphataseConjugated Second Antibody Proteins were separated on 12% SDS–PAGE as described and then transferred by electroblotting onto nitrocellulose membranes. To block unspecific antibody binding sites, the blots were placed in a solution of 5% bovine serum albumin, 10 mM Tris–HC1, 150 mM NaC1, and 0.1% Tween-20 (TBS-T) for 30 min. The blots were then transferred to the primary antibody solution diluted 1:250 overnight at 4°C with shaking. The blots were washed with TBS-T four times for 15 min each after which the blots were placed into alkaline phosphatase-conjugated anti-rabbit antibody, 1:5000 (Jackson ImmunoResearch Laboratories, West Grove, PA) in TBS-T. The blots were washed with TBS-T four times for 15 min each and were placed in a mixture of 33 ml of 50 mg/ml nitro blue tetrazolium and 16.5 ml of 50 mg/ml of 5-bromo-1-chloro-3-indolyl phosphate prepared in 100 mM Tris, pH 9.5, containing 100 mM NaC1 and 5 mM MgC12 and incubated with shaking until the purple color of positive bands appeared. The color development reaction was stopped by placing the blots in 20 mM Tris, pH 8, containing 5 mM EDTA.
338
YANTIRI ET AL. TABLE I
NADH Oxidase Activity of Urine Samples from Laboratory Volunteers and Patients with Cancer NADH oxidase activity, units/100 ml urine Patient Normal volunteer
Breast cancer Colon cancer Lymphoma Multiple myeloma Colon cancer
(F) (F) (M) (M) (F) (M) (M) (F) (F)
No Addition (1)
11 mM capsaicin (2)
Ratio 2/1
100 mM capsaicin (3)
Ratio 3/1
0.2 0.13 0.15 0.2 0.15 0.1 0.15 0.4 0.15
0.2 0.13 0.15 0.2 0.2 0.03 0.2 0.6 0.3
1.0 1.0 1.0 1.0 1.33 0.3 1.33 1.5 2.0
0.2 0.13 0.15 0.2 0.1 0.05 0.1 0.6 0.65
1.0 1.0 1.0 1.0 0.67 0.5 0.67 1.5 4.3
Note. Assays were for 5 min. One unit 5 0.3 nmole NADH oxidized min21.
Protein Determinations
RESULTS
Proteins were determined by the bichinchoninic acid/copper assay (9) obtained from Sigma using bovine serum albumin as standard.
Urine from cancer patients exhibited an NADH oxidase of low specific activity either inhibited or stimulated by capsaicin (Table I). The NADH oxidase activity from urine of healthy volunteers was unaffected by capsaicin. The stimulation or inhibition was not a function of gender (e.g., with colon cancer) or tumor type and was given by both solid (colon, mammary, multiple myeloma) cancers and a lymphoma (Table I). The activity was concentrated 25-fold by ammonium sulfate fractionation and was stable to both heat and digestion with proteinase K (Table II). With the urine from a multiple myeloma patient used in these experiments, both total activity and the activity stimulated by capsaicin increased upon heating to 80°C for 15 min or heating to 100°C for 10 min. Also, while the total activity was reduced somewhat by treatment with proteinase K, the activity stimulated by capsaicin was increased (Table II). Ammonium sulfate fractionation was utilized to concentrate the capsaicin-responsive activity from the urine (Table III). Solid ammonium sulfate added to 70% of saturation recovered approximately 70% of the capsaicin-stimulated activity with the bulk of the capsaicin-stimulated activity being present in the 50 to 70% ammonium sulfate fraction. As the capsaicin-stimulated NADH oxidase activity appeared to resist proteinase K treatment, the resuspended material following ammonium sulfate precipitation was digested for 2 h at 37° with 200 mg/ml proteinase K. When separated by FPLC (Fig. 1), two regions contained capsaicin-responsive NADH oxidase activity following the proteinase K treatment. A region with activity stimulated by capsaicin included about 10 fractions between fractions 60 and 70 (Fig. 1, inset) collected just ahead of the two main absorbance peaks. The latter represented the bulk of the peptide frag-
Cyanogen Bromide Digestion and Methylation For cyanogen bromide digestion, the proteins separated by 10 or 12% SDS–PAGE were electroblotted to ProBlott PVDF membrane according to Towbin et al. (8). SDS–PAGE were preelectrophoresed with 0.1 mM thioglycolate to block free radicals. Regions containing proteins of interest on ProBlott PVDF membrane were cut into small pieces and placed in 10 mg/ml CNBr (dissolved in 70% formic acid) overnight in the dark at room temperature. The digested solutions were then collected and dried under nitrogen. The residues were rewetted with water and dried three times to reduce acid residues. The dried peptide fragments were resuspended in sample buffer and the pH of the samples was adjusted to neutrality with 1 M Tris. The digested fragments were separated on a discontinuous 15% acrylamide peptide gel as described by Scha¨ger and Von Jagow (10) and Stone et al. (11). Alternatively, gel slices were digested with CNBr directly (12). In addition, the CNBr digestion was carried out using methylated material. For methylation, three different protocols were followed. Methylation in solution. Fractions were desalted using Amicon centrifugal concentrations. Then the fractions were dried and resuspended in 5 mM EDTA purged with nitrogen gas prior to use, and 5 ml of NH4OH was added. The mixture was purged with nitrogen for 5 min after which methyl iodide was added under nitrogen (13) in an approximate ratio of 5 mmole methyl iodide/mmole of protein. After 2 h, the reaction was stopped by drying the reaction mixture under oxygen. The dried sample was resuspended in distilled water and volatiles were removed by four drying and resuspension cycles. Methylation in the gel. The gel slices were cut into small pieces and dried under nitrogen. EDTA (5 mM) was purged with nitrogen and the methyl iodide (5 mmole/mmole protein) was added. Gel pieces were covered with the methylation mixture and incubated for several hours under nitrogen at room temperature. Then they were dried, resuspended in distilled water, and carried through a total of four resuspension and drying cycles as above. Methylation on PVDF The proteins of interest were localized on PVDF by staining with Coomassie blue. Bands were excised and incubated in 5 mM EDTA containing 5 mmole methyl iodide/mmole protein for 1 h (14).
339
CAPSAICIN-RESPONSIVE NADH OXIDASES FROM HUMAN URINE TABLE II
Response of NADH Oxidase Activity of Urine from a Female Patient with Late-Stage Multiple Myeloma to Heat and Proteinase K NADH oxidase activity, units/100 ml urine Treatment
No Addition
1 mM capsaicin
D
100 mM capsaicin
D
None 80° 15 min 100° 10 min Proteinase K a
0.5 0.55 0.7 0.71
0.45 0.3 0.55 0.8
20.05 20.25 20.15 10.09
0.7 0.95 1.05 1.05
10.2 10.4 10.35 10.34
Note. Assays were for 5 min. One unit 5 0.3 nmole NADH oxidized min21. 200 mg proteinase K/ml at 37° C for 2 h.
a
ments generated by proteinase K shown by SDS– PAGE to exhibit apparent molecular masses of 33 kDa and less than 14 kDa. A second peak of NADH oxidase activity was encountered in fraction 36. This activity was inhibited by capsaicin (Fig. 1, inset). When analyzed by SDS–PAGE, fraction 60 of the activity stimulated by capsaicin exhibited a single major protein band at about 33 kDa and several minor bands of less than 14 kDa (Fig. 2). When assayed for NADH oxidase activity following FPLC, these fractions required 100 mM capsaicin and 0.03% hydrogen peroxide in order to demonstrate significant activity. The capsaicin and hydrogen peroxide could be added in either order but greatest stimulations were obtained if the hydrogen peroxide was added first. The material of fraction 36 corresponded to a native molecular weight of between 60 and 80 kDa (Fig. 1) and contained several protein bands upon analysis by
SDS–PAGE with the bulk of the material at about 45 kDa (Fig. 2). However, the 45-kDa proteins were widely distributed over the first absorbance peak of
TABLE III
Concentration of Capsaicin-Stimulated NADH Oxidase Activity of Urine by Ammonium Sulfate Precipitation
Fraction
Capsaicin-stimulated NADH oxidase activity, units/100 ml
Starting urine 0–70% solid ammonium sulfate pellet 0–70% solid ammonium sulfate supernatant Total (pellet 1 supernatant) 0–30% solid ammonium sulfate pellet 0–30% solid ammonium sulfate supernatant 0–50% ammonium sulfate pellet 0–50% ammonium sulfate supernatant 30–50% solid ammonium sulfate pellet 30–50% solid ammonium sulfate supernatant 50–70% solid ammonium sulfate pellet 50–70% solid ammonium sulfate supernatant
100 70 30 100 20 125 10 75 10 100 65 20
Note. The urine was from a female patient with late-stage multiple myeloma. Assays were for 5 min. One unit 5 0.3 nmole NADH oxidized min21.
FIG. 1. FPLC trace of separation of 30 to 70% ammonium sulfate precipitate from urine of a female patient with late-stage multiple myeloma. The urine was treated for 2 h with 200 mg/ml proteinase K at 34°. The activity profiles of the fractions are given in the inset. The NADH oxidase activities were measured in the absence of capsaicin or in the presence of 1 mM capsaicin or 100 mM capsaicin. Activities inhibited by capsaicin were observed in fraction 36. Capsaicin-stimulated activity was seen between fractions 55 and 75. The peptide molecular mass associated with the capsaicin-inhibited activity was estimated to be between 60 and 70 kDa whereas that of the capsaicin-stimulated activity was estimated to be between 30 and 40 kDa.
340
YANTIRI ET AL.
FIG. 2. SDS–PAGE (12% acrylamide) of fractions from the FPLC separation of Fig. 1. Activity in fraction 36 correlated with the presence of a 66-kDa band (arrow) not seen in fractions 35 and 37. Activity in fraction 60 correlated with a 33-kDa band (arrow) but this fraction lacked the 66-kDa component.
Fig. 1 and present in fractions lacking the capsaicininhibited NADH oxidase. The latter correlated with the presence of a band at 66 kDa (Fig. 2, Fraction 36, arrow) corresponding approximately to twice the molecular weight of the serum form of the capsaicininhibited NADH oxidase (5). The material of the 33-kDa band was N-terminally blocked and, by analogy to the serum form, would be expected to resist further digestion with proteases (4). When separated on preparative SDS–PAGE from the other proteins present following proteinase K digestion and digested in the gel with cyanogen bromide, the 33-kDa band was resistant to further digestion. However when treated with methyl iodide to methylate cysteines, followed by a second cyanogen bromide digestion, cleavage did occur with the generation of a 27-kDa component (Fig. 3). The 27-kDa fragment was resistant to further cyanogen bromide cleavage both before and after methylation (not shown). The 33-kDa band cross-reacted with antisera to the drug-responsive serum form of the NADH oxidase as shown by electrochemical luminescence detection (Fig. 4). The isolation protocol as described for Fig. 4 was carried out in parallel with urine from a healthy volunteer and from a patient with proteinuria (76-yearold female with 100 mg/dl urinary protein) in addition to urine from the myeloma patient (Fig. 5) with detection of immunoreactive bands by means of alkaline phosphatase-linked second antibody and NBT-BCIP. These analyses failed to reveal either a NADH oxidase activity inhibited by capsaicin or a band immunoreactive with the antisera to the cancer-specific serum form of the NADH oxidase with the sera from the healthy volunteer (Fig. 5B) or from the patient with proteinurea (Fig. 5C). For the myeloma patient, an immunoreactive band, very similar to that of Fig. 4, was observed (Fig. 5A).
DISCUSSION
A capsaicin-inhibited (or stimulated) NADH oxidase of the cancer cell surface was first described for plasma membrane vesicles of HeLa cells (1). The activity was seen with sealed, right side-out vesicles and, since NADH is an impermeant substrate, was assumed to be external (2). The activity also was shed both into culture media conditioned by growth of HeLa cells (3) and into sera with autochthonous tumors in cancer patients (5). The present findings demonstrate a capsaicin-responsive activity also present in urine of cancer patients. As
FIG. 3. Partial cleavage by cyanogen bromide following methylation of the 33-kDa protein cut from a preparative SDS–PAGE gel. The resultant cleavage product consisted of a band at 27 kDa. Also present was a band at 52 kDa not present in the starting material and the uncleaved starting material at 33-kDa.
CAPSAICIN-RESPONSIVE NADH OXIDASES FROM HUMAN URINE
FIG. 4. Reaction of the 33-kDa component (arrow) of Fraction 60 of Fig. 2 with a monoclonal antibody to the drug-responsive NADH oxidase isolated from sera of cancer patients (4, 17). Detection was by electrochemiluminescence. Immunoprecipitated proteins were separated by SDS–PAGE and transferred to nitrocellulose membrane by electroblotting. The nitrocellulose membranes were blocked in a solution of 10% non-fat milk prepared in PBS/0.05% (v/v) Tween 20 containing 1% (w/v) blocking reagent 1096 1Z6 (Boehringer Mannheim) and probed with peroxidase-conjugated second antibody. Immunoreactive proteins were detected by chemiluminescence using an ECL kit (Amersham Corp.) following the manufacturer’s instructions.
with sera, NADH oxidase activity of urine of healthy volunteers was unresponsive to capsaicin. The NADH oxidase was of low specific activity in urine, but urine might represent both a useful starting material for isolation of the protein(s) involved and for diagnostic purposes due to the ease of collection and ready availability in quantity. In order to begin to investigate these possibilities, the NADH oxidase was concentrated from urine by ammonium sulfate fractionation and proteinase K digestion. The resistance of the NADH oxidase activity from urine to proteinase K was not unexpected based on work with HeLa cells (15) and results with serum (4). Here the activity was found to resist digestion by a variety of proteases, including not only proteinase K but trypsin, chymotrypsin, V-8, subtilisin, lysylaminopeptidase C, and pronase. The protein also resisted digestion with cyanogen bromide and was N-terminally blocked to frustrate attempts to generate either N-terminal or internal amino acid sequence. The serum (4) and urine forms as well as the activity from
341
HeLa cells (15) were resistant to heat as well as to proteinase digestion. A single urine sample collected from a female patient with multiple myeloma was utilized to isolate the capsaicin-stimulated activity. Following proteinase K digestion, most of the urine proteins were digested and the next step was to remove these fragments by FPLC. The FPLC separation yielded two large, broad absorbance peaks with molecular masses of less than 20 kDa which contained the bulk of these fragments. The recovery of activity stimulated by capsaicin was estimated as 70% after ammonium sulfate precipitation and proteinase K and about 40% after FPLC (Table IV). The capsaicin-stimulated activity was recovered from the FPLC separation in a fraction collected just ahead of the first of the two large absorbance peaks containing fragments from proteinase K digestion (Fig. 1). The activity correlated with a 33-kDa band that cross-reacted with a monoclonal antibody to the 33.5-kDa form of the protein from human serum. This activity was low in the absence of capsaicin and required both capsaicin and dilute hydrogen peroxide for activity. Even though the urine sample fractionated initially contained predominantly an activity stimulated by capsaicin, a second activity peak was observed in the trailing portion of a small absorbance peak in the 60- to 80-kDa range of the FPLC separation. This activity peak contained several components in the 45-kDa molecular mass range plus a component at about 66-kDa. The NADH oxidase activity of the second peak was inhibited rather than stimulated by both 1 mM and 100 mM capsaicin as was the activity from culture media conditioned by growth of HeLa cells (3) and the activity
FIG. 5. Western blot analysis of the 33-kDa component (arrow) from FPLC separations comparing three urine samples analyzed in parallel. Following separation on 12% SDS–PAGE as in Fig. 2, proteins were electroblotted to nitrocellulose and incubated overnight at 4°C with 1:250 diluted monoclonal antibody to the drug-responsive NADH oxidase as in Fig. 4. Detection was with alkaline phosphatase-conjugated second antibody diluted 1:5000 followed by incubation with NBT-BCIP. (A) Urine from the female multiple myeloma patient. (B) Urine from a healthy male volunteer. (C) Urine from a female patient with proteinurea.
342
YANTIRI ET AL. TABLE IV
Recovery of 100 mM Capsaicin-Stimulated Activity and Protein from Urine of a Female Patient with Late Stage Multiple Myeloma
Fraction Original urine Ammonium sulfate pellet (30–70%) Pellet after proteinase K, 200 mg/ml, 2 h Pellet after FPLC
Capsaicin-stimulated NADH oxidase activity, units/100 ml
disulfide reductase whose physiological function has been suggested to be protein–disulfide thiol interchange (18). The NADH oxidase activity from HeLa cells inhibited by antitumor sulfonylurea exhibits a sulfonylurea-inhibited protein disulfide interchange activity (19) also inhibited by capsaicin.
Protein, mg/100 ml
100
90
73
47 a
70 40
Note. Values are for 100 ml of urine. Similar results were obtained for a female patient with late stage metastatic colon cancer. One unit 5 0.3 nmole NADH oxidized min21. a The total recovery of protein after ammonium sulfate fractionation was 85%.
from the majority of the serum samples when assayed under oxidizing conditions (5). The resistance of the urine and serum NADH oxidase proteins to both heat and protease treatment as well as to cyanogen bromide cleavage is unusual and has thus far precluded the generation of primary amino acid sequence. One or more active site cysteines in the NADH oxidase protein has been inferred from studies with thiol reagents (16). In order to facilitate cyanogen bromide cleavage, the active site cysteines were first methylated with methyl iodide followed by cyanogen bromide. Methylated cysteines often behave as methionines following cyanogen bromide treatment to introduce additional potential cyanogen bromide cleavage sites in the protein (13). With the 33-kDa urine form, methylation followed by cyanogen bromide did appear to introduce a cyanogen bromide cleavage site with the generation of a ca 27-kDa fragment. However, the 27-kDa protein was resistant to cyanogen bromide both before and after methylation and no useful amino acid sequence resulted. These findings may suggest the presence of a cysteine located approximately 50 amino acids from the C-terminus since the resultant peptide was still blocked to N-terminal sequencing. The function of the cancer-specific NADH oxidase even prior to shedding from cancer cells remains unknown. Based on studies with HeLa cells and the properties of the monoclonal antibody (17) some relationship of the membrane-located form to growth is indicated. Certainly since the NADH site is at the external cell surface and since an external source of cellular NADH is unlikely, any role in NADH oxidation is assumed to be artifactual and only a convenient method of assay. Also, oxygen may not be the natural electron acceptor even for NADH oxidation (18). A comparable activity from plants stimulated by growth hormones has properties of an NADH: protein
ACKNOWLEDGMENTS We thank Connie Chalko and Plymouth Community Hospital, Plymouth, Indiana, Dr. Lawrence Helson, Chappaqua, NY, and Richard Bentlage, Saint Elizabeth Hospital, Lafayette, Indiana, for providing patient urine samples and for helpful discussions. Yanxia Yang provided technical assistance. This work was supported in part by a gift from Leonard P. Shaykin.
REFERENCES 1. Morre´, D. J., Chueh, P.-J. and Morre´, D. M. (1995) Proc. Natl. Acad. Sci. USA 92, 1831–1835. 2. Morre´, D. J., Wilkinson, F. E., Kim, C., Cho, N., Lawrence, J., Morre´, D. M., and McClure, D. (1996) Biochim. Biophys. Acta 1280, 197–206. 3. Wilkinson, F., Kim, C., Cho, N., Chueh, P.-J., Leslie, S., MoyaCamarena, S., Wu, L.-Y., Morre´, D. M., and Morre´, D. J. (1996) Arch. Biochem. Biophys. 336, 275–282. 4. Chueh, P.-J., Morre´, D. J., Wilkinson, F. E., Gibson, T., and Morre´, D. M. (1997) Arch. Biochem. Biophys. 342, 38 – 47. 5. Morre´, D. J., Caldwell, S., Mayorga, A., Wu, L.-Y., and Morre´, D. M. (1997) Arch. Biochem. Biophys. 342, 224 –230. 6. Laemmli, U. K. (1990) Nature 227, 680 – 685. 7. Butcher, L. A., and Tomkins, J. K. (1985) Anal. Biochem. 148, 384 –388. 8. Towbin, H., Staehelin, T., and Gordon, J. (1979) Proc. Natl. Acad. Sci. USA 76, 4350 – 4354. 9. Smith, P. K., Krohm, R. I., Hermanson, G. T., Mallia, A. K., Gartner, F. H., Provenzano, E. K., Fujimoto, E. K., Goeke, N. M., Olson, B. J., and Klenk, D. C. (1985) Anal. Biochem. 100, 76 – 85. 10. Scha¨ger, H., and Von Jagow, G. (1987) Anal. Biochem. 166, 368 –379. 11. Stone, K. L., La Presti, M. B., Williams, N. D., Crawford, J. M., De Angelis, R., and William, K. R. (1989) Techniques in Protein Chemistry, pp. 51–53, Academic Press, New York. 12. Jahnen, W., Ward, L. D, Reid, G. E., Moritz, R. L., and Simpson, R. J. (1990) Biochem. Biophys. Res. Commun. 166, 139 –145. 13. Awad, W., and Wilcox, P. E. (1964) Biochem. Biophys. Res. Commun. 17, 709 –715. 14. Ploug, M., Stoffer, B., and Jensen, A. L. (1992) Electrophoresis 13, 148 –153. 15. del Castillo-Olivares, A., Yantiri, F., Chueh, P.-J., Wang, S., Sweeting, M., Sedlak, D., Morre´, D. M., Burgess, J., and Morre´, D. J. Arch. Biochem. Biophys., in press. 16. Morre´, D. J., and Morre´, D. M. (1995) J. Bioenerg. Biomemb. 24, 137–144. 17. Cho, N., Chueh, P.-J., Caldwell, S., Wu, L.-Y., Werderitsh, D. A., Morre´, D. M., and Morre´, D. J. (1998) Proc. Amer. Assoc. Cancer Res. 39, 437 (abs). 18. Chueh, P.-J., Morre´, D. M., Penel, C., DeHahn, T., and Morre´, D. J. (1997) J. Biol. Chem. 272, 11221–11227. 19. Morre´, D. J., Jacobs, E., Sweeting, M., deCabo, R., and Morre´, D. M. (1997) Biochim. Biophys. Acta 1325, 117–125.
Vol. 358, No. 2, October 15, pp. 336 –342, 1998 Article No. BB980877
Capsaicin-Responsive NADH Oxidase Activities from Urine of Cancer Patients Ferda Yantiri,* D. James Morre´,*,1 Kader Yagiz,* Silvia Barogi,* Sui Wang,† Pin-Ju Chueh,† NaMi Cho,* Dagmar Sedlak,* and Dorothy M. Morre´† *Department of Medicinal Chemistry and Molecular Pharmacology and †Department of Foods and Nutrition, Purdue University, West Lafayette, Indiana 47907
Received March 30, 1998, and in revised form July 22, 1998
NADH oxidases of low specific activities from urine of cancer patients were found to be inhibited or stimulated by the vanilloid capsaicin (8-methyl-N-vanillyl-6-noneamide). Similar activities, inhibited or stimulated by capsaicin, were reported previously for sera of cancer patients but not for sera of normal volunteers or for patients with disorders other than cancer. Like those from sera, the activities from urine were resistant to heat and to digestion with proteinase K. Two different fractions with capsaicin-responsive NADH oxidase activities were obtained by FPLC. One fraction in which the 33-kDa band was the major component exhibited NADH oxidase activity stimulated by capsaicin. Another fraction in which 66-kDa and 45-kDa bands were major components exhibited NADH oxidase activities inhibited by capsaicin. A monoclonal antibody generated to a ca 34-kDa form of the NADH oxidase from sera reacted with a urine protein of a ca 33-kDa band in the capsaicin-stimulated fraction. The 33-kDa protein was of low abundance and was estimated to be present in amounts between 5 and 100 mg/L, depending on the particular patient. © 1998 Academic Press
In a series of studies from our laboratory, we have described an NADH oxidase activity from the external surface of plasma membranes of HeLa cells that was inhibited by the vanilloid, capsaicin (1). A shed form of the activity was found in culture media conditioned by the growth of HeLa cells (2). A 33.5-kDa protein with capsaicin-inhibited NADH oxidase activity has been purified from this source (3) and from sera of cancer 1 To whom correspondence should be addressed at Department of Medicinal Chemistry and Molecular Pharmacology, 1333 HANS Life Science Research Building, Purdue University, West Lafayette, IN 47907-1333. (Fax) 765-494-4007.
336
patients (4). The serum form of the activity was widely distributed among cancer patients encompassing a range of tumor types, including leukemias, lymphomas, and solid tumors (breast, prostate, lung, colon, ovarian, and others) (5). When sera of cancer patients were assayed under oxidizing conditions in the presence of 0.03% hydrogen peroxide, most were inhibited by capsaicin but in about 10% of the sera, the activity was stimulated by capsaicin. The NADH activity from sera of healthy volunteers was unaffected by capsaicin. With pooled sera stored frozen, two fractions with antisera cross-reactive forms of the activity, resistant to proteinase K and heat, and with molecular mass corresponding to 26 –29 kDa were resolved by ion exchange and hydrophobic interaction chromatography (S. Barogi, Purdue University, unpublished results). The 26- to 29-kDa form in one fraction was inhibited by capsaicin while the 26- to 29-kDa form in the other fraction was stimulated by capsaicin. In this report, an activity similar to that of sera was found in urine of cancer patients but not in urine of healthy volunteers. The capsaicin responses were more nearly equally distributed among inhibition and stimulation even under oxidizing conditions. When fractionated by FPLC, two fractions were obtained. In one fraction, the activity was inhibited by capsaicin. In the other fraction, the activity was stimulated by capsaicin. Capsaicin inhibition was associated with a larger complex including 66- and 45-kDa proteins. The fraction with capsaicin-stimulated activity contained only the single 33-kDa band. The findings point to the possibility of noninvasive cancer detection using the capsaicin-responsive NADH oxidase from urine as the test parameter. Additionally, the findings indicate two forms of the activity. As for sera, one was stimulated by capsaicin and the other was inhibited by capsaicin. 0003-9861/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.
CAPSAICIN-RESPONSIVE NADH OXIDASES FROM HUMAN URINE
MATERIALS AND METHODS
Urine Samples Urine samples were from the patient population of Plymouth Community Hospital, Plymouth, Indiana, were patient samples contributed by Dr. Lawrence Hellson, Chappaqua, New York, or were from healthy laboratory volunteers collected at Purdue University. Urine was analyzed fresh or frozen following collection and stored frozen at 220°C. Informed consent was obtained and confidentiality of medical records was assured by assigning a number to each urine sample. Patients were confirmed as having been diagnosed with active, advanced disease (metastatic for breast and colon) at the time of urine collection and cancers were identified only as to organ site. All patients had received standard radiation and/or chemotherapy appropriate to their disease conditions.
NADH Oxidation NADH oxidase activity was determined as the disappearance of NADH measured at 340 nm with 430 nm as reference using an SLM Aminco DW-2000 spectrophotometer in the dual-wavelength mode of operation or a Hitachi U3210 spectrophotometer at 340 nm with continuous recording over 5- or 10-min intervals once steady-state rates were obtained. The reaction mixture contained 50 mM Tris– Mes buffer (pH 7.0), 2 mM KCN to inhibit any potential mitochondrial oxidase activity, and 150 mM NADH in a total volume of 2.5 ml. Assay was at 37°C with constant stirring. With purified fractions, 1 mM reduced glutathione was added to reduce the protein in the presence of substrate. After 10 min, 0.03% hydrogen peroxide (or 10 mM oxidized glutathione) was added to reoxidize the protein under renaturing conditions and in the presence of substrate to start the reaction. A millimolar extinction coefficient of 6.22 was used to determine NADH disappearance. To determine capsaicin inhibition, capsaicin in DMSO2 was added to a final concentration of 1 mM (0.1% DMSO) and the rate of NADH oxidation determined over 10 min. Capsaicin at a final concentration of 100 mM (0.2% DMSO) was then added and the assay was continued for an additional 10 min.
Ammonium Sulfate Precipitation To between 100 and 1000 ml of urine, sufficient solid ammonium sulfate was added and dissolved to achieve 30% of saturation. After equilibration on ice with stirring for 30 min, the precipitated protein was collected by centrifugation at 10,000g for 10 min. To the 30% of saturation ammonium sulfate supernatant was added solid ammonium sulfate to 70% of saturation and the precipitation and collection steps were repeated. In some experiments, other ammonium sulfate fractions were collected in a similar manner.
Proteinase K Digestion The 30 to 70% of saturation ammonium sulfate pellet or the 50 to 75% of saturation ammonium sulfate pellet was resuspended in deionized water in a ratio of 4 ml per 100 ml starting urine (25-fold concentration). To each milliliter of resuspended pellet was added 200 mg proteinase K (Sigma) and the mixture was incubated at 37°C for 2 h. The supernatant following centrifugation at 10,000g for 30 min was used in subsequent steps.
FPLC The proteinase K digest (12 ml) was applied to a Pharmacia HiLoad 26/60 Superdex 200 pg FPLC Gel Filtration column. The
2
Abbreviations used: DMSO, dimethyl sulfoxide; PVDF, polyvinylidene fluoride.
337
eluant was 50 mM NaCl plus 20 mM Tris–HCI, pH 7. The flow rate was 2.65 ml/min.
Analytical SDS–PAGE Electrophoresis Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS– PAGE) was with the buffer system of Laemmli (6) on acrylamide slab gels. Proteins were denatured in sample buffer by boiling for 3 min and analyzed by SDS–PAGE (12% acrylamide). The standard marker proteins (Sigma markers, low molecular weight) were a mixture of bovine serum albumin (66 kDa), chicken egg ovalbumin (45 kDa), glyceraldehyde-3-phosphate dehydrogenase (36 kDa), bovine erythrocyte carbonic anhydrase (29 kDa), bovine pancreas trypsinogen (24 kDa), soybean trypsin inhibitor (20 kDa) and a-lactalbumin from bovine milk (14.2 kDa) or prestained molecular weight markers (low molecular weight range, Sigma) for protein staining or immunoblotting, respectively. The gels were stained for protein using 0.1% Coomassie brilliant blue R-250 or silver (7).
Preparative SDS–PAGE Electrophoresis The gels were either stained with Coomassie blue to visualize proteins and bands excised directly from the gel or transferred to ProBlott PVDF membrane (8) and then stained with Coomassie blue. The preparative gels were 12% acrylamide, 1.5 mm thick. The gels were preelectrophoresed with 0.1 mM thioglycolate to block free radicals. Proteins were denatured in sample buffer by boiling for 5 min and cooled on ice for 10 min. A Hoefer (High-Resolution Standard 16 cm SLAB Gel System) dual-slab unit preparative system with cooling was used. Separation was at 200 V for 4 h and 45 min. The buffer system of Laemmli (6) was used.
Western Blot Analysis with Detection by Electrochemiluminescence (ECL) Proteins were transferred onto nitrocellulose (0.2 mm) and blocked in 10% nonfat milk in TBS-T for 1 h. The nitrocellulose was washed twice with TBS-T and incubated with primary antibody 1:25,000 overnight at 4°C. The nitrocellulose was incubated with horseradish peroxidase-conjugated second antibody for 1 h at room temperature after washing with TBS-T four times for 15 min each. The immunoreactive bands were visualized by incubation for 1 min in detection agent purchased from Amersham (Arlington Heights, IL). The detection agent contains luminol which is oxidized in alkaline conditions by horseradish peroxidase/hydrogen peroxide. Phenols were present as enhancers.
Western Blot Analysis with Detection by Color Development Using Alkaline PhosphataseConjugated Second Antibody Proteins were separated on 12% SDS–PAGE as described and then transferred by electroblotting onto nitrocellulose membranes. To block unspecific antibody binding sites, the blots were placed in a solution of 5% bovine serum albumin, 10 mM Tris–HC1, 150 mM NaC1, and 0.1% Tween-20 (TBS-T) for 30 min. The blots were then transferred to the primary antibody solution diluted 1:250 overnight at 4°C with shaking. The blots were washed with TBS-T four times for 15 min each after which the blots were placed into alkaline phosphatase-conjugated anti-rabbit antibody, 1:5000 (Jackson ImmunoResearch Laboratories, West Grove, PA) in TBS-T. The blots were washed with TBS-T four times for 15 min each and were placed in a mixture of 33 ml of 50 mg/ml nitro blue tetrazolium and 16.5 ml of 50 mg/ml of 5-bromo-1-chloro-3-indolyl phosphate prepared in 100 mM Tris, pH 9.5, containing 100 mM NaC1 and 5 mM MgC12 and incubated with shaking until the purple color of positive bands appeared. The color development reaction was stopped by placing the blots in 20 mM Tris, pH 8, containing 5 mM EDTA.
338
YANTIRI ET AL. TABLE I
NADH Oxidase Activity of Urine Samples from Laboratory Volunteers and Patients with Cancer NADH oxidase activity, units/100 ml urine Patient Normal volunteer
Breast cancer Colon cancer Lymphoma Multiple myeloma Colon cancer
(F) (F) (M) (M) (F) (M) (M) (F) (F)
No Addition (1)
11 mM capsaicin (2)
Ratio 2/1
100 mM capsaicin (3)
Ratio 3/1
0.2 0.13 0.15 0.2 0.15 0.1 0.15 0.4 0.15
0.2 0.13 0.15 0.2 0.2 0.03 0.2 0.6 0.3
1.0 1.0 1.0 1.0 1.33 0.3 1.33 1.5 2.0
0.2 0.13 0.15 0.2 0.1 0.05 0.1 0.6 0.65
1.0 1.0 1.0 1.0 0.67 0.5 0.67 1.5 4.3
Note. Assays were for 5 min. One unit 5 0.3 nmole NADH oxidized min21.
Protein Determinations
RESULTS
Proteins were determined by the bichinchoninic acid/copper assay (9) obtained from Sigma using bovine serum albumin as standard.
Urine from cancer patients exhibited an NADH oxidase of low specific activity either inhibited or stimulated by capsaicin (Table I). The NADH oxidase activity from urine of healthy volunteers was unaffected by capsaicin. The stimulation or inhibition was not a function of gender (e.g., with colon cancer) or tumor type and was given by both solid (colon, mammary, multiple myeloma) cancers and a lymphoma (Table I). The activity was concentrated 25-fold by ammonium sulfate fractionation and was stable to both heat and digestion with proteinase K (Table II). With the urine from a multiple myeloma patient used in these experiments, both total activity and the activity stimulated by capsaicin increased upon heating to 80°C for 15 min or heating to 100°C for 10 min. Also, while the total activity was reduced somewhat by treatment with proteinase K, the activity stimulated by capsaicin was increased (Table II). Ammonium sulfate fractionation was utilized to concentrate the capsaicin-responsive activity from the urine (Table III). Solid ammonium sulfate added to 70% of saturation recovered approximately 70% of the capsaicin-stimulated activity with the bulk of the capsaicin-stimulated activity being present in the 50 to 70% ammonium sulfate fraction. As the capsaicin-stimulated NADH oxidase activity appeared to resist proteinase K treatment, the resuspended material following ammonium sulfate precipitation was digested for 2 h at 37° with 200 mg/ml proteinase K. When separated by FPLC (Fig. 1), two regions contained capsaicin-responsive NADH oxidase activity following the proteinase K treatment. A region with activity stimulated by capsaicin included about 10 fractions between fractions 60 and 70 (Fig. 1, inset) collected just ahead of the two main absorbance peaks. The latter represented the bulk of the peptide frag-
Cyanogen Bromide Digestion and Methylation For cyanogen bromide digestion, the proteins separated by 10 or 12% SDS–PAGE were electroblotted to ProBlott PVDF membrane according to Towbin et al. (8). SDS–PAGE were preelectrophoresed with 0.1 mM thioglycolate to block free radicals. Regions containing proteins of interest on ProBlott PVDF membrane were cut into small pieces and placed in 10 mg/ml CNBr (dissolved in 70% formic acid) overnight in the dark at room temperature. The digested solutions were then collected and dried under nitrogen. The residues were rewetted with water and dried three times to reduce acid residues. The dried peptide fragments were resuspended in sample buffer and the pH of the samples was adjusted to neutrality with 1 M Tris. The digested fragments were separated on a discontinuous 15% acrylamide peptide gel as described by Scha¨ger and Von Jagow (10) and Stone et al. (11). Alternatively, gel slices were digested with CNBr directly (12). In addition, the CNBr digestion was carried out using methylated material. For methylation, three different protocols were followed. Methylation in solution. Fractions were desalted using Amicon centrifugal concentrations. Then the fractions were dried and resuspended in 5 mM EDTA purged with nitrogen gas prior to use, and 5 ml of NH4OH was added. The mixture was purged with nitrogen for 5 min after which methyl iodide was added under nitrogen (13) in an approximate ratio of 5 mmole methyl iodide/mmole of protein. After 2 h, the reaction was stopped by drying the reaction mixture under oxygen. The dried sample was resuspended in distilled water and volatiles were removed by four drying and resuspension cycles. Methylation in the gel. The gel slices were cut into small pieces and dried under nitrogen. EDTA (5 mM) was purged with nitrogen and the methyl iodide (5 mmole/mmole protein) was added. Gel pieces were covered with the methylation mixture and incubated for several hours under nitrogen at room temperature. Then they were dried, resuspended in distilled water, and carried through a total of four resuspension and drying cycles as above. Methylation on PVDF The proteins of interest were localized on PVDF by staining with Coomassie blue. Bands were excised and incubated in 5 mM EDTA containing 5 mmole methyl iodide/mmole protein for 1 h (14).
339
CAPSAICIN-RESPONSIVE NADH OXIDASES FROM HUMAN URINE TABLE II
Response of NADH Oxidase Activity of Urine from a Female Patient with Late-Stage Multiple Myeloma to Heat and Proteinase K NADH oxidase activity, units/100 ml urine Treatment
No Addition
1 mM capsaicin
D
100 mM capsaicin
D
None 80° 15 min 100° 10 min Proteinase K a
0.5 0.55 0.7 0.71
0.45 0.3 0.55 0.8
20.05 20.25 20.15 10.09
0.7 0.95 1.05 1.05
10.2 10.4 10.35 10.34
Note. Assays were for 5 min. One unit 5 0.3 nmole NADH oxidized min21. 200 mg proteinase K/ml at 37° C for 2 h.
a
ments generated by proteinase K shown by SDS– PAGE to exhibit apparent molecular masses of 33 kDa and less than 14 kDa. A second peak of NADH oxidase activity was encountered in fraction 36. This activity was inhibited by capsaicin (Fig. 1, inset). When analyzed by SDS–PAGE, fraction 60 of the activity stimulated by capsaicin exhibited a single major protein band at about 33 kDa and several minor bands of less than 14 kDa (Fig. 2). When assayed for NADH oxidase activity following FPLC, these fractions required 100 mM capsaicin and 0.03% hydrogen peroxide in order to demonstrate significant activity. The capsaicin and hydrogen peroxide could be added in either order but greatest stimulations were obtained if the hydrogen peroxide was added first. The material of fraction 36 corresponded to a native molecular weight of between 60 and 80 kDa (Fig. 1) and contained several protein bands upon analysis by
SDS–PAGE with the bulk of the material at about 45 kDa (Fig. 2). However, the 45-kDa proteins were widely distributed over the first absorbance peak of
TABLE III
Concentration of Capsaicin-Stimulated NADH Oxidase Activity of Urine by Ammonium Sulfate Precipitation
Fraction
Capsaicin-stimulated NADH oxidase activity, units/100 ml
Starting urine 0–70% solid ammonium sulfate pellet 0–70% solid ammonium sulfate supernatant Total (pellet 1 supernatant) 0–30% solid ammonium sulfate pellet 0–30% solid ammonium sulfate supernatant 0–50% ammonium sulfate pellet 0–50% ammonium sulfate supernatant 30–50% solid ammonium sulfate pellet 30–50% solid ammonium sulfate supernatant 50–70% solid ammonium sulfate pellet 50–70% solid ammonium sulfate supernatant
100 70 30 100 20 125 10 75 10 100 65 20
Note. The urine was from a female patient with late-stage multiple myeloma. Assays were for 5 min. One unit 5 0.3 nmole NADH oxidized min21.
FIG. 1. FPLC trace of separation of 30 to 70% ammonium sulfate precipitate from urine of a female patient with late-stage multiple myeloma. The urine was treated for 2 h with 200 mg/ml proteinase K at 34°. The activity profiles of the fractions are given in the inset. The NADH oxidase activities were measured in the absence of capsaicin or in the presence of 1 mM capsaicin or 100 mM capsaicin. Activities inhibited by capsaicin were observed in fraction 36. Capsaicin-stimulated activity was seen between fractions 55 and 75. The peptide molecular mass associated with the capsaicin-inhibited activity was estimated to be between 60 and 70 kDa whereas that of the capsaicin-stimulated activity was estimated to be between 30 and 40 kDa.
340
YANTIRI ET AL.
FIG. 2. SDS–PAGE (12% acrylamide) of fractions from the FPLC separation of Fig. 1. Activity in fraction 36 correlated with the presence of a 66-kDa band (arrow) not seen in fractions 35 and 37. Activity in fraction 60 correlated with a 33-kDa band (arrow) but this fraction lacked the 66-kDa component.
Fig. 1 and present in fractions lacking the capsaicininhibited NADH oxidase. The latter correlated with the presence of a band at 66 kDa (Fig. 2, Fraction 36, arrow) corresponding approximately to twice the molecular weight of the serum form of the capsaicininhibited NADH oxidase (5). The material of the 33-kDa band was N-terminally blocked and, by analogy to the serum form, would be expected to resist further digestion with proteases (4). When separated on preparative SDS–PAGE from the other proteins present following proteinase K digestion and digested in the gel with cyanogen bromide, the 33-kDa band was resistant to further digestion. However when treated with methyl iodide to methylate cysteines, followed by a second cyanogen bromide digestion, cleavage did occur with the generation of a 27-kDa component (Fig. 3). The 27-kDa fragment was resistant to further cyanogen bromide cleavage both before and after methylation (not shown). The 33-kDa band cross-reacted with antisera to the drug-responsive serum form of the NADH oxidase as shown by electrochemical luminescence detection (Fig. 4). The isolation protocol as described for Fig. 4 was carried out in parallel with urine from a healthy volunteer and from a patient with proteinuria (76-yearold female with 100 mg/dl urinary protein) in addition to urine from the myeloma patient (Fig. 5) with detection of immunoreactive bands by means of alkaline phosphatase-linked second antibody and NBT-BCIP. These analyses failed to reveal either a NADH oxidase activity inhibited by capsaicin or a band immunoreactive with the antisera to the cancer-specific serum form of the NADH oxidase with the sera from the healthy volunteer (Fig. 5B) or from the patient with proteinurea (Fig. 5C). For the myeloma patient, an immunoreactive band, very similar to that of Fig. 4, was observed (Fig. 5A).
DISCUSSION
A capsaicin-inhibited (or stimulated) NADH oxidase of the cancer cell surface was first described for plasma membrane vesicles of HeLa cells (1). The activity was seen with sealed, right side-out vesicles and, since NADH is an impermeant substrate, was assumed to be external (2). The activity also was shed both into culture media conditioned by growth of HeLa cells (3) and into sera with autochthonous tumors in cancer patients (5). The present findings demonstrate a capsaicin-responsive activity also present in urine of cancer patients. As
FIG. 3. Partial cleavage by cyanogen bromide following methylation of the 33-kDa protein cut from a preparative SDS–PAGE gel. The resultant cleavage product consisted of a band at 27 kDa. Also present was a band at 52 kDa not present in the starting material and the uncleaved starting material at 33-kDa.
CAPSAICIN-RESPONSIVE NADH OXIDASES FROM HUMAN URINE
FIG. 4. Reaction of the 33-kDa component (arrow) of Fraction 60 of Fig. 2 with a monoclonal antibody to the drug-responsive NADH oxidase isolated from sera of cancer patients (4, 17). Detection was by electrochemiluminescence. Immunoprecipitated proteins were separated by SDS–PAGE and transferred to nitrocellulose membrane by electroblotting. The nitrocellulose membranes were blocked in a solution of 10% non-fat milk prepared in PBS/0.05% (v/v) Tween 20 containing 1% (w/v) blocking reagent 1096 1Z6 (Boehringer Mannheim) and probed with peroxidase-conjugated second antibody. Immunoreactive proteins were detected by chemiluminescence using an ECL kit (Amersham Corp.) following the manufacturer’s instructions.
with sera, NADH oxidase activity of urine of healthy volunteers was unresponsive to capsaicin. The NADH oxidase was of low specific activity in urine, but urine might represent both a useful starting material for isolation of the protein(s) involved and for diagnostic purposes due to the ease of collection and ready availability in quantity. In order to begin to investigate these possibilities, the NADH oxidase was concentrated from urine by ammonium sulfate fractionation and proteinase K digestion. The resistance of the NADH oxidase activity from urine to proteinase K was not unexpected based on work with HeLa cells (15) and results with serum (4). Here the activity was found to resist digestion by a variety of proteases, including not only proteinase K but trypsin, chymotrypsin, V-8, subtilisin, lysylaminopeptidase C, and pronase. The protein also resisted digestion with cyanogen bromide and was N-terminally blocked to frustrate attempts to generate either N-terminal or internal amino acid sequence. The serum (4) and urine forms as well as the activity from
341
HeLa cells (15) were resistant to heat as well as to proteinase digestion. A single urine sample collected from a female patient with multiple myeloma was utilized to isolate the capsaicin-stimulated activity. Following proteinase K digestion, most of the urine proteins were digested and the next step was to remove these fragments by FPLC. The FPLC separation yielded two large, broad absorbance peaks with molecular masses of less than 20 kDa which contained the bulk of these fragments. The recovery of activity stimulated by capsaicin was estimated as 70% after ammonium sulfate precipitation and proteinase K and about 40% after FPLC (Table IV). The capsaicin-stimulated activity was recovered from the FPLC separation in a fraction collected just ahead of the first of the two large absorbance peaks containing fragments from proteinase K digestion (Fig. 1). The activity correlated with a 33-kDa band that cross-reacted with a monoclonal antibody to the 33.5-kDa form of the protein from human serum. This activity was low in the absence of capsaicin and required both capsaicin and dilute hydrogen peroxide for activity. Even though the urine sample fractionated initially contained predominantly an activity stimulated by capsaicin, a second activity peak was observed in the trailing portion of a small absorbance peak in the 60- to 80-kDa range of the FPLC separation. This activity peak contained several components in the 45-kDa molecular mass range plus a component at about 66-kDa. The NADH oxidase activity of the second peak was inhibited rather than stimulated by both 1 mM and 100 mM capsaicin as was the activity from culture media conditioned by growth of HeLa cells (3) and the activity
FIG. 5. Western blot analysis of the 33-kDa component (arrow) from FPLC separations comparing three urine samples analyzed in parallel. Following separation on 12% SDS–PAGE as in Fig. 2, proteins were electroblotted to nitrocellulose and incubated overnight at 4°C with 1:250 diluted monoclonal antibody to the drug-responsive NADH oxidase as in Fig. 4. Detection was with alkaline phosphatase-conjugated second antibody diluted 1:5000 followed by incubation with NBT-BCIP. (A) Urine from the female multiple myeloma patient. (B) Urine from a healthy male volunteer. (C) Urine from a female patient with proteinurea.
342
YANTIRI ET AL. TABLE IV
Recovery of 100 mM Capsaicin-Stimulated Activity and Protein from Urine of a Female Patient with Late Stage Multiple Myeloma
Fraction Original urine Ammonium sulfate pellet (30–70%) Pellet after proteinase K, 200 mg/ml, 2 h Pellet after FPLC
Capsaicin-stimulated NADH oxidase activity, units/100 ml
disulfide reductase whose physiological function has been suggested to be protein–disulfide thiol interchange (18). The NADH oxidase activity from HeLa cells inhibited by antitumor sulfonylurea exhibits a sulfonylurea-inhibited protein disulfide interchange activity (19) also inhibited by capsaicin.
Protein, mg/100 ml
100
90
73
47 a
70 40
Note. Values are for 100 ml of urine. Similar results were obtained for a female patient with late stage metastatic colon cancer. One unit 5 0.3 nmole NADH oxidized min21. a The total recovery of protein after ammonium sulfate fractionation was 85%.
from the majority of the serum samples when assayed under oxidizing conditions (5). The resistance of the urine and serum NADH oxidase proteins to both heat and protease treatment as well as to cyanogen bromide cleavage is unusual and has thus far precluded the generation of primary amino acid sequence. One or more active site cysteines in the NADH oxidase protein has been inferred from studies with thiol reagents (16). In order to facilitate cyanogen bromide cleavage, the active site cysteines were first methylated with methyl iodide followed by cyanogen bromide. Methylated cysteines often behave as methionines following cyanogen bromide treatment to introduce additional potential cyanogen bromide cleavage sites in the protein (13). With the 33-kDa urine form, methylation followed by cyanogen bromide did appear to introduce a cyanogen bromide cleavage site with the generation of a ca 27-kDa fragment. However, the 27-kDa protein was resistant to cyanogen bromide both before and after methylation and no useful amino acid sequence resulted. These findings may suggest the presence of a cysteine located approximately 50 amino acids from the C-terminus since the resultant peptide was still blocked to N-terminal sequencing. The function of the cancer-specific NADH oxidase even prior to shedding from cancer cells remains unknown. Based on studies with HeLa cells and the properties of the monoclonal antibody (17) some relationship of the membrane-located form to growth is indicated. Certainly since the NADH site is at the external cell surface and since an external source of cellular NADH is unlikely, any role in NADH oxidation is assumed to be artifactual and only a convenient method of assay. Also, oxygen may not be the natural electron acceptor even for NADH oxidation (18). A comparable activity from plants stimulated by growth hormones has properties of an NADH: protein
ACKNOWLEDGMENTS We thank Connie Chalko and Plymouth Community Hospital, Plymouth, Indiana, Dr. Lawrence Helson, Chappaqua, NY, and Richard Bentlage, Saint Elizabeth Hospital, Lafayette, Indiana, for providing patient urine samples and for helpful discussions. Yanxia Yang provided technical assistance. This work was supported in part by a gift from Leonard P. Shaykin.
REFERENCES 1. Morre´, D. J., Chueh, P.-J. and Morre´, D. M. (1995) Proc. Natl. Acad. Sci. USA 92, 1831–1835. 2. Morre´, D. J., Wilkinson, F. E., Kim, C., Cho, N., Lawrence, J., Morre´, D. M., and McClure, D. (1996) Biochim. Biophys. Acta 1280, 197–206. 3. Wilkinson, F., Kim, C., Cho, N., Chueh, P.-J., Leslie, S., MoyaCamarena, S., Wu, L.-Y., Morre´, D. M., and Morre´, D. J. (1996) Arch. Biochem. Biophys. 336, 275–282. 4. Chueh, P.-J., Morre´, D. J., Wilkinson, F. E., Gibson, T., and Morre´, D. M. (1997) Arch. Biochem. Biophys. 342, 38 – 47. 5. Morre´, D. J., Caldwell, S., Mayorga, A., Wu, L.-Y., and Morre´, D. M. (1997) Arch. Biochem. Biophys. 342, 224 –230. 6. Laemmli, U. K. (1990) Nature 227, 680 – 685. 7. Butcher, L. A., and Tomkins, J. K. (1985) Anal. Biochem. 148, 384 –388. 8. Towbin, H., Staehelin, T., and Gordon, J. (1979) Proc. Natl. Acad. Sci. USA 76, 4350 – 4354. 9. Smith, P. K., Krohm, R. I., Hermanson, G. T., Mallia, A. K., Gartner, F. H., Provenzano, E. K., Fujimoto, E. K., Goeke, N. M., Olson, B. J., and Klenk, D. C. (1985) Anal. Biochem. 100, 76 – 85. 10. Scha¨ger, H., and Von Jagow, G. (1987) Anal. Biochem. 166, 368 –379. 11. Stone, K. L., La Presti, M. B., Williams, N. D., Crawford, J. M., De Angelis, R., and William, K. R. (1989) Techniques in Protein Chemistry, pp. 51–53, Academic Press, New York. 12. Jahnen, W., Ward, L. D, Reid, G. E., Moritz, R. L., and Simpson, R. J. (1990) Biochem. Biophys. Res. Commun. 166, 139 –145. 13. Awad, W., and Wilcox, P. E. (1964) Biochem. Biophys. Res. Commun. 17, 709 –715. 14. Ploug, M., Stoffer, B., and Jensen, A. L. (1992) Electrophoresis 13, 148 –153. 15. del Castillo-Olivares, A., Yantiri, F., Chueh, P.-J., Wang, S., Sweeting, M., Sedlak, D., Morre´, D. M., Burgess, J., and Morre´, D. J. Arch. Biochem. Biophys., in press. 16. Morre´, D. J., and Morre´, D. M. (1995) J. Bioenerg. Biomemb. 24, 137–144. 17. Cho, N., Chueh, P.-J., Caldwell, S., Wu, L.-Y., Werderitsh, D. A., Morre´, D. M., and Morre´, D. J. (1998) Proc. Amer. Assoc. Cancer Res. 39, 437 (abs). 18. Chueh, P.-J., Morre´, D. M., Penel, C., DeHahn, T., and Morre´, D. J. (1997) J. Biol. Chem. 272, 11221–11227. 19. Morre´, D. J., Jacobs, E., Sweeting, M., deCabo, R., and Morre´, D. M. (1997) Biochim. Biophys. Acta 1325, 117–125.