AMERICAN JOURNAL OF REPRODUCTIVE IMMUNOLOGY 27:163-166 © 1992 MUNKSGAARD
A'ftJoducltve Tumor Immuno1oNl
Characterization of the Anti-Cancer Activity of Transferrin-Adriamycin Conjugates JOSEPH A. SIZENSKY, KLARA BARABAS, AND W. PAGE FAULK Center for Reproduction and Transplantation Immunology, Methodist Hospital, Indianapolis, Indiana (J A.S., WPF.); Institute ofBiophysics, Biological Research Center, Hungarian Academy of Sciences, H -6701, Szeged, Hungary (K.B.J; Department of Medicinal Chemistry, Purdue University, West Lafayette, Indiana (WPF)
The anthracycline anti-cancer drug adriamycin (Adr) was coupled to human transferrin (Trf) by using a glutaraldehyde technique. The effect ofTrf-Adr conjugates and unconjugated Adr on human cells was determined by using normal peripheral blood mononuclear cells and chronic myelogenous K562 cells. Cytotoxicity was determined by using an assay that measures the conversion of a tetrazolium salt (MIT) into a purple product (formazan) by mitochondrial dehydrogenases in viable cells. We found that free Adr at a concentration of 1 x 10 -7 had little effect on K562 cells, while Trf-Adr conjugates inhibited 75% of cellular activity. When normal peripheral blood mononuclear cells were tested against Trf-Adr conjugates, the 50% inhibitory concentration was found to be 1.4-1.7 x 10- 6 M, at which concentration >85% ofK562 cells were inhibited. Interactions of Trf-Adr conjugates with plasma membrane energy-producing systems are the proposed mechanisms of cytotoxicity. (Am J Reprod Immunol. 1992; 27:163-166.) Key words: Transferrin receptors, transferrin-adriamycin conjugates, MIT assay, cancer, drug targeting. ABSTRACT:
INTRODUCTION
Adriamycin (Adr) is an anthracycline antineoplastic agent used widely in the treatment of many neoplasms, althouyh its use is limited by toxic dose-related side effects. ,2 It has been proposed that Adr coupled to tumorspecific monoclonal antibodies could eliminate many of the toxicities of Adr by delivering drug directly to tumor cells," However, the clinical use ofantibody-drug complexes is limited by problems, including the development of anaphylactic reactions." One promising approach that may overcome these problems is the use of Adr conjugates of transferrin (Trf), a normal plasma protein and the natural ligand of the Trf receptor. 5 ,6 Trf receptors are upregulated on the plasma membranes of malignant/-" and other proliferating9-11 cells, but are absent from most normal adult resting cells, suggesting that Trf-Adr conjugates should be capable of targeting drug directly to tumor cells." Indeed, this approach has been used successfully against tumor cells both in vitro l 2-14 and in vivo. 15 In this article we report the characterization of the antitumor effect of Tf-Adr conjugates on human leukemic K562 cells. Cytotoxicity was determined by using an assay that measures the conversion of a tetrazolium salt (MIT) into a purple product (formazan) by mitochondrial dehydrogenases in viable cells." The appearance of purple
Submitted December 12, 1991; accepted March 23, 1992. Address reprint requests to Prof. W. Page Faulk, Center for Reproduction and Transplantation Immunology, Methodist Hospital, 1701 N. Senate Boulevard, Indianapolis, IN 46202.
color is observed as a change in the absorbance at 570 nm and is directly proportional to cell number. Trf-Adr conjugates were found to be significantly more cytotoxic than free Adr, and conjugates inhibited tumor cells more efficiently than normal cells. MATERIALS AND METHODS
Conjugate Preparation Trf-Adr conjugates were prepared by using a modification of the glutaraldehyde crosslinking method ofYeh and Faulk. 12 To a mixture (1:1 ml) of8.5 mM of Adr (Cetus, Emoryville, CA) and 0.5 mM of human Trf(98% ironsaturated or 98% iron-free as determined spectrophotometrically'"; Sigma, St. Louis, MO) in 0.9% NaCI was added dropwise 1 ml of 0.9% NaCI containing 21.5 mM of glutaraldehyde (Kodak, Rochester, NY) over 5 min with gentle stirring. The coupling reaction was terminated by adding 0.8 ml of 0.9% NaCI containing 37 mM of ethanolamine (Sigma) (pH 8), and the mixture was dialyzed in HBS, pH 7.4. The turbid mixture was centrifuged at 2,500 rpm for 15 min, and the clear supernatant was chromatographed through a 34 x 2.6 em column of Sepharose CL-4B (Pharmacia, Piscataway, NJ). Spectrophotometric readings of column fractions were taken at 280 nm and 495 nm to identify protein and Adr peaks, respectively. Trf-Adr conjugates were dialyzed overnight and molar concentrations ofTrfand Adr were determined spectrophotometrically. Cell Culture Human chronic myelogenous leukemia K562 cells were grown in suspension in a humidified atmosphere of95% (v/v) air and 5% (v/v) CO 2 at 37°C in 90% (v/v) RPMI-1640 at pH 7:0 in 25 mM Hepes (Sigma) and 10% (v/v) heatinactivated fetal bovine serum (FBS) (Bioproducts for Science, Indianapolis, IN) supplemented with 102 units/ml penicillin G, 1021J-g/ml streptomycin, and 1201J-g/ml L-glutamine (Whittaker Bioproducts, Waltersville, MD). Normal human peripheral blood mononuclear cells were isolated from citrated whole blood by flotation on Ficoll-hypa~ue (Pharmacia, Piscataway, NJ) as described by Boyum. 8
Drug Treatment To individual wells of a 96-well tissue culture plate (Corning, Corning, NY) 50 IJ-I of culture medium were added containing double-strength dilutions of the appropriate drug and equilibrated at 37°C in a humidified 95% air, 5% CO 2 atmosphere. K562 cells and normal mononuclear cells were harvested and washed twice with serum-free culture medium and resuspended in complete medium to final concentrations of2 x 105 cells/ml and 2 x 10 6 cells/ml, respectively. To the appropriate wells of the pre-equilibrated 96-well plates were added 50 IJ-I of cells and the plate was incubated for 48 h at 37°C in a humidified 95% air, 5% CO 2 atmosphere. In experiments
SIZENSKY ET AL.
164
including iron-free Trf-Adr conjugates, desferroxamine (CIBA-GEIGY, Summit, NJ) was included at a final concentration of 100 fJ-M during the incubation.
Cytotoxicity Assay The CellTiter-96 nonradioactive cell proliferation/cytotoxicity assay was performed according to manufacturer's instructions (Promega, Madison, WI). Briefly, 20 fJ-I of dye solution was added to each well followed by incubation at 37°C for 4 h in a humidified 95% air, 5% CO2 atmosphere to allow for formation of formazan crystals. Formazan crystals were solubilized by the addition oflOO fJ-I of solubilization solution to each well followedby overnight incubation of the sealed plates. Absorbance was measured by using a TiterTek Multiskan Plus plate reader (ICNlFlow, Costa Mesa, CA)equipped with a 570-nm fixed wavelength filter (ICN/Flow).
100
...J
~
~
80
0
a:
I-
z
0
0
u..
60
0
z
0
i= iii J:
40
~
~ 0
20
RESULTS
Glutaraldehyde crosslinking of Adr to Trf produces Trf-Adr conjugates with a heterogeneous range of conjugation numbers (i.e., an average molar ratio of Adr:Trf ofbetween -1-9) and lowaggregation ofTrf. In this study,
0 0
2
3
4
5
[ADR], IJM
Fig. 2. Cytotoxicity of Trf-Adr conjugates on K562 cells and normal peripheral blood mononuclear cells. Aliquots of! x 104K562cells(0,e) or 1 x lOs normal cells (~,.) were incubated with the appropriate concentration of Trf-Adr conjugates with high conjugation numbers (solid shapes) or 'Iif-Adrconjugates with lowconjugation numbers (open shapes) in 100 ILl RPMI-1640plus 25 mMHepes (pH 7.0) plus 10% FBS for 48 h at 37°C in a humidified atmosphere of 95% air, 5% CO2 in a 96-well plate. Otherwise as in Fig. 1.
100
80 ...J
0
a:
I-
z
0
60
0
u..
0
Z
0
i=
40
fQ
J:
~
0~
20
o o
1
2
3
4
5
[ADR], IJM
Fig. 1. Cytotoxicity of free Adr and Trf-Adr conjugates on K562 cells. Aliquots of 1 x 104 K562 cells were incubated with the appropriate concentration offree Adr (.), high conjugation number Trf-Adr conjugates (e), or low conjugation number ofTrf-Adr conjugates (0) in 100 ILl of RPMI-1640 plus 25 mM Hepes (pH 7.0) plus 10% FBS for 48 h at 37°C in a humidified atmosphere of 95% air, 5% CO 2 in a 96-well plate. After addition of 20 ILl dye solution per well and incubation for a further 4 h, 100 ILl of solubilization solution was added. The plate was sealed and incubated overnight to completely solubilize formazan crystals. Results are presented as % inhibition ofODs7o formation of drug-treated wells compared with untreated controls (i.e., 0% inhibition) after background correction. Each point is calculated from the average of quadruplicate wells in each of two independent experiments. The standard deviation between replicate wells and individual experiments never exceeded 8% of the mean.
conjugate fractions with an average conjugation number of0.8-1.1 (predominantlyTrfmonomers) and 6.0-8.0 (predominantly Trf dimers) were pooled separately and referred to as the low conjugation number pool and the high conjugation number pool, respectively. No detectable drug was released from transferrin as determined by spectrophotometry after dialysis. The inhibitory effects of Adr and Trf-Adr on K562 cells are shown in Fig. 1. Free Adr had little effect on K562 cells at the lowest concentration used (i.e., below 1 x 10- 7 M), but conjugates ofTrf-Adr inhibited cellular activity by more than 75% at this concentration (Fig. 1). In addition, the concentration at which 50% inhibition of cellular activity (ICso)was achieved was found to be 6 x 10- 7 M for free Adr, but conjugates at that concentration retained over 80% inhibition. No activity was detected in cells at concentrations greater than 5 x 10 - 6 M for either free or conjugated Adr. The effect of conjugation number (i.e., the number of Adr molecules per Trf molecule) on the cytotoxicity of Trf-Adr was determined by incubating K562 cells with conjugates oflow and high conjugation numbers. As shown in Figs. 1 and 2, Trf-Adr conjugates had very similar growth inhibition effects despite significant differences in conjugation numbers. This observation suggests that, when molecules ofTrf-Adr conjugate bind cell surfaces, only a limited number of Adr moieties can interact directly with the plasma membrane at anyone time. Results of cytotoxicity studies done with Trf-Adr con-
ANTI-CANCER ACTIVITYOF TRANSFERRIN CONJUGATES
100
...J
o a:
80
I-
z
o o II.. o
60
Z
o
i=
iii i:
40
~
~
20
o
o
1
2
3
4
5
[ADR], IJM
Fig. 3. Cytotoxicity of Trf-Adr conjugates on K562 cells. Aliquots of 1 x 104 K562 cells were incubated with the appropriate concentration of98% iron-saturated Trf-Adr conjugates (e) or 98% iron-free Trf-Adr conjugates (0) in 100 ....1RPMI-1640 containing 25 mM Hepes (pH 7.0), 10%FBS, and 100 ....M desferroxamine for 48 h at 37°C in a humidified atmosphere of95% air, 5%CO2 in a 96-wellplate. Otherwise as in Fig. 1.
jugates on normal peripheral blood mononuclear cells are depicted in Fig. 2. Trf-Adr conjugates had little effect on normal cells at concentrations below 1 x 10 -7 M, but this concentration of conjugate inhibited K562 cells by >75%. The IC 50for Trf-Adr conjugates on normal peripheral blood mononuclear cells was found to be 1.4 x 10 - 6 M and 1. 7 x 10 - 6 M for conjugates of high and low conjugation numbers, respectively. However, at these concentrations,Trf-AdrconjugatesinhibitedK562 cells by >85%. Cell proliferation is dependent upon iron derived from transferrin. 19-22 The effect of iron saturation in Trf-Adr conjugates on the cytotoxicity of K562 cells is shown in Fig. 3. When the Trfin Trf-Adr conjugates was 98% ironsaturated, K562 cells were inhibited similarly to the data in Figs. 1 and 2, but somewhat less inhibition was observed when the Trf in Trf-Adr conjugates was 98% iron-free. In this case, the cytotoxicity was comparable with iron-saturated conjugates at concentrations
A'ftJoducltve Tumor Immuno1oNl
Characterization of the Anti-Cancer Activity of Transferrin-Adriamycin Conjugates JOSEPH A. SIZENSKY, KLARA BARABAS, AND W. PAGE FAULK Center for Reproduction and Transplantation Immunology, Methodist Hospital, Indianapolis, Indiana (J A.S., WPF.); Institute ofBiophysics, Biological Research Center, Hungarian Academy of Sciences, H -6701, Szeged, Hungary (K.B.J; Department of Medicinal Chemistry, Purdue University, West Lafayette, Indiana (WPF)
The anthracycline anti-cancer drug adriamycin (Adr) was coupled to human transferrin (Trf) by using a glutaraldehyde technique. The effect ofTrf-Adr conjugates and unconjugated Adr on human cells was determined by using normal peripheral blood mononuclear cells and chronic myelogenous K562 cells. Cytotoxicity was determined by using an assay that measures the conversion of a tetrazolium salt (MIT) into a purple product (formazan) by mitochondrial dehydrogenases in viable cells. We found that free Adr at a concentration of 1 x 10 -7 had little effect on K562 cells, while Trf-Adr conjugates inhibited 75% of cellular activity. When normal peripheral blood mononuclear cells were tested against Trf-Adr conjugates, the 50% inhibitory concentration was found to be 1.4-1.7 x 10- 6 M, at which concentration >85% ofK562 cells were inhibited. Interactions of Trf-Adr conjugates with plasma membrane energy-producing systems are the proposed mechanisms of cytotoxicity. (Am J Reprod Immunol. 1992; 27:163-166.) Key words: Transferrin receptors, transferrin-adriamycin conjugates, MIT assay, cancer, drug targeting. ABSTRACT:
INTRODUCTION
Adriamycin (Adr) is an anthracycline antineoplastic agent used widely in the treatment of many neoplasms, althouyh its use is limited by toxic dose-related side effects. ,2 It has been proposed that Adr coupled to tumorspecific monoclonal antibodies could eliminate many of the toxicities of Adr by delivering drug directly to tumor cells," However, the clinical use ofantibody-drug complexes is limited by problems, including the development of anaphylactic reactions." One promising approach that may overcome these problems is the use of Adr conjugates of transferrin (Trf), a normal plasma protein and the natural ligand of the Trf receptor. 5 ,6 Trf receptors are upregulated on the plasma membranes of malignant/-" and other proliferating9-11 cells, but are absent from most normal adult resting cells, suggesting that Trf-Adr conjugates should be capable of targeting drug directly to tumor cells." Indeed, this approach has been used successfully against tumor cells both in vitro l 2-14 and in vivo. 15 In this article we report the characterization of the antitumor effect of Tf-Adr conjugates on human leukemic K562 cells. Cytotoxicity was determined by using an assay that measures the conversion of a tetrazolium salt (MIT) into a purple product (formazan) by mitochondrial dehydrogenases in viable cells." The appearance of purple
Submitted December 12, 1991; accepted March 23, 1992. Address reprint requests to Prof. W. Page Faulk, Center for Reproduction and Transplantation Immunology, Methodist Hospital, 1701 N. Senate Boulevard, Indianapolis, IN 46202.
color is observed as a change in the absorbance at 570 nm and is directly proportional to cell number. Trf-Adr conjugates were found to be significantly more cytotoxic than free Adr, and conjugates inhibited tumor cells more efficiently than normal cells. MATERIALS AND METHODS
Conjugate Preparation Trf-Adr conjugates were prepared by using a modification of the glutaraldehyde crosslinking method ofYeh and Faulk. 12 To a mixture (1:1 ml) of8.5 mM of Adr (Cetus, Emoryville, CA) and 0.5 mM of human Trf(98% ironsaturated or 98% iron-free as determined spectrophotometrically'"; Sigma, St. Louis, MO) in 0.9% NaCI was added dropwise 1 ml of 0.9% NaCI containing 21.5 mM of glutaraldehyde (Kodak, Rochester, NY) over 5 min with gentle stirring. The coupling reaction was terminated by adding 0.8 ml of 0.9% NaCI containing 37 mM of ethanolamine (Sigma) (pH 8), and the mixture was dialyzed in HBS, pH 7.4. The turbid mixture was centrifuged at 2,500 rpm for 15 min, and the clear supernatant was chromatographed through a 34 x 2.6 em column of Sepharose CL-4B (Pharmacia, Piscataway, NJ). Spectrophotometric readings of column fractions were taken at 280 nm and 495 nm to identify protein and Adr peaks, respectively. Trf-Adr conjugates were dialyzed overnight and molar concentrations ofTrfand Adr were determined spectrophotometrically. Cell Culture Human chronic myelogenous leukemia K562 cells were grown in suspension in a humidified atmosphere of95% (v/v) air and 5% (v/v) CO 2 at 37°C in 90% (v/v) RPMI-1640 at pH 7:0 in 25 mM Hepes (Sigma) and 10% (v/v) heatinactivated fetal bovine serum (FBS) (Bioproducts for Science, Indianapolis, IN) supplemented with 102 units/ml penicillin G, 1021J-g/ml streptomycin, and 1201J-g/ml L-glutamine (Whittaker Bioproducts, Waltersville, MD). Normal human peripheral blood mononuclear cells were isolated from citrated whole blood by flotation on Ficoll-hypa~ue (Pharmacia, Piscataway, NJ) as described by Boyum. 8
Drug Treatment To individual wells of a 96-well tissue culture plate (Corning, Corning, NY) 50 IJ-I of culture medium were added containing double-strength dilutions of the appropriate drug and equilibrated at 37°C in a humidified 95% air, 5% CO 2 atmosphere. K562 cells and normal mononuclear cells were harvested and washed twice with serum-free culture medium and resuspended in complete medium to final concentrations of2 x 105 cells/ml and 2 x 10 6 cells/ml, respectively. To the appropriate wells of the pre-equilibrated 96-well plates were added 50 IJ-I of cells and the plate was incubated for 48 h at 37°C in a humidified 95% air, 5% CO 2 atmosphere. In experiments
SIZENSKY ET AL.
164
including iron-free Trf-Adr conjugates, desferroxamine (CIBA-GEIGY, Summit, NJ) was included at a final concentration of 100 fJ-M during the incubation.
Cytotoxicity Assay The CellTiter-96 nonradioactive cell proliferation/cytotoxicity assay was performed according to manufacturer's instructions (Promega, Madison, WI). Briefly, 20 fJ-I of dye solution was added to each well followed by incubation at 37°C for 4 h in a humidified 95% air, 5% CO2 atmosphere to allow for formation of formazan crystals. Formazan crystals were solubilized by the addition oflOO fJ-I of solubilization solution to each well followedby overnight incubation of the sealed plates. Absorbance was measured by using a TiterTek Multiskan Plus plate reader (ICNlFlow, Costa Mesa, CA)equipped with a 570-nm fixed wavelength filter (ICN/Flow).
100
...J
~
~
80
0
a:
I-
z
0
0
u..
60
0
z
0
i= iii J:
40
~
~ 0
20
RESULTS
Glutaraldehyde crosslinking of Adr to Trf produces Trf-Adr conjugates with a heterogeneous range of conjugation numbers (i.e., an average molar ratio of Adr:Trf ofbetween -1-9) and lowaggregation ofTrf. In this study,
0 0
2
3
4
5
[ADR], IJM
Fig. 2. Cytotoxicity of Trf-Adr conjugates on K562 cells and normal peripheral blood mononuclear cells. Aliquots of! x 104K562cells(0,e) or 1 x lOs normal cells (~,.) were incubated with the appropriate concentration of Trf-Adr conjugates with high conjugation numbers (solid shapes) or 'Iif-Adrconjugates with lowconjugation numbers (open shapes) in 100 ILl RPMI-1640plus 25 mMHepes (pH 7.0) plus 10% FBS for 48 h at 37°C in a humidified atmosphere of 95% air, 5% CO2 in a 96-well plate. Otherwise as in Fig. 1.
100
80 ...J
0
a:
I-
z
0
60
0
u..
0
Z
0
i=
40
fQ
J:
~
0~
20
o o
1
2
3
4
5
[ADR], IJM
Fig. 1. Cytotoxicity of free Adr and Trf-Adr conjugates on K562 cells. Aliquots of 1 x 104 K562 cells were incubated with the appropriate concentration offree Adr (.), high conjugation number Trf-Adr conjugates (e), or low conjugation number ofTrf-Adr conjugates (0) in 100 ILl of RPMI-1640 plus 25 mM Hepes (pH 7.0) plus 10% FBS for 48 h at 37°C in a humidified atmosphere of 95% air, 5% CO 2 in a 96-well plate. After addition of 20 ILl dye solution per well and incubation for a further 4 h, 100 ILl of solubilization solution was added. The plate was sealed and incubated overnight to completely solubilize formazan crystals. Results are presented as % inhibition ofODs7o formation of drug-treated wells compared with untreated controls (i.e., 0% inhibition) after background correction. Each point is calculated from the average of quadruplicate wells in each of two independent experiments. The standard deviation between replicate wells and individual experiments never exceeded 8% of the mean.
conjugate fractions with an average conjugation number of0.8-1.1 (predominantlyTrfmonomers) and 6.0-8.0 (predominantly Trf dimers) were pooled separately and referred to as the low conjugation number pool and the high conjugation number pool, respectively. No detectable drug was released from transferrin as determined by spectrophotometry after dialysis. The inhibitory effects of Adr and Trf-Adr on K562 cells are shown in Fig. 1. Free Adr had little effect on K562 cells at the lowest concentration used (i.e., below 1 x 10- 7 M), but conjugates ofTrf-Adr inhibited cellular activity by more than 75% at this concentration (Fig. 1). In addition, the concentration at which 50% inhibition of cellular activity (ICso)was achieved was found to be 6 x 10- 7 M for free Adr, but conjugates at that concentration retained over 80% inhibition. No activity was detected in cells at concentrations greater than 5 x 10 - 6 M for either free or conjugated Adr. The effect of conjugation number (i.e., the number of Adr molecules per Trf molecule) on the cytotoxicity of Trf-Adr was determined by incubating K562 cells with conjugates oflow and high conjugation numbers. As shown in Figs. 1 and 2, Trf-Adr conjugates had very similar growth inhibition effects despite significant differences in conjugation numbers. This observation suggests that, when molecules ofTrf-Adr conjugate bind cell surfaces, only a limited number of Adr moieties can interact directly with the plasma membrane at anyone time. Results of cytotoxicity studies done with Trf-Adr con-
ANTI-CANCER ACTIVITYOF TRANSFERRIN CONJUGATES
100
...J
o a:
80
I-
z
o o II.. o
60
Z
o
i=
iii i:
40
~
~
20
o
o
1
2
3
4
5
[ADR], IJM
Fig. 3. Cytotoxicity of Trf-Adr conjugates on K562 cells. Aliquots of 1 x 104 K562 cells were incubated with the appropriate concentration of98% iron-saturated Trf-Adr conjugates (e) or 98% iron-free Trf-Adr conjugates (0) in 100 ....1RPMI-1640 containing 25 mM Hepes (pH 7.0), 10%FBS, and 100 ....M desferroxamine for 48 h at 37°C in a humidified atmosphere of95% air, 5%CO2 in a 96-wellplate. Otherwise as in Fig. 1.
jugates on normal peripheral blood mononuclear cells are depicted in Fig. 2. Trf-Adr conjugates had little effect on normal cells at concentrations below 1 x 10 -7 M, but this concentration of conjugate inhibited K562 cells by >75%. The IC 50for Trf-Adr conjugates on normal peripheral blood mononuclear cells was found to be 1.4 x 10 - 6 M and 1. 7 x 10 - 6 M for conjugates of high and low conjugation numbers, respectively. However, at these concentrations,Trf-AdrconjugatesinhibitedK562 cells by >85%. Cell proliferation is dependent upon iron derived from transferrin. 19-22 The effect of iron saturation in Trf-Adr conjugates on the cytotoxicity of K562 cells is shown in Fig. 3. When the Trfin Trf-Adr conjugates was 98% ironsaturated, K562 cells were inhibited similarly to the data in Figs. 1 and 2, but somewhat less inhibition was observed when the Trf in Trf-Adr conjugates was 98% iron-free. In this case, the cytotoxicity was comparable with iron-saturated conjugates at concentrations
