IN VITRO Volume 12, No. 11, 1976
GENETIC CHARACTERIZATION OF METHOTREXATE-RESISTANT CHINESE HAMSTER OVARY CELLS 1 WAYNE F. FLINTOFF, 2SUSAN M. SPINDLER, ANDLOUIS SIMINOVITCH
Department of Medical Genetics, University of Toronto, Toronto, Ontario, Canada M5S 1A8
SUMMARY In a previous report, we described the selection and partial characterization of three methotrexate ~Mtx)-resistant Chinese hamster ovary cells ICHO) ~1~. Class I cells contained an apparent structural alteration in dihydrofolate reductase. Class II cells had an alteration affecting the permeability of the drug. Class I I I cells, selected from Class I cells, had an increased activity of the altered enzyme. In the work described here, it has been shown that the spontaneous mutation rate to Class I resistance is in the order of 2 )< 10 -9 mutations per locus per generation and that in single-step mutagenized selections the number of resistant colonies of Classes I and II are about equal. Class I and Class I I ! resistance is expressed codominantly in somatic cell hybrids, whereas the Class II resistant marker is a recessive trait.
Key words: somatic cell genetics; methotrexate-resistance; Chinese hamster ovary cells; cell-cell hybridization. INTRODUCTION In a previous report, we showed that three distinct stable phenotypes which are resistant to methotrexate IMtx~ can be isolated from Chinese hamster ovary {CHO~ cells {1). Class I cells, which were selected in a single-step, are resistant because of an apparent structural alteration in dihydrofolate reductase. Class II cells, which were also obtained in a single-step selection, were altered in the ability to transport the drug. Both Class I and II ceils contained unaltered levels of enzyme activity. Class I I I cells, selected in a second-step from Class I cells, contained increased activity of the structurally altered reductase. In order to fully exploit these ntx-resistant markers, it is important to characterize them genetically. Thus their applicability in other cell systems depends on knowledge of mutation rates and their phenotypic behavior in respect to dominance or recessiveness in somatic cell hybrids. This latter information is also important in at-
tempts to map the markers and for studies on the function of wild-type and mutant enzymes in the same cell. In this communication, we show that in somatic cell hybrids between Mtx-sensitive and resistant cells, the Class I and Class I I I markers are expressed as codominant traits and the Class II marker as a recessive characteristic. In addition, estimates are made of the spontaneous mutation rate and the distribution of Class I and Class II resistant phenotypes in a single-step selection. MATERIALS AND METHODS
Cell lines, media and culture conditions. The parental CHO cell line used was obtained from Dr. W. C. Dewey and is auxotrophic for proline (Pro-); Pro-3 and Pro-4 refer to subclones of this line obtained in our laboratory. The Mtx-resistant cell lines were those previously described (1) and are designated in the following manner: Pro-3 n t x RI refers to a Mtx-resistant cell, selected from Pro-3, with Class I resistant properties. Subclones of this line are designated as Pro-3 MtxR~ 1-x. ~Presented in the formal symposium on Somatic Cell Similarly, isolate~ of Class II and Class I I I resisGenetics at the 27th Annual Meeting of the Tissue Cul- tant cells are designated, for example, as Pro-3 ture Association, Philadelphia, Pennsylvania, June Mtx RI1 and Pro-3 n t x RIII, respectively. Cell lines 7-10, 1976. 2Present address: Department of Bacteriology and that are both Mtx-resistant and ouabain (Oua)Immunology, University of Western Ontario, London, resistant are referred to, for example, as Pro-3 Ontario N6A 5C1. Mtxm OuaR1. 749
75O
FLINTOFF, SPINDLER, AND SIMINOVITCH
cell lines to the drugs was expressed by the Dlo value, the drug concentration that reduces cell survival to 10%. These values were determined either by complete dose response curves or by plating 50 and 500 ceils at various drug concentrations in 24-well Linbro trays as previously described (1). After 8 days incubation at 37°C, colonies were stained with 1% methylene blue in 70% isopropanol. Colonies of > 50 cells were used in the determination of relative plating efficiencies. Relative resistance is expressed as the ratio of the D,o value for the resistant line to the D,o value for the sensitive line. Dihydrofolate reductase assay. The preparation of cell extracts and the spectrophotometric assay for dihydrololate reductase were as previously described ~1). Protein concentrations were determined by the method of Lowry (6), using BSA as a standard. Enzyme activity is expressed as nmol dihydrofolate reduced per mg per min at 37°C. Karyotype analysis. Ceils growing exponentially in suspension culture were incubated with 0.25 t~g per ml colcemid {Grand Island Biological Co.) for 1.5 to 2 hr at 37°C. The ceils were washed with hypotonic and fixing solutions. Slides were prepared, stained with Giemsa, and Selection of Oua-resistant and Pro* cells for hy- 20 representative chromosome spreads were bridization. Mtx-resistant cells were selected for counted. Determination of mutation rate. The spontaneresistance to the codominant Oua marker 14) in the following manner: Cells were plated at 10' per ous forward mutation rate was determined by 100-mm tissue culture dish in a-medium contain- measuring the rate of mutant accumulation in ing 3 mM Oua {Sigma Chemical Co. ) and supple- nonselective medium. For this purpose, Pro-3 mented with 10% FBS. After 8 days at 37°C, sur- cells were cloned and one clone was grown up for viving colonies were picked, grown up, cloned, 30 generations of undiluted growth. Subsetested for resistance and maintained under non- quently, 1 to 5 splits were made. At various times, cells were plated in selective medium ta-special selective conditions. For the particular hybridization cross which medium supplemented with 40 gg per ml Lwas to be used, we required wild-type cells re- proline and 10% DFBS} containing 1).
I0 -~ E_
2] 0-
10 ' F
I 10 -~
I
I LIIIII]
L t lllllll 10 -s
I 10 ~
I IIIIH]
I 10 -~
Methotrexate!Mi FI6. 1. Dose-response curves for hybrids formed between ser~itive cells and Class I cells. Exponentially growing cells were plated in selectivemedium with various concentrations of Mtx. After 8 days incubation at 37°C, plates were stained and resistant coloniescounted. O--O, Pro*5; • - - • , Pro-3 MtxRI OuaR3-6; A--A, HybR[2; A--A, HybR[3. The dose-response curve for a hybrid, HybRU2, formed from the fusion of Pro*5 and the Class II resistant cell Pro-3 Mtx nn OuaR5-3 is presented in Fig. 2. It is clear that the survival of this hybrid in Mtx was similar to that of the sensitive parent or of the control hybrid, Hyb2. The D,o value for HybRn2 was 2.1 X 10-s M for the Pro*5 sensitive parent, and 5.2 X 10-s for the control hybrid Hyb2, 1.5 X 10-s M for the Pro*5 sensitive parent, and 5.2 X 10-' n for the resistant parent Pro-3 Mtx RII OuaR5-3. In this particular hybrid, the Class II Mtx-resistant marker behaves as a recessive trait. The survival curves for two hybrids, HybRm2 and HybRIn3, formed between Pro*5 and Pro-3 MtxRm OuaR3-5, the resistant parent, indicate that their survival in the drug was similar to that of the resistant parent (Fig. 3}. The D,o values for HybRIII2 and HybRIII3 were 3.5 X 10-6 M, and 1.2 X 10-6 M, respectively, compared to 6.2 X 10-6 n for the resistant parent. These data suggest that Class I I I Mtx-resistance is a codominant trait in hybrids. Characterization of the dihydrofolate reductase in hybrids. Since the Class I pseudodiploid cells
have an apparent structural alteration in the dihydrofolate reductase and Class I I I ceils have an increased activity of this particular enzyme tl }, it was of interest next to determine the nature of the enzyme and its activity level in the various hybrids. Cell extracts were prepared and the enzyme was assayed in the presence of various concentrations of Mtx. Fig. 4 shows the enzyme titration curves for two hybrids, HybRtl and Hybm2, formed from the fusion of a sensitive cell and a Class I resistant ceil. The enzyme titration curve in the hybrids lay between the curves for the enzymes from the parental sensitive and resistant cells. An equal mixture of the parental sensitive and resistant reductases titrated in a similar manner. This suggests that in the hybrids both sensitive and resistant enzymes are being expressed, which is consistent with a codominant expression as indicated by the survival curves. The enzyme titration curve for a hybrid, HybRm2, formed between a sensitive and a Class I I I resistant cell is shown in Fig. 5. The enzyme expressed in the MtxRm hybrid was of the resistant type. However, the level of the enzyme activity was intermediate between the activities of the enzymes from the parental lines. The nybRIII2 cell had a relative activity of 3.9 compared to 1.0 for the sensitive parental line Pro÷5, and 7.7 for
10 °
I0-'
T, 10 -~ E_
2, 10-
10-~]~ 10 9
I
L-J-iLIIII
I L AIIIHI
~I~LIH] 10-~
10 ~
I LI 10 -~
Methotrexate[M]
FIG. 2. Dose-response curves for a hybrid formed between two sensitive cells and between a sensitive and a Class H resistant cell. Plating conditions were as in Fig.
1. O--O, Pro'5; A--A, Pro-3 MtxRII OuaR5-3; • - - • , HybRU2; A--A, Hyb2 (control hybridL
754
D L~
FLINTOFF, SPINDLER, AND SIMINOVITCH
!! / 10-9
10-s
10 -~
10 -~
10 -~
10-'
MethotrexateIM]
FxG. 3. Dose-response curves for hybrids/ormed between sensitive cells and Class I H resistant cells. Plating conditions were as in Fig. 1. O - - O , Pro*5; • - - • , Pro-3 Mtx RIII Oua R 3-5; A - - A , HybRIIl2; A--A, HybRlII3.
the resistant parental line Pro-3MtxRlII OuaR3-5. T h e intermediate level of activity in the Class I I I hybrid is consistent with a c o d o m i n a n t expression as indicated by the dose-response curves. Fig. 5 also contains the enzyme titration curves for a hybrid, HybnU2, f o r m e d f r o m a sensitive line a n d a Class I I resistant line, a n d for a control hybrid, Hyb2. T h e enzyme found in these particular hybrids was of the sensitive type. T h i s was expected for the control hybrid and for the Class I I hybrid since the parental Class II resistant line does not contain any a p p a r e n t structural alterations in the reductase enzyme ~1 ). It was i m p o r t a n t to determine if other independ e n t hybrids b e h a v e d similarly in respect to the d o m i n a n c e or recessiveness of the M t x - r e s i s t a n t markers. T h u s the resistance of several different hybrid cell lines a n d the response of the reduetase enzyme in crude cell extracts of these hybrid cells to inhibition by M t x was d e t e r m i n e d . T a b l e 3 summarizes the resistance level a n d the nature of
TABLE 3 CELLULAR RESISTANCE AND ENZYME DATA FOR PSEUDODIPLOID LINES USED IN HYBRIDIZATION Relative l),oa qMI
A. Sensitive (Wild Type) Pro-3 tsH1 Pro+4 Pro'5 Pro'2 Pro-30ua R 1-2 Pro-40ua R 2-4
21 21 21 21 21 21
1.2X l0 -s 2X 10-s 1.5)< 10-s 1.8)
GENETIC CHARACTERIZATION OF METHOTREXATE-RESISTANT CHINESE HAMSTER OVARY CELLS 1 WAYNE F. FLINTOFF, 2SUSAN M. SPINDLER, ANDLOUIS SIMINOVITCH
Department of Medical Genetics, University of Toronto, Toronto, Ontario, Canada M5S 1A8
SUMMARY In a previous report, we described the selection and partial characterization of three methotrexate ~Mtx)-resistant Chinese hamster ovary cells ICHO) ~1~. Class I cells contained an apparent structural alteration in dihydrofolate reductase. Class II cells had an alteration affecting the permeability of the drug. Class I I I cells, selected from Class I cells, had an increased activity of the altered enzyme. In the work described here, it has been shown that the spontaneous mutation rate to Class I resistance is in the order of 2 )< 10 -9 mutations per locus per generation and that in single-step mutagenized selections the number of resistant colonies of Classes I and II are about equal. Class I and Class I I ! resistance is expressed codominantly in somatic cell hybrids, whereas the Class II resistant marker is a recessive trait.
Key words: somatic cell genetics; methotrexate-resistance; Chinese hamster ovary cells; cell-cell hybridization. INTRODUCTION In a previous report, we showed that three distinct stable phenotypes which are resistant to methotrexate IMtx~ can be isolated from Chinese hamster ovary {CHO~ cells {1). Class I cells, which were selected in a single-step, are resistant because of an apparent structural alteration in dihydrofolate reductase. Class II cells, which were also obtained in a single-step selection, were altered in the ability to transport the drug. Both Class I and II ceils contained unaltered levels of enzyme activity. Class I I I cells, selected in a second-step from Class I cells, contained increased activity of the structurally altered reductase. In order to fully exploit these ntx-resistant markers, it is important to characterize them genetically. Thus their applicability in other cell systems depends on knowledge of mutation rates and their phenotypic behavior in respect to dominance or recessiveness in somatic cell hybrids. This latter information is also important in at-
tempts to map the markers and for studies on the function of wild-type and mutant enzymes in the same cell. In this communication, we show that in somatic cell hybrids between Mtx-sensitive and resistant cells, the Class I and Class I I I markers are expressed as codominant traits and the Class II marker as a recessive characteristic. In addition, estimates are made of the spontaneous mutation rate and the distribution of Class I and Class II resistant phenotypes in a single-step selection. MATERIALS AND METHODS
Cell lines, media and culture conditions. The parental CHO cell line used was obtained from Dr. W. C. Dewey and is auxotrophic for proline (Pro-); Pro-3 and Pro-4 refer to subclones of this line obtained in our laboratory. The Mtx-resistant cell lines were those previously described (1) and are designated in the following manner: Pro-3 n t x RI refers to a Mtx-resistant cell, selected from Pro-3, with Class I resistant properties. Subclones of this line are designated as Pro-3 MtxR~ 1-x. ~Presented in the formal symposium on Somatic Cell Similarly, isolate~ of Class II and Class I I I resisGenetics at the 27th Annual Meeting of the Tissue Cul- tant cells are designated, for example, as Pro-3 ture Association, Philadelphia, Pennsylvania, June Mtx RI1 and Pro-3 n t x RIII, respectively. Cell lines 7-10, 1976. 2Present address: Department of Bacteriology and that are both Mtx-resistant and ouabain (Oua)Immunology, University of Western Ontario, London, resistant are referred to, for example, as Pro-3 Ontario N6A 5C1. Mtxm OuaR1. 749
75O
FLINTOFF, SPINDLER, AND SIMINOVITCH
cell lines to the drugs was expressed by the Dlo value, the drug concentration that reduces cell survival to 10%. These values were determined either by complete dose response curves or by plating 50 and 500 ceils at various drug concentrations in 24-well Linbro trays as previously described (1). After 8 days incubation at 37°C, colonies were stained with 1% methylene blue in 70% isopropanol. Colonies of > 50 cells were used in the determination of relative plating efficiencies. Relative resistance is expressed as the ratio of the D,o value for the resistant line to the D,o value for the sensitive line. Dihydrofolate reductase assay. The preparation of cell extracts and the spectrophotometric assay for dihydrololate reductase were as previously described ~1). Protein concentrations were determined by the method of Lowry (6), using BSA as a standard. Enzyme activity is expressed as nmol dihydrofolate reduced per mg per min at 37°C. Karyotype analysis. Ceils growing exponentially in suspension culture were incubated with 0.25 t~g per ml colcemid {Grand Island Biological Co.) for 1.5 to 2 hr at 37°C. The ceils were washed with hypotonic and fixing solutions. Slides were prepared, stained with Giemsa, and Selection of Oua-resistant and Pro* cells for hy- 20 representative chromosome spreads were bridization. Mtx-resistant cells were selected for counted. Determination of mutation rate. The spontaneresistance to the codominant Oua marker 14) in the following manner: Cells were plated at 10' per ous forward mutation rate was determined by 100-mm tissue culture dish in a-medium contain- measuring the rate of mutant accumulation in ing 3 mM Oua {Sigma Chemical Co. ) and supple- nonselective medium. For this purpose, Pro-3 mented with 10% FBS. After 8 days at 37°C, sur- cells were cloned and one clone was grown up for viving colonies were picked, grown up, cloned, 30 generations of undiluted growth. Subsetested for resistance and maintained under non- quently, 1 to 5 splits were made. At various times, cells were plated in selective medium ta-special selective conditions. For the particular hybridization cross which medium supplemented with 40 gg per ml Lwas to be used, we required wild-type cells re- proline and 10% DFBS} containing 1).
I0 -~ E_
2] 0-
10 ' F
I 10 -~
I
I LIIIII]
L t lllllll 10 -s
I 10 ~
I IIIIH]
I 10 -~
Methotrexate!Mi FI6. 1. Dose-response curves for hybrids formed between ser~itive cells and Class I cells. Exponentially growing cells were plated in selectivemedium with various concentrations of Mtx. After 8 days incubation at 37°C, plates were stained and resistant coloniescounted. O--O, Pro*5; • - - • , Pro-3 MtxRI OuaR3-6; A--A, HybR[2; A--A, HybR[3. The dose-response curve for a hybrid, HybRU2, formed from the fusion of Pro*5 and the Class II resistant cell Pro-3 Mtx nn OuaR5-3 is presented in Fig. 2. It is clear that the survival of this hybrid in Mtx was similar to that of the sensitive parent or of the control hybrid, Hyb2. The D,o value for HybRn2 was 2.1 X 10-s M for the Pro*5 sensitive parent, and 5.2 X 10-s for the control hybrid Hyb2, 1.5 X 10-s M for the Pro*5 sensitive parent, and 5.2 X 10-' n for the resistant parent Pro-3 Mtx RII OuaR5-3. In this particular hybrid, the Class II Mtx-resistant marker behaves as a recessive trait. The survival curves for two hybrids, HybRm2 and HybRIn3, formed between Pro*5 and Pro-3 MtxRm OuaR3-5, the resistant parent, indicate that their survival in the drug was similar to that of the resistant parent (Fig. 3}. The D,o values for HybRIII2 and HybRIII3 were 3.5 X 10-6 M, and 1.2 X 10-6 M, respectively, compared to 6.2 X 10-6 n for the resistant parent. These data suggest that Class I I I Mtx-resistance is a codominant trait in hybrids. Characterization of the dihydrofolate reductase in hybrids. Since the Class I pseudodiploid cells
have an apparent structural alteration in the dihydrofolate reductase and Class I I I ceils have an increased activity of this particular enzyme tl }, it was of interest next to determine the nature of the enzyme and its activity level in the various hybrids. Cell extracts were prepared and the enzyme was assayed in the presence of various concentrations of Mtx. Fig. 4 shows the enzyme titration curves for two hybrids, HybRtl and Hybm2, formed from the fusion of a sensitive cell and a Class I resistant ceil. The enzyme titration curve in the hybrids lay between the curves for the enzymes from the parental sensitive and resistant cells. An equal mixture of the parental sensitive and resistant reductases titrated in a similar manner. This suggests that in the hybrids both sensitive and resistant enzymes are being expressed, which is consistent with a codominant expression as indicated by the survival curves. The enzyme titration curve for a hybrid, HybRm2, formed between a sensitive and a Class I I I resistant cell is shown in Fig. 5. The enzyme expressed in the MtxRm hybrid was of the resistant type. However, the level of the enzyme activity was intermediate between the activities of the enzymes from the parental lines. The nybRIII2 cell had a relative activity of 3.9 compared to 1.0 for the sensitive parental line Pro÷5, and 7.7 for
10 °
I0-'
T, 10 -~ E_
2, 10-
10-~]~ 10 9
I
L-J-iLIIII
I L AIIIHI
~I~LIH] 10-~
10 ~
I LI 10 -~
Methotrexate[M]
FIG. 2. Dose-response curves for a hybrid formed between two sensitive cells and between a sensitive and a Class H resistant cell. Plating conditions were as in Fig.
1. O--O, Pro'5; A--A, Pro-3 MtxRII OuaR5-3; • - - • , HybRU2; A--A, Hyb2 (control hybridL
754
D L~
FLINTOFF, SPINDLER, AND SIMINOVITCH
!! / 10-9
10-s
10 -~
10 -~
10 -~
10-'
MethotrexateIM]
FxG. 3. Dose-response curves for hybrids/ormed between sensitive cells and Class I H resistant cells. Plating conditions were as in Fig. 1. O - - O , Pro*5; • - - • , Pro-3 Mtx RIII Oua R 3-5; A - - A , HybRIIl2; A--A, HybRlII3.
the resistant parental line Pro-3MtxRlII OuaR3-5. T h e intermediate level of activity in the Class I I I hybrid is consistent with a c o d o m i n a n t expression as indicated by the dose-response curves. Fig. 5 also contains the enzyme titration curves for a hybrid, HybnU2, f o r m e d f r o m a sensitive line a n d a Class I I resistant line, a n d for a control hybrid, Hyb2. T h e enzyme found in these particular hybrids was of the sensitive type. T h i s was expected for the control hybrid and for the Class I I hybrid since the parental Class II resistant line does not contain any a p p a r e n t structural alterations in the reductase enzyme ~1 ). It was i m p o r t a n t to determine if other independ e n t hybrids b e h a v e d similarly in respect to the d o m i n a n c e or recessiveness of the M t x - r e s i s t a n t markers. T h u s the resistance of several different hybrid cell lines a n d the response of the reduetase enzyme in crude cell extracts of these hybrid cells to inhibition by M t x was d e t e r m i n e d . T a b l e 3 summarizes the resistance level a n d the nature of
TABLE 3 CELLULAR RESISTANCE AND ENZYME DATA FOR PSEUDODIPLOID LINES USED IN HYBRIDIZATION Relative l),oa qMI
A. Sensitive (Wild Type) Pro-3 tsH1 Pro+4 Pro'5 Pro'2 Pro-30ua R 1-2 Pro-40ua R 2-4
21 21 21 21 21 21
1.2X l0 -s 2X 10-s 1.5)< 10-s 1.8)
