a-FETOPROTEIN AND ALBUMIN SYNTHESIS DURING THE CELL CYCLE Yutaka Tsukada and Hidematsu Hirai Department of Biochemistry Hokkaido University School of Medicine Sapporo, Japan
A n ascites hepatoma of the rat AH66 cell line was cultivated in vitro and the synthesis of a-fetoprotein ( A F P ) and serum albumin 1-3 was observed during the cell cycle. MATERIAL A N D METHODS Culture of Cell Lines Ascites hepatoma, AH66 cell line and its clonal cells were cultivated as previously described.' The cells were cultivated and maintained in culture tubes at 37" C. The culture was started with 5 X 10' cells in a plate containing 1 ml Eagles MEM (Nissui Seiyaku Co., Ltd.) supplemented with 20% heatinactivated calf serum. Clones of A H 6 6 cell line were isolated in soft agar method.& T h e C-4 clone and A-1 clone are the clones used in this experiment. Purification and Deterrniriutiori of A F P and Albutnin Rat A F P was purified by the method of Nishi and Watabe.;.'; Rat albumin was purified by the method of Schreiber.; These purified proteins were used for immunization of rabbit or horse. T h e amount of A F P and albumin in the supernatant of culture medium was determined by the Mancini's radial immunodiffusion technique.R In order to detect A F P at a lower concentration than 5-10 pgIml, which is the sensitivity limit of the gel-precipitation techniques, we used radioimmunoassay established by Nishi et al.!' based on the sandwich technique. The principle of the method was described previously by Ceska et al.lo for the determination of IgE. Irnrnurioarrtorcidiography
T h e synthesis of A F P and albumin by cell suspension was also determined by assessing the incorporation of radioactive amino acid into the proteins during incubation iti vitro according to the method of Hochwald.ll Each culture was incubated with 1 pCi of [lC]leucine for 48 hours at 37" C. The cell-free culture fluid dialyzed against 0.015 M phosphate buffer ( p H 7.4) for 72 hours was lyophilized and used for the analysis. The precipitation lines were developed for 24 hours at room temperature. The plates were thoroughly washed with
37
38
Annals New York Academy of Sciences
phosphate-buffered saline and dried. Autoradiography was made at room temperature by touching the film (Fuji, R-150) to the dried plate. The exposure time ranged from 2 to 3 weeks. Synchronous Culture
Synchronized cells at the onset of S phase were obtained by the double treatment of 3 mM thymidine.12 When the excess thymidine was removed and replenished with new culture fluid the cells started to proliferate synchronously from S phase to G2, M, and G1 phases. We incubated 1 x 10" synchronized cells in 1 ml culture medium with 1 pCi of [3H]thymidine for 30 minutes at 37" C. After the incubation the cells were thoroughly washed. The radioactivity was counted in a liquid scintillation counter. From the data of growth curve, [3H]thymidine uptake and mitotic index, the cell cycle ( G l , S, G2, M ) was c a l c ~ l a t e d . ~ ~ AFP and albumin secreted into the culture media during the synchronized cultivation were measured as previously described except for 1 hour incubation time. Immunofluorescent Staining
Double staining of cells with the fluorescent antibodies to AFP and albumin was performed basically according to the method of Coons et al." The tumor cells, fixed in acetone for 10 minutes, were treated with the horse anti-AFP serum for 30 minutes at 3 7 ° C and then with a fluorescein isothiocyanate (FITC) conjugated yG fraction isolated from rabbit antiserum against horse yG. After washing the slide, the tumor cells were stained with tetramethyl-rhodamine isothiocyanate (RITC) conjugated fl fraction isolated from the rabbit antiserum against rat albumin for 60 minutes at 37" C. The specificity of the positive fluorescence was tested by "blocking" procedures. The slides were pretreated with the @ fractions of the same antisera and thus blocked slides were taken as controls. RESULTS Growth Curve of Clones
FIGURE 1 shows the growth curve of two clonal cell lines, AH66 C-4 and A-I. The cell multiplication rate of the A-1 clone is much slower than that of the C-4 clone.
A F P and Albumin Synthesis These clonal cells synthesize AFP and albumin. Labeling of an AFP or albumin line indicate the synthesis of these proteins in vitro (FIGURES 2, 3). Many precipitin lines formed with antiserum to rat serum proteins were carved by the carrier proteins.
T s u k a d a & Hirai: AFP a n d Albumin Synthesis
39
G.4
50.
FIGURE1. Cell growth curve. The original inoculum was 5x 10' cells/ml. Circles show the average number of duplicate cultures.
A-1
0
24
48
72
96
120 Hours
TABLE 1 shows the amount of AFP and albumin synthesized by the C-4 and A-1 clones. For example, from FIGURE1 the increase of number of cells of C-4 in the logarithmic phase (between 24 and 72 hours) was 3.9 X 1Oj and during this phase 5.4 pg AFP was synthesized. T h e amount of AFP synthesized per cell per hour was calculated as 5.4 pg/3.9 x 102/48 hours = 2.9 X lo-' p g . The amount of AFP synthesized by the C-4 clone per cell per hour was approximately 10 times higher than albumin synthesis by the same clone. On
FIGURE2. Serum protein synthesis iri virro (AH66). Immunoelectrophoresis (A) and autoradiography ( B ) of original AH66 cell line. Single precipitin line in a,region indicates AFP. aSm Rabbit anti-rat serum; aaf . Horse anti-rat AFP serum; details are shown in the text.
Annals New York Academy of Sciences
40 AFP
AND
TABLEI ALBUMINSYNTHESIZED BY CULTURED TUMORCELLS
AFP * (pg)
AFP Fg/ cell/hour i
Albumin
*
(fig)
Albumin pg/ cell/our i
AH66 C-4
5.4
2.9 x
0.52
0.28 x lo-'
AH66 A-1
1.1
2.5 x lo-'
0.78
1.8x 10.'
* AFP and albumin synthesized during the
t The proteins synthesized
logarithmic phase (24-72 hours).
(pg) /increased number of cells/48 hours.
the other hand, the production of AFP by the A-1 clone was not so different from that of albumin.
AFP and Albumin Synthesis During the Cell Cycle As shown in TABLE 2, the mean generation times of the C-4 and A-1 clones were 20.5 and 50.7 hours, respectively. The G-1 phase of the A-1 clone (36 hours) was 4 times as long as that of the C-4 clone (9 hours), although the G2 phase of these clones had the same length ( 4 hours). T h e M phase of both clones was less than one hour, and the S phase of A-1 clone (10 hours) is longer than that of the C-4 clone. The C-4 clone synthesized AFP from late GI phase to the end of the S phase for 9 hours and albumin from the midphase of S to the beginning of the G2 phase for 4 hours (FIGURE 4). The A-1 clone synthesized AFP from late G1 phase t o late G 2 phase for 25 hours and albumin from midphase of S to late G2 phase for 9 hours (FIGURE
5).
The duration of albumin synthesis was found to be short compared with that of AFP synthesis.
FIGURE 3. Serum protein synthesis in virro (AH66). Immunoelectrophoresis ( A ) and autoradiography ( B ) of the AH66, A-1 clone. Clear single precipitin line indicates albumin. aSm Rabbit anti-rat serum; aAlb Rabbit anti-rat albumin serum. Details are shown in the text.
Tsukada & Hirai: AFP and Albumin Synthesis
41
FIGURE 4. AFP and albumin synthesis of AH66 C-4 clone (synchronous culture).
Double Staining with Fluorescent Antibodies to AFP and Albumin
Double staining of cells with the fluorescence antibody technique was carried out, using synchronized cells of C-4 and A-1 clone. The cells at the late S phase were stained with either F I T C giving green staining, o r R l T C giving red staining. Green stained cells and red stained cells were clearly separated and no intermediate cells were observed. These results indicated that AFP and albumin were produced by two different populations of cells.
DISCUSSION It has been reported previously that clonal cells isolated from ascites hepatoma, AH66 cell line, produced varying amount of AFP.'. The present paper shows that among these clones, the C-4 clone produces a TABLE2 CELLCYCLE* Clone
GI
S
GI1
C-4 clone
9
7
4
A-1 clone
36
10
4
M 0.5 0.5-1.0
Generation Time 20.5 50.7
~~
* Synchronized cells were obtained at the onset of S phase by double thymidine treatment. Details are shown in the text.
Annals New York Academy of Scienm
42
high level of AFP but low level of albumin and the A-1 clone produces a low level of AFP but a relatively high level of albumin (TABLE 1). The mean generation times of the C-4 and A-1 clones were 20.5 and 50.7 hours, respectively (TABLE2 ) . The difference of generation time depends chiefly upon the length of G1 phase, which is 4 times longer in the A-1 clone than in the C-4 clone. The phase during which AFP is synthesized extends from late G1 phase to the terminal stage of S phase (C-4 clone) or to the middle period of G2 phase (A-1 clone). In contrast, albumin is synthesized from the middle period of S phase to early G2 phase (C-4 clone) or to late 0 2 phase (A-1 clone). Thus, albumin
S+
IG- IG
sl 0-0
o-e
AFP Albumin
2 00
0
10
20
30
Lo
Hours
FIGURE5 . AFP and albumin synthesis of AH66 A-1 clone (synchronous culture).
synthesis occurs at a later stage of the cell cycle and lasts for a much shorter period of time than AFP synthesis. Several authors, using synchronous cultures, have demonstrated the synthesis of tyrosine aminotransferase l 5 and immunoglobulin l i are in relation to the cell cycle. These proteins are inducible chiefly from mid G2 phase to the end of S or early G2 phase. Our data agree well with theirs. Double staining of the cells at the late S phase with fluorescent-conjugated antibodies to AFP and albumin showed that AFP and albumin are probably synthesized by different cells. The change in the number of stained cells appeared to occur slightly earlier than the phase of AFP and albumin synthesis. The delayed increase in the protein levels in the medium may be due to the time necessary for secretion of these proteins. When using the immunofluorescence technique for AFP we usually find that only a few cells in sections or smears of hepatoma are stained. This fact
Tsukada & Hirai: AFP and Albumin Synthesis
43
indicates that hepatomas consist of a mixed cell population. Attention should be paid to the cell cycle, however, since some cells are assumed to be in a non-AFP-producing phase of the cycle during which they are not stained with the immunofluorescence. The present data indicate that even a clonal cell line consists of different populations such as AFP-producing, albumin-producing or nonproducing cells. The serum levels of AFP in hepatoma patients vary from very low to very high. This wide distribution may be explained by the fact that the hepatoma consists of a mixed cell population. If the nonproducing cells are the majority of the population the patients show very low level of AFP, and vice versa. Cells of the AH66 line are capable of synthesizing AFP and albumin; however, the synthesis of other serum proteins was not observed as shown in FIGURES 2 and 3. As previously reported by Tsukada,18 AH60C, 66G, and 49WH were demonstrated to be capable of synthesizing transferrin. Becker et al.19 also reported that rat hepatomas induced by 2FAA synthesized serum protein in vitro. It might be noteworthy to indicate that these hepatoma cell lines are malignant enough to kill rats in 1-2 weeks, but the normal function of liver cells are nevertheless still maintained in the cancer cells.
SUMMARY The synthesis of a-fetoprotein and albumin in two clones of AH66 hepatoma was studied. (1) Amounts of AFP and albumin synthesized by the C-4 clone were 2.9 and 0.28 X 10-7 pg per cell per hour, respectively. AFP and albumin amounts synthesized by the A-1 clone were 2.5 and 1.8 X I@' pg per cell per hour, respectively. (2) The cell cycles of the C-4 and A-1 clones were as follows: C-4 clone: mean generation time 20.5 hours: G1, 10 hours; S, 7 hours; G2, 4 hours; and M, 30 minutes. A-1 clone: mean generation time 50.7 hours: G1, 36 hours; S, 10 hours; G2, 4 hours; and M, 30-60 minutes. (3) AFP was found to be synthesized from late G1 phase to the end of the S phase for 9 hours in C-4 and 25 hours in A-1. The albumin production was from late S phase to the beginning of the G2 phase for 4 hours in C-4 and 9 hours in A-1, which were approximately half or one-third of the time spent on AFP production. ( 4 ) The double staining with fluorescent-conjugated antibodies to AFP and albumin demonstrated that AFP and albumin are probably synthesized by different cells. REFERENCES Y . , M. MIKUN & H. HIRAI. 1974. I n vifro cloning of a rat ascites 1. TSUKADA, hepatoma cell line, AH66, with special reference to alpha-fetoprotein synthesis. Int. J. Cancer 13: 196-202. 2. TSUKADA, Y . & H. HIRAI. 1973. / ) I vifro cloning of rat ascites hepatoma cell line, with reference to alpha-fetoprotein synthesis. Tumor Res. 8: 88-93. 3. HIRAI,H., S. NISHI,H. WATABE & Y. TSUKADA. 1973. Some chemical, experi-
44
Annals New York Academy of Sciences
mental, and clinical investigations of a-fetoprotein. Gann Monogr. Cancer Res. 14: 19-34. 4. GOTO, M. & H. SATO. 1965. Studies on tissue culture of ascites tumors. 111. Colony formation of Yoshida sarcoma cells in agar medium. Rep. Res. Inst. Tuberc. Leprosy Cancer, Tohoku Univ. 12: 319-323. 5. NISHI, S. 1970. Isolation and characterization of a human fetal a-globulin from the sera of fetuses and a hepatoma patient. Cancer Res. 30: 2507-2513. 6. WATABE,H. 1974. Purification and chemical characterization of a-fetoprotein from rat and mouse. Int. J. Cancer 13: 377-388. G., H. M. ROTERMUND, H. MAENO,K. WEIGAND & R. LESCH.1969. 7. SCHREIBER, The proportion of the incorporation of leucine into albumin to that into total protein in rat liver and hepatoma Morris 5123 TC. Eur. J . Biochem. 10: 355361. 8. MANCINI,G., A. 0. CARBONARA & J. F. HEREMANS.1965. Immunochemical quantitation of antigens by single radial immunodiffusion. Imrnunochemistry 2: 235-254. 9. NISHI, S., K. KOBAYASHI & H. HIRAI. 1974. Proceedings of the 33rd Annual Meeting of the Japanese Cancer Association. : 161. 1972. A new and simple radioimmunoassay method 10. CESKA,M. & V. LUNDKVIST. for the determination of IgE. Immunochemistry 9: 1021-1030. 11. HOCHWALD, G. M., G. J. THORBECKE & R. ASOFSKY.1961. Immunoradioautography of yglobulin synthesized in vitro. J . Exp. Med. 114: 459-467. G. & D. BOOTSMA.1966. Synchronization of mammalian cells in vifro 12. GALVAZI, by inhibition of DNA synthesis. 11. Population dynamics. Exp. Cell Res. 41: 438-452. 13. PAUL,J. 1970. Cell and tissue culture. : 30-32. 14. COONS,A.H.,H. J. CREECH,R. N. JONES& E. BERLINER.1942. The demonstration of pneumococcal antigen in tissues by the use of fluorescent antibody. J. Immunol. 45: 159-167. 15. MARTIN,D.JR., G. M. TOMKINS & D. GRANNER.1969. Synthesis and induction of tyrosine aminotransferase in synchronized hepatoma cells in culture. Biochemistry 45: 248-255. I., J. L. FAHEY & H. GRANGER.1967. Synthesis of immunoglobulins 16. FINEGOLD, by human cell lines in tissue culture. J. Imrnunol. 99: 839-852. M., Y. YAGI,G. E. MOORE& D. PRESSMAN.1969. Immunoglobulin 17. TAKAHASHI, production in synchronized cultures of human hematopoietic cell lines. I. Variation of cellular immunoglobulin level with the generation cycle. J. Immunol. 103: 834-843. Y., M. MIKUNI& H. HIRAI. 1972. Proceedings of the 31st Annual 18. TSUKADA, Meeting of the Japanese Cancer Association. : 81. 19. BECKER,F. F.,K. M. KLEIN& R. ASOFSKY.1972. Plasma protein synthesis by N-2-fluorenylacetamide-inducedprimary hepatocellular carcinomas and hepatic nodules. Cancer Res. 32: 914-920.
A n ascites hepatoma of the rat AH66 cell line was cultivated in vitro and the synthesis of a-fetoprotein ( A F P ) and serum albumin 1-3 was observed during the cell cycle. MATERIAL A N D METHODS Culture of Cell Lines Ascites hepatoma, AH66 cell line and its clonal cells were cultivated as previously described.' The cells were cultivated and maintained in culture tubes at 37" C. The culture was started with 5 X 10' cells in a plate containing 1 ml Eagles MEM (Nissui Seiyaku Co., Ltd.) supplemented with 20% heatinactivated calf serum. Clones of A H 6 6 cell line were isolated in soft agar method.& T h e C-4 clone and A-1 clone are the clones used in this experiment. Purification and Deterrniriutiori of A F P and Albutnin Rat A F P was purified by the method of Nishi and Watabe.;.'; Rat albumin was purified by the method of Schreiber.; These purified proteins were used for immunization of rabbit or horse. T h e amount of A F P and albumin in the supernatant of culture medium was determined by the Mancini's radial immunodiffusion technique.R In order to detect A F P at a lower concentration than 5-10 pgIml, which is the sensitivity limit of the gel-precipitation techniques, we used radioimmunoassay established by Nishi et al.!' based on the sandwich technique. The principle of the method was described previously by Ceska et al.lo for the determination of IgE. Irnrnurioarrtorcidiography
T h e synthesis of A F P and albumin by cell suspension was also determined by assessing the incorporation of radioactive amino acid into the proteins during incubation iti vitro according to the method of Hochwald.ll Each culture was incubated with 1 pCi of [lC]leucine for 48 hours at 37" C. The cell-free culture fluid dialyzed against 0.015 M phosphate buffer ( p H 7.4) for 72 hours was lyophilized and used for the analysis. The precipitation lines were developed for 24 hours at room temperature. The plates were thoroughly washed with
37
38
Annals New York Academy of Sciences
phosphate-buffered saline and dried. Autoradiography was made at room temperature by touching the film (Fuji, R-150) to the dried plate. The exposure time ranged from 2 to 3 weeks. Synchronous Culture
Synchronized cells at the onset of S phase were obtained by the double treatment of 3 mM thymidine.12 When the excess thymidine was removed and replenished with new culture fluid the cells started to proliferate synchronously from S phase to G2, M, and G1 phases. We incubated 1 x 10" synchronized cells in 1 ml culture medium with 1 pCi of [3H]thymidine for 30 minutes at 37" C. After the incubation the cells were thoroughly washed. The radioactivity was counted in a liquid scintillation counter. From the data of growth curve, [3H]thymidine uptake and mitotic index, the cell cycle ( G l , S, G2, M ) was c a l c ~ l a t e d . ~ ~ AFP and albumin secreted into the culture media during the synchronized cultivation were measured as previously described except for 1 hour incubation time. Immunofluorescent Staining
Double staining of cells with the fluorescent antibodies to AFP and albumin was performed basically according to the method of Coons et al." The tumor cells, fixed in acetone for 10 minutes, were treated with the horse anti-AFP serum for 30 minutes at 3 7 ° C and then with a fluorescein isothiocyanate (FITC) conjugated yG fraction isolated from rabbit antiserum against horse yG. After washing the slide, the tumor cells were stained with tetramethyl-rhodamine isothiocyanate (RITC) conjugated fl fraction isolated from the rabbit antiserum against rat albumin for 60 minutes at 37" C. The specificity of the positive fluorescence was tested by "blocking" procedures. The slides were pretreated with the @ fractions of the same antisera and thus blocked slides were taken as controls. RESULTS Growth Curve of Clones
FIGURE 1 shows the growth curve of two clonal cell lines, AH66 C-4 and A-I. The cell multiplication rate of the A-1 clone is much slower than that of the C-4 clone.
A F P and Albumin Synthesis These clonal cells synthesize AFP and albumin. Labeling of an AFP or albumin line indicate the synthesis of these proteins in vitro (FIGURES 2, 3). Many precipitin lines formed with antiserum to rat serum proteins were carved by the carrier proteins.
T s u k a d a & Hirai: AFP a n d Albumin Synthesis
39
G.4
50.
FIGURE1. Cell growth curve. The original inoculum was 5x 10' cells/ml. Circles show the average number of duplicate cultures.
A-1
0
24
48
72
96
120 Hours
TABLE 1 shows the amount of AFP and albumin synthesized by the C-4 and A-1 clones. For example, from FIGURE1 the increase of number of cells of C-4 in the logarithmic phase (between 24 and 72 hours) was 3.9 X 1Oj and during this phase 5.4 pg AFP was synthesized. T h e amount of AFP synthesized per cell per hour was calculated as 5.4 pg/3.9 x 102/48 hours = 2.9 X lo-' p g . The amount of AFP synthesized by the C-4 clone per cell per hour was approximately 10 times higher than albumin synthesis by the same clone. On
FIGURE2. Serum protein synthesis iri virro (AH66). Immunoelectrophoresis (A) and autoradiography ( B ) of original AH66 cell line. Single precipitin line in a,region indicates AFP. aSm Rabbit anti-rat serum; aaf . Horse anti-rat AFP serum; details are shown in the text.
Annals New York Academy of Sciences
40 AFP
AND
TABLEI ALBUMINSYNTHESIZED BY CULTURED TUMORCELLS
AFP * (pg)
AFP Fg/ cell/hour i
Albumin
*
(fig)
Albumin pg/ cell/our i
AH66 C-4
5.4
2.9 x
0.52
0.28 x lo-'
AH66 A-1
1.1
2.5 x lo-'
0.78
1.8x 10.'
* AFP and albumin synthesized during the
t The proteins synthesized
logarithmic phase (24-72 hours).
(pg) /increased number of cells/48 hours.
the other hand, the production of AFP by the A-1 clone was not so different from that of albumin.
AFP and Albumin Synthesis During the Cell Cycle As shown in TABLE 2, the mean generation times of the C-4 and A-1 clones were 20.5 and 50.7 hours, respectively. The G-1 phase of the A-1 clone (36 hours) was 4 times as long as that of the C-4 clone (9 hours), although the G2 phase of these clones had the same length ( 4 hours). T h e M phase of both clones was less than one hour, and the S phase of A-1 clone (10 hours) is longer than that of the C-4 clone. The C-4 clone synthesized AFP from late GI phase to the end of the S phase for 9 hours and albumin from the midphase of S to the beginning of the G2 phase for 4 hours (FIGURE 4). The A-1 clone synthesized AFP from late G1 phase t o late G 2 phase for 25 hours and albumin from midphase of S to late G2 phase for 9 hours (FIGURE
5).
The duration of albumin synthesis was found to be short compared with that of AFP synthesis.
FIGURE 3. Serum protein synthesis in virro (AH66). Immunoelectrophoresis ( A ) and autoradiography ( B ) of the AH66, A-1 clone. Clear single precipitin line indicates albumin. aSm Rabbit anti-rat serum; aAlb Rabbit anti-rat albumin serum. Details are shown in the text.
Tsukada & Hirai: AFP and Albumin Synthesis
41
FIGURE 4. AFP and albumin synthesis of AH66 C-4 clone (synchronous culture).
Double Staining with Fluorescent Antibodies to AFP and Albumin
Double staining of cells with the fluorescence antibody technique was carried out, using synchronized cells of C-4 and A-1 clone. The cells at the late S phase were stained with either F I T C giving green staining, o r R l T C giving red staining. Green stained cells and red stained cells were clearly separated and no intermediate cells were observed. These results indicated that AFP and albumin were produced by two different populations of cells.
DISCUSSION It has been reported previously that clonal cells isolated from ascites hepatoma, AH66 cell line, produced varying amount of AFP.'. The present paper shows that among these clones, the C-4 clone produces a TABLE2 CELLCYCLE* Clone
GI
S
GI1
C-4 clone
9
7
4
A-1 clone
36
10
4
M 0.5 0.5-1.0
Generation Time 20.5 50.7
~~
* Synchronized cells were obtained at the onset of S phase by double thymidine treatment. Details are shown in the text.
Annals New York Academy of Scienm
42
high level of AFP but low level of albumin and the A-1 clone produces a low level of AFP but a relatively high level of albumin (TABLE 1). The mean generation times of the C-4 and A-1 clones were 20.5 and 50.7 hours, respectively (TABLE2 ) . The difference of generation time depends chiefly upon the length of G1 phase, which is 4 times longer in the A-1 clone than in the C-4 clone. The phase during which AFP is synthesized extends from late G1 phase to the terminal stage of S phase (C-4 clone) or to the middle period of G2 phase (A-1 clone). In contrast, albumin is synthesized from the middle period of S phase to early G2 phase (C-4 clone) or to late 0 2 phase (A-1 clone). Thus, albumin
S+
IG- IG
sl 0-0
o-e
AFP Albumin
2 00
0
10
20
30
Lo
Hours
FIGURE5 . AFP and albumin synthesis of AH66 A-1 clone (synchronous culture).
synthesis occurs at a later stage of the cell cycle and lasts for a much shorter period of time than AFP synthesis. Several authors, using synchronous cultures, have demonstrated the synthesis of tyrosine aminotransferase l 5 and immunoglobulin l i are in relation to the cell cycle. These proteins are inducible chiefly from mid G2 phase to the end of S or early G2 phase. Our data agree well with theirs. Double staining of the cells at the late S phase with fluorescent-conjugated antibodies to AFP and albumin showed that AFP and albumin are probably synthesized by different cells. The change in the number of stained cells appeared to occur slightly earlier than the phase of AFP and albumin synthesis. The delayed increase in the protein levels in the medium may be due to the time necessary for secretion of these proteins. When using the immunofluorescence technique for AFP we usually find that only a few cells in sections or smears of hepatoma are stained. This fact
Tsukada & Hirai: AFP and Albumin Synthesis
43
indicates that hepatomas consist of a mixed cell population. Attention should be paid to the cell cycle, however, since some cells are assumed to be in a non-AFP-producing phase of the cycle during which they are not stained with the immunofluorescence. The present data indicate that even a clonal cell line consists of different populations such as AFP-producing, albumin-producing or nonproducing cells. The serum levels of AFP in hepatoma patients vary from very low to very high. This wide distribution may be explained by the fact that the hepatoma consists of a mixed cell population. If the nonproducing cells are the majority of the population the patients show very low level of AFP, and vice versa. Cells of the AH66 line are capable of synthesizing AFP and albumin; however, the synthesis of other serum proteins was not observed as shown in FIGURES 2 and 3. As previously reported by Tsukada,18 AH60C, 66G, and 49WH were demonstrated to be capable of synthesizing transferrin. Becker et al.19 also reported that rat hepatomas induced by 2FAA synthesized serum protein in vitro. It might be noteworthy to indicate that these hepatoma cell lines are malignant enough to kill rats in 1-2 weeks, but the normal function of liver cells are nevertheless still maintained in the cancer cells.
SUMMARY The synthesis of a-fetoprotein and albumin in two clones of AH66 hepatoma was studied. (1) Amounts of AFP and albumin synthesized by the C-4 clone were 2.9 and 0.28 X 10-7 pg per cell per hour, respectively. AFP and albumin amounts synthesized by the A-1 clone were 2.5 and 1.8 X I@' pg per cell per hour, respectively. (2) The cell cycles of the C-4 and A-1 clones were as follows: C-4 clone: mean generation time 20.5 hours: G1, 10 hours; S, 7 hours; G2, 4 hours; and M, 30 minutes. A-1 clone: mean generation time 50.7 hours: G1, 36 hours; S, 10 hours; G2, 4 hours; and M, 30-60 minutes. (3) AFP was found to be synthesized from late G1 phase to the end of the S phase for 9 hours in C-4 and 25 hours in A-1. The albumin production was from late S phase to the beginning of the G2 phase for 4 hours in C-4 and 9 hours in A-1, which were approximately half or one-third of the time spent on AFP production. ( 4 ) The double staining with fluorescent-conjugated antibodies to AFP and albumin demonstrated that AFP and albumin are probably synthesized by different cells. REFERENCES Y . , M. MIKUN & H. HIRAI. 1974. I n vifro cloning of a rat ascites 1. TSUKADA, hepatoma cell line, AH66, with special reference to alpha-fetoprotein synthesis. Int. J. Cancer 13: 196-202. 2. TSUKADA, Y . & H. HIRAI. 1973. / ) I vifro cloning of rat ascites hepatoma cell line, with reference to alpha-fetoprotein synthesis. Tumor Res. 8: 88-93. 3. HIRAI,H., S. NISHI,H. WATABE & Y. TSUKADA. 1973. Some chemical, experi-
44
Annals New York Academy of Sciences
mental, and clinical investigations of a-fetoprotein. Gann Monogr. Cancer Res. 14: 19-34. 4. GOTO, M. & H. SATO. 1965. Studies on tissue culture of ascites tumors. 111. Colony formation of Yoshida sarcoma cells in agar medium. Rep. Res. Inst. Tuberc. Leprosy Cancer, Tohoku Univ. 12: 319-323. 5. NISHI, S. 1970. Isolation and characterization of a human fetal a-globulin from the sera of fetuses and a hepatoma patient. Cancer Res. 30: 2507-2513. 6. WATABE,H. 1974. Purification and chemical characterization of a-fetoprotein from rat and mouse. Int. J. Cancer 13: 377-388. G., H. M. ROTERMUND, H. MAENO,K. WEIGAND & R. LESCH.1969. 7. SCHREIBER, The proportion of the incorporation of leucine into albumin to that into total protein in rat liver and hepatoma Morris 5123 TC. Eur. J . Biochem. 10: 355361. 8. MANCINI,G., A. 0. CARBONARA & J. F. HEREMANS.1965. Immunochemical quantitation of antigens by single radial immunodiffusion. Imrnunochemistry 2: 235-254. 9. NISHI, S., K. KOBAYASHI & H. HIRAI. 1974. Proceedings of the 33rd Annual Meeting of the Japanese Cancer Association. : 161. 1972. A new and simple radioimmunoassay method 10. CESKA,M. & V. LUNDKVIST. for the determination of IgE. Immunochemistry 9: 1021-1030. 11. HOCHWALD, G. M., G. J. THORBECKE & R. ASOFSKY.1961. Immunoradioautography of yglobulin synthesized in vitro. J . Exp. Med. 114: 459-467. G. & D. BOOTSMA.1966. Synchronization of mammalian cells in vifro 12. GALVAZI, by inhibition of DNA synthesis. 11. Population dynamics. Exp. Cell Res. 41: 438-452. 13. PAUL,J. 1970. Cell and tissue culture. : 30-32. 14. COONS,A.H.,H. J. CREECH,R. N. JONES& E. BERLINER.1942. The demonstration of pneumococcal antigen in tissues by the use of fluorescent antibody. J. Immunol. 45: 159-167. 15. MARTIN,D.JR., G. M. TOMKINS & D. GRANNER.1969. Synthesis and induction of tyrosine aminotransferase in synchronized hepatoma cells in culture. Biochemistry 45: 248-255. I., J. L. FAHEY & H. GRANGER.1967. Synthesis of immunoglobulins 16. FINEGOLD, by human cell lines in tissue culture. J. Imrnunol. 99: 839-852. M., Y. YAGI,G. E. MOORE& D. PRESSMAN.1969. Immunoglobulin 17. TAKAHASHI, production in synchronized cultures of human hematopoietic cell lines. I. Variation of cellular immunoglobulin level with the generation cycle. J. Immunol. 103: 834-843. Y., M. MIKUNI& H. HIRAI. 1972. Proceedings of the 31st Annual 18. TSUKADA, Meeting of the Japanese Cancer Association. : 81. 19. BECKER,F. F.,K. M. KLEIN& R. ASOFSKY.1972. Plasma protein synthesis by N-2-fluorenylacetamide-inducedprimary hepatocellular carcinomas and hepatic nodules. Cancer Res. 32: 914-920.
