Int. Archs Allergy appl. Immun. 60: 346-35? (1979)
Murine Lymphoma Alkaline Phosphatase: A Cell Membrane Carcinofetal Enzyme R. A. Floyd, V. G. Tisdale and J. R. Lumb' Biology Department. Atlanta University. Atlanta. Ga.
Introduction Alkaline phosphatase (APase) has been associated with murine thymic lymphomas [3. 4, 16. 23, 24], Lagerlof and Kaplan [11] and Wilson el al. [26] have correlated the appearance of APase with malignant trans formation and have ruled out the possibili ties that APase is associated with cellular proliferation or with the leukemia virus in the absence of malignant transformation. Contradictory results were obtained using a 1 Research Career Development Awardee (CA 0067).
less sensitive histochemical procedure [211. APase activity has also been associated with lymphoblastic fetal thymus cells up to 16 days gestation, but it is not found on thymic lymphocytes in a normal adult [11]. This APase activity of murine lymphomas has been shown to be similar to the APase that occurs at a low level in adult spleen [ 131. The observation of a placental-type APase in a variety of human tumors [7, 25] has suggested that the comparison of this mu rine lymphoma APase be made with that of the murine placenta. Therefore, a compari son of the murine lymphoma, fetal thymus.
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/5/2018 12:11:13 AM
Abstract. Alkaline phosphatase (APase) has been shown to have a membrane-bound localization in the murine fetal thymus, in murine thymic lymphoma and in adult spleen. Since it was suggested from these previous experiments that the lymphoma APase might represent an embryonic function, a detailed biochemical comparison of the lymphoma APase with the fetal thymus, placenta, fetus and spleen APases was performed. The para meters investigated were pH optimum, activation, inhibition, heat inactivation, substrate ratios. Michaelis constant, and electrophoretic analysis in the presence and absence of neuraminidase with the substrates «-naphthyl phosphate, /(-glycerophosphate, and p-nitrophenyl phosphate. The results indicate that the lymphoma APase is very similar to the fetal thymus, placenta and spleen APases. Furthermore, these results lend support to the hypothesis that the APase activity which appears in thymic lymphoma might represent a derepressed embryonic function. Thus, the murine lymphoma APase may be termed a cell membrane carcinofetal enzyme.
Comparison of Alkaline Phosphatases
Materials and Methods Source of Tissues Inbred C57BI/Ka mice were originally ob tained from Dr. Henry S. Kaplan at Stanford Uni versity. They were maintained by brother-sister mating. Thymic lymphomas were induced in these mice by urethan treatment of C57LV as previous ly described [13]. The lymphomatous tissues were taken from thymus, spleen, and lymph nodes. These tissues were pooled after histological and histochemical examination confirmed that these tissues were leu kemic. The spleens were taken from adlut C57BI/Ka males. The fetal thymuses, whole fetuses, and pla centas were taken from 16-day pregnant C57B1/ Ka mice. Preparation of Extracts For most experiments crude extracts were pre pared by homogenizing the tissue in a glass homogenizer (Corning) in 0.01 M Tris at pH 7. For electrophoresis experiments, the homogenized ex tract was centrifuged at 75,000 p for 30 min. The precipitate was treated with l°/o Noniodet P-40 (NP40; Shell Oil, Inc.), incubated at 37 C for 15 min and centrifuged at 75,000 g for 30 min. The NP40 treatment was repeated twice on the precipitates and the three resulting supernatants were pooled. The pooled supernatants and the re maining precipitate were then treated with 30°/o
butanol and centrifuged at 75,000 g for 30 min. The resulting aqueous layer was dialyzed to re move butanol. The fetal thymus and lymphoma extracts were treated with 1,0°/o NP40 and incu bated at 37 C for 15 min. One of the fetal thy mus extracts contained 10°/o NP40 instead of P/o because it was further concentrated. The fetal thy mus extracts were labelled F, (l°/o NP40) and F2 (10°/o NP40). The substrates used were NP (0.052 m/W). GP (50 mM), and pNPP (3.38 mM). As buffers, bor ate. 2-amino-2-methyl-l,3-propanediol (ammediol), Tris, and diethanolamine (DEA) were used at varying concentrations. Spectrophotometric Assays for APase The enzyme was preincubated with buffer at 37 °C for 5 min except for experiments to deter mine EDTA inhibition in which a 20-min preincu bation was required [Floyd and Lamb, unpub lished observations]. The substrate was added re sulting in a 1-ml total reaction volume which was allowed to incubate for 30 min at 37 °C. After the 30-min incubation the reaction was stopped with 0.2 M EDTA in 0.5 N NaOH for measurement of p-nitrophenol at 400 nm [13], For reactions using the substrate GP, phosphate was assayed by the method of Lecocq and fnesi [12]. Flttoriinetric Assays for APase The fluorimetric assay was employed when us ing the substrate NP (17). The buffer (1 M DEA) was preincubated at 37 °C for 5 min in a 4-mI re action volume at pH 10. The extract and substrate NP were added and the formation of «-naphthol was recorded by fluorescence for 2-5 min at con stant temperature. For the experiments to test the level of inhibition with EDTA, the extract was preincubated for 20 min with the buffer [Floyd and Lump, unpublished observations] and the re action initiated by the addition of substrate. The optimum pH of the extracts was deter mined using the appropriate substrate with the pH of the buffer varying from 9 to 10.5. Heat inacti vation rate was determined by heating tubes pre pared for assay, including 5 mM magnesium chlo ride, for varying times at 55 °C prior to assay. The heat inactivation rate is described in the equa tion A = A„e'kt
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/5/2018 12:11:13 AM
placenta, whole fetus, and adult spleen has been made using the substrates «-naphthyl phosphate (NP), /f-glycerophosphate (GP), and p-nitrophenyl phosphate (pNPP). The parameters chosen for this characterization included pH optimum, activation by magne sium and nucleophilic buffers, inhibition by amino acids and ethylenediaminetetraacetic acid (EDTA), heat inactivation, substrate ratios, Michaelis constant (K.m), and electro phoretic analysis. Partial characterization of the fetal thymus APase activity has been re ported [ 14].
347
34N
Floyd/Tisdale/Lumh
Results Figures I and 2 show the pH optima for the tissues and substrates tested. These data show that spleen, lymphoma, placenta, and whole fetus extracts all have an APase ac tivity which has an optimum pH around 10 using the substrates pNPP and GP. Table I shows the magnesium activation for the tissues and substrates tested. All ex tracts show a similar level of activation by magnesium. With the exception of the pla centa there appears to be no significant dif ferences among the tissues. These data show much higher variability and. therefore, are
Big. 1. The pH optimum of APase activities of placenta ( • ) . spleen (/\), lympoma (O), and whole fetus ( ) using the substrate pNPP.
pH
Fig. 2. The pH optimum of APase activities of placenta ( • ) , spleen (¿\), lymphoma (O), and whole fetus ( ) using the substrate GP. Table I. Level of APase activation by magnesium Tissue
Lymphoma Spleen Placenta Whole fetus Fetal thymus
Percent activation />NPP
GP
NP
30.2 ± 2.4 27.3 ± 6.3 45.5 ± 9.1 25.8 ± 4.3 no data
39.3 ± 0.6 39.3 ± 0.6 66.8 ± 11.1 41.3 ± 6.8 no data
46 ± 0.21 46 ± 0.16 46 ± 0.31 no data 47 ± 0.17
Mean ± SD were determined from at least three experiments.
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/5/2018 12:11:13 AM
where A,, is the unheated activity, A is the activity remaining after time t and k is the heat inactiva tion rate constant. The rate was determined by Least Square analysis of the data expressed as the natural log of activity vs. time at 55 C. Kinetics: The enzyme reaction was carried out as indicated above using varying concentrations of NP as substrate. A Lineweaver-Bruke double-reci procal plot of the results was used to determine the K|„. Substrate ratio was calculated as the activ ity of the enzyme extract with either NP or GP divided by the activtiy with pNPP. Activation and inhibition: Experiments with magnesium chloride (5 mM), EDTA (0.1 mM), L-phenylalanine (10 mM). L-homoarginine 10 mM), Tris buffer (0.5 M), ammediol buffer (0.5 M), DBA (1.0 M). and borate (0.5 M) were performed along with the proper controls to determine the percent activation or inhibition of APase activity. Statistical analysis: All data arc reported as mean and standard deviation of at least three ex periments. A Students t test was performed to de termine significance using the facilities of the At lanta University Computing Center. Electrophoresis was performed according to Fislwtan [5]. APase activity was localized using two different substrates. For /;NPP the enzyme lo calization was indicated by the yellow p-nitrophcnyl which diffused quickly. An azo dye was used as a coupling agent for the «-naphthol prod uct with NP as substrate (15|.
Comparison of Alkaline Phosphatases
349
not considered significant. The placenta shows a lower level of endogenous magne sium and a much higher APasc specific ac tivity [Lumb and Yahya, unpublished]. It is suggested this may account for this high variability. The «-naphthyl phosphatase (NPase) ac tivity of lymphoma, spleen, placenta and fe tal thymus extracts was increased by the presence of the nucleophilic buffers DEA and ammediol. while NPase activity was in hibited by the presence of the neutral buff er, borate (table II). Table III indicates the level of inhibition by EDTA using the three substrates with the five tissue extracts. All five extracts show almost complete inhibition with all three substrates. The level of APase inhibition by the amino acids L-phenylalanine and L-homoarginine is shown in tables IV and V. L-Homoarginine has a greater inhibitory ef fect on APase activity than does L-phenylalanine.
Table II. Effect of butTer concentration on APase activity Ammediol1* DEA3
Lymphoma Spleen Placenta Fetal thymus
1.24 ± 1.91 ± 1.65 ± 1.63 ±
0.29 0.14 0.26 0.64
2.88 ± 0.32 2.81 ±0.11 2.78 ±0.18 2.87 ± 0.43
Tissue
Percent inhibition />NPP
GP
85.7 ± 7.6 Lymphoma 96.5 ± 0.7 Spleen 95.0 ±0.1 Placenta 96.0 ±0.1 Whole fetus Fetal thymus no data
NP
90.8 ± 6.9 84.7 ± 1.5 94.5 ± 0.7 85.5 ± 0.7 no data
98.57 ± 0.78 100 98.47 ± 0.86 no data 100
Mean±SD were determined from at least three experiments.
Table IV. Level of APase inhibition by amin acids Tissue
Percent inhibition L-phenylalanine L-homoarginine
Lymphoma Spleen Placenta Fetal thymus
45.66 39.82 48.20 49.36
± ± ± ±
4.63 7.08 5.01 6.93
100 100 100 100
At least three experiments were performed using the substrate NP using 0.05 M DEA buffer at pH 10.
Borate3 0.37 0.31 0.37 0.40
± ± ± ±
0.08 0.02 0.05 0.07
1 The activity in 0.5 M ammediol divided by the activity in 0.05 M ammediol expressed as mean ± SD of at least three experiments. - The activity in I M DEA divided by the activity in 0.05 M DEA expressed as mean ± SD of at least three experiments. 3 The activity in 0.5 M borate divided by the ac tivity in 0.05 M borate expressed as mean ± SD of at least three experiments.
Tabic V. Level of APase inhibition by L-phenylalanine Tissue
Lymphoma Spleen Placenta Whole fetus Fetal thymus
Percent inhibition />NPP
GP
31.6 ± 6.6 27.2 ± 6.8 25.6 ±6.6 31.8 ± 8.2 31.3 ± 4.1
24.7 ± 2.9 24.0 ± 2.0 19.8 ±11.2 30.8 ± 2.6 no data
Mean ± SD were determined from at least three experiments.
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/5/2018 12:11:13 AM
Tissue
Table III. Level of APase inhibition by EDTA
Floyd/Tisdale/Lumh
350
Figures 3 and 4 show typical heat inacti vation patterns of all APase activities using the three substrates. It is clear that all tis sues have a similar pattern of inactivation. Time in M inutes o t 55 °C
A statistical analysis of these heat inactiva tion rates verifies this conclusion (table VI). Substrate ratios indicate that the pH 10 APase activities of lymphoma, fetal thymus, placenta, spleen and whole fetus hydrolyze GP and pNPP with approximately the same level of activity (table VII). The km for the NPase activities of lym phoma, spleen, placenta, and fetal thymus show that all APases have a similar affinity for the substrate NP (table VIII). Figure 5 shows the electrophoretic pat terns of APases from all five tissues using Table VI. Heat inactivation rate of APase activities Tissue
Heat inactivation rate pNPP
Fig. 3. Heat inactivation of APase activities of lymphoma (A), spleen (A), placenta (■-), and whole fetus (■ ) using the substrate pNPP. The natural logarithm of the activity remaining is plot ted against the time incubated at 55 'C in order to determine the heat inactivation rate constant. Tim e in M inutes o t 5 5 °C
Lymphoma Spleen Placenta Whole fetus Fetal thymus
0.13 ± 0.11 ± 0.13 ± 0.15 ± 0.11 ±
NP
GP 0.04 0.16 ± 0.14 0.01 0.14 ± 0.08 0.03 0.10 ± 0.02 0.03 0.09 ± 0.04 0.02 no data
0.15 ± 0.01 0.17 ± 0.04 0.15 ± 0.02 no data 0.15 ± 0.02
Mean ± SD were determined from at least three experiments.
Table VII. Substrate ratio of APase activity Substrate ratio1 NP Lymphoma Spleen Placenta Fetal thymus
Fig. 4. Heat inactivation of APase activities of lymphoma (A)> spleen (A), placenta (Q, and whole fetus (■ ) using the substrate GP. For fur ther explanation see figure 3.
1.03 0.76 0.91 1.01
GP ± 0.16 ±0.17 ± 0.01 ± 0.16
1.29 ± 0.42 1.30 ± 0.21 0.85 ± 0.04 no data
Mean ± SD were determined from at least three experiments. 1 Substrate ratio is the ratio of activity using the substrate NP or GP divided by the activity using the substrate pNPP.
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/5/2018 12:11:13 AM
Tissue
Comparison of Alkaline Phosphatases
351
the substrate NP. The placenta, spleen, lym phoma and fetal thymus APase have simi lar electrophoretic mobilities, while the lymphoma and fetal thymus APases are identical. The whole fetus definitely has a distinct electrophoretic mobility. The lym phoma APase and faster migrating APase of
Table VIII. Michaelis constant for APase Tissue
Km X 10 5 M
Lymphoma Spleen Placenta Fetal thymus
2.4 3.4 3.0 2.1
± ± ± ±
0.7 0.4 1.6 0.7
Mean ± St) were obtained from at least three experiments. These experiments were performed using the substrate NP.
the whole fetus are slowed by neuraminidase (fig. 6). The neuraminidase treatment signifi cantly retards the tailing in the placenta and spleen extracts which may represent an ac tion on the APase in these extracts although the total migration of APase is not retarded by these extracts. The fetal thymus APase is significantly slowed by the neuraminidase treatment (fig. 7) as in the lymphoma APase (fig. 6). All tissues have p-nitrophenyl phos phatase (pNPPase) activity (fig. 8). Figure 8 compares the electrophoretic mobility of the NPase and pNPPase activities of all tissues. Each tissue APase activity appears to have NPase and pNPPase activities with identical electrophoretic mobilities. The fetal thymus APase appears to have a different mobility from the lymphoma APase in figures 7 and 8. This difference in mobility is probably due to the higher detergent content in this fetal thymus extract since the mobilities were always identical when the detergent concentrations were the same.
P S
W
L
Pn S
Sn W Wn L
Ln
F,
Fig. 5. Acrylamide gel electrophoresis of APase activity using NP as substrate. P Placenta; S spleen; W whole fetus; 1. lymphoma, and F, fetal thymus.
Fig. 6. Acrylamide gel electrophoresis of APase activity using NP as substrate in the presence and absence of neuraminidase. P placenta: S spleen: W whole fetus; I. lymphoma, and n neuraminidase treated.
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/5/2018 12:11:13 AM
P
352
Floyd Tisdale t.umb
F2
F2n
Fig. 7. Acrylamide gel electrophoresis of APase showing NPase of the fetal thymus in the presence and absence of neuraminidase. F, Fetal thymus: n neuraminidase treated.
Pa Pb Sa Sb Wa Wb La Lb F2a F2b Fig. 8. Acrylamide gel electrophosphoresis of APase comparing NPase and /;NPPase activities. P Placenta: S spleen; W whole fetus: I lymphoma: Fa fetal thymus; a pNPPase. and b NP.
Lwnb and Docll 1131 have compared the lymphoma APase to that of the adult mouse liver, spleen, and duodenum. Liver and duodenum were shown to be different from each other as well as the spleen in pH opti mum. heat inactivation, ratio of activity to ward /;NPP, and GP and electrophoretic mobility, while the spleen and lymphoma APases were shown to be similar by these same characteristics. This characterization has included more parameters and tissues. The marine placenta was included in the comparison since it has a very high specific activity and the placental APase has been shown in the human system to be similar to APase activities in human tumors [25). Be cause the size of the fetal thymus has limit ed the numbers of experiments that can be performed, the whole fetus at that same age was included in the comparison. Partial characterization of the fetal thymus includ ing pH optimum, heat inactivation, EDTA inhibition, and substrate ratio has been re ported 114). By comparison of these data with that previously reported, the following conclusions are drawn. All APase activities tested had an optimum pH around 10. Mag nesium and the nucleophilic buffers used ac tivated the APase activities. Each APase ac tivity' was inhibited by EDTA. L-phenylalanine. /,-homoarginine, and borate, ap proximately 97, 35, 100 and 36" 0 . respec tively. The electrophoretic mobility of each APase was identical with the substrates NP and /tNPP indicating that they are hydro lyzed by the same enzyme. The moiblity of the APase of the lymphoma, fetal thymus, spleen, placenta, and whole fetus extracts was slowed by neuraminidase. Thus, the APase of all the extracts tested is a glyco
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/5/2018 12:11:13 AM
Discussion
Comparison of Alkaline Phosphatases
353
protein. It can be concluded that the APase of all tissues is similar with the exception of the whole fetus. The whole fetus had a dif ferent electrophoretic mobility from the oth er APascs, even though it was effected by neuraminidase and appears to be the same enzyme which hydrolyzes the substrates NP and pNPP. It is not surprising that the APases in extracts of the whole fetus at 16 days gestation show differences since it is com posed of several tissues which have been shown to have distinct APase isozymes in the mature adult. The more appropriate comparison is with the fetal thymus APase activity. Present data show no differences between the fetal thymus APase and the lymphoma APase. Our results are in general agreement with two other reports of biochemical char acterization of murine leukemic APase [2,
19]. Cory et al. [2] have identified a new APase activity in leukemic livers which was ‘Mg** independent’ by comparison to the normal liver APase which is activated three fold by magnesium. However, their ‘Mg**-independent’ isozyme was activated 46°/o, roughly the same level of activation that we observed in all four tissues. Neu mann et al. \ 19] stated in the summary that magnesium was inhibitory but showed inhi bition with 60 mM magnesium chloride and variable (30-80°/o) activation by 1 mM magnesium chloride. Both reports show a similar heat inactivation pattern which is in agreement with our data, that is, approxi mately half the activity is destroyed by 55 C in 5 min. Both Cory et al. [2] and Neumann et al. [19] report a lower pH opti mum using Tris buffer. Since Tris does not buffer well above pH 9.0, we have em-
Table IX. Properties of placenta and tumor APase from murine and human sources Sources
/.-Phenylalanine %
/.-Homoarginine %
Heat inactivation
Cellular localization
EDTA Km, mM %
Murine lymphoma Murine placenta Regan isoenzyme Human placenta Nagao Regan variant HeLa parent HeLa TCRC-I HeLA TCRC-2
46
100
labile
cell membrane
99
48
100
labile
cell membrane of 99 0.030 fetal porion [8] intramembrane space 10 [10] 0.55 [6] 1.0(6] of mitochondria [20] 0 [18] 0.51 [6] 2.2(18]
11 [22]
stable [7]
75 [7]
11 (22]
stable [7]
90(18] 75 [9]
0 [9]
77 [22]
37 [22]
stable [18] moderately stable [9] stable [22]
73 [22]
11 [22]
stable [22]
0(22]
78 [22]
labile [22]
-
50 [18] 34 [9]
1.6(9]
0.26 [18] U [9]
controversial [22] intramembrane space of mitochondria [20] cell membrane [20] Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/5/2018 12:11:13 AM
72 [7]
0.024
Floyd/Tisdale/Lumb
354
Acknowledgement This work has been supported by Research Corporation, Grant GY-10465 from the National Science Foundation, Grant BC-112 from the American Cancer Society, and Grant RR 8006-05 from the Division of Research Resources, Nation al Cancer Institute. This publication is listed as Atlanta University Center Science Research In stitute Publication 68.
References 1 Coggin. J. H„ jr. and Anderson, N. G.: Can cer. differentiation and embryonic antigens. Some central problems. Adv. Cancer Res. 19: 105-165 (1973). 2 Cory, J. G.: Whitford, T.W., and Rich, M. A.: Alkaline phosphatase activity in virus-induced murine leukemia. Biochem. Med. 5: 465-471 (1971). 3 De Thé, G.: Association of enzymes: ATPase and alkaline phosphatase with the virions of murine leukemia. Natn. Cancer Inst. Monogr. 22: 169-189 (1966). 4 Fey, F. and Zelms, B.: Histochemical detection of alkaline phosphatase in hematopoietic tis sues of normal mice of different inbred strains. Acta biol. mcd. Germ. 12: 195-202 (1964). 5 Fishman, L.: Acrylamide disc gel electropho resis of alkaline phosphatase of human tissues, serum and ascites fluid using triton X-100 in the sample and the gel matrix. Biochem. Med. 9: 309-315 (1974). 6 Fishman, W. H.: Immunologic and biochemi cal approaches to alkaline phosphatase isoen zyme analysis: the regan isoenzyme. Ann. N.Y. Acad. Sci. 166: 745-759 (1969). 7 Fishman, W. H.; Inglis, N. I.; Stolback. L. L., and Kraut, M. J.: A scrum alkaline phospha tase isoenzyme of human neoplastic cell origin. Cancer Res. 28: 150-154 (1968). 8 Floyd. R. A.: A comparative electrophoretic study of normal and leukemic tissues; MS the sis Atlanta University (1973). 9 Higashino, K.; Hashinotsume. M.; Kang, K.Y.; Takahashi, Y„ and Yamamura. Y.: Studies on a variant alkaline phosphatase in sera of pa
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/5/2018 12:11:13 AM
ployed ammediol which has a higher pK.,. Undoubtedly the differences in ionic strength also affect the activity of the en zyme. The murine lymphoma differs from the human Regan isoenzyme in its heat stabili ty, inhibition by L-phcnylalanine, L-homoarginine and EDTA, km, and histochemical localization. A comparison of the mu rine lymphoma to human tumor APases is seen in table IX. The APase activity in the TCRC-1 subline of HeLa cells appears very similar to the Regan isoenzyme while the TCRC-2 subline APase is surprisingly simi lar to the murine lymphoma APase. Both APases have similar heat lability and histochemical localization. In fact, the amino acid inhibition patterns are similar in high molarity nucleophilic buffers [Floyd and Lumb, in preparation]. Despite the detailed differences between murine lymphoma APase and human pla centa and tumor APases, the analogy be tween the two systems is evident. In both systems an enzyme which is normally found in the placenta is also found in the tumor. The inescapable conclusion is that the APase represents the reactivation of a gene which is usually active only during fetal stages. Mounting evidence in many systems which indicates that embryonic and fetal proteins, isoenzymes, and antigens appear in malig nant cells has been recently reviewed [I]. In the murine lymphoma system this relation ship is made clearer by the appearance of a similar isoenzyme in the fetal thymus and in a few cells in the thymic-dependent areas of the spleen. This provides a system in which a normal dedifferentiation process, the im mune response, can be compared to the ab normal dedifferentiation process, malignant transformation.
Comparison of Alkaline Phosphatases
19 Neumann, H.; Hauck-Granoth, R., and Haran-Ghera, N.: A comparative biochemical study of alkaline phosphatases in normal and leukemic mice. Cancer Res. 31: 1657-1701 (1971). 20 Sasaki. W. and Fishman, W. H.: Ultrastructural studies on Regan and non-Regan isoenzyme of alkaline phosphatase in human ovarian can cer cells. Cancer Res. 33: 3008-3018 (1973). 21 Sieglcr, R. and Rich, M. A.: Significance of in creased alkaline phosphatase in viral-induced thymic lymphoma. Proc. Soc. exp. Biol. Med. 125: 868-871 (1967). 22 Singer, R. M. and Fishman, W. H.: Character ization of two HeLa sublines: TCRC-1 pro duced Regan isoenzyme and TCRC-2 non-Re gan isoenzyme. J. Cell Biol. 60: 777-780 (1974). 23 Smith, C.: Studies of the thymus of the mam mal. XII. Histochemistry of the thymus of C57B1 and AKR mice. J. natn. Cancer Inst. 26: 389-404 (1961). 24 Smith, C.: Studies of the thymus of the mam mal. XIII. Histochemistry of irradiated thy muses of C57BI strain mice. J. natn. Cancer Inst. 29: 375-386 (1962). 25 Stolbach, L. L.; Krant, M. J., and Fishman. W. H.: Ectopic production of an alkaline phospha tase in cancer patients. New Engl. J. Med. 281: 757-762 (1969). 26 Wilson, K. J.; Neumann, H., and Haran-Ghera, N.: Chronological appearance of alkaline phosphatase activity in virus-induced thymic lymphomas in C57B1/6 mice. Cancer Res. 31: 1702-1705 (1971).
Correspondence to: Dr. J. R. Lumb. Biology Department, Atlanta University, Atlanta. GA 30314 (USA)
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/5/2018 12:11:13 AM
tients with hepatocellular carcinoma. Clinica chim. Acta 40: 67-81 (1972). 10 Higashino, K.; Otani, R., and Kudo, S.: Hepa tocellular carcinoma and a variant alkaline phosphatase. Ann. intern. Med. 83: 74-78 (1975). 11 Lagcriôf, B. A. M. and Kaplan, H. S.: Specif icity of the relationship between thymic alka line phosphatase activity and lymphoma devel opment in strain C57BI mice. J. natn. Cancer Inst. 38: 437-458 (1967). 12 Lecocq, J. and Inesi, G.: An assay for phos phate in the presence of ATP. Anat. Biol. 15: 160-163 (1966). 13 Lumb, J. R. and Doell. R. G.: The biochemi cal characterization of alkaline phosphatase from chemical and viral induced thymic lym phoma of C57B1 mice. Cancer Res. 30: 1391-1396 (1970). 14 Lumb, J. R. and Floyd. R. A.: Placental alka line phosphatase appearance in murine lym phoma. Proc. 5th Meet. Int. Res. Group on Carcinoembryonic Proteins, Copenhagen (1977). 15 McKay, O. G.; Hertig, A. T.; Adams, E. C., and Danzigler, S.: Histochemical observations on the germ cells of human embryos. Anat. Rec. 117: 201-219 (1953). 16 Metcalf, D.; Sparrow, N., and Wyllie, R.: Al kaline phosphatase activity in mouse lym phoma tissue. Aust. J. exp. Biol. mcd. Sci. 40: 215 (1962). 17 Moss, D. W.: Kinetics of phosphatase action of naphthyl phosphate determined by a highly sensitive spectrofluorimetric technique. Biochem. J. 76: 32 (1960). 18 Nakayama, T.; Yoshida, M.. and Kitamura. M.: /.-leucine sensitive heat-stable alkaline phosphatase isoenzyme detected in a patient with pleuritis carcinoniatosa. Clinica chim. Acta 30: 546-548 (1970).
355
Murine Lymphoma Alkaline Phosphatase: A Cell Membrane Carcinofetal Enzyme R. A. Floyd, V. G. Tisdale and J. R. Lumb' Biology Department. Atlanta University. Atlanta. Ga.
Introduction Alkaline phosphatase (APase) has been associated with murine thymic lymphomas [3. 4, 16. 23, 24], Lagerlof and Kaplan [11] and Wilson el al. [26] have correlated the appearance of APase with malignant trans formation and have ruled out the possibili ties that APase is associated with cellular proliferation or with the leukemia virus in the absence of malignant transformation. Contradictory results were obtained using a 1 Research Career Development Awardee (CA 0067).
less sensitive histochemical procedure [211. APase activity has also been associated with lymphoblastic fetal thymus cells up to 16 days gestation, but it is not found on thymic lymphocytes in a normal adult [11]. This APase activity of murine lymphomas has been shown to be similar to the APase that occurs at a low level in adult spleen [ 131. The observation of a placental-type APase in a variety of human tumors [7, 25] has suggested that the comparison of this mu rine lymphoma APase be made with that of the murine placenta. Therefore, a compari son of the murine lymphoma, fetal thymus.
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/5/2018 12:11:13 AM
Abstract. Alkaline phosphatase (APase) has been shown to have a membrane-bound localization in the murine fetal thymus, in murine thymic lymphoma and in adult spleen. Since it was suggested from these previous experiments that the lymphoma APase might represent an embryonic function, a detailed biochemical comparison of the lymphoma APase with the fetal thymus, placenta, fetus and spleen APases was performed. The para meters investigated were pH optimum, activation, inhibition, heat inactivation, substrate ratios. Michaelis constant, and electrophoretic analysis in the presence and absence of neuraminidase with the substrates «-naphthyl phosphate, /(-glycerophosphate, and p-nitrophenyl phosphate. The results indicate that the lymphoma APase is very similar to the fetal thymus, placenta and spleen APases. Furthermore, these results lend support to the hypothesis that the APase activity which appears in thymic lymphoma might represent a derepressed embryonic function. Thus, the murine lymphoma APase may be termed a cell membrane carcinofetal enzyme.
Comparison of Alkaline Phosphatases
Materials and Methods Source of Tissues Inbred C57BI/Ka mice were originally ob tained from Dr. Henry S. Kaplan at Stanford Uni versity. They were maintained by brother-sister mating. Thymic lymphomas were induced in these mice by urethan treatment of C57LV as previous ly described [13]. The lymphomatous tissues were taken from thymus, spleen, and lymph nodes. These tissues were pooled after histological and histochemical examination confirmed that these tissues were leu kemic. The spleens were taken from adlut C57BI/Ka males. The fetal thymuses, whole fetuses, and pla centas were taken from 16-day pregnant C57B1/ Ka mice. Preparation of Extracts For most experiments crude extracts were pre pared by homogenizing the tissue in a glass homogenizer (Corning) in 0.01 M Tris at pH 7. For electrophoresis experiments, the homogenized ex tract was centrifuged at 75,000 p for 30 min. The precipitate was treated with l°/o Noniodet P-40 (NP40; Shell Oil, Inc.), incubated at 37 C for 15 min and centrifuged at 75,000 g for 30 min. The NP40 treatment was repeated twice on the precipitates and the three resulting supernatants were pooled. The pooled supernatants and the re maining precipitate were then treated with 30°/o
butanol and centrifuged at 75,000 g for 30 min. The resulting aqueous layer was dialyzed to re move butanol. The fetal thymus and lymphoma extracts were treated with 1,0°/o NP40 and incu bated at 37 C for 15 min. One of the fetal thy mus extracts contained 10°/o NP40 instead of P/o because it was further concentrated. The fetal thy mus extracts were labelled F, (l°/o NP40) and F2 (10°/o NP40). The substrates used were NP (0.052 m/W). GP (50 mM), and pNPP (3.38 mM). As buffers, bor ate. 2-amino-2-methyl-l,3-propanediol (ammediol), Tris, and diethanolamine (DEA) were used at varying concentrations. Spectrophotometric Assays for APase The enzyme was preincubated with buffer at 37 °C for 5 min except for experiments to deter mine EDTA inhibition in which a 20-min preincu bation was required [Floyd and Lamb, unpub lished observations]. The substrate was added re sulting in a 1-ml total reaction volume which was allowed to incubate for 30 min at 37 °C. After the 30-min incubation the reaction was stopped with 0.2 M EDTA in 0.5 N NaOH for measurement of p-nitrophenol at 400 nm [13], For reactions using the substrate GP, phosphate was assayed by the method of Lecocq and fnesi [12]. Flttoriinetric Assays for APase The fluorimetric assay was employed when us ing the substrate NP (17). The buffer (1 M DEA) was preincubated at 37 °C for 5 min in a 4-mI re action volume at pH 10. The extract and substrate NP were added and the formation of «-naphthol was recorded by fluorescence for 2-5 min at con stant temperature. For the experiments to test the level of inhibition with EDTA, the extract was preincubated for 20 min with the buffer [Floyd and Lump, unpublished observations] and the re action initiated by the addition of substrate. The optimum pH of the extracts was deter mined using the appropriate substrate with the pH of the buffer varying from 9 to 10.5. Heat inacti vation rate was determined by heating tubes pre pared for assay, including 5 mM magnesium chlo ride, for varying times at 55 °C prior to assay. The heat inactivation rate is described in the equa tion A = A„e'kt
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/5/2018 12:11:13 AM
placenta, whole fetus, and adult spleen has been made using the substrates «-naphthyl phosphate (NP), /f-glycerophosphate (GP), and p-nitrophenyl phosphate (pNPP). The parameters chosen for this characterization included pH optimum, activation by magne sium and nucleophilic buffers, inhibition by amino acids and ethylenediaminetetraacetic acid (EDTA), heat inactivation, substrate ratios, Michaelis constant (K.m), and electro phoretic analysis. Partial characterization of the fetal thymus APase activity has been re ported [ 14].
347
34N
Floyd/Tisdale/Lumh
Results Figures I and 2 show the pH optima for the tissues and substrates tested. These data show that spleen, lymphoma, placenta, and whole fetus extracts all have an APase ac tivity which has an optimum pH around 10 using the substrates pNPP and GP. Table I shows the magnesium activation for the tissues and substrates tested. All ex tracts show a similar level of activation by magnesium. With the exception of the pla centa there appears to be no significant dif ferences among the tissues. These data show much higher variability and. therefore, are
Big. 1. The pH optimum of APase activities of placenta ( • ) . spleen (/\), lympoma (O), and whole fetus ( ) using the substrate pNPP.
pH
Fig. 2. The pH optimum of APase activities of placenta ( • ) , spleen (¿\), lymphoma (O), and whole fetus ( ) using the substrate GP. Table I. Level of APase activation by magnesium Tissue
Lymphoma Spleen Placenta Whole fetus Fetal thymus
Percent activation />NPP
GP
NP
30.2 ± 2.4 27.3 ± 6.3 45.5 ± 9.1 25.8 ± 4.3 no data
39.3 ± 0.6 39.3 ± 0.6 66.8 ± 11.1 41.3 ± 6.8 no data
46 ± 0.21 46 ± 0.16 46 ± 0.31 no data 47 ± 0.17
Mean ± SD were determined from at least three experiments.
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/5/2018 12:11:13 AM
where A,, is the unheated activity, A is the activity remaining after time t and k is the heat inactiva tion rate constant. The rate was determined by Least Square analysis of the data expressed as the natural log of activity vs. time at 55 C. Kinetics: The enzyme reaction was carried out as indicated above using varying concentrations of NP as substrate. A Lineweaver-Bruke double-reci procal plot of the results was used to determine the K|„. Substrate ratio was calculated as the activ ity of the enzyme extract with either NP or GP divided by the activtiy with pNPP. Activation and inhibition: Experiments with magnesium chloride (5 mM), EDTA (0.1 mM), L-phenylalanine (10 mM). L-homoarginine 10 mM), Tris buffer (0.5 M), ammediol buffer (0.5 M), DBA (1.0 M). and borate (0.5 M) were performed along with the proper controls to determine the percent activation or inhibition of APase activity. Statistical analysis: All data arc reported as mean and standard deviation of at least three ex periments. A Students t test was performed to de termine significance using the facilities of the At lanta University Computing Center. Electrophoresis was performed according to Fislwtan [5]. APase activity was localized using two different substrates. For /;NPP the enzyme lo calization was indicated by the yellow p-nitrophcnyl which diffused quickly. An azo dye was used as a coupling agent for the «-naphthol prod uct with NP as substrate (15|.
Comparison of Alkaline Phosphatases
349
not considered significant. The placenta shows a lower level of endogenous magne sium and a much higher APasc specific ac tivity [Lumb and Yahya, unpublished]. It is suggested this may account for this high variability. The «-naphthyl phosphatase (NPase) ac tivity of lymphoma, spleen, placenta and fe tal thymus extracts was increased by the presence of the nucleophilic buffers DEA and ammediol. while NPase activity was in hibited by the presence of the neutral buff er, borate (table II). Table III indicates the level of inhibition by EDTA using the three substrates with the five tissue extracts. All five extracts show almost complete inhibition with all three substrates. The level of APase inhibition by the amino acids L-phenylalanine and L-homoarginine is shown in tables IV and V. L-Homoarginine has a greater inhibitory ef fect on APase activity than does L-phenylalanine.
Table II. Effect of butTer concentration on APase activity Ammediol1* DEA3
Lymphoma Spleen Placenta Fetal thymus
1.24 ± 1.91 ± 1.65 ± 1.63 ±
0.29 0.14 0.26 0.64
2.88 ± 0.32 2.81 ±0.11 2.78 ±0.18 2.87 ± 0.43
Tissue
Percent inhibition />NPP
GP
85.7 ± 7.6 Lymphoma 96.5 ± 0.7 Spleen 95.0 ±0.1 Placenta 96.0 ±0.1 Whole fetus Fetal thymus no data
NP
90.8 ± 6.9 84.7 ± 1.5 94.5 ± 0.7 85.5 ± 0.7 no data
98.57 ± 0.78 100 98.47 ± 0.86 no data 100
Mean±SD were determined from at least three experiments.
Table IV. Level of APase inhibition by amin acids Tissue
Percent inhibition L-phenylalanine L-homoarginine
Lymphoma Spleen Placenta Fetal thymus
45.66 39.82 48.20 49.36
± ± ± ±
4.63 7.08 5.01 6.93
100 100 100 100
At least three experiments were performed using the substrate NP using 0.05 M DEA buffer at pH 10.
Borate3 0.37 0.31 0.37 0.40
± ± ± ±
0.08 0.02 0.05 0.07
1 The activity in 0.5 M ammediol divided by the activity in 0.05 M ammediol expressed as mean ± SD of at least three experiments. - The activity in I M DEA divided by the activity in 0.05 M DEA expressed as mean ± SD of at least three experiments. 3 The activity in 0.5 M borate divided by the ac tivity in 0.05 M borate expressed as mean ± SD of at least three experiments.
Tabic V. Level of APase inhibition by L-phenylalanine Tissue
Lymphoma Spleen Placenta Whole fetus Fetal thymus
Percent inhibition />NPP
GP
31.6 ± 6.6 27.2 ± 6.8 25.6 ±6.6 31.8 ± 8.2 31.3 ± 4.1
24.7 ± 2.9 24.0 ± 2.0 19.8 ±11.2 30.8 ± 2.6 no data
Mean ± SD were determined from at least three experiments.
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/5/2018 12:11:13 AM
Tissue
Table III. Level of APase inhibition by EDTA
Floyd/Tisdale/Lumh
350
Figures 3 and 4 show typical heat inacti vation patterns of all APase activities using the three substrates. It is clear that all tis sues have a similar pattern of inactivation. Time in M inutes o t 55 °C
A statistical analysis of these heat inactiva tion rates verifies this conclusion (table VI). Substrate ratios indicate that the pH 10 APase activities of lymphoma, fetal thymus, placenta, spleen and whole fetus hydrolyze GP and pNPP with approximately the same level of activity (table VII). The km for the NPase activities of lym phoma, spleen, placenta, and fetal thymus show that all APases have a similar affinity for the substrate NP (table VIII). Figure 5 shows the electrophoretic pat terns of APases from all five tissues using Table VI. Heat inactivation rate of APase activities Tissue
Heat inactivation rate pNPP
Fig. 3. Heat inactivation of APase activities of lymphoma (A), spleen (A), placenta (■-), and whole fetus (■ ) using the substrate pNPP. The natural logarithm of the activity remaining is plot ted against the time incubated at 55 'C in order to determine the heat inactivation rate constant. Tim e in M inutes o t 5 5 °C
Lymphoma Spleen Placenta Whole fetus Fetal thymus
0.13 ± 0.11 ± 0.13 ± 0.15 ± 0.11 ±
NP
GP 0.04 0.16 ± 0.14 0.01 0.14 ± 0.08 0.03 0.10 ± 0.02 0.03 0.09 ± 0.04 0.02 no data
0.15 ± 0.01 0.17 ± 0.04 0.15 ± 0.02 no data 0.15 ± 0.02
Mean ± SD were determined from at least three experiments.
Table VII. Substrate ratio of APase activity Substrate ratio1 NP Lymphoma Spleen Placenta Fetal thymus
Fig. 4. Heat inactivation of APase activities of lymphoma (A)> spleen (A), placenta (Q, and whole fetus (■ ) using the substrate GP. For fur ther explanation see figure 3.
1.03 0.76 0.91 1.01
GP ± 0.16 ±0.17 ± 0.01 ± 0.16
1.29 ± 0.42 1.30 ± 0.21 0.85 ± 0.04 no data
Mean ± SD were determined from at least three experiments. 1 Substrate ratio is the ratio of activity using the substrate NP or GP divided by the activity using the substrate pNPP.
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/5/2018 12:11:13 AM
Tissue
Comparison of Alkaline Phosphatases
351
the substrate NP. The placenta, spleen, lym phoma and fetal thymus APase have simi lar electrophoretic mobilities, while the lymphoma and fetal thymus APases are identical. The whole fetus definitely has a distinct electrophoretic mobility. The lym phoma APase and faster migrating APase of
Table VIII. Michaelis constant for APase Tissue
Km X 10 5 M
Lymphoma Spleen Placenta Fetal thymus
2.4 3.4 3.0 2.1
± ± ± ±
0.7 0.4 1.6 0.7
Mean ± St) were obtained from at least three experiments. These experiments were performed using the substrate NP.
the whole fetus are slowed by neuraminidase (fig. 6). The neuraminidase treatment signifi cantly retards the tailing in the placenta and spleen extracts which may represent an ac tion on the APase in these extracts although the total migration of APase is not retarded by these extracts. The fetal thymus APase is significantly slowed by the neuraminidase treatment (fig. 7) as in the lymphoma APase (fig. 6). All tissues have p-nitrophenyl phos phatase (pNPPase) activity (fig. 8). Figure 8 compares the electrophoretic mobility of the NPase and pNPPase activities of all tissues. Each tissue APase activity appears to have NPase and pNPPase activities with identical electrophoretic mobilities. The fetal thymus APase appears to have a different mobility from the lymphoma APase in figures 7 and 8. This difference in mobility is probably due to the higher detergent content in this fetal thymus extract since the mobilities were always identical when the detergent concentrations were the same.
P S
W
L
Pn S
Sn W Wn L
Ln
F,
Fig. 5. Acrylamide gel electrophoresis of APase activity using NP as substrate. P Placenta; S spleen; W whole fetus; 1. lymphoma, and F, fetal thymus.
Fig. 6. Acrylamide gel electrophoresis of APase activity using NP as substrate in the presence and absence of neuraminidase. P placenta: S spleen: W whole fetus; I. lymphoma, and n neuraminidase treated.
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/5/2018 12:11:13 AM
P
352
Floyd Tisdale t.umb
F2
F2n
Fig. 7. Acrylamide gel electrophoresis of APase showing NPase of the fetal thymus in the presence and absence of neuraminidase. F, Fetal thymus: n neuraminidase treated.
Pa Pb Sa Sb Wa Wb La Lb F2a F2b Fig. 8. Acrylamide gel electrophosphoresis of APase comparing NPase and /;NPPase activities. P Placenta: S spleen; W whole fetus: I lymphoma: Fa fetal thymus; a pNPPase. and b NP.
Lwnb and Docll 1131 have compared the lymphoma APase to that of the adult mouse liver, spleen, and duodenum. Liver and duodenum were shown to be different from each other as well as the spleen in pH opti mum. heat inactivation, ratio of activity to ward /;NPP, and GP and electrophoretic mobility, while the spleen and lymphoma APases were shown to be similar by these same characteristics. This characterization has included more parameters and tissues. The marine placenta was included in the comparison since it has a very high specific activity and the placental APase has been shown in the human system to be similar to APase activities in human tumors [25). Be cause the size of the fetal thymus has limit ed the numbers of experiments that can be performed, the whole fetus at that same age was included in the comparison. Partial characterization of the fetal thymus includ ing pH optimum, heat inactivation, EDTA inhibition, and substrate ratio has been re ported 114). By comparison of these data with that previously reported, the following conclusions are drawn. All APase activities tested had an optimum pH around 10. Mag nesium and the nucleophilic buffers used ac tivated the APase activities. Each APase ac tivity' was inhibited by EDTA. L-phenylalanine. /,-homoarginine, and borate, ap proximately 97, 35, 100 and 36" 0 . respec tively. The electrophoretic mobility of each APase was identical with the substrates NP and /tNPP indicating that they are hydro lyzed by the same enzyme. The moiblity of the APase of the lymphoma, fetal thymus, spleen, placenta, and whole fetus extracts was slowed by neuraminidase. Thus, the APase of all the extracts tested is a glyco
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/5/2018 12:11:13 AM
Discussion
Comparison of Alkaline Phosphatases
353
protein. It can be concluded that the APase of all tissues is similar with the exception of the whole fetus. The whole fetus had a dif ferent electrophoretic mobility from the oth er APascs, even though it was effected by neuraminidase and appears to be the same enzyme which hydrolyzes the substrates NP and pNPP. It is not surprising that the APases in extracts of the whole fetus at 16 days gestation show differences since it is com posed of several tissues which have been shown to have distinct APase isozymes in the mature adult. The more appropriate comparison is with the fetal thymus APase activity. Present data show no differences between the fetal thymus APase and the lymphoma APase. Our results are in general agreement with two other reports of biochemical char acterization of murine leukemic APase [2,
19]. Cory et al. [2] have identified a new APase activity in leukemic livers which was ‘Mg** independent’ by comparison to the normal liver APase which is activated three fold by magnesium. However, their ‘Mg**-independent’ isozyme was activated 46°/o, roughly the same level of activation that we observed in all four tissues. Neu mann et al. \ 19] stated in the summary that magnesium was inhibitory but showed inhi bition with 60 mM magnesium chloride and variable (30-80°/o) activation by 1 mM magnesium chloride. Both reports show a similar heat inactivation pattern which is in agreement with our data, that is, approxi mately half the activity is destroyed by 55 C in 5 min. Both Cory et al. [2] and Neumann et al. [19] report a lower pH opti mum using Tris buffer. Since Tris does not buffer well above pH 9.0, we have em-
Table IX. Properties of placenta and tumor APase from murine and human sources Sources
/.-Phenylalanine %
/.-Homoarginine %
Heat inactivation
Cellular localization
EDTA Km, mM %
Murine lymphoma Murine placenta Regan isoenzyme Human placenta Nagao Regan variant HeLa parent HeLa TCRC-I HeLA TCRC-2
46
100
labile
cell membrane
99
48
100
labile
cell membrane of 99 0.030 fetal porion [8] intramembrane space 10 [10] 0.55 [6] 1.0(6] of mitochondria [20] 0 [18] 0.51 [6] 2.2(18]
11 [22]
stable [7]
75 [7]
11 (22]
stable [7]
90(18] 75 [9]
0 [9]
77 [22]
37 [22]
stable [18] moderately stable [9] stable [22]
73 [22]
11 [22]
stable [22]
0(22]
78 [22]
labile [22]
-
50 [18] 34 [9]
1.6(9]
0.26 [18] U [9]
controversial [22] intramembrane space of mitochondria [20] cell membrane [20] Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/5/2018 12:11:13 AM
72 [7]
0.024
Floyd/Tisdale/Lumb
354
Acknowledgement This work has been supported by Research Corporation, Grant GY-10465 from the National Science Foundation, Grant BC-112 from the American Cancer Society, and Grant RR 8006-05 from the Division of Research Resources, Nation al Cancer Institute. This publication is listed as Atlanta University Center Science Research In stitute Publication 68.
References 1 Coggin. J. H„ jr. and Anderson, N. G.: Can cer. differentiation and embryonic antigens. Some central problems. Adv. Cancer Res. 19: 105-165 (1973). 2 Cory, J. G.: Whitford, T.W., and Rich, M. A.: Alkaline phosphatase activity in virus-induced murine leukemia. Biochem. Med. 5: 465-471 (1971). 3 De Thé, G.: Association of enzymes: ATPase and alkaline phosphatase with the virions of murine leukemia. Natn. Cancer Inst. Monogr. 22: 169-189 (1966). 4 Fey, F. and Zelms, B.: Histochemical detection of alkaline phosphatase in hematopoietic tis sues of normal mice of different inbred strains. Acta biol. mcd. Germ. 12: 195-202 (1964). 5 Fishman, L.: Acrylamide disc gel electropho resis of alkaline phosphatase of human tissues, serum and ascites fluid using triton X-100 in the sample and the gel matrix. Biochem. Med. 9: 309-315 (1974). 6 Fishman, W. H.: Immunologic and biochemi cal approaches to alkaline phosphatase isoen zyme analysis: the regan isoenzyme. Ann. N.Y. Acad. Sci. 166: 745-759 (1969). 7 Fishman, W. H.; Inglis, N. I.; Stolback. L. L., and Kraut, M. J.: A scrum alkaline phospha tase isoenzyme of human neoplastic cell origin. Cancer Res. 28: 150-154 (1968). 8 Floyd. R. A.: A comparative electrophoretic study of normal and leukemic tissues; MS the sis Atlanta University (1973). 9 Higashino, K.; Hashinotsume. M.; Kang, K.Y.; Takahashi, Y„ and Yamamura. Y.: Studies on a variant alkaline phosphatase in sera of pa
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/5/2018 12:11:13 AM
ployed ammediol which has a higher pK.,. Undoubtedly the differences in ionic strength also affect the activity of the en zyme. The murine lymphoma differs from the human Regan isoenzyme in its heat stabili ty, inhibition by L-phcnylalanine, L-homoarginine and EDTA, km, and histochemical localization. A comparison of the mu rine lymphoma to human tumor APases is seen in table IX. The APase activity in the TCRC-1 subline of HeLa cells appears very similar to the Regan isoenzyme while the TCRC-2 subline APase is surprisingly simi lar to the murine lymphoma APase. Both APases have similar heat lability and histochemical localization. In fact, the amino acid inhibition patterns are similar in high molarity nucleophilic buffers [Floyd and Lumb, in preparation]. Despite the detailed differences between murine lymphoma APase and human pla centa and tumor APases, the analogy be tween the two systems is evident. In both systems an enzyme which is normally found in the placenta is also found in the tumor. The inescapable conclusion is that the APase represents the reactivation of a gene which is usually active only during fetal stages. Mounting evidence in many systems which indicates that embryonic and fetal proteins, isoenzymes, and antigens appear in malig nant cells has been recently reviewed [I]. In the murine lymphoma system this relation ship is made clearer by the appearance of a similar isoenzyme in the fetal thymus and in a few cells in the thymic-dependent areas of the spleen. This provides a system in which a normal dedifferentiation process, the im mune response, can be compared to the ab normal dedifferentiation process, malignant transformation.
Comparison of Alkaline Phosphatases
19 Neumann, H.; Hauck-Granoth, R., and Haran-Ghera, N.: A comparative biochemical study of alkaline phosphatases in normal and leukemic mice. Cancer Res. 31: 1657-1701 (1971). 20 Sasaki. W. and Fishman, W. H.: Ultrastructural studies on Regan and non-Regan isoenzyme of alkaline phosphatase in human ovarian can cer cells. Cancer Res. 33: 3008-3018 (1973). 21 Sieglcr, R. and Rich, M. A.: Significance of in creased alkaline phosphatase in viral-induced thymic lymphoma. Proc. Soc. exp. Biol. Med. 125: 868-871 (1967). 22 Singer, R. M. and Fishman, W. H.: Character ization of two HeLa sublines: TCRC-1 pro duced Regan isoenzyme and TCRC-2 non-Re gan isoenzyme. J. Cell Biol. 60: 777-780 (1974). 23 Smith, C.: Studies of the thymus of the mam mal. XII. Histochemistry of the thymus of C57B1 and AKR mice. J. natn. Cancer Inst. 26: 389-404 (1961). 24 Smith, C.: Studies of the thymus of the mam mal. XIII. Histochemistry of irradiated thy muses of C57BI strain mice. J. natn. Cancer Inst. 29: 375-386 (1962). 25 Stolbach, L. L.; Krant, M. J., and Fishman. W. H.: Ectopic production of an alkaline phospha tase in cancer patients. New Engl. J. Med. 281: 757-762 (1969). 26 Wilson, K. J.; Neumann, H., and Haran-Ghera, N.: Chronological appearance of alkaline phosphatase activity in virus-induced thymic lymphomas in C57B1/6 mice. Cancer Res. 31: 1702-1705 (1971).
Correspondence to: Dr. J. R. Lumb. Biology Department, Atlanta University, Atlanta. GA 30314 (USA)
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/5/2018 12:11:13 AM
tients with hepatocellular carcinoma. Clinica chim. Acta 40: 67-81 (1972). 10 Higashino, K.; Otani, R., and Kudo, S.: Hepa tocellular carcinoma and a variant alkaline phosphatase. Ann. intern. Med. 83: 74-78 (1975). 11 Lagcriôf, B. A. M. and Kaplan, H. S.: Specif icity of the relationship between thymic alka line phosphatase activity and lymphoma devel opment in strain C57BI mice. J. natn. Cancer Inst. 38: 437-458 (1967). 12 Lecocq, J. and Inesi, G.: An assay for phos phate in the presence of ATP. Anat. Biol. 15: 160-163 (1966). 13 Lumb, J. R. and Doell. R. G.: The biochemi cal characterization of alkaline phosphatase from chemical and viral induced thymic lym phoma of C57B1 mice. Cancer Res. 30: 1391-1396 (1970). 14 Lumb, J. R. and Floyd. R. A.: Placental alka line phosphatase appearance in murine lym phoma. Proc. 5th Meet. Int. Res. Group on Carcinoembryonic Proteins, Copenhagen (1977). 15 McKay, O. G.; Hertig, A. T.; Adams, E. C., and Danzigler, S.: Histochemical observations on the germ cells of human embryos. Anat. Rec. 117: 201-219 (1953). 16 Metcalf, D.; Sparrow, N., and Wyllie, R.: Al kaline phosphatase activity in mouse lym phoma tissue. Aust. J. exp. Biol. mcd. Sci. 40: 215 (1962). 17 Moss, D. W.: Kinetics of phosphatase action of naphthyl phosphate determined by a highly sensitive spectrofluorimetric technique. Biochem. J. 76: 32 (1960). 18 Nakayama, T.; Yoshida, M.. and Kitamura. M.: /.-leucine sensitive heat-stable alkaline phosphatase isoenzyme detected in a patient with pleuritis carcinoniatosa. Clinica chim. Acta 30: 546-548 (1970).
355
