Biochunica et Biophysica Acta, I (190 ( 199 ! ) 3 i 1-316 © 1991 Elsevier Science Publishers B.V. All rights reserved 0167-4781/91/$03.50
311
BBAEXP 923111
Sequential activation of genes for heme pathway enzymes during erythroid differentiation of mouse Friend virus-transformed erythroleukemia cells Hiroyoshi Fujita ~, Masayuki Yamamoto 2, Takashi Yamagami 2, Norio Hayashi 2, Terry R. Bishop 3, Hubert De Verneuil 4, Takeo Yoshinaga 5, Shigeki Shibahara 2, Richard Morimoto 6 and Shigeru Sassa i t The Rockefelh'r Uniremity, New York, NY (U. S.A.), 2 Departments o.f Biochemisoy and Apldied l'hy.~iology, Tohoku Unit'emit.v, Semlai (Japan), ~ Deparnnent of Medical Gem'tic.~', The Johns Hopkins Uniremity, Bahimore, MD (U.S.A.), 4 I.'aculte ~h' Medecim,, X. Bichat, Paris (Fram'e), "~Department of Public Ih,alth, Kyoto Unil'erviO, Kyoto (JapatD arid r, Dei~arlmenL~'of Molecuho" Bioh~,q' and ('ell BiohJgy, Northwe.~'n'rn Unirer~'ity, l'h'amlon, IL (U.S.A.)
(Received 14 May 1991)
Key words: lleme synthesis; Erythroleukemia cell; c$-Aminolevulinate synthase; ! leme oxygenase; Sequential activation
Changes in the level of transcripts encoding enzymes of the heme biosynthetic pathway as well as those encoding ubiquitous proteins were examined in murine Friend virus-transformed erythroleukemia cells during erythroid cell differentiation induced by chemicals including dimethyl sulfoxide (DMSO). Early changes following DMSO treatment were marked decreases in mRNAs for three ubiquitous proteins, i.e., a 70 kDa heat shock protein ( < 6 h), heme oxygenase and nonspecific ~-aminolevulinate synthase (ALAS) ( < 12 h). These changes were followed by sequential increases in mRNAs for enzymes in the heme biosynthetic pathway. Namely, mRNAs for the erythroidspecific ALAS, 8-aminolevtllinate dehydratase, porphobilinogen deaminase and uroporphyrinogen decarboxylase started to increase at 12, 18, 18-24 and 24 h, respectively. Nuclear runoff studies revealed that these changes are largely transcriptional. Treatments with otI~er inducers of erythroid differentiation, e.g., hexamethylene bisacetamide, n-butyric acid and N'-methylnicotinamide, also showed simi!ar effects on mRNAs as those fo:lowing DMSO. These findings suggest that both suppression of ubiquitous genes and activation of heme pathway enzyme genes are associated with erythroid differentiation, and the former occurs preceding changes in the latter.
Introduction
Murine Friend virus-transformed crythroleukemia (MEL) cells are erythroid precursor cells transformed by the Friend leukemia virus complex, which serve as a useful system for in vitro studies of erythroid cell differentiation [1]. These cells can be induced, by a variety of agents including DMSO, to initiate the pro-
Abbreviations: ALAD, ~5-aminolevulinate dehydratase; ALAS, 6aminolevulinate synthase; ALAS-E, the erythroid-specific ALAS; ALAS-N, the non-specific ALAS; BA, butyric acid; DMSO, dimethyl sulfoxide; HMBA, hexamethylene bisaeetamide; HO, heme oxygenase; hsp70, heat-shock protein 70; MEL, murine erythroleukemia; NMN, N'-methylnicotinamide; PBGD, porphobilinogen deaminase; UROD, uroporphyrinogen decarboxylase. Correspondence: S. Sassa, The Rockefeller University, New York, NY 10021, U.S.A.
gram of terminal erythroid differentiation [2]. Prior to the irreversible commitment to differentiation, a number of metabolic changes occur, e.g., alterations in membrane permeability [3], a transient increase in cAMP concentration [4], an increase in membrane-associated protein kinase C activity [5], a decrease in heme oxygenase (HO) (EC 1.14.99.3) mRNA [6], and the modulation of a number of genes, including c-myb, c-myc, c-los, p53 and the 70 kDa heat-shock protein (hsp70) [7-9]. When MEL cells are incubated with an inducing chemical beyond the latent period, they become irreversibly committed to undergo erythroid differentiation [10], accumulate a- and /3-globin mRNA [11,12], increase enzymatic activities of the heme b,iosynthetic pathway [13], and exhibit erythroid-specific membrane antigens [14]. Increases in activities of enzymes in the heine biosynthetic pathway occur sequentially in the order as they appear in their biosynthetic sequence, i.e., ~-aminolevulinate synthase (ALAS) (EC
312 2.3.1.37) being the first, and take place substantially earlier than the accumulation of heine [13,15]. Recently, it has been shown that two separate genes exist for ALAS in chicken reticulocytes, one encoding the erythroid-specific form (ALAS:E) and the other encoding the hepatic, or the non-specific form (ALAS-N) [16]. Although increases in enzymatic activities of heine pathway enzymes and their time-courses were examined, little is known about their gene activation in MEL cells, particularly that of the ALAS-N isozyme. There was also no study in which changes in mRNA levels for several key enzymes in the heme biosynthetic and the catabolic pathways were examined in a systematic manner. in this study, we examined increases in the levels of transcripts of the genes for certain enzymes in the heme biosynthetic pathway which show early and characteristic increases during crythroid differentiation of MEL cells. These genes include those for ALAS-E, 8-aminolevulinate dehydratase (ALAD) (EC 4.2.1.24), porphobilinogen deaminase (PBGD) (EC 4.3.1.8) and uroporphyrinogen decarboxylase (UROD) (EC 4.1.1.37). We also examined the changes in the levels of mRNAs for three ubiquitously expressed genes, i.e., ALAS-N, HO and hsp70. Materials and Methods
Cell cultures. A clone of MEL cells, DS-19 [17], was a generous gift from Dr. R.A. Rifkind, Memorial SIoan.Kettering Cancer Center, New York, NY. Cells were grown in suspension in a modified Ham's FI2 medium [18] containing 10% (v/v) heat-inactivated fetal bovine serum. Cultures were routinely split every 3-4 days, to yield a cell density of 5.104 cells per ml, to maintain a logarithmic growth. In each experiment, cells from a 24-48 h old culture were suspended in a fresh medium at a cell density of 5.104 cells per ml, and incubated for 16 h, prior to the addition of chemicals. Incubations were then continued for various periods as indicated in the figures. eDNA probes. The eDNA probe for ALAS-N used was a rat ALAS-N eDNA pKRA2cA [19]. The eDNA probe for ALAS-E used was a 1.2 kb rat ALAS-E eDNA fragment cloned from a eDNA library constructed in Agtll from rat anemic spleen poly(A) + RNA [20]. eDNA was screened by the antibody against the rat ALAS-N which also recognizes the rat ALAS-E [20]. The ALAS-E eDNA fragment corresponds to a region starting from 733 bp to 1888 bp of the reported nucleotide sequence of the mouse ALAS-E eDNA [21]. The nucleotide sequence of the probe was 96%, 65% and 75% similar to that of the corresponding nucleotide in mouse ALAS-E [21], chicken ALAS-E eDNA [16] and rat ALAS-N eDNA [19], respectively. In addition, rat heine oxygenase cDNA (pRHO-1) [22],
rat ALAD eDNA (pALAD-1) [23], rat PBGD eDNA (p44SB-1), human UROD eDNA (pUD3) [24] and human hsp70 (pH2.3) [25] were also used for the quantification of mRNA concentrations. Each cDNA fragment was inserted into a pGEM4z vector (Promega, Madison, Wi) for the preparation of an RNA probe, according to the method of Melton et al. [26]. Chicken /3-actin eDNA was nick translated and used as an internal control. Northern blot analysis. Total RNA was isolated from ! • 1 0 7 cells according to the method of Cathala et al. [27]. 20/~g of total RNA were analyzed by Northern blot analysis as described previously [28]. After blotting to a sheet of Zeta-probe blotting membrane (Bio-Rad, Richmond, CA), RNA samples were hybridized with a radioactively labeled specific RNA probe for each protein, which were then treated with RNase A (1/~g/ml), followed by washing under a stringent condition. Levels of mRNAs were quantitated by densitometry using a LKB Ultroscan XL densitometer. The level of/3-actin mRNA was used as a reference control in each experiment. Since the levels of mRNAs for some heine pathway enzymes in untreated control cells showed significant changes as a function of incubation periods, compared with insignificant changes in /3-actin controls, mRNA contents in induced cells were expressed as ratios of those observed in the concurrent untreated controls. Nuclear runoff transcription assay. Nuclei were isolated from 4.107 uninduced or DMSO-induced cells (24 h treatment). Runoff transcription assay was performed as described previously [29]. The yield of 32p. labeled RNA transcripts was 1.9.10 7 cpm and 1.5.10 7 cpm for the control and the induced cells, respectively. An equivalent amount of radioactivity (1.1.107 cpm) from both RNA samples was hybridized to DNA immobilized on a strip of nitrocellulose according to the method described previously [29]. Quantification was made by densitometry using a LKB Ultroscan XL densitometer. Chemicals. DMSO (spectre grade) was purchased from Fisher, Fair Lawn, NJ; hexamethylene bisacetamide (HMBA) was from Aldrich, Milwaukee, Wl; n-butyric acid (BA) and N'-methylnicotinamide (NMN) were from Sigma, St. Louis, MI; and all tissue culture materials were from Gibco, Grand Island, NY. Results
Expression of ALAS-N and ALAS-E mRNAs in uninduced MEL cells Northern blot analysis of total RNA from DS-19 cells using RNA probes for ALAS-N and ALAS-E is shown in Fig. 1. mRNAs encoding ALAS-N and ALAS-E were both detectable in untreated cells, without any cross hybridization. The size of the transcript
313 hybridized with the ALAS-E RNA probe was smaller by = 200 bases than that of the transcript hybridized with the ALAS-N RNA probe (2.3 kb vs. 2.5 kb). A similar finding was reported for the two ALAS mRNAs in chicken reticulocytes [16].
~)
1.0
m
i-. o
O 0 0
oo
Time-courses of changes in mRN,4s Changes in the level of mRNAs encoding heme pathway enzymes and mRNAs for three ubiquitous proteins, i.e., hsp70, ALAS-N and HO, were examined as a function of time in MEL cells after treatment with DMSO. The first change was a rapid and marked decline in hsp70 mRNA (to a 36% of the control) at 6 h, which then gradually increased to the control level by 24 h (Fig. 2). Next changes were the reduction in the level of ALAS-N and HO mRNAs, both of which occurred between 6 h and 12 h. At 6 h, the levels of ALAS-N and HO mRNAs were already lower than the untreated control (80-88% of the control; n.s.), and both mRNAs were markedly decreased at 12 h ( = 40% of the control). They continued to decline to reach the nadir at 18 h ( < 30% of the control), and then returned to a 50-60% level of the control at 24 h. Transcripts encoding heme biosynthetic pathway enzymes were found to increase following DMSO treatment, but much later than the changes in mRNAs for the three ubiquitous proteins (Figs. 2 and 3). The first significant increase occurred at 12 h, 18 h, 18-24 h and 24 h, for mRNAs encoding ALAS-E, ALAD, PBGD and UROD, respectively. These mRNAs continued to increase thereafter during the experimental period, and their levels at 48 h were 14.8, 11.6 4.7 and 4.4-fold that of the control, for ALAS-E, ALAD, PBGD and UROD, respectively (Fig. 3). No significant changes in fl-actip mRNA concentrations were observed during the experimental period (Fig. 3).
e o
s. =.
0.5
0
0
•
I
I
I
6
12
18
24
Time (hi
1
'
!
28S---..~
O ~ ....
ALAS.N
mm
18S
iiili
28S HO 18S
I
I
28S
JW70 18S"~Dl
I
t
I
I
I
I
I
I
28S.----~ B-Actin y,
.~
,
I
I)MS0
-
+
I
-
1-
I
I
-
t-
,
I -
1-
Fig. 2. Time-courses of changes in mRNA concentrations for hsp70,
HO and ALAS-N in MEL cells. M E t cells were incubated for 24 h, either in the absence or presence of i.5% (v/v) DMSO. o, hsp70; o, HO; O, ALAS-N.
Changes in newly transcribed mRNAs 28S----~
18 S ~
I
i
i
;:~,
:~."" "
I,..,~:?. ,-. ,'~
"
,::: •" .~ "
.':~,,r; ,~. ,~>,~,
,.
Fig. 1. Northern blot analysis of ALAS mRNAs. Northern blot analysis of total cellular RNA was carried out as described in Materials and Methods. A sheet of Zeta-probe blotting membrane was divided longitudinally into two-halves, and each piece was hybridized, either (A) with a 3zP-labeled ALAS-N RNA probe, or (B) with a 32p-labeled ALAS-E RNA probe.
To examine whether the observed increases in mRNA concentrations are due to the activation of their genes, runoff transcription analysis was performed. The levels of the newly synthesized mRNAs in DMSO-induced cells were > 8, 4, 1.6 and 1.5-fold of the untreated control, for ALAS-E, ALAD, PBGD and UROD, respectively (Fig. 4). Relative concentrations of newly synthesized mRNAs in the induced cells were 1:6.4:77:6.6 for ALAS-E, ALAD, PBG and UROD, respectively. In contrast, the levels of newly transcribed mRNAs for ALAS-N and HO were 45 + 20% (n = 3J and 75 _+ 15% (n = 3) of the untreated control, respectively, whereas those for hsp70 and /~actin showed little changes.
314
ALAD
i! ,0
j,s
4~
i
I
i
I
i
6
12
18
24
48
HO
" "
PBGD
hsp70
~ .,.~11
UROD
8-Actin
m
DMSO
0
~'~
ALAS-N
ALAS-E
-
-
+
+
Fig. 4, Runoff transcription assays for enzymes in the heme biosynthetic pathway and for ubiquitous proteins. M E L cells were incubated for 24 h, either in the absence or presence of 1.5% (v/v~ DMSO. Runoff transcription assay was carried out as described in Materials and Methods.
ALAS-E, ALAD, PBGD and UROD mRNAs by these inducers were in the ranges of 8.5-33.5-fold, 4.5-8-fold, 1.4-3.4-fold and 1.3-1.9-fold of the untreated control,
I IN
Al~lbll tM
0.8 Z 14,1
ALAD
0.4
c~ ,o
0.2
.J
118
OI
0
m PlaiD
0.8 188
O.e g
Im
0.4 0.2 0
UROO tN
,.oi
0 1 IN
0.8
I~otln
0.6 tim
[
Fig, 3, Time-courses of changes in mRNA concentrations for ALASE, Ai..AD, PW3D and UROD in MEL cells. MEL cells were incubated for 48 h, either in the absence or presence of 1.5% (v/v)
0.4 0.2
0
l
0
DMSO, e, ALAS-E: o, A L A D : ~ , PBGD; II, UROD. 0.8
= o.o
Effects o/carious inducers of erythroid differentiation on gene actication Effects of four chemically unrelated inducers of erythroid differentiation on the changes in mRNA levels are shown in Fig. 5. Treatment of MEL cells with either DMSO [1], HMBA [30], BA [31] or NMN [32] for 24 h, increased the level of mRNAs for ALASE, ALAD, PBGD and UROD. These findings suggest that increases in these mRNAs are characteristic of erythroid differentiation, and not restricted to treatment with DMSO. The magnitudes of induction of
|
0.4 0.2 0
0
m
2 ¢~
Q
Z
Z •
CJ a
Z
=l=m
Fig. 5. Effects of various chemicals on the level of mRNAs encoding enzymes in the heme biosynthetic pathway and ubiquitous proteins in MEL cells. MEL cells were incubated with the chemical for 24 h. Northern blot analysis of total cellular RNA was carried out as described in Materials and Methods.
315 respectively. In contrast, the level of mRNAs for ubiquitous enzymes such as ALAS-N, HO and hsp70 showed early and marked decreases to 0.12-0.35-fold, 0.06-0.16-fold and 0.31-0.65-fold of the control, respectively (Fig. 5). Thus, decreases in the ubiquitous mRNAs were also observed in cells treated by each of the four chemicals, suggesting that they are also associated with the cell differentiation event. Discussion
Our findings in this study demonstrate that significant increases in the accumulation of mRNAs encoding ALAS.E, ALAD, PBGD and UROD occur during erythroid differentiation of MEL cells (Fig. 3). It is known that treatment of MEL cells with actinomycin D prevents the DMSO-mediated increases in enzyme activities in the heme biosynthetic pathway [13], suggesting that the sequential increases in the enzymatic activities are the result of transcriptional activation of genes for these enzymes. Direct demonstration of gene activation was subsequently shown for ALAS-E [21], PBGD [33] and U ROD [33], and de novo synthesis of ALAS [34] and ALAD [35] were also demonstrated in MEL cells following DMSO treatment. The magnitudes of increases in mRNAs for these enzymes were also similar to those of the newly transcribed mRNAs (Fig. 4). These findings suggest that increases in mRNAs encoding ALAS-E, ALAD, PBGD and UROD are largely due to the transcriptional activation of genes for these enzymes. Increases in mRNAs encoding enzymes in the heme biosynthetic pathway took place sequentially in the order as these enzymes appear in their biosynthetic sequence, similar to sequential increases in their enzymatic activities [13,15]. Sequential increases in heme biosynthetic enzymes were also reported during erythroid differentiation of K562 human erythroleukemia cells [36], while others did not find sequential increases of these enzymes in MEL cells [33]. Inconsistencies among these findings are probably in part clue to the fact that these changes were reported for each mRNA or enzyme activity separately by different laboratories. Since there are significant differences in time-courses of these changes depending on culture conditions, it is difficult to compare a finding reported by one laboratory to those reported by others, or from one experiment to another. Our findings on the sequential activation of heme pathway genes are especially important in this respect that the level of each mRNA at a given time point was determined by the analysis of the identical RNA extract from a single time-course of a representative experiment, thus all changes are comparable to each other. In our study, changes in mRNA concentrations were corrected against the time-dependent changes in mRNAs in untreated cells using conowrrent
untreated control cells, thus all data can be compared with each other. It is clear from our findings that there are sequential increases in mRNAs encoding heme biosynthetic enzymes, and ALAS-E mRNA is the first to increase. The commitment period for these cells was reported to be 12-16 h [37], thus the onset of increases in ALAS-E mRNA appears to occur at around the commencement of the commitment. In contrast to ALAS-E mRNA, or mRNAs for other heme biosynthetic pathway enzymes, which showed marked increases in induced MEL cells, the level of ALAS-N mRNA rapidly declined during the latent period of differentiation, i.e., < 12 h, and remained below the untreated control level throughout the incubation period up to 48 h. A similar change in ALAS-N mRNA was also observed in cells treated with HMBA, BA or NMN Thus, the rapid and marked decline of ALAS-N mRNA is also a characteristic finding occurring during the latent period of cell differentiation. In addition to ALAS-N mRNA, mRNAs for both HO and hsp70 showed an early decline (Fig. 2). All these genes are ubiquitously expressed in many tissues, and regulated not only by their specific inducers, but also by various other factors [7,38]. In this respect, they are different from the genes in the heme biosynthetic pathway which show specific increases only in developing erythroid cells [39]. It is significant to note that the earliest change in induced MEL cells were the suppression of these ubiquitous gene expression. HO is a rate-limiting enzyme for heme catabolism, which is inducible in various cells following treatment with hemin and a variety of chemicals [40]. In MEL cells, however, its activity was shown t~ !,,~ ~ r k e d l y decreased, and its reduction in activity is considered to conserve heme for hemoglobin formation [41]. A similar finding was also reported in the human !(562 erythroleukemia cells [42]. Recently, it has been demonstrated that HO itself is a 32 kDa heat shock protein [43]. In this respect, it is interesting to note that both HO and hsp70 showed a rapid and marked decline during the latent period after DMSO treatment [6]. Time-courses of ttO and hsp70 mRNAs, however, were not closely related to each other (Fig. 2). This finding suggests that, although both proteins are heat-shock proteins, there must be certain differences in the gene expression of HO and hsp70, as has been observed in other systems [44]. In contrast, the levels of ALAS-N mRNA and HO mRNA declined, and recovered in a very similar fashion (Fig. 2). This finding suggests that there may be a coordinate regulation between ALAS-N and HO, the rate-limiting enzyme of heine biosynthesis and catabolism in non-erythroid tissues, respectively, in MEL cells undergoing erythroid differentiation. Thus, our findings in this study indicate that the inducer of erythroid differentiation first suppresses the expression of uhiquitou~ genes during the latent pe-
316 riod, which is then followed by sequential increases in the mRNAs for enzymes in the heme biosynthetic pathway. Induction of ALAS-E mRNA coincides approximately with the commitment period of these cells for erythroid differentiation. Since cellular differentiation is associated not only with the expression of specific genes of the differentiated cells, but also with the loss of housekeeping proteins, the observed early decreases in mRNAs encoding ubiquitous proteins may be an important determinant involved in the differentiation process of MEL cells. At the present time, the mechanism for such rapid decreases in the expression of ubiquitous genes is not clear, but studies toward elucidation of this question are under way in our laboratory. Decreased expression of HO will also contribute to the conservation of heine for increased hemoglobin synthesis in the differentiated cells [41,42].
Acknowledgements This work was supported in part by a grant from USPHS DK-32890. The excellent technical assistance of Ms. Luba Garbaczewski and Mr. Jotham Lefford is gratefully acknowledged.
References 1 Friend, C.. Scher, W., Holland, J.G. and Sato, T, (1971) Proc. Natl. Acad. Sci. USA 68. 378-383. 2 Marks, P,A, and Rifkind, R.A, (1978) Annu. Rev. Biochem. 47, 419-448, 3 Lannigan, D,A, and Knauf, P,A. (1985) J, Biol, Chem. 260, 7322-7324. 4 Gazitt, Y,, Reuben, R,C,, Deitch, A,D,, Marks, P,A. and Rifkind, R,A, (1978)Cancer Res. 38, 3779-3783, 5 Melloni, E., Pontremolli, S., Michctti, M., Sacco, 0., Cakiroglu, A.G,, Jackson, J,F., Rifkind, R.A, and Marks, P.A. (1987) Prec. Natl, Acad, Sci. USA 84, 5282-5286. 6 Fujita, H, and Sa~a, S, (1989) Br, J. Haematol. 73, 557-560. 7 Hensold, J.O, and Housman, D. (1988) Mol. Cell Biol. 8, 22192223, 8 Ram~y, R,G,, ikeda, K., Rifkind, R,A. and Marks, P.A. (1986) Pmc. Natl, Acad. Sci, USA 83, 6849-6853. 9 Tadokoro, K. and ikav.a, Y. (1986) Biochem. Biophys. Res. Commun, 135, 1112-1118. 10 Gusella, J., Geller, R,, Clarke, B., Weeks, V. and Housman, D. (1976) Cell 9, 221-229, I! Ross, J,, Ikawa, Y. and Leder, P. (1972) Prec. Natl. '~cad. Sci. USA 69, 3620-3623, 12 Ohta, Y,, Tanal~a, M., Terada, M., Miller, O.J., Bank. A.. Marks, P,A. and Rifkind, R.A, (1976, Proc, Natl. Acad. Sci. USA 73, 1232- ! 236. 13 Sassa, S, (1975)J, Exp, Med, 143, 305-315,
14 Ikawa, Y., Aida, M. and Saito, T. (1975) Gann 66, 583-584. 15 Rutherford, T., Thompson, G.G. and Moore, M.R. (1979) Prec. Natl. Acad. Sci. USA 76, 833-836. 16 Riddle, R.D., Yamamoto, M. and Engel, J.D. (1989) Prec. Natl. Acad. Sci. USA 86, 792-796. 17 Singer, D., Cooper, M., Maniatis, G., Marks, P.A. and Rifkind, R.A. (1974) Prec. N-,.tl. Acad. Sci. USA 71, 2668-2670. 18 Sassa, S. and Kappas, A. (1977) J. Biol. Chem. 252, 2428-2436. 19 Yamamoto, M., Kure, S, Engel, J.D. and Hiraga, K. (1988) J. Biol. Chem. 263, 15973-15979. 20 Yamagami, T., Yamamoto, M., Kosakai, M., Munakata, H. and Hayashi, N. (1989) Seikagaku 61,817(Abs). 21 Schoenhaut, D.S. and Curtis, P.J. (1986) Gene 48, 55-63. 22 Shibahara, S., Muller, R., Tagushi, H. and Yoshida, T. (1985) Prec. Natl. Acad. Sci. USA 82, 7865-7869. 23 Bishop, T.R., Cohen, P.J., Boyer, S.H., Noyes, A.N. and Frelin, L.P. (1986) Prec. Natl. Acad. Sci. USA 83, 5568-5572. 24 De Verneuil, H., Grandchamp, B., Romeo, P.-H., Reich, N., Beaumont, C., Goosens, M., Nicolas, H. and Nordmann, Y. (1986) J. Clin Invest 77, 431-435. 25 Wu, B., Hunt, C. and Morimoto, R. (1~5) Mol. Cell Biol. 5, 330-341. 26 Melton, D.A., Krieg, P.A., Rebagliati, M.R., Maniatis, T., Zinn, K. and Green, M.R. (1984) Nucleic Acids Res. 12, 7035-7056. 27 Cathala, G., Savouret, J-F., Mcndez, B. et al. (1983) DNA 2, 329-335. 28 Maniatis, T., Fritsch, E.F. and Sambrook, J. (1982) Molecular Cloning, A laboratory manual, Cold Spring Harbor Laboratory, Cold Spring Harbor. 29 Ausubel, F.M., Brent, R., Kingston, R.E. et al. (1990) Current Protocols in Molecular Biology, John Wiley, New York. 30 Reuben, R.C., Khanna, P.L., Gazitt, Y., Breslow, R., Rifkind, R.A. and Marks, P.A. (1978) J. Biol. Chem. 253, 4214-4218. 31 Leder, A. and Leder, P. (1975) Cell 5, 319-322. 32 Terada, M., Fujiki, H., Marks, P.A. and Sugimura, T. (1979) Prec. Natl. Acad. Sci. USA 76, 6411-6414. 33 Grandchamp, B., Beaumont, C., De Verneuil, H. and Nordmann, Y. (1985)J. Biol. Cllem. 260, 9630-9635. 34 EIferink, C.J., Sassa, S. and May, B.K. (1988) J. Biol. Chem. 263, 13012-13016. 35 Chang, C.S. and Sassa, S. (1984) Blood 64, 64-70. 36 Hoffman, R., lbrahim, N., Murnane, M.J., Diamond, A., Forget, B.G. and Levere, R.D. (1980) Blood 56, 567-570. 37 Fibach, E., Gambari, R., Shaw, P.A., Maniatis, G., Reuben, R.C., Sassa, S., Rifkind, R.A. and Marks, P.A. (1979) Prec. Natl. Acad. Sci. USA 76, 1906-1910. 38 Lindquist, S. (1986) Annu. Rev. Biochem. 55, 1151-1191. 39 Sassa, S. (1990) Int. J. Cell Cloning 8, 10-26. 40 Shibahara, S. (1988) Semin. Hematol. 25, 370-376. 41 Sassa, S. (1983) in Regulation of Hemoglobin Biosynthesis (Goldwasser, E., ed.), pp. 359-383, Elsevier, Amsterdam. 42 Abraham, N.G., Mitrion¢, S.M., Hodgson, W.J., Levere, R.D. and Shibahara, S. (1988) Adv. Exp. Med. Biol. 241, 97-116. 43 Shibahara, S., Muller, R.M. and Taguchi, H. (1987) J. Biol. Chem. 262, 12889-12892. 44 Mitani, K., Fujita, H., Sassa, S. and Kappas, A. (1990) Biochem. Biophys. Res. Comm 166, 1429-1434.
311
BBAEXP 923111
Sequential activation of genes for heme pathway enzymes during erythroid differentiation of mouse Friend virus-transformed erythroleukemia cells Hiroyoshi Fujita ~, Masayuki Yamamoto 2, Takashi Yamagami 2, Norio Hayashi 2, Terry R. Bishop 3, Hubert De Verneuil 4, Takeo Yoshinaga 5, Shigeki Shibahara 2, Richard Morimoto 6 and Shigeru Sassa i t The Rockefelh'r Uniremity, New York, NY (U. S.A.), 2 Departments o.f Biochemisoy and Apldied l'hy.~iology, Tohoku Unit'emit.v, Semlai (Japan), ~ Deparnnent of Medical Gem'tic.~', The Johns Hopkins Uniremity, Bahimore, MD (U.S.A.), 4 I.'aculte ~h' Medecim,, X. Bichat, Paris (Fram'e), "~Department of Public Ih,alth, Kyoto Unil'erviO, Kyoto (JapatD arid r, Dei~arlmenL~'of Molecuho" Bioh~,q' and ('ell BiohJgy, Northwe.~'n'rn Unirer~'ity, l'h'amlon, IL (U.S.A.)
(Received 14 May 1991)
Key words: lleme synthesis; Erythroleukemia cell; c$-Aminolevulinate synthase; ! leme oxygenase; Sequential activation
Changes in the level of transcripts encoding enzymes of the heme biosynthetic pathway as well as those encoding ubiquitous proteins were examined in murine Friend virus-transformed erythroleukemia cells during erythroid cell differentiation induced by chemicals including dimethyl sulfoxide (DMSO). Early changes following DMSO treatment were marked decreases in mRNAs for three ubiquitous proteins, i.e., a 70 kDa heat shock protein ( < 6 h), heme oxygenase and nonspecific ~-aminolevulinate synthase (ALAS) ( < 12 h). These changes were followed by sequential increases in mRNAs for enzymes in the heme biosynthetic pathway. Namely, mRNAs for the erythroidspecific ALAS, 8-aminolevtllinate dehydratase, porphobilinogen deaminase and uroporphyrinogen decarboxylase started to increase at 12, 18, 18-24 and 24 h, respectively. Nuclear runoff studies revealed that these changes are largely transcriptional. Treatments with otI~er inducers of erythroid differentiation, e.g., hexamethylene bisacetamide, n-butyric acid and N'-methylnicotinamide, also showed simi!ar effects on mRNAs as those fo:lowing DMSO. These findings suggest that both suppression of ubiquitous genes and activation of heme pathway enzyme genes are associated with erythroid differentiation, and the former occurs preceding changes in the latter.
Introduction
Murine Friend virus-transformed crythroleukemia (MEL) cells are erythroid precursor cells transformed by the Friend leukemia virus complex, which serve as a useful system for in vitro studies of erythroid cell differentiation [1]. These cells can be induced, by a variety of agents including DMSO, to initiate the pro-
Abbreviations: ALAD, ~5-aminolevulinate dehydratase; ALAS, 6aminolevulinate synthase; ALAS-E, the erythroid-specific ALAS; ALAS-N, the non-specific ALAS; BA, butyric acid; DMSO, dimethyl sulfoxide; HMBA, hexamethylene bisaeetamide; HO, heme oxygenase; hsp70, heat-shock protein 70; MEL, murine erythroleukemia; NMN, N'-methylnicotinamide; PBGD, porphobilinogen deaminase; UROD, uroporphyrinogen decarboxylase. Correspondence: S. Sassa, The Rockefeller University, New York, NY 10021, U.S.A.
gram of terminal erythroid differentiation [2]. Prior to the irreversible commitment to differentiation, a number of metabolic changes occur, e.g., alterations in membrane permeability [3], a transient increase in cAMP concentration [4], an increase in membrane-associated protein kinase C activity [5], a decrease in heme oxygenase (HO) (EC 1.14.99.3) mRNA [6], and the modulation of a number of genes, including c-myb, c-myc, c-los, p53 and the 70 kDa heat-shock protein (hsp70) [7-9]. When MEL cells are incubated with an inducing chemical beyond the latent period, they become irreversibly committed to undergo erythroid differentiation [10], accumulate a- and /3-globin mRNA [11,12], increase enzymatic activities of the heme b,iosynthetic pathway [13], and exhibit erythroid-specific membrane antigens [14]. Increases in activities of enzymes in the heine biosynthetic pathway occur sequentially in the order as they appear in their biosynthetic sequence, i.e., ~-aminolevulinate synthase (ALAS) (EC
312 2.3.1.37) being the first, and take place substantially earlier than the accumulation of heine [13,15]. Recently, it has been shown that two separate genes exist for ALAS in chicken reticulocytes, one encoding the erythroid-specific form (ALAS:E) and the other encoding the hepatic, or the non-specific form (ALAS-N) [16]. Although increases in enzymatic activities of heine pathway enzymes and their time-courses were examined, little is known about their gene activation in MEL cells, particularly that of the ALAS-N isozyme. There was also no study in which changes in mRNA levels for several key enzymes in the heme biosynthetic and the catabolic pathways were examined in a systematic manner. in this study, we examined increases in the levels of transcripts of the genes for certain enzymes in the heme biosynthetic pathway which show early and characteristic increases during crythroid differentiation of MEL cells. These genes include those for ALAS-E, 8-aminolevulinate dehydratase (ALAD) (EC 4.2.1.24), porphobilinogen deaminase (PBGD) (EC 4.3.1.8) and uroporphyrinogen decarboxylase (UROD) (EC 4.1.1.37). We also examined the changes in the levels of mRNAs for three ubiquitously expressed genes, i.e., ALAS-N, HO and hsp70. Materials and Methods
Cell cultures. A clone of MEL cells, DS-19 [17], was a generous gift from Dr. R.A. Rifkind, Memorial SIoan.Kettering Cancer Center, New York, NY. Cells were grown in suspension in a modified Ham's FI2 medium [18] containing 10% (v/v) heat-inactivated fetal bovine serum. Cultures were routinely split every 3-4 days, to yield a cell density of 5.104 cells per ml, to maintain a logarithmic growth. In each experiment, cells from a 24-48 h old culture were suspended in a fresh medium at a cell density of 5.104 cells per ml, and incubated for 16 h, prior to the addition of chemicals. Incubations were then continued for various periods as indicated in the figures. eDNA probes. The eDNA probe for ALAS-N used was a rat ALAS-N eDNA pKRA2cA [19]. The eDNA probe for ALAS-E used was a 1.2 kb rat ALAS-E eDNA fragment cloned from a eDNA library constructed in Agtll from rat anemic spleen poly(A) + RNA [20]. eDNA was screened by the antibody against the rat ALAS-N which also recognizes the rat ALAS-E [20]. The ALAS-E eDNA fragment corresponds to a region starting from 733 bp to 1888 bp of the reported nucleotide sequence of the mouse ALAS-E eDNA [21]. The nucleotide sequence of the probe was 96%, 65% and 75% similar to that of the corresponding nucleotide in mouse ALAS-E [21], chicken ALAS-E eDNA [16] and rat ALAS-N eDNA [19], respectively. In addition, rat heine oxygenase cDNA (pRHO-1) [22],
rat ALAD eDNA (pALAD-1) [23], rat PBGD eDNA (p44SB-1), human UROD eDNA (pUD3) [24] and human hsp70 (pH2.3) [25] were also used for the quantification of mRNA concentrations. Each cDNA fragment was inserted into a pGEM4z vector (Promega, Madison, Wi) for the preparation of an RNA probe, according to the method of Melton et al. [26]. Chicken /3-actin eDNA was nick translated and used as an internal control. Northern blot analysis. Total RNA was isolated from ! • 1 0 7 cells according to the method of Cathala et al. [27]. 20/~g of total RNA were analyzed by Northern blot analysis as described previously [28]. After blotting to a sheet of Zeta-probe blotting membrane (Bio-Rad, Richmond, CA), RNA samples were hybridized with a radioactively labeled specific RNA probe for each protein, which were then treated with RNase A (1/~g/ml), followed by washing under a stringent condition. Levels of mRNAs were quantitated by densitometry using a LKB Ultroscan XL densitometer. The level of/3-actin mRNA was used as a reference control in each experiment. Since the levels of mRNAs for some heine pathway enzymes in untreated control cells showed significant changes as a function of incubation periods, compared with insignificant changes in /3-actin controls, mRNA contents in induced cells were expressed as ratios of those observed in the concurrent untreated controls. Nuclear runoff transcription assay. Nuclei were isolated from 4.107 uninduced or DMSO-induced cells (24 h treatment). Runoff transcription assay was performed as described previously [29]. The yield of 32p. labeled RNA transcripts was 1.9.10 7 cpm and 1.5.10 7 cpm for the control and the induced cells, respectively. An equivalent amount of radioactivity (1.1.107 cpm) from both RNA samples was hybridized to DNA immobilized on a strip of nitrocellulose according to the method described previously [29]. Quantification was made by densitometry using a LKB Ultroscan XL densitometer. Chemicals. DMSO (spectre grade) was purchased from Fisher, Fair Lawn, NJ; hexamethylene bisacetamide (HMBA) was from Aldrich, Milwaukee, Wl; n-butyric acid (BA) and N'-methylnicotinamide (NMN) were from Sigma, St. Louis, MI; and all tissue culture materials were from Gibco, Grand Island, NY. Results
Expression of ALAS-N and ALAS-E mRNAs in uninduced MEL cells Northern blot analysis of total RNA from DS-19 cells using RNA probes for ALAS-N and ALAS-E is shown in Fig. 1. mRNAs encoding ALAS-N and ALAS-E were both detectable in untreated cells, without any cross hybridization. The size of the transcript
313 hybridized with the ALAS-E RNA probe was smaller by = 200 bases than that of the transcript hybridized with the ALAS-N RNA probe (2.3 kb vs. 2.5 kb). A similar finding was reported for the two ALAS mRNAs in chicken reticulocytes [16].
~)
1.0
m
i-. o
O 0 0
oo
Time-courses of changes in mRN,4s Changes in the level of mRNAs encoding heme pathway enzymes and mRNAs for three ubiquitous proteins, i.e., hsp70, ALAS-N and HO, were examined as a function of time in MEL cells after treatment with DMSO. The first change was a rapid and marked decline in hsp70 mRNA (to a 36% of the control) at 6 h, which then gradually increased to the control level by 24 h (Fig. 2). Next changes were the reduction in the level of ALAS-N and HO mRNAs, both of which occurred between 6 h and 12 h. At 6 h, the levels of ALAS-N and HO mRNAs were already lower than the untreated control (80-88% of the control; n.s.), and both mRNAs were markedly decreased at 12 h ( = 40% of the control). They continued to decline to reach the nadir at 18 h ( < 30% of the control), and then returned to a 50-60% level of the control at 24 h. Transcripts encoding heme biosynthetic pathway enzymes were found to increase following DMSO treatment, but much later than the changes in mRNAs for the three ubiquitous proteins (Figs. 2 and 3). The first significant increase occurred at 12 h, 18 h, 18-24 h and 24 h, for mRNAs encoding ALAS-E, ALAD, PBGD and UROD, respectively. These mRNAs continued to increase thereafter during the experimental period, and their levels at 48 h were 14.8, 11.6 4.7 and 4.4-fold that of the control, for ALAS-E, ALAD, PBGD and UROD, respectively (Fig. 3). No significant changes in fl-actip mRNA concentrations were observed during the experimental period (Fig. 3).
e o
s. =.
0.5
0
0
•
I
I
I
6
12
18
24
Time (hi
1
'
!
28S---..~
O ~ ....
ALAS.N
mm
18S
iiili
28S HO 18S
I
I
28S
JW70 18S"~Dl
I
t
I
I
I
I
I
I
28S.----~ B-Actin y,
.~
,
I
I)MS0
-
+
I
-
1-
I
I
-
t-
,
I -
1-
Fig. 2. Time-courses of changes in mRNA concentrations for hsp70,
HO and ALAS-N in MEL cells. M E t cells were incubated for 24 h, either in the absence or presence of i.5% (v/v) DMSO. o, hsp70; o, HO; O, ALAS-N.
Changes in newly transcribed mRNAs 28S----~
18 S ~
I
i
i
;:~,
:~."" "
I,..,~:?. ,-. ,'~
"
,::: •" .~ "
.':~,,r; ,~. ,~>,~,
,.
Fig. 1. Northern blot analysis of ALAS mRNAs. Northern blot analysis of total cellular RNA was carried out as described in Materials and Methods. A sheet of Zeta-probe blotting membrane was divided longitudinally into two-halves, and each piece was hybridized, either (A) with a 3zP-labeled ALAS-N RNA probe, or (B) with a 32p-labeled ALAS-E RNA probe.
To examine whether the observed increases in mRNA concentrations are due to the activation of their genes, runoff transcription analysis was performed. The levels of the newly synthesized mRNAs in DMSO-induced cells were > 8, 4, 1.6 and 1.5-fold of the untreated control, for ALAS-E, ALAD, PBGD and UROD, respectively (Fig. 4). Relative concentrations of newly synthesized mRNAs in the induced cells were 1:6.4:77:6.6 for ALAS-E, ALAD, PBG and UROD, respectively. In contrast, the levels of newly transcribed mRNAs for ALAS-N and HO were 45 + 20% (n = 3J and 75 _+ 15% (n = 3) of the untreated control, respectively, whereas those for hsp70 and /~actin showed little changes.
314
ALAD
i! ,0
j,s
4~
i
I
i
I
i
6
12
18
24
48
HO
" "
PBGD
hsp70
~ .,.~11
UROD
8-Actin
m
DMSO
0
~'~
ALAS-N
ALAS-E
-
-
+
+
Fig. 4, Runoff transcription assays for enzymes in the heme biosynthetic pathway and for ubiquitous proteins. M E L cells were incubated for 24 h, either in the absence or presence of 1.5% (v/v~ DMSO. Runoff transcription assay was carried out as described in Materials and Methods.
ALAS-E, ALAD, PBGD and UROD mRNAs by these inducers were in the ranges of 8.5-33.5-fold, 4.5-8-fold, 1.4-3.4-fold and 1.3-1.9-fold of the untreated control,
I IN
Al~lbll tM
0.8 Z 14,1
ALAD
0.4
c~ ,o
0.2
.J
118
OI
0
m PlaiD
0.8 188
O.e g
Im
0.4 0.2 0
UROO tN
,.oi
0 1 IN
0.8
I~otln
0.6 tim
[
Fig, 3, Time-courses of changes in mRNA concentrations for ALASE, Ai..AD, PW3D and UROD in MEL cells. MEL cells were incubated for 48 h, either in the absence or presence of 1.5% (v/v)
0.4 0.2
0
l
0
DMSO, e, ALAS-E: o, A L A D : ~ , PBGD; II, UROD. 0.8
= o.o
Effects o/carious inducers of erythroid differentiation on gene actication Effects of four chemically unrelated inducers of erythroid differentiation on the changes in mRNA levels are shown in Fig. 5. Treatment of MEL cells with either DMSO [1], HMBA [30], BA [31] or NMN [32] for 24 h, increased the level of mRNAs for ALASE, ALAD, PBGD and UROD. These findings suggest that increases in these mRNAs are characteristic of erythroid differentiation, and not restricted to treatment with DMSO. The magnitudes of induction of
|
0.4 0.2 0
0
m
2 ¢~
Q
Z
Z •
CJ a
Z
=l=m
Fig. 5. Effects of various chemicals on the level of mRNAs encoding enzymes in the heme biosynthetic pathway and ubiquitous proteins in MEL cells. MEL cells were incubated with the chemical for 24 h. Northern blot analysis of total cellular RNA was carried out as described in Materials and Methods.
315 respectively. In contrast, the level of mRNAs for ubiquitous enzymes such as ALAS-N, HO and hsp70 showed early and marked decreases to 0.12-0.35-fold, 0.06-0.16-fold and 0.31-0.65-fold of the control, respectively (Fig. 5). Thus, decreases in the ubiquitous mRNAs were also observed in cells treated by each of the four chemicals, suggesting that they are also associated with the cell differentiation event. Discussion
Our findings in this study demonstrate that significant increases in the accumulation of mRNAs encoding ALAS.E, ALAD, PBGD and UROD occur during erythroid differentiation of MEL cells (Fig. 3). It is known that treatment of MEL cells with actinomycin D prevents the DMSO-mediated increases in enzyme activities in the heme biosynthetic pathway [13], suggesting that the sequential increases in the enzymatic activities are the result of transcriptional activation of genes for these enzymes. Direct demonstration of gene activation was subsequently shown for ALAS-E [21], PBGD [33] and U ROD [33], and de novo synthesis of ALAS [34] and ALAD [35] were also demonstrated in MEL cells following DMSO treatment. The magnitudes of increases in mRNAs for these enzymes were also similar to those of the newly transcribed mRNAs (Fig. 4). These findings suggest that increases in mRNAs encoding ALAS-E, ALAD, PBGD and UROD are largely due to the transcriptional activation of genes for these enzymes. Increases in mRNAs encoding enzymes in the heme biosynthetic pathway took place sequentially in the order as these enzymes appear in their biosynthetic sequence, similar to sequential increases in their enzymatic activities [13,15]. Sequential increases in heme biosynthetic enzymes were also reported during erythroid differentiation of K562 human erythroleukemia cells [36], while others did not find sequential increases of these enzymes in MEL cells [33]. Inconsistencies among these findings are probably in part clue to the fact that these changes were reported for each mRNA or enzyme activity separately by different laboratories. Since there are significant differences in time-courses of these changes depending on culture conditions, it is difficult to compare a finding reported by one laboratory to those reported by others, or from one experiment to another. Our findings on the sequential activation of heme pathway genes are especially important in this respect that the level of each mRNA at a given time point was determined by the analysis of the identical RNA extract from a single time-course of a representative experiment, thus all changes are comparable to each other. In our study, changes in mRNA concentrations were corrected against the time-dependent changes in mRNAs in untreated cells using conowrrent
untreated control cells, thus all data can be compared with each other. It is clear from our findings that there are sequential increases in mRNAs encoding heme biosynthetic enzymes, and ALAS-E mRNA is the first to increase. The commitment period for these cells was reported to be 12-16 h [37], thus the onset of increases in ALAS-E mRNA appears to occur at around the commencement of the commitment. In contrast to ALAS-E mRNA, or mRNAs for other heme biosynthetic pathway enzymes, which showed marked increases in induced MEL cells, the level of ALAS-N mRNA rapidly declined during the latent period of differentiation, i.e., < 12 h, and remained below the untreated control level throughout the incubation period up to 48 h. A similar change in ALAS-N mRNA was also observed in cells treated with HMBA, BA or NMN Thus, the rapid and marked decline of ALAS-N mRNA is also a characteristic finding occurring during the latent period of cell differentiation. In addition to ALAS-N mRNA, mRNAs for both HO and hsp70 showed an early decline (Fig. 2). All these genes are ubiquitously expressed in many tissues, and regulated not only by their specific inducers, but also by various other factors [7,38]. In this respect, they are different from the genes in the heme biosynthetic pathway which show specific increases only in developing erythroid cells [39]. It is significant to note that the earliest change in induced MEL cells were the suppression of these ubiquitous gene expression. HO is a rate-limiting enzyme for heme catabolism, which is inducible in various cells following treatment with hemin and a variety of chemicals [40]. In MEL cells, however, its activity was shown t~ !,,~ ~ r k e d l y decreased, and its reduction in activity is considered to conserve heme for hemoglobin formation [41]. A similar finding was also reported in the human !(562 erythroleukemia cells [42]. Recently, it has been demonstrated that HO itself is a 32 kDa heat shock protein [43]. In this respect, it is interesting to note that both HO and hsp70 showed a rapid and marked decline during the latent period after DMSO treatment [6]. Time-courses of ttO and hsp70 mRNAs, however, were not closely related to each other (Fig. 2). This finding suggests that, although both proteins are heat-shock proteins, there must be certain differences in the gene expression of HO and hsp70, as has been observed in other systems [44]. In contrast, the levels of ALAS-N mRNA and HO mRNA declined, and recovered in a very similar fashion (Fig. 2). This finding suggests that there may be a coordinate regulation between ALAS-N and HO, the rate-limiting enzyme of heine biosynthesis and catabolism in non-erythroid tissues, respectively, in MEL cells undergoing erythroid differentiation. Thus, our findings in this study indicate that the inducer of erythroid differentiation first suppresses the expression of uhiquitou~ genes during the latent pe-
316 riod, which is then followed by sequential increases in the mRNAs for enzymes in the heme biosynthetic pathway. Induction of ALAS-E mRNA coincides approximately with the commitment period of these cells for erythroid differentiation. Since cellular differentiation is associated not only with the expression of specific genes of the differentiated cells, but also with the loss of housekeeping proteins, the observed early decreases in mRNAs encoding ubiquitous proteins may be an important determinant involved in the differentiation process of MEL cells. At the present time, the mechanism for such rapid decreases in the expression of ubiquitous genes is not clear, but studies toward elucidation of this question are under way in our laboratory. Decreased expression of HO will also contribute to the conservation of heine for increased hemoglobin synthesis in the differentiated cells [41,42].
Acknowledgements This work was supported in part by a grant from USPHS DK-32890. The excellent technical assistance of Ms. Luba Garbaczewski and Mr. Jotham Lefford is gratefully acknowledged.
References 1 Friend, C.. Scher, W., Holland, J.G. and Sato, T, (1971) Proc. Natl. Acad. Sci. USA 68. 378-383. 2 Marks, P,A, and Rifkind, R.A, (1978) Annu. Rev. Biochem. 47, 419-448, 3 Lannigan, D,A, and Knauf, P,A. (1985) J, Biol, Chem. 260, 7322-7324. 4 Gazitt, Y,, Reuben, R,C,, Deitch, A,D,, Marks, P,A. and Rifkind, R,A, (1978)Cancer Res. 38, 3779-3783, 5 Melloni, E., Pontremolli, S., Michctti, M., Sacco, 0., Cakiroglu, A.G,, Jackson, J,F., Rifkind, R.A, and Marks, P.A. (1987) Prec. Natl, Acad, Sci. USA 84, 5282-5286. 6 Fujita, H, and Sa~a, S, (1989) Br, J. Haematol. 73, 557-560. 7 Hensold, J.O, and Housman, D. (1988) Mol. Cell Biol. 8, 22192223, 8 Ram~y, R,G,, ikeda, K., Rifkind, R,A. and Marks, P.A. (1986) Pmc. Natl, Acad. Sci, USA 83, 6849-6853. 9 Tadokoro, K. and ikav.a, Y. (1986) Biochem. Biophys. Res. Commun, 135, 1112-1118. 10 Gusella, J., Geller, R,, Clarke, B., Weeks, V. and Housman, D. (1976) Cell 9, 221-229, I! Ross, J,, Ikawa, Y. and Leder, P. (1972) Prec. Natl. '~cad. Sci. USA 69, 3620-3623, 12 Ohta, Y,, Tanal~a, M., Terada, M., Miller, O.J., Bank. A.. Marks, P,A. and Rifkind, R.A, (1976, Proc, Natl. Acad. Sci. USA 73, 1232- ! 236. 13 Sassa, S, (1975)J, Exp, Med, 143, 305-315,
14 Ikawa, Y., Aida, M. and Saito, T. (1975) Gann 66, 583-584. 15 Rutherford, T., Thompson, G.G. and Moore, M.R. (1979) Prec. Natl. Acad. Sci. USA 76, 833-836. 16 Riddle, R.D., Yamamoto, M. and Engel, J.D. (1989) Prec. Natl. Acad. Sci. USA 86, 792-796. 17 Singer, D., Cooper, M., Maniatis, G., Marks, P.A. and Rifkind, R.A. (1974) Prec. N-,.tl. Acad. Sci. USA 71, 2668-2670. 18 Sassa, S. and Kappas, A. (1977) J. Biol. Chem. 252, 2428-2436. 19 Yamamoto, M., Kure, S, Engel, J.D. and Hiraga, K. (1988) J. Biol. Chem. 263, 15973-15979. 20 Yamagami, T., Yamamoto, M., Kosakai, M., Munakata, H. and Hayashi, N. (1989) Seikagaku 61,817(Abs). 21 Schoenhaut, D.S. and Curtis, P.J. (1986) Gene 48, 55-63. 22 Shibahara, S., Muller, R., Tagushi, H. and Yoshida, T. (1985) Prec. Natl. Acad. Sci. USA 82, 7865-7869. 23 Bishop, T.R., Cohen, P.J., Boyer, S.H., Noyes, A.N. and Frelin, L.P. (1986) Prec. Natl. Acad. Sci. USA 83, 5568-5572. 24 De Verneuil, H., Grandchamp, B., Romeo, P.-H., Reich, N., Beaumont, C., Goosens, M., Nicolas, H. and Nordmann, Y. (1986) J. Clin Invest 77, 431-435. 25 Wu, B., Hunt, C. and Morimoto, R. (1~5) Mol. Cell Biol. 5, 330-341. 26 Melton, D.A., Krieg, P.A., Rebagliati, M.R., Maniatis, T., Zinn, K. and Green, M.R. (1984) Nucleic Acids Res. 12, 7035-7056. 27 Cathala, G., Savouret, J-F., Mcndez, B. et al. (1983) DNA 2, 329-335. 28 Maniatis, T., Fritsch, E.F. and Sambrook, J. (1982) Molecular Cloning, A laboratory manual, Cold Spring Harbor Laboratory, Cold Spring Harbor. 29 Ausubel, F.M., Brent, R., Kingston, R.E. et al. (1990) Current Protocols in Molecular Biology, John Wiley, New York. 30 Reuben, R.C., Khanna, P.L., Gazitt, Y., Breslow, R., Rifkind, R.A. and Marks, P.A. (1978) J. Biol. Chem. 253, 4214-4218. 31 Leder, A. and Leder, P. (1975) Cell 5, 319-322. 32 Terada, M., Fujiki, H., Marks, P.A. and Sugimura, T. (1979) Prec. Natl. Acad. Sci. USA 76, 6411-6414. 33 Grandchamp, B., Beaumont, C., De Verneuil, H. and Nordmann, Y. (1985)J. Biol. Cllem. 260, 9630-9635. 34 EIferink, C.J., Sassa, S. and May, B.K. (1988) J. Biol. Chem. 263, 13012-13016. 35 Chang, C.S. and Sassa, S. (1984) Blood 64, 64-70. 36 Hoffman, R., lbrahim, N., Murnane, M.J., Diamond, A., Forget, B.G. and Levere, R.D. (1980) Blood 56, 567-570. 37 Fibach, E., Gambari, R., Shaw, P.A., Maniatis, G., Reuben, R.C., Sassa, S., Rifkind, R.A. and Marks, P.A. (1979) Prec. Natl. Acad. Sci. USA 76, 1906-1910. 38 Lindquist, S. (1986) Annu. Rev. Biochem. 55, 1151-1191. 39 Sassa, S. (1990) Int. J. Cell Cloning 8, 10-26. 40 Shibahara, S. (1988) Semin. Hematol. 25, 370-376. 41 Sassa, S. (1983) in Regulation of Hemoglobin Biosynthesis (Goldwasser, E., ed.), pp. 359-383, Elsevier, Amsterdam. 42 Abraham, N.G., Mitrion¢, S.M., Hodgson, W.J., Levere, R.D. and Shibahara, S. (1988) Adv. Exp. Med. Biol. 241, 97-116. 43 Shibahara, S., Muller, R.M. and Taguchi, H. (1987) J. Biol. Chem. 262, 12889-12892. 44 Mitani, K., Fujita, H., Sassa, S. and Kappas, A. (1990) Biochem. Biophys. Res. Comm 166, 1429-1434.
