The activities of aminolaevulinate synthetase, aminolaevulinate dehydratase and haem synthetase have been examined in short-term cultures of embryonic mouse liver.

Although synthesis of haemoglobin was induced by erythropoietin in these cultures no increase in activity was detected in any of the three enzymes over 24 h in culture. In each case, however, enzyme activity was higher when erythropoietin was present than in its absence.

The significance of these findings is discussed in relation to control of haemoglobin synthesis and it is concluded that enzyme activity is not rate limiting during induction of haemoglobin synthesis in vitro.

It has been shown that three haem synthesizing enzymes - aminolaevulinate synthetase (ALAS), aminolaevulinate dehydratase (ALAD), and haem synthetase - fluctuate in mouse foetal liver during the hepatic phase of erythropoiesis (Freshney & Paul, 1971). The activities of these enzymes increase sequentially from the 13th to the 17th days of gestation when the foetal liver is erythropoietic, and may be related to the rapid increase in the rate of synthesis of haemoglobin between the 13th and 15th days. Since haemoglobin synthesis in cell cultures from mouse foetal livers 13 days post-fertilization can be stimulated by erythropoietin (Cole & Paul, 1966), this provides a suitable system for investigating the participation of haem synthesizing enzymes in the regulation of hepatic erythropoiesis.

Levere & Granick (1965) and Wainwright & Wainwright (1970) reported that ALAS activity might be rate-limiting in the synthesis of haemoglobin by chick blastoderm. Moreover, Wilt (1968) suggested that RNA synthesis was not ratelimiting at the onset of haemoglobin synthesis in chick blastoderm. Haem synthesis may also regulate the synthesis of haemoglobin in other erythroid systems (Levere, Kappas & Granick, 1967 ; Grayzel, Hörchner & London, 1966). Hence, using the same system as Cole & Paul (1966), haemoglobin synthesis has been measured before and after treatment with erythropoietin. The activities of ALAS, ALAD and haem synthetase have been measured in replicate samples and fluctuations in their activities compared to that of haemoglobin synthesis.

Porton white Swiss mice were killed by cervical fracture late on the 13th day of pregnancy (counting the detection of a vaginal plug as day zero). Livers were dissected from the embryos under aseptic conditions, disaggregated and cultured by the method of Cole & Paul (1966). The dispersed cells were grown at about 106/ml in Waymouth’ s medium MB 752/1 in medical flat bottles, 4 oz or 32 oz (approx. 0-1 or 11), or in test-tubes, depending on the number of cells required for each enzyme determination.

To determine the rate of haemoglobin synthesis 59Fe, preincubated for 18 h with mouse serum (as a source of transferrin), was added to aliquots of the cultures for 1 h. 59Fe-labelled haem was extracted in butanone from an acidified Drabkin’ s lysate of the washed cells and counted on glass-fibre discs in a scintillation counter (after Cole & Paul, 1966).

ALAS activity was measured in a hypotonic lysate of washed cells by the method of Freshney & Paul (1970, 1971). ALAD activity was measured in a 0-15M-KC1 homogenate by spectrophotometric determination of porphobilinogen produced from aminolaevulinate in anaerobic conditions (Shemin, 1962). Haem synthetase activity was measured in a 0-15M-KC1 homogenate, containing 0·4 °/0 (v/v) Tween 20, by measuring incorporation of 59Fe into haem under anaerobic conditions (Freshney & Paul, 1971).

Erythropoietin was prepared from human urine and was a gift from the National Heart and Lung Institute of the U.S.A. It was procured by the Department of Physiology, University of the North East, Corrientes, Argentina, and processed by the Haematology Research Laboratories, Children’ s Hospital of Los Angeles, for distribution by the National Heart Institute under Research Grant HE-1O88O. It was authorized for distribution by the Committee on Erythropoietin of the National Heart Institute.

Haemoglobin synthesis

The stimulation by erythropoietin of 59Fe incorporation into haem was demonstrated by Cole & Paul (1966) and was confirmed in the present series of experiments (Figs. 1–3). A similar experiment was performed in which the acidification of the Drabkin’ s lysate was omitted. This resulted in a negligible recovery of 59Fe-labelled haem (Table 1) and, since the acidification is necessary for the cleavage of haem from globin, this implies that the haem labelled in this assay is normally protein-bound. There was a decline in the recovery of free haem between 0 and 24 h, suggesting that there might have been a decrease in haem synthesis relative to globin synthesis over this period.

Table 1.

Recovery of59Fe from free haem after culture with [59Fe]transferrin

Recovery of59Fe from free haem after culture with [59Fe]transferrin
Recovery of59Fe from free haem after culture with [59Fe]transferrin
Fig. 1.

Suspensions of 13-day embryonic liver cells, prepared as in Materials and Methods, were inoculated into medical flat bottles. Cultures of 107 cells/10 ml were used for 59Fe incorporation in 4 oz (approx. 0.1 1) bottles; they were pulsed for 1 h before sampling with 0.5 ml 59Fe/transferrin, 13 µCi per ml, 504 µCi per µmole. 5 × 107 cells in 50 ml were cultured in 32 oz (approx. 1 1) bottles for ALAS assay. Duplicate sample bottles were removed at the times indicated and the cells washed in balanced salt solution and assayed immediately as indicated in Materials and Methods. The values for 59Fe incorporation per 10® cells were low in this experiment and may have been due to the low concentration of 69Fe used (one third of the normal). Broken lines are 59Fe incorporation, solid lines ALAS activities. Open symbols are controls, closed symbols samples cultured in the presence of human urinary erythropoietin, 0.6 units per ml.

Fig. 1.

Suspensions of 13-day embryonic liver cells, prepared as in Materials and Methods, were inoculated into medical flat bottles. Cultures of 107 cells/10 ml were used for 59Fe incorporation in 4 oz (approx. 0.1 1) bottles; they were pulsed for 1 h before sampling with 0.5 ml 59Fe/transferrin, 13 µCi per ml, 504 µCi per µmole. 5 × 107 cells in 50 ml were cultured in 32 oz (approx. 1 1) bottles for ALAS assay. Duplicate sample bottles were removed at the times indicated and the cells washed in balanced salt solution and assayed immediately as indicated in Materials and Methods. The values for 59Fe incorporation per 10® cells were low in this experiment and may have been due to the low concentration of 69Fe used (one third of the normal). Broken lines are 59Fe incorporation, solid lines ALAS activities. Open symbols are controls, closed symbols samples cultured in the presence of human urinary erythropoietin, 0.6 units per ml.

Aminolaevulinate synthetase

The activity of ALAS in trypsinized suspensions was 50-70 % lower than in fresh liver. After culture for 24 h the activity declined a further 75 % relative to the initial activity (Fig. 1). In the presence of erythropoietin the decrease was less (about 60 % of the initial value). In this and in two other similar experiments there was no indication of any increase of ALAS activity following treatment with erythropoietin.

Aminolaevulinate dehydratase

ALAD activity was not reduced by overnight trypsinization. The activity of this enzyme declined during culture though not as rapidly as that of ALAS (Fig. 2). Induction of haemoglobin synthesis could be produced regularly (five separate experiments) without induction of ALAD activity. After culture in the presence of erythropoietin ALAD activity was slightly higher than in the absence of erythropoietin.

Fig. 2.

One ml of a cell suspension containing 1 · 37 × 106 cells per ml from 13-day embryonic liver were inoculated into each of a number of Pyrex test-tubes. Before collection, duplicate samples were labelled for 30 min by the addition of 0· 15 ml [59Fe]transferrin, 13µCi per ml, 448μCi per μmole. At the times indicated, duplicate 59Fe-labelled samples and unlabelled samples were removed, washed twice in ice-cold Hanks’ s balanced salt solution, and stored frozen. Broken lines indicate 59Fe incorporation and solid lines ALAD activity. Open symbols represent control samples, the closed symbols represent samples cultured with 0 ·6 units per ml human urinary erythropoietin.

Fig. 2.

One ml of a cell suspension containing 1 · 37 × 106 cells per ml from 13-day embryonic liver were inoculated into each of a number of Pyrex test-tubes. Before collection, duplicate samples were labelled for 30 min by the addition of 0· 15 ml [59Fe]transferrin, 13µCi per ml, 448μCi per μmole. At the times indicated, duplicate 59Fe-labelled samples and unlabelled samples were removed, washed twice in ice-cold Hanks’ s balanced salt solution, and stored frozen. Broken lines indicate 59Fe incorporation and solid lines ALAD activity. Open symbols represent control samples, the closed symbols represent samples cultured with 0 ·6 units per ml human urinary erythropoietin.

Haem synthetase

The activity of haem synthetase was not reduced by overnight trypsinization. In the presence of erythropoietin virtually no decline in haem synthetase activity was detected over 24 h in culture, while in the absence of erythropoietin the activity fell to less than half of the initial value (Fig. 3). This result was obtained in two separate experiments.

Fig. 3.

One ml aliquots of an embryonic liver-cell suspension containing 0 · 81 × 106 cells per ml were incubated in test-tubes. 59Fe incorporation was measured in duplicate tubes by incubation for 1 h prior to sampling with 0.15 ml [59Fe]transferrin, 13 µCi per ml, 510 µCi per µmole. The samples were then washed in ice-cold balanced salt solution and stored frozen in 1 ml Drabkin’ s solution. For haem synthetase estimations, 10 test-tube cultures were pooled, the cells centrifuged at 1000 g for 5 min, washed twice in balanced salt solution and stored frozen at —70 °C. Broken lines are 59Fe incorporation into haemoglobin in culture, solid lines haem synthetase activity of extracts. Solid symbols are samples cultured with 0·6 units per ml human urinary erythropoietin, open symbols are sample controls.

Fig. 3.

One ml aliquots of an embryonic liver-cell suspension containing 0 · 81 × 106 cells per ml were incubated in test-tubes. 59Fe incorporation was measured in duplicate tubes by incubation for 1 h prior to sampling with 0.15 ml [59Fe]transferrin, 13 µCi per ml, 510 µCi per µmole. The samples were then washed in ice-cold balanced salt solution and stored frozen in 1 ml Drabkin’ s solution. For haem synthetase estimations, 10 test-tube cultures were pooled, the cells centrifuged at 1000 g for 5 min, washed twice in balanced salt solution and stored frozen at —70 °C. Broken lines are 59Fe incorporation into haemoglobin in culture, solid lines haem synthetase activity of extracts. Solid symbols are samples cultured with 0·6 units per ml human urinary erythropoietin, open symbols are sample controls.

None of the enzymes studied demonstrated any induction comparable to the extent of the induction of haemoglobin synthesis. This suggests that increased synthesis of these enzymes is not necessary for the induction of haemoglobin synthesis by erythropoietin in 13-day-mouse embryo liver cultures. This is in contrast to the report by Bottomley & Smithee (1969), who showed that ALAS was induced in marrow cultures stimulated by erythropoietin, although Hrinda & Goldwasser (1968) were unable to demonstrate ALAS induction with this material. The regulation of haemoglobin synthesis by ALAS activity (Levere & Granick, 1965; Wilt, 1968; Wainwright & Wainwright, 1970) may be restricted to yolk-sac erythropoiesis and may not be important in hepatic erythropoiesis.

However, although enzyme activities were never higher than at the beginning of culture, they were always relatively higher in cells incubated for 24 h in the presence of erythropoietin. Some possible explanations are as follows:

  1. Whilst most cells demonstrate a reduction of enzyme activity, a small population sensitive to erythropoietin may exhibit increased activity; induction in this small population may be masked by the more general pattern of loss of enzyme activity in the remainder.

  2. Erythropoietin may induce, directly, the synthesis of enzyme at the translational level or the synthesis of messenger at the transcriptional level but this may be counteracted by degradation of the enzyme. Bottomley & Smithee (1969) have suggested that erythropoietin induces ALAS at the transcriptional level though they did not report any induction of haem synthetase.

  3. The rate of enzyme degradation may be influenced by erythropoietin. Schimke & Doyle (1970) have stressed the importance of enzyme degradation as a controlling step in the induction of arginase and tryptophan pyrrolase in rat liver. The higher enzyme activities in erythropoietin-treated cultures may represent a reduced rate of degradation of enzymes. Since Pitot, Peraino, Lamar & Kennan (1965) have shown that template stability may be subject to regulatory changes, and, since an increase in template stability has been proposed during hepatic erythropoiesis (Djaldetti, Chui, Marks & Rifkind, 1970), the observed differences in enzyme activity could be due to an increase in the stabilities of the messenger RNA templates for these enzymes.

  4. Induction of haemoglobin synthesis by erythropoietin may bring about a reduction in the haem pool. Since haem may be a co-repressor in the transcriptional control of ALAS synthesis (Kappas & Granick, 1968), any reduction in haem would cause derepression of ALAS synthesis. The same situation may also hold for the other two enzymes. Moreover, haem may directly inhibit the activity of these enzymes. In particular there have been reports that haem inhibits ALAS activity (Scholnick, Hammaker & Marver, 1969), although this was not confirmed by Kappas & Granick (1968). Hence any fluctuation in the haem pool produced by elevated haemoglobin synthesis might be reflected in a change in enzyme activity.

  5. ALAS activity in Rhodopseudomonas spheroides is influenced by the concentration of an intracellular inhibitor (Marriot, Neuberger & Tait, 1969; Tuboi, Kim & Kikuchi, 1969). If a similar inhibitor existed in erythroid cells then it might be influenced by erythropoietin.

With the possible exception of the ALAS assay, where the concentration of the enzyme extract is high, cellular factors are diluted so greatly in the assay mixtures that it seems unlikely that haem or other inhibitors operate in these assay conditions.

Since ALAS declines very soon after the 13th day in vivo a decline in activity in culture of late 13-day liver might be expected. ALAD and haem synthetase, however, would be expected to rise prior to the induction of haemoglobin synthesis, if the phenomena observed in vivo were precisely reproduced in culture.

As they are not, the induction of haemoglobin synthesis in vitro by erythropoietin is not strictly comparable to the 13-to 15-day induction of haemoglobin synthesis in the liver. The in vitro observations are of short duration (24 h); prolonged hepatic erythropoiesis as seen in vivo may require stimulation of haem-synthesizing enzymes.

In conclusion, these results imply that the induction of haemoglobin synthesis by erythropoietin is not a consequence of induction of haem-synthesizing enzymes, although erythropoietin does have some effect on them. The major effect of erythropoietin must be presumed to result from another mechanism.

The authors wish to express their gratitude to Mrs Sheila Wilson for expert technical assistance. This work was supported by grants from the Cancer Research Campaign and the Medical Research Council. Erythropoietin was procured by the Department of Physiology, University of the Northeast, Corrientes, Argentina, and processed by the Haematology Research Laboratories, Children’ s Hospital of Los Angeles, for distribution by the National Heart Institute under Research Grant HE-10880. It was authorized for distribution by the Committee on Erythropoietin of the National Heart Institute.

Bottomley
,
S. S.
&
Smithee
,
G. A.
(
1969
).
Effect of erythropoietin on bone marrowy-aminolevulinic acid synthetase and heme synthetase
.
J. Lab. clin. Med
.
14
,
445
452
.
Cole
,
R. J.
&
Paul
,
J.
(
1966
).
The effects of erythropoietin on haem synthesis in mouse yolk sac and cultured foetal liver cells
.
J. Embryol. exp. Morph
.
15
,
245
260
.
Djaldetti
,
M.
,
Chui
,
D.
,
Marks
,
P. A.
&
Rifkind
,
R. A.
(
1970
).
Erythroid cell development in fetal mice: stabilization of the haemoglobin synthetic capacity
.
J. molec. Biol
.
50
,
345
359
.
Freshney
,
R. I.
&
Paul
,
J.
(
1970
).
Measurement of aminolaevulinate synthetase activity in normal mouse liver with (2-14C) glycine
.
Biochim. biophys. Acta
220
,
594
601
.
Freshney
,
R. I.
&
Paul
,
J.
(
1971
).
The activities of three haem-synthesizing enzymes during hepatic erythropoiesis in the mouse
.
J. Embryol. exp. Morph
.
26
,
313
322
.
Grayzel
,
A. L
,
Hórchner
,
P.
&
London
,
I. M.
(
1966
).
The stimulation of globin synthesis by heme
.
Proc. natn. Acad. Sci. U.S.A
.
55
,
650
655
.
Hrinda
,
M. E.
&
Goldwasser
,
E.
(
1968
).
Studies on the control of haemoglobin formation in marrow cells
.
Ann. N. Y. Acad. Sci
.
149
, Article 1,
412
415
.
Kappas
,
A.
&
Granick
,
S.
(
1968
).
Steroid induction of porphyrin synthesis in liver cell culture. I. The effects of haem, uridine diphosphate glucoronic acid, and inhibitors of nucleic acid and protein synthesis on the induction process
.
J. biol. Chem
.
243
,
346
352
.
Levere
,
R. D.
&
Granick
,
S.
(
1965
).
Control of haemoglobin in synthesis in the cultured chick blastoderm by δ-aminolevulinic acid synthetase: increase in the rate of haemoglobin formation with δ-amino levulinic acid
.
Proc. natn. Acad. Sci. U.S.A
.
54
,
134
137
.
Levere
,
R. D.
,
Kappas
,
A.
&
Granick
,
S.
(
1967
).
Stimulation of haemoglobin synthesis in chick blastoderms by certain 5/?-androstane and 5/?-pregnane steroids
.
Proc. natn. Acad. Sci. U.S.A
.
58
,
985
990
.
Marriot
,
J.
,
Neuberger
,
A.
&
Tait
,
G. H.
(
1969
).
Control of aminolaevulinate synthetase activity in Rhodopseudomonas spheroides
.
Biochem. J
.
III
,
385
394
.
Pitot
,
H. C.
,
Peraino
,
C.
,
Lamar
,
C.
&
Kennan
,
A. L.
(
1965
).
Template stability of some enzymes in rat liver and hepatoma
.
Proc. natn. Acad. Sci. U.S.A
.
54
,
845
851
.
Schimke
,
R. T.
&
Doyle
,
D.
(
1970
).
Control of enzyme levels in animal tissues
.
A. Rev. Biochem
.
39
,
929
976
.
Scholnick
,
P. L.
,
Hammaker
,
L. E.
&
Mar Ver
,
H.
S
. (
1969
).
Soluble hepatic Æ-amino-levulinic acid synthetase: end-product inhibition of the partially purified enzyme
.
Proc, natn. Acad. Sci. U.S.A
.,
63
,
65
70
.
Shemin
,
D.
(
1962
).
J-Aminolaevulinic acid dehydratase from Rhodopseudomonas spheroides
.
Meth. Enzym
.
5
,
882
.
Tuboi
,
S.
,
Kim
,
H. J.
&
Kikuchi
,
G.
(
1969
).
Occurrence of a specific and reversible inhibitor of delta-aminolevulinate synthetase in extracts of Rhodopseudomonas spheroides
.
Archs Biochem. Biophys
.
130
,
92
100
.
Wainwright
,
S. D.
&
Wainwright
,
L. K.
(
1970
).
A kinetic study of the effects of Æ-amino-levulinic acid upon the synthesis of embryonic and fetal haemoglobins in the blood island of the developing chick blastodisc
.
Can. J. Biochem
.
48
,
400
406
.
Wilt
,
F. H.
(
1968
).
Heme and regulation of embryonic hemoglobin synthesis
.
Biochem. biophys. Res. Commun
.
33
,
113
117
.