P U R I F I C A T I O N A N D T R A N S L A T I O N OF O V A L B U M I N , CONALBUMIN, OVOMUCOID AND LYSOZYME MESSENGER RNA RICHARD D. PALMITER* AND LYNNE T. SMITH
Dept. of Biochemistry, Searle Research Laboratories, High Wycombe, Bucks., England (Received 13 June 1973; in revised form 15 July 1973) ABSTRACT. Messenger RNAs for 4 egg white proteins (ovalbumin, conalbumin, ovomucoid and lysozyme) were assayed in a cell-free, protein-synthesizing system derived from rabbit reticulocytes. Each of these messengers was purified about 33-fold from hen oviduct polysomal RNA using oligo(dT) cellulose. The apparent minimal translational efficiencies varied from 22 translations for each ovalbumin mRNA to less than 1 for lysozyme mRNA. These messengers are not polycistronic with each other since they have distinct sedimentation values of: lysozyme, 8.5 S; ovomucoid, 11 S; ovalbumin, 15 S; and conalbumin 18 S. Several aspects of this system indicate that it will be valuable for dissecting the fine control of mRNA metabolism. I. INTRODUCTION Synthesis of egg white proteins is controlled by three steroid hormones: estrogen, progesterone, and testosterone [1]. Investigations on the role of each of these hormones in the oviduct reveal that the rates of synthesis of four of the egg white proteins are controlled independently of each other [2], despite the fact that they are all synthesized in the same tubular gland cells [3]. Moreover, combinations of these hormones (e.g. estrogen plus progesterone, or estrogen plus testosterone) produce synergistic effects on the rates of synthesis of some of the egg white proteins, while not affecting others [4, 5]. Since the rate of protein synthesis is determined by the rate of polypeptide initiation and elongation, as well as the mRNA concentration, we have been developing techniques to distinguish among these possibilities in order to ascertain how these steroids regulate egg white protein synthesis [4, 6]. This report shows that four egg white protein messengers, which together constitute over 70~o of the total mRNA in oviduct magnum of laying hens or hormonetreated chicks [1], can be assayed, and potentially quantitated, using a cell-free, protein-synthesizing system. Results obtained using this assay support the hypothesis that these steroids control specific protein synthesis primarily by regulating the concentration of translatable mRNA rather than the efficiency of mRNA translation [4, 7]. II. METHODS Polysomes were prepared from magnum of hen oviduct essentially as described [6], except that the homogenization buffer contained heparin at 0.9 mg ml-1 to inhibit RNase. RNA was ex* Present Address: The Biological Laboratories, Harvard University, Cambridge, Massachusetts 02138, U.S.A.
129 Molecular Biology Reports 1 (1973) 129-134. All Rights Reserved Copyright 9 1973 by D. Reidel Publishing Company, Dordrecht-Holland
tracted as described in Method IVa [8] and dissolved in 20 mM HEPES (N-2-hydroxyethylpiperazine-N'-2-ethane sulfonic acid), pH 7.5. For sucrose gradient centrifugation, 0.75 ml of RNA (1 mg m1-1) was layered over 5 to 20~o sucrose gradients in 10 mM HEPES, 1 mM EDTA, pH 7.5, and centrifuged 10.5 hr at 280000 9 at 2 ~ (SW 41 rotor, Beckman). Polysomal RNA (1.6 mg) was hybridized to 32 mg of oligo(dT) cellulose (Searle Products for Research) in 0.5 M KCI, 10 mM HEPES, pH 7.5. The unbound RNA was removed batchwise with three 8 ml washes of 0.1 M KC1, 10 mM HEPES; mRNA was then eluted with four successive 150 lal washes of distilled water. The second to fourth washes were pooled (yield was 34 lag) and used in the experiments described for purified RNA. All purification procedures were performed at 0 to 4 ~ and the solutions were sterilized with diethyl pyrocarbonate [8] prior to use. m R N A activity was measured by adding aliquots of RNA to a protein-synthesizing system derived from rabbit reticulocytes. The assay conditions were as described [8] and [aH]isoleucine (Radiochemical Center) was used at 10 laCi ml- x of reaction mixture. After a 90 min incubation at 26 ~ the samples were mixed with a solution containing detergent and carrier antigen [8] and centrifuged for 10 min at 13000 g. Aliquots of the supernatant were incubated for 60 min with a 1.5- to 2-fold excess of the appropriate specific antibody; the immunoprecipitate was collected by centrifuging it through sucrose [8], except that 60 lal of 1 M sucrose was overlaid with 40 lal of 0.5 M sucrose and the centrifugation time was increased to 3 min. These steps help to reduce the background which was 150 to 200 cpm in these experiments. The isotope incorporation into globin was determined as described [8]. The immunological specificity of each of the antibodies used was demonstrated as before [2] and a further characterization of the radioactivity which was precipitated from the lysate by each antibody will be presented elsewhere [7]. The radioactivity in ovalbumin, conalbumin, and lysozyme immunoprecipitates migrated as a single peak nearly coincident with authentic proteins synthesized in vivo when subjected to SDS acrylamide gel electrophoresis (e.g. see [9]); ovomucoid migrated predominantly as a single peak with an apparent molecular weight of about 22000 (a value close to that expected for the polypeptide portion of this protein which contains about 25 ~ carbohydrate), but there was another peak with a mobility similar to that of authentic ovomucoid [7]. The most important step insuring immunological specificity is the purification of the antigens used for immunization [2]. III. RESULTS AND DISCUSSION The sedimentation properties of the messengers for 4 of the egg white proteins was obtained by sucrose gradient centrifugation as shown in Fig. 1. Approximate sedimentation values for the 4 messengers are: lysozyme, 8.5 S, ovomucoid, 11 S; ovalbumin, 15 S; conalbumin 18 S. The S values for these messengers increase in the same order as the molecular weight of the polypeptides for which they encode (lysozyme, 14 500, ovomucoid, 22 000; ovalbumin, 42 000; conalbumin, 76 000). Of these messengers, only ovalbumin had previously been assayed [9-11] and its sedimentation value was similar to that shown here [9, 12]. This value may be a poor indication of ovalbumin m R N A molecular weight, since analysis on polyacrylamide gels in both native and denaturing conditions reveals that this m R N A has a molecular weight of approximately 875000 or about 1.5 times that estimated from sucrose gradients [13]. Similar discrepancies between S values and molecular weight may thus exist for the other messengers as well. One hypothesis which might explain why ovalbumin m R N A is more than twice the size of the translated sequence is that it is polycistronic; however, the experiment shown in Fig. 1 argues against this possibility since the other messengers which we assayed are the ones most likely to be polycistronic with ovalbumin mRNA. 130
=""
i
i
'=
!
O~o
~o CON I
2es
.... 11o'= . . . .
ov I 1 \ Mu
0
r
l~.
;' 2
/I /l I
I r
o
I~
v
~rDY ~6s\
4"
,,~
I'\
-.
,,,,,,~
0 MU LY8
T
~a
2
~
l liio! con i ilti! ~4
C
5
10 Fraction
15
20
Fig. 1. Sedimentation rate of conalbumin (CON), ovalbumin (OV), ovomucoid (MU), and lysozyme (LYS) mRNA. Hen polysomal R N A (750 lag) and 3H-labeled E. coll. rRNA (9 lag) were centrifuged together on a sucrose gradient as described in Section II. E. eoli tRNA was centrifuged on a separate gradient. Fractions (0.6 ml) were collected and 50 lal aliquots were assayed in a 250 lal reaction mixture to determine ovalbumin, eonalbumin, and ovomucoid m R N A activity. To measure lysozyme mRNA, aliquots of 350 pl were mixed with 59 lag of carrier chick embryo rRNA, precipitated with 2 volumes of ethanol, dissolved in 50 pl of 20 mM HEPES, and then assayed as above. Since only twothirds of the reaction mixture was used for immunoprecipitation, all values refer to the m R N A activity in 33 lal of each fraction and the incorporation into lysozyme was also corrected to this value. Other aliquots (150 lad were precipitated with acid and counted to determine the location of the E. eoli rRNA. 131
Messenger R N A was purified from total polysomal R N A by hybridization to oligo(dT) cellulose. Fig. 2 shows the effect of adding increasing amounts of either the unpurified (Fig. 2a) or purified R N A (Fig. 2b) to a rabbit reticulocyte lysate. In both cases there was a linear increase in the amount of specific product synthesized with the lower concentrations of R N A but at higher concentrations the response was not proportional. Although globin synthesis was depressed about 20 ~o at the highest R N A concentrations, total [aH]isoleucine incorporation into protein was either constant (Fig. 2a) or increased (Fig. 2b), suggesting that these messengers compete with globin m R N A for translation factors. We estimate that 3 ~tg of purified m R N A (Fig. 2b) is equivalent to
7.5
L
~
I ,,"
?
...." ~ .......... ....
g /
o/
:,P"I
,.
//_/
i .s I II
I
! Li I'#/'-co. ,,,,' I
zs
jI
I
.._ I
I I,.
Dept. of Biochemistry, Searle Research Laboratories, High Wycombe, Bucks., England (Received 13 June 1973; in revised form 15 July 1973) ABSTRACT. Messenger RNAs for 4 egg white proteins (ovalbumin, conalbumin, ovomucoid and lysozyme) were assayed in a cell-free, protein-synthesizing system derived from rabbit reticulocytes. Each of these messengers was purified about 33-fold from hen oviduct polysomal RNA using oligo(dT) cellulose. The apparent minimal translational efficiencies varied from 22 translations for each ovalbumin mRNA to less than 1 for lysozyme mRNA. These messengers are not polycistronic with each other since they have distinct sedimentation values of: lysozyme, 8.5 S; ovomucoid, 11 S; ovalbumin, 15 S; and conalbumin 18 S. Several aspects of this system indicate that it will be valuable for dissecting the fine control of mRNA metabolism. I. INTRODUCTION Synthesis of egg white proteins is controlled by three steroid hormones: estrogen, progesterone, and testosterone [1]. Investigations on the role of each of these hormones in the oviduct reveal that the rates of synthesis of four of the egg white proteins are controlled independently of each other [2], despite the fact that they are all synthesized in the same tubular gland cells [3]. Moreover, combinations of these hormones (e.g. estrogen plus progesterone, or estrogen plus testosterone) produce synergistic effects on the rates of synthesis of some of the egg white proteins, while not affecting others [4, 5]. Since the rate of protein synthesis is determined by the rate of polypeptide initiation and elongation, as well as the mRNA concentration, we have been developing techniques to distinguish among these possibilities in order to ascertain how these steroids regulate egg white protein synthesis [4, 6]. This report shows that four egg white protein messengers, which together constitute over 70~o of the total mRNA in oviduct magnum of laying hens or hormonetreated chicks [1], can be assayed, and potentially quantitated, using a cell-free, protein-synthesizing system. Results obtained using this assay support the hypothesis that these steroids control specific protein synthesis primarily by regulating the concentration of translatable mRNA rather than the efficiency of mRNA translation [4, 7]. II. METHODS Polysomes were prepared from magnum of hen oviduct essentially as described [6], except that the homogenization buffer contained heparin at 0.9 mg ml-1 to inhibit RNase. RNA was ex* Present Address: The Biological Laboratories, Harvard University, Cambridge, Massachusetts 02138, U.S.A.
129 Molecular Biology Reports 1 (1973) 129-134. All Rights Reserved Copyright 9 1973 by D. Reidel Publishing Company, Dordrecht-Holland
tracted as described in Method IVa [8] and dissolved in 20 mM HEPES (N-2-hydroxyethylpiperazine-N'-2-ethane sulfonic acid), pH 7.5. For sucrose gradient centrifugation, 0.75 ml of RNA (1 mg m1-1) was layered over 5 to 20~o sucrose gradients in 10 mM HEPES, 1 mM EDTA, pH 7.5, and centrifuged 10.5 hr at 280000 9 at 2 ~ (SW 41 rotor, Beckman). Polysomal RNA (1.6 mg) was hybridized to 32 mg of oligo(dT) cellulose (Searle Products for Research) in 0.5 M KCI, 10 mM HEPES, pH 7.5. The unbound RNA was removed batchwise with three 8 ml washes of 0.1 M KC1, 10 mM HEPES; mRNA was then eluted with four successive 150 lal washes of distilled water. The second to fourth washes were pooled (yield was 34 lag) and used in the experiments described for purified RNA. All purification procedures were performed at 0 to 4 ~ and the solutions were sterilized with diethyl pyrocarbonate [8] prior to use. m R N A activity was measured by adding aliquots of RNA to a protein-synthesizing system derived from rabbit reticulocytes. The assay conditions were as described [8] and [aH]isoleucine (Radiochemical Center) was used at 10 laCi ml- x of reaction mixture. After a 90 min incubation at 26 ~ the samples were mixed with a solution containing detergent and carrier antigen [8] and centrifuged for 10 min at 13000 g. Aliquots of the supernatant were incubated for 60 min with a 1.5- to 2-fold excess of the appropriate specific antibody; the immunoprecipitate was collected by centrifuging it through sucrose [8], except that 60 lal of 1 M sucrose was overlaid with 40 lal of 0.5 M sucrose and the centrifugation time was increased to 3 min. These steps help to reduce the background which was 150 to 200 cpm in these experiments. The isotope incorporation into globin was determined as described [8]. The immunological specificity of each of the antibodies used was demonstrated as before [2] and a further characterization of the radioactivity which was precipitated from the lysate by each antibody will be presented elsewhere [7]. The radioactivity in ovalbumin, conalbumin, and lysozyme immunoprecipitates migrated as a single peak nearly coincident with authentic proteins synthesized in vivo when subjected to SDS acrylamide gel electrophoresis (e.g. see [9]); ovomucoid migrated predominantly as a single peak with an apparent molecular weight of about 22000 (a value close to that expected for the polypeptide portion of this protein which contains about 25 ~ carbohydrate), but there was another peak with a mobility similar to that of authentic ovomucoid [7]. The most important step insuring immunological specificity is the purification of the antigens used for immunization [2]. III. RESULTS AND DISCUSSION The sedimentation properties of the messengers for 4 of the egg white proteins was obtained by sucrose gradient centrifugation as shown in Fig. 1. Approximate sedimentation values for the 4 messengers are: lysozyme, 8.5 S, ovomucoid, 11 S; ovalbumin, 15 S; conalbumin 18 S. The S values for these messengers increase in the same order as the molecular weight of the polypeptides for which they encode (lysozyme, 14 500, ovomucoid, 22 000; ovalbumin, 42 000; conalbumin, 76 000). Of these messengers, only ovalbumin had previously been assayed [9-11] and its sedimentation value was similar to that shown here [9, 12]. This value may be a poor indication of ovalbumin m R N A molecular weight, since analysis on polyacrylamide gels in both native and denaturing conditions reveals that this m R N A has a molecular weight of approximately 875000 or about 1.5 times that estimated from sucrose gradients [13]. Similar discrepancies between S values and molecular weight may thus exist for the other messengers as well. One hypothesis which might explain why ovalbumin m R N A is more than twice the size of the translated sequence is that it is polycistronic; however, the experiment shown in Fig. 1 argues against this possibility since the other messengers which we assayed are the ones most likely to be polycistronic with ovalbumin mRNA. 130
=""
i
i
'=
!
O~o
~o CON I
2es
.... 11o'= . . . .
ov I 1 \ Mu
0
r
l~.
;' 2
/I /l I
I r
o
I~
v
~rDY ~6s\
4"
,,~
I'\
-.
,,,,,,~
0 MU LY8
T
~a
2
~
l liio! con i ilti! ~4
C
5
10 Fraction
15
20
Fig. 1. Sedimentation rate of conalbumin (CON), ovalbumin (OV), ovomucoid (MU), and lysozyme (LYS) mRNA. Hen polysomal R N A (750 lag) and 3H-labeled E. coll. rRNA (9 lag) were centrifuged together on a sucrose gradient as described in Section II. E. eoli tRNA was centrifuged on a separate gradient. Fractions (0.6 ml) were collected and 50 lal aliquots were assayed in a 250 lal reaction mixture to determine ovalbumin, eonalbumin, and ovomucoid m R N A activity. To measure lysozyme mRNA, aliquots of 350 pl were mixed with 59 lag of carrier chick embryo rRNA, precipitated with 2 volumes of ethanol, dissolved in 50 pl of 20 mM HEPES, and then assayed as above. Since only twothirds of the reaction mixture was used for immunoprecipitation, all values refer to the m R N A activity in 33 lal of each fraction and the incorporation into lysozyme was also corrected to this value. Other aliquots (150 lad were precipitated with acid and counted to determine the location of the E. eoli rRNA. 131
Messenger R N A was purified from total polysomal R N A by hybridization to oligo(dT) cellulose. Fig. 2 shows the effect of adding increasing amounts of either the unpurified (Fig. 2a) or purified R N A (Fig. 2b) to a rabbit reticulocyte lysate. In both cases there was a linear increase in the amount of specific product synthesized with the lower concentrations of R N A but at higher concentrations the response was not proportional. Although globin synthesis was depressed about 20 ~o at the highest R N A concentrations, total [aH]isoleucine incorporation into protein was either constant (Fig. 2a) or increased (Fig. 2b), suggesting that these messengers compete with globin m R N A for translation factors. We estimate that 3 ~tg of purified m R N A (Fig. 2b) is equivalent to
7.5
L
~
I ,,"
?
...." ~ .......... ....
g /
o/
:,P"I
,.
//_/
i .s I II
I
! Li I'#/'-co. ,,,,' I
zs
jI
I
.._ I
I I,.
