THE JOURNALOF BIOLOGICAL
CHEMISTRY
Vol. 256, No. 18, Issue of September 25, pp. 9551-9557, 1981
Printed in U.S.A.
Alteration in the Protein Components
of Catecholamine-sensitive
of Rat Reticulocytes*
Adenylate Cyclase during Maturation
(Received for publication, February 11, 1981)
Andrew C. Larner andElliott M. Ross$
From the Departments of Pharmacology and Biochemistry, University of Virginia School of Medicine, Charlottesville,
Virginia 22908
The maturing rat reticulocyte was used as a model
system in which to studydevelopmental changes in the
protein components of hormone-sensitive adenylate cyclase. Plasma membranes from rat erythrocytes display 10 to 20% of the adenylate cyclase activity and 30
to 50% of the/?-adrenergic receptors which are measured inmembranes from rat reticulocytes, as noted by
others. Reticulocyte membranes also display equal activities in response to (-)-isoproterenol in thepresence
of either GTP or GTPyS, whereas erythrocyte membrane adenylate cyclase is twice as active in the presence of isoproterenol plus GTPyS as in thepresence of
isoproterenol plus GTP. We have studied this system
in greater detail by developing or applying independent
assays for the catalytic protein
(C) and the guanine
nucleotide-binding regulatory protein (G/F) of adenylate cyclase.
C was assayed in membranes by its intrinsic Mn2+stimulated activity. It was also measured by reconstituting membranes with saturating amountsof GTPySactivated G/F, yielding an operationally defined Vmlu
for the catalyst. By either assay, reticulocytes display
about &fold greater C activity than do erythrocytes.
G/F wasassayed by its ability to confer GTPyS-stimdated activity upon C (which was supplied by membranes of cyc- S49 lymphoma cells). This assay indicates that reticulocyte membranes contain about 3
times as much G/F as do erythrocyte membranes. Cholera toxin and [s2P]NADwere used to [32P]ADP-ribosylate the45,000- and 52,000-dalton subunits of G/F. Total
decreased 3- to
incorporation of 32Pinto these subunits
4-fold with reticulocyte maturation. The ratio o f label
in the 52,000-dalton peptide to thatin the 45,000-dalton
peptide decreased from 0.29 in reticulocyte membranes
to 0.14 in erythrocyte membranes.
The apparently coordinate decrease in the amounts
of C, G/F, and /%adrenergicreceptors suggest that the
stoichiometry between these components is maintained
during maturation, and may account for the decrease
in adenylate cyclase in the membranes. However, the
qualitative changes in responsiveness to hormones in
the presence of GTP or GTPyS may be related to loss or
proteolysis of the 52,000-dalton subunit of G/F.
* This work was supported by United States Public Health Service
Grants GM 26445 and AM22125. A preliminary report of these
findings was presented at the 1981 meeting of the American Society
of Pharmacology and Experimental Therapeutics (I). The costs of
publication of this article were defrayed in part by the payment of
page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18U.S.C. Section 1734 solely to indicate
this fact.
f An Established Investigator of the American Heart Association.
To whom communications should be addressed at: Department of
Pharmacology, University of Texas Health Science Center, DaUas,
TX 75235.
Several investigators have observed that both the ability of
intact cells to synthesize cyclic AMP and the activity of
catecholamine-stimulated adenylate cyclase in plasma membrane preparations are much higher in reticulocytes than in
erythrocytes (2-8). Catecholamine-stimulated adenylate cyclase activity was also observed to decrease as raterythrocytes
age in the peripheral circulation (9, 10). Qualitative changes
in the coupling of P-adrenergic receptors with adenylate cyclase and alterations in the effects of guanine nucleotides on
the binding of adrenergic agonists have also been noted during
reticulocyte maturation (6, 8). There is, however, little information available on the molecular nature of these alterations
in the adenylate cyclase system.
The @-adrenergicadenylate cyclase system is known to
consist of at least three protein components: the ,5-adrenergic
receptor, the adenylate cyclase catalytic protein, and a guanine nucleotide-binding regulatory protein (11-13). The decrease in adenylate cyclase activity during reticulocyte maturation might thus be caused by decreases in one or more of
these components. The qualitative changes in the regulation
of the enzyme might be caused by changes in the stoichiometry of these proteins, a modification of one or more of them,
or an alteration of the structure or composition of the plasma
membrane in which they interact. Thisproblem was initially
approached by Bilezikian and co-workers (2, 6), who showed
that plasma membranes from reticulocytes contain 2- to 3fold more P-adrenergic receptors than do erythrocyte membranes, and this observation has been c o n f i i e d by others (3,
8). Limbird et al. (8) have attemptedtoquantitate
G/F’
during reticulocyte maturation by measuring [’*P]ADP-ribosylation of the 45,000-dalton subunit of that protein. They
found that reticulocytes contain about 3times as much of the
32
P-labeled product as do erythrocyte membranes, and hypothesized that this difference in the amount of G/F was
related to the changes which were observed in the regulation
of the enzyme. These investigators also observed that reticulocyte membranes, but not erythrocytemembranes, display a
decreased affinity for P-adrenergic agonists in the presence of
GTP, suggesting either adifference in the amount of the G/F
in the two membranes or inits properties. However, Bilezikian
et al. (6) did not observe such a difference between the two
cell types. No attempts toquantitate changes in the catalytic
components of adenylate cyclase have been reported. It thus
remained unclear how G/F and C change in concentration
’
The abbreviations used are: G/F, GTP-binding protein of adenylate cyclase; C, catalytic protein of adenylate cyclase; GTPyS,
guanosine 5’-0-(3-thiotriphosphate);APH, 1-acetyl-2-phenylhydrazine; Hepes, 4-(2-hydroxyethyl)-1-piperazineethanesulfonate;
Gpp(NH)p,guanyl-5’-ylimidodiphosphate;
R020-1724,4-(3-butoxyl-4methoxybenzyl)2-imidazolidinone;buffer A,20 mM Hepes, 1 mM
EDTA, 2 mM MgC1, (pH 8.0); SDS, sodium dodecyl sulfate.
9551
9552
Adenylate Cyclase Proteins in Reticulocyte Maturation
and in stoichiometry with respect to receptors as reticulocytes
mature.
In the present study, we have developed several independent methods to assay p-adrenergic receptors, G/F, and C in
rat erythrocyte and reticulocyte membranes in order to determine whether changes in the relative stoichiometry of these
three proteins occurs with changes in adenylate cyclase activity. p-Adrenergic receptors were measured by the binding of
the selective ligand [3H]dihydroalprenolo1 (14). G/F activity
in detergent extracts of membranes was assayed by reconstitution according to its ability to confer upon free C an increased activity in the presence of GTPyS and Mg’+ (12, 13,
15). This technique provides a method to measure the presumed physiologic activity of G/F. G/F was also quantitated
by cholera toxin-catalyzed ADP-ribosylation using [32P]NAD
(16-18). An assay for C was developed which consisted of
adding saturating amounts of exogenous GTPyS-activated
hepatic G/F toerythrocyte or reticulocyte membranes. Under
these conditions, total catalytic activity in the membranes is
expressed. C was also assayed in membranes using Mn2+-ATP
as a substrate (12).
The results obtained from these studies indicate that all
three components of the 8-adrenergic adenylate cyclase system decrease by aboutthe same amountas reticulocytes
mature. Therefore, the stoichiometry between the components appears to be maintained. However, evidence is presented which indicates the stateof the G/F protein is altered
with maturation, and this alteration in G/Fmay account for
changes in hormonal sensitivity of reticulocyte and erythrocyte membranes.
EXPERIMENTAL PROCEDURES AND RESULTS’
DISCUSSION
The activity of adenylate cyclase and its regulation by
hormones are frequently altered during the development of a
tissue or in response to other metabolic or endocrinologic
events. Since hormone-sensitive adenylate cyclase activity
reflects the activities and interactions of at least three distinct
proteins (13), it is reasonable to study such ontogenetic and
endocrinologic changes in the system by investigating quantitative and qualitative alterations in its individual protein
components. We have used the maturing rat reticulocyte as a
model system to test such an approach. Because there are no
direct molecular assays for C or G/F, we have employed
multiple enzymologic assays for each protein.
Previous publications demonstrated that rat erythrocyte
membranes display 5- to 10-fold less catecholamine-stimulated adenylate cyclase activity than do reticulocyte membranes (1-8). It was also shown that erythrocytes have onehalf to one-third the number of &-adrenergic receptors as
reticulocytes (2, 3, 5, 8). Until recently, however, no direct
evidence had been presented to quantitate either G/For C as
a function of reticulocyte maturation. Using cholera toxincatalyzed ADP-ribosylation of G/F, Limbird et al. (8) suggested that there is approximately 3 times as much G/F in rat
reticulocyte membranes as in erythrocyte membranes. In
addition, these investigators provided evidence that the padrenergic receptors of reticulocyte membranes are more sen-
’
Portions of this paper(including“ExperimentalProcedures,”
“Results,” Figs. 1-9, and Table 111) are presented in miniprint at the
end of this paper. Miniprint is easily read with the aid of a standard
magnifyingglass. Full size photocopies are available
from the Journal
of Biological Chemistry, 9650 Rockville Pike, Bethesda, MD 20814.
Request Document No 81M-313, cite author(s), and include a check
or money order for$8.40 per set of photocopies.Full sized photocopies
are also included in the microfilm edition of the Journal that is
available from Waverly Press.
sitive to regulation by guanine nucleotides than were the
receptors on erythrocytes. Bilezikian et al. (6) did not observe
an effect of guanine nucleotides on agonist a f f i t i e s in either
erythrocyte or reticulocyte membranes. In contrast, Fleming
and Ross have observed this effect in appropriately washed
membranes from either cell type.3This discrepancy is probably caused by differences in the procedures used by different
investigators to prepare and assay the membranes.
In the present work, we studied these alterations in the
activity of the adenylate cyclase system by measuring individual changes in the p-adrenergic receptor, C, and G/F which
take place during maturation. According to the assays which
were used, each of the three proteins decline in activity to
about the same extent.
The catalyst was measured using two distinct enzymatic
assays. The simplest was to measure Mn”-stimulated adenylate cyclase activity. We and others have shown that C is
stimulated directly by Mn’+ in the absence of G/F (12, 32),
and high concentrations of Mn2+inhibit ligand-mediated regulatory interactions of C and G/F. While the concentration of
Mn” used here was not high enough to uncouple such interactions had they occurred, unliganded G/F would not be
expected to alter Mn2+-stimulatedactivity (33). Mn’+-stimulated activity is thus a useful assay for C in soluble preparations and in reconstituted mixtures of C, G/F, and phospholipid (12, 33, 34). Using this assay, we found a 3- to 6-fold
decrease in C during the maturation of reticulocytes (Fig. 7).
However, the absolute specific activities were not easily reproducible. We therefore applied a novel assay for C based
upon the ability of a saturating amount of GTPyS-activated
G/F tomaximally stimulate the enzyme. This assay, suggested
by the work of Gilman and co-workers (23, 30), requires the
use of a highly concentrated source of at least partially purified
G/F so that the maximal activity which is observed reflects
true saturation of C rather than inhibition of activity by
detergent or other proteins when large amounts of G/F are
added to the assays (Fig. 8). This assay for C indicates a 3fold decrease in activity during maturation (Fig. 9). We believe
that this is a more reliable estimate than the2- to 6-fold range
suggested by the Mn2+-stimulated assay. It is reassuring,
however, that an assay which depends upon the productive
interaction of C with G/F essentially agrees with an assay
which probably depends upon the absence of active regulatory
interactions between C and other proteins.
The reconstitutive assay of G/F used in these studies measures G/F according to itsability to confer guanine nucleotidestimulated activity upon C in the presence ofMg’+. It was
originallyused in the demonstration of G/F as a distinct
protein (12, 15), and has since been used by ourselves and
others to monitor the fractionation and purification of G/F
(20, 23, 36). As a source of C, we have used a purified
preparation of plasma membranes of cyc- mutant S49 lymphoma cells (21), which are genetically deficient in G/F (12,
15). An alternative source of C is a chromatographically
purified fraction which is resolved from G/F after detergent
extraction of liver or brain plasma membranes (33, 34). Regardless of the source of C, the experiment depicted in Fig. 4
points out that this assay is valid only when a saturating
amount of C is present. At suboptimal amounts of C, reconstituted activities can vary widely, and it is likely that theuse
of insufficient amounts of C have led to the determination of
artifactually low activities of G/F (15,37,38). Using the assay
described here, we have obtained consistent and reproducible
activities in different G/F preparations over more than 1 year,
in contrast tothe day-to-day variability reported by Farfel et
al. (37).
J. W. Fleming and E. M. Ross, unpublished observations.
Adenylate
Reticulocyte
Cyclase
Proteins
inMaturation
The other assay for G/F is not dependent upon its regulatory activity, but rather upon its being a substrate for the
ADP-ribosyltransferase activity of cholera toxin. While the
toxin is not absolutely specific for the 52,000- and 45,000dalton subunits of G/F (39),it is selective enough so that they
are theprincipal labeled products when E3’P]NAD and the AI
subunit of the toxin are incubated with plasma membranes
under appropriate conditions (Fig. 6). This procedure has
been used by several groups to estimate the amount of G/F
present in a membrane (16, 17, 20, 31), and Limbird et al. (8)
have used this assay in the study of reticulocyte maturation.
The reconstitutive assay suggests that reticulocyte membranes contain about 3 times as much G/F as do304
erythrocyte
membranes (Fig. 5).This estimate
wasconfirmed
956
790 by the
ADP-ribosylation assay (Fig. 6, Table I). The ADP-ribosylation assay does not detect the 35,000-dalton subunit of G/F
(23), and thus it isnot easy to determine how this subunit is
altered during reticulocyte maturation. However, purified G/
F contains roughly as much of the 35,000-dalton subunit as
the sum of the larger two subunits, suggesting that a molecule
of G/F contains one 35,000-dalton subunit andone subunit of
either 52,000 or 45,000 daltons. This implies that the amount
of the 35,000-dalton subunit also decreases by about %fold
during reticulocyte development.
The data discussed above indicate that the plasma membrane concentrations ofC, G/F, and &adrenergic receptors
decreased by a factor of three as reticulocytes mature into
erythrocytes. Since our data and the data of others (2-8)
indicate a 3- to 10-fold decrease in adenylate cyclase activities
during maturation, it is likely that the parallel loss of the
component proteins accounts for the loss of activity in the
complete system. The discrepancy between 10-fold decreases
in some activities and the %fold decreases in C and G/Fmay
be rationalized by the argument that the regulation of adenylate cyclase activity depends on a bimolecular interaction
of the two proteins within the plasma membrane bilayer.
More involved explanations could also be invoked.
It is more difficult to understand the qualitative alterations
in the regulation of adenylate cyclase as erythrocytes mature.
Bilezikian and co-workers (6) first described differences in the
response of reticulocyte and erythrocyte adenylate cyclases to
hormones in the presence of different guanine nucleotides (g),
a fiidingwhich we have c o n f i i e d (Table 11).In reticulocyte
membranes, isoproterenol plus GTP produced as high an
activity of adenylate cyclase as did isoproterenol plus GTPyS.
Inmature erythrocytes, however, isoproterenol plus GTP
were only half as efficacious at stimulating the system. BileTABLEI
Changes in cholera toxin-catalyzed [32 PJADP-ribosylationof G / F
as a function of reticulocytepopulation
Membranes derived from blood containing increasing fractionsof
reticulocytes were labeled with C3’P]NAD in the presence of cholera
toxin. Proteins were fractionated usingSDS-polyacrylamidegel electrophoresis as described under “Experimental Procedures.” Autoradiographs of gels were scannedat 660 nm and the absorbance of those
bandscorresponding to the 45,000- and52,000-daltonsubunits of
G/F was recorded. Absorbance was linear as a function of protein
which was subjected to electrophoresis overthe ranges measured.
Reticulocyte
2.2
14
32
50
72
Absorbance at 660 n m
45,000-dalton 52,000-ddton
subunit
subunit
0.67
0.72
0.88
1.47
1.12
0.096
0.112
0.176
0.408
0.320
Ratio
.~
52,000-dalton
absorbance
45.000-daltonabsorbance
~
0.14
0.16
0.20
0.28
0.29
9553
TABLE11
Differential responsiveness of adenylate cyclase to isoproterenol in
the presence of GTP or GTP$
Membranes were prepared from rats injected
with increasing doses
of APH. Adenylate cyclase activities weremeasured as described
under “Experimental Procedures.” Adenylatecyclase activity is constant over the 10-min assay period in the presence either of isoproterenol (INE) plus GTP or isoproterenol plusGTPyS.
Adenylate cyclase
Reticulocytes
I
2.2
14
32
50
1171
991 72
1NE plus GTP
‘z?$$
Activitv ratio
INE plus GTP
INE plus GTPyS
pmol/min/mg
240
444
502
1120
979
0.54
0.61
0.83
0.96
0.99
zikian et al. (6)characterized this decreased responsiveness to
hormone plus GTP relative to hormone plus Gpp(NH)p as an
uncoupling of the system. This view was supported by Limbird
that the affinity of
and co-workers (8), whoalsoshowed
binding of adrenergic agonists to reticulocyte receptors was
decreased by guanine nucleotides whereas nucleotides had no
effect on the affinity of agonists for erythrocyte receptors.
It is difficult to rationalize these changes in the responsiveness of red blood cells to guanine nucleotides simply on the
basis of their complements of C, G/F, and receptor, which
decrease in parallel. Instead, we suggest that these differences
in coupling efficiency may be caused by differences in the
subunit composition of the G/F molecules in these two membranes. The data of Table I indicate that as reticulocytes
mature, the relative amount of the 52,000-dalton subunit of
G/F decreases. This decrease may reflect its cleavage to a
45,000-dalton proteolytic product (as would beconsistent with
the similarity of the peptide profiles of the two subunits, see
Ref. 40). Studies of purified, unlabeled preparations of G/F
(23) and of [32PJADP-ribosylatedG/F indicate that most
mammalian G/F generally contains some of the 52,000-dalton
subunit, while G/F from avian erythrocytes contains only the
45,000-dalton protein (16, 17). Thus, while the amount of
52,000-dalton G/F polypeptide in rat reticulocytes is typical
of other mammalian tissues, the low amount of this subunitin
mature erythrocytesis similar to thesituation in avian erythrocytes. Avian erythrocytes also display a negligible response
to hormone in the presence of GTP and adiminished guanine
nucleotide-mediated decrease in receptor affinity for agonists
(41-43). Comparative studies of G/F from turkey erythrocytes
and wild type S49 murine lymphoma cells (44) suggest that at
least some of these characteristic patterns of responsiveness
are conferred by G/F, rather than
by C, the hormone receptor,
or themembrane bilayer. Based upon these correlations, it is
tempting to hypothesize that thealterations in responsiveness
to guanine nucleotides which are observed as reticulocytes
mature are directly caused by the cleavage of larger subunit
of G/F from 52,000 to 45,000 daltons.
This suggestion is also in agreement with the finding of
Gilman’s group that fractions of rabbit hepatic G/F which are
enriched in either the52,000-dalton polypeptide or the 45,000dalton polypeptide display appropriately different regulatory
responses to nucleotides and cation^.^ We are currently pursuing the kinetics of activation of erythrocyte and reticulocyte
G/F with respect to cations and guanine nucleotides in order
to support our contention that the regulatory responses are
characteristic of altered G/F molecules. We are also attempting to proteolyze the 52,000-dalton subunit of reticulocyte G/
P. C. Sternweis, J. K. Northup,and
communication.
A. G. Gilman,personal
9554
Adenylate Cyclase Proteins
in
Reticulocyte Maturation
F in order to change its responses to those characteristic of 18. Moss, J., and Vaughan, M. (1979) Annu. Rev. Biochem. 48, 581600
the erythrocyte protein.
The data reported here support the idea that the loss of 19. Johnson, R. A., and Walseth, T. F. (1979)Adv. Cyclic Nucleotide
Res. 10, 135-167
adenylate cyclase activities which occurs as rat reticulocytes 20. Cassel,
D., and Pfeuffer, T. (1978) Proc. Natl. Acad.Sci. U. S. A.
mature is related to the parallel loss of C , G/F, and P-adre75,2669-2673
nergicreceptors. The qualitative changes in responses to 21. Ross, E. M., Maguire, M. E., Sturgill, T. W., Biltonen, R. L., and
guanine nucleotides whichoccurduringmaturation (6) are
Gilman, A. G. (1977) J. Biol. Chem. 252,5761-5775
more likely attributable to the 52,000-dalton subunitof G/F. 22. Ross, E. M., and Schatz, G . (1978) Methods Enzymol. 53, 222229
A more important result of this work is the suggestion that
J. K., Sternweis, P. C., Schleifer, L. S., Smigel, M. D.,
reasonable assays for the individual proteins components of 23. Northup,
Ross, E. M., and Gilman, A.G. (1980) Proc. Natl. Acad. Sci.
adenulate cyclase may be applied productively
to the study of
U.S.A. 77,6516-6520
ontogenetic and endocrinologic regulation of this important 24. Brecker, G. (1949) Am. J. Clin. Pathol. 19,895-896
enzyme system.
25. Solomon, Y., Londos, C., and Rodbell, M. (1974) Anal. Biochem.
Acknowledgments-We would like to thank Drs. J. W. Fleming
and J. K. Northup for advice and discussion, and Dr. Northup for
providing purified G/F.
REFERENCES
1. Lamer, A. C., and Ross, E. M. (1981) Fed. Proc. 40,653
2. Bilezikian, J. P., Spiegel, A. M., Brown, M., and Aurbach, G. D.
(1977) Mol. Pharrnacol. 13, 775-784
3. Charness, M. E., Bylund, D. B., Beckman, B. S., Hollenberg, M.
D., and Snyder, S. H. (1976) Life Sci. 19,243-250
4. Kaiser, G., Wiemer, G., Kremer, G., Dietz, J., Hellwich, M., and
Palm, D. (1978) Eur. J. Pharrnacol. 48,255-262
5. Beckman, B. S., and Hollenberg, M. D. (1979) Biochem. Pharmacol. 28,239-248
6. Bilezikian, J. P., Spiegel, A. M., Gammon, D. E., and Aurbach, G.
D. (1977) Mol. Pharrnacol. 13, 786-795
7. Bilezikian, J. P.(1978) Biochim. Biophys. Acta 542,263-273
8. Limbird, L. E., Gill, D. M., Stadel, J. M., Hickey, A. R., and
Lefkowitz, R. J. (1980) J. Biol. Chem. 255,1854-1861
9. Pfeffer, S. R., and Swislocki, N.I. (1976)Arch. Biochem. Biophys.
177,117-122
10. Bylund, D. B., Tellez-Iiion, M. T., and Hollenberg, M. D. (1977)
Life Sci. 21, 403-410
11. Pfeuffer, T.(1977) J. Biol. Chem. 252,7224-7234
12. Ross, E. M., Howlett, A. C., Ferguson, K. M., and Gilman, A. G.
(1978) J. Biol. Chem. 253,6401-6412
13. Ross, E. M., and Gilman, A. G. (1980) Annu. Rev. Biochem. 49,
533-564
14. Mukherjee, C., Caron, M. G., Coverstone, M., and Lefkowitz, R.
J. (1975) J. Biol. Chem. 250,4869-4876
15. Ross, E. M., and Gilman, A. G. (1977) J. Biol. Chem. 252,69666969
16. Johnson, G. L., Kaslow, H. R., and Bourne, H. R. (1978) J. Biol.
Chem. 253,7120-7123
17. Gill, D. M., and Meren, R. (1978) Proc. Natl. Acad.Sci. U. S. A .
75,3050-3054
58,541-548
26. Fleming, J. W., and Ross, E. M. (1980) J. Cyclic Nucleotide Res.
6,407-419
27. Laemmli, U. K. (1970) Nature (Lond.)227,680-685
28. Swanstrom, R., and Shank,P. R. (1978) Anal. Biochem. 86, 184192
29. Lowry, 0.H., Rosebrough, N. J., Fan, A. L., and Randall, R. J.
(1951) J. Biol. Chem. 193,265-275
30. Sternweis, P. C., and Gilman, A. G . (1979) J. Bwl. Chem. 254,
3333-3340
31. Schleifer, L. S., Garrison, J. C., Sternweis, P. C., Northup, J. K.,
and Gilman, A. G. (1980) J. Biol. Chem. 255,2641-2644
32. Limbird, L. E., Hickey, A. R., and Lefkowitz, R. J . (1979) J. Biol.
Chem. 254,2677-2683
33. Ross, E. M. (1981) J. Biol. Chem. 256, 1949-1953
34. Strittmatter, S., and Neer, E. J. (1980) Proc. Natl. Acad. Sci.
U.S.A. 77,6344-6348
35. Jain, S. K., and Hochstein, P. (1980) Arch. Biochem. Biophys.
201,683-687
36, Kaslow, H. R., Johnson, G. L., Brothers, V. M., and Bourne, H.
R. (1980) J. Biol. Chem. 255,3736-3741
37. Farfel, Z., Brickman, A. S., Kaslow, H. R., Brothers, V. M., and
Bourne, H. R. (1980) N. Engl. J. Med. 303,237-241
38.
Bhat, M. K., Iyengar, R., Abramowitz, J., Bordelon-Riser, M. E.,
and Birnbaumer. L. (1980) Proc. Natl. Acad. Sci. U.S.A. 77,
3836-3840
39. Moss, J., Stanley, S. J., Watkins, P. A., and Vaughan, M. (1980)
J. Biol. Chem. 255, 7835-7837
40. Hudson, T. H., and Johnson, G.L. (1980) J. Biol. Chem. 255,
7480-7486
41. Tolkovsky, A. M., and Levitzki, A. (1978) Biochemistry 17,37953810
42. Abramowitz, J., Iyengar, R., and Birnbaumer, L. (1980) J. Biol.
Chem. 255,8259-8265
43. Iyengar, R., Abramowitz, J., Bordelon-Riser, M. E., Blume, A. J.,
and Birnbaumer, L. (1980) J. Bwl. Chem. 255, 10312-10321
44. Kaslow, H. R., Farfel, Z., Johnson, G. L., and Bourne, H. R. (1979)
Mol, Pharrnacol. 15,472-483
Adenylate
Cyclase
Proteins
in Reticulocyte
Maturation
9555
Alteration in the Protein Componentsof CatecholsmineSensitive Adenylate Cyclase During Maturation of Rat RetiCUlOCYteEi
Experimental Details
h d r e w C. Lamer and Elliott M. ROES
EXPERIMENTAL PROCEDURES
58 sDs/0.258 2-mtrfiptoeChanol/l2.51 glyf.pr~1/0.058
Iples were heated et 100' C for 5 min and
P-labelad
MALsLhu
Iu3zPlATP
[3BlDihydro.lprenolol
vas obtained from New England
was eynthallired as demxibed by Johnnon and Walseth 119). Nuc+far.
I PINAD was syn( 2 0 ) . Cholera enterotoxin W ~ I I
thesized a s described by Cassel and Pfeuffsr
purchased from Schwarz/Mann. A11 matariala for gel electrophoresis were
purchased from Bio-Rad. R020-1724 wae a gift from Dr. Herbert bheppard (Hoffman-LaRoehe). The growth of cyc- s49 lymphoma cells and preparation Of
plasma membranes therefrom have been described (21). Cholate wan purchased
from Sigma and purified as described (22). Rabbit hepatic G/P, purified a s
described ( 2 3 ) . wae a gift Of Dr. J.K. Northup. APE was obtained from Sigma
and GTP-7-S from Bochringer.
Male Sprague-Dawley rate ( 2 0 0 g) yere given intramuscular injections of
MPI for three consecutive days.
Blood v a ~
collected on the seventh day after
the first injection. The range of APH doses used to produce
increasing degrees of retiCU1OCytOsiB was 10 to 6 0 mg/kg body weight. The fraction of the
red cell population composed of reticulocytes W ~ Bdetermined after staining
with Wen Methylene Blue 124) and is erpresaed as the percentage of reticulocytes in a sample of red cells. The standard t z r m s of the means of this
Wrcentage ranged from 0.58 reticulocytes (in blood containing 2-58 retifulocyteat to 101 reticulocytes (in blood containing 70-901 reticulocytes).
asMmtkmaf"
Blood was withdrawn by cardiac puncture
and collected in at least one
volume of 2 0 mu KPi/5 mM EGTA/Z m H EDTA/l40 mM NaC1 (pH
7.4) at 4%. The blood
filtered through twola OTS of cheesecloth and centrifuged at 1200 x 9 for
5 min. The supernatant and tuffy coat were removed and the pellet was TCBUSd~/l
pended in lyeis buffer I20 m H Tris-Cl/O.l mM p h e n y l ~ e t h y l s u l f o n y l f l ~ ~ ~ imu
EDTA (pH 7.4)) to g i v e about a sixteen-fold dilution Of packed cells. AdditlOnal P h e n y l m e t h y l ~ u l f o n y l f l v o r i d e (0.5 M in dimethylsulforide) was added to
a final Concentration of 0.5 mn.
The euspension vas forcefully d r a m through
a 14 gauge. 5-1/4- needle prior to being centrifuged at9000 I g for 15 mi".
The membranes were washed repeatedly with lysis bufferuntil white. The final
wash Of the membranesWas done in 20 m H RepeB/2 m H IgCl2/1 m u EDTA (pH 8.0)
(buffer A). After the addition Of dithiothreitol to a concentration of 0.5 mu
membranes were frozen in a bath of dry ice and 2-mtthoryethsnol and stored at
Was
-850
RESULTS
~ " i n " d "
Many investigators have noted that membranes obtained from rat zeticulocytes display much higher adenylate cyclase aCtivitie8 than do those obtained
from cat erythrocytes (1-3.5.81.
We have confirmed these findings (Fig.1) using several different sti~ylatorsof adenylate cyclaee. Banal activity and the
activity stimulated by Mn , P-, or ieoproterenol plus either GTP or GTP-7-8
a11 decrease during maturation. The magnitude
of thiB decrease rangesfrom
two- to ten-fold, depending upon the activators w e d in the ansay.
we consistently find specific activities of adenylate cycla8e am much an
ten-fold higher than thoee reported by others
12,3,8).
Notably, r e find
~ Bmembranes
~
from mature
easily measurable activities
of adenylate E Y C ~ in
even after correction fer the contribution of the IetiCuloCytes
eryth&cytes.
present in normal blood 12-48). TO examine this discrepancy. erythrocyte
membranes were prepared according t o Bilezikisn +f nL 12). The edenylhte
cyclase activities in this preparation were ~ i m i 1 r . r to tho*= reported by those
InVestigatorB (Table 1111. but the preparation warn bright red. When theee
membranes were further purified by the method we describe, the Bpcific activities of adenylate cyc1a.e increased three- to four-fold (Table 111). BaBed
on these data, we ascribe our increased specific activities t o increased
purity of the membrane preparation.
c.
The same procedure was w e d
t o prepare membranes from the blood of rats
norever, memwhich had been injected with APB to induce reticulocytosis.
branes from blood with lncreasing fractiona of reticuloytes were increasingly
brownish-red in COlOC. By BpectroBcopiC Criteria, this color is not due only
to hemoglobinor cytochromes, and the colored material Cannot
be Beparated
from the membranes by centrifuging the membranes through continuous sucrose
gradients 158-401) in buffer A.
Ye BpeCUlDte that the color is caused by the
treatment withAPE, and that it is perhapa a heme 0 1 hemoglobin derivative
covalently bound to the membranes. It is not observed in membranes of reticulocyten which were induced by bleeding.
Protein was determined by the method of Lowry
& (29).
f
s
t
&
&
Y
&
l
a
R
e
t
i
c
v
l
O
c
v
r
e
.
e
n
b
l
.
n
c
s
Reticulocytes (%)
G/F was assayed br ita ability to restoreGTP-7-S-Btimuldted activity to
C in membranes of CYC S49 lymphoms cells ( 1 2 ) . Red blood c e l l Plasma m e m branes (10 mg/ml) were solubilized by stirringf o r 1 h at 4' C in buffer A
containing 6.3 mM sodium cholate and 0.85 I ammonium eulfate. The mixture was
centrifuged at 2 0 0 , 0 0 0 x 9 for 1 h at 4' C. The supernatants contained0.81.0 m g / m 1 protein for erythrocyte membranes and1.5
2 ng/ml protein for
reticulocyte membranes.
It is estimated that this proeedvre Bolubilizee 80%
to 858 of the G/P activity in the membranes of either cell type since a
second attempt to solubilize activity from the first pellet yielded only 102 0 8 ae much activity as did the firat so1ubilization. Thie cholate extract
was diluted
withan equal volume Of 19 mu cholate in buffer A. Diluted extract
115 u1) wan added to 25ul Of 5 mg/ml Cyc membranes in buffer A containing
0.5 mM dithiothreitol.
If less than 7.5 p 1 undiluted extract was t o be used
in a n aaeay, the diluted Supernatantwan further diluted with 12
o n cholate
Plus 0.425 I! ammonium sulfate in buffer A. After a 15 min Incubation at +a C
s 8.0)/15 m u Mgcl / o m mg/mi
20 el of reconetitution buffer (150 mu ~ a ~ e p e(pn
pyruvate kinase/o.3 mg/ml BSA/l8 mM phosphoenolpyruvate/1.25 m & A T P / 3 o o p
GTP-,-SI
was added t o each tube. This mixture wae incubated 80 m i n at 30 c
before the addition Of 40 el OF adenylate cyclase assay buffer, (125 mM
NaHepes IPH 8.0)/10 m u MgC12/0.025 mg/ml PYruvate k 4 p s e / 0 . 2 5 gg/ml bovine
BeCUm albumin/l5 mll phoephoenclpyruv~te/0.625 mll Ic PIATP (10 cpml/2.5
mu
EWTW'O 25 m R020-1724/100 U M GTP-Y-SI.
Assays were atopped after 10 m i n at
30' C ;nd ;j2Pieyclic AIIP was quantified as described by Salomon et & 125).
Subatantiation of this assay ie dimcussed below.
F i g u r e 1. Adenylate Cyclase activities in membranes prepared from rat red
blood cells. Male Sprague-Dawley rats were injected with varylng Concentratione of APH and membraneswere prepared a8 described in the Experimental
Procedures. Adenylate Cyclaee activity was essayed in the premnce of 0.1 .X
GTP plus I uu propranolol (-1, 0.1 mM GTP-I-s (01. 10 DM INE plum GTP
IO) o r
ActiYities stimulated by GTP Plus Propranolol and
INE p l u s GTP-1-8
GTP-Y-S alone are displayed on the scale on the right.
( A).
Table I11
-
C was assayed after m a x i m a l activation by the addition of saturating
amounts Of GTP-~-s-IICtivated G/P
For these all*dy(l rabbit hepatic G/P was
activated by incubation at 30' C fbr 3 h in the presehce of 10 mu NaHepes (pa
8.0)/10 mu MgC12/1 mll W/l 6 M EDTIVO.1) LubrOl 12A9/10 PM G T W - S ,
Reconstitution Of activated G/P into erythrocyte or reticulocyte membranes was accam limhed hy adding 10 yl G/F in 10 rnll NaHepea (pH 8 0)/1 mM EQTA/l mM
DTT70.0258 Lubrol 1219 to 30 el membranes at 2 mg/ml.
M t e r a 20 min incubation on i c e , 6 0 u1 adenylate c y s l ~ s cd s e q buffer wan added, anda s * a y s were
continued for 10 min at 30° C.
Adenylate Cyclase ActiVities of Rat Erythrocyte We.branea
Prepared by Different Pr'ocedumB
Membranes were prepared from rat erythrocytes either according to B i l e r i k i a n
(2) (0) or as described in Experimental Procedures ( A ) .
A portion of
the material prepared according to B i l e z i k i a n mf & was then further purified
by the protocol we described IC). Each preparation was essayed for adenylate
cyclase in the premence of the activators .horn.
N ~ P
was used at a conoentra-
+f nL
+'"
^C
.
I
-I
Preparative Technique
"__"__"__"_"_" " " " " " " " " "
"_"""""~
Adenylate Cyclase Activity
(pmol/min/mg
'
INEIGTP
INE+Gpp(NE)p
NaP
245
A
319
654
B
16
112
59
C
59
379
152
Adenylate Cyclase Proteins
in
9556
Reticulocyte Maturation
i s L systematic over-estimation Of the fractionof reticulocytes in highly
IetiCuloCytOtiC blood, since reticulocyte Counting in reproducible among individuals.
.
0 12
0 10
0 08
0 06
0 04
0 02
1 0 0
Reticulocytes (%I
Membrane Extract
(PI)
F i g u r e 2.
Adenylate cyclase activities in membranes from blood containing
inmeasins fractione of xeticulowtes. Blood from cats injected with large
Figure 4.
Reconstitution of adenylate cyclase activity in cyc- SI9 Cell
plasma membranes with increasing amounts of cholate-solubilized tcythroeyte
G/P.
Erythrocyte membranes were solubilized with cholate a~ described and the
supernatants were diluted with equal volumes of buffer A containing 19 mM
cho&ate. Diluted supernatants (15 p l l were added to the indicqted amounts of
cyc membranes contained in 25 p l , and reconstitutiOn waB performed 111 deln reconstitutions that contained l e s s
scribed in Experimental Procedures.
than 7.5 p l of the original undiluted supernatant. the Concentrations
Of
cholate and (NHdI SO4 were kept constant by dilution with buffer A COnt*inin9
12 m I cholate and 0.425 I INHII~SO&.
/
6oo(
-0
20
40
60
80
Reticulocytes (%)
P-Adrsnergic receptors in membranes derived from blood conyining
increasing fractions Of retICUloEyte~. The @-adrenergic antagonist 1 HlDHA
used to manay B-receptors as described in Experimental Procedures. Iembranes were prepared from rats injected with increasing doses Of APE lo) or
from vacying mixtures of rlllimally retiCulocytOtiC and control blood 1. I.
PigUrC 3.
Was
0
20
40
60
Reficulocyies (74
ELPA€tlYuYinervthrocvredndRrriculoevreIn order to monitor changes in the amount Of G/P in red blood cell
membranes during maturation. IC initially aeaayed its activity according t o
it8 ability to reconstitute a stimulatory rcsponae to GTP-I-S in a preparation
containing the Catalytic protein C. Plasma membranes of the cyc- mutant 649
as being deficient in G/F
lymphoma cell have previously been described
(12 15) but are an easily obtainable source of relatively concentrated C. G/F
wad d u b i l i z e d from erythrocyte and reticulocyte membranes using cholate at
high ionic strength, and it8 activity was assayed according to its abilityto
reconatltute GTP-Y-S-Btimulated adenylate cyclaee activity in cyc- membranes.
This assay i a demonstrated in F i g . 4. In this experiment, erythrocyte membrsnes were treated with 6.3 m n Cholate plus 0.85 M ammonium sulfate. condiIncreasing amounts of this
t i o m which yield maximal solubilization Of G/F.
detergent extract were reconetituted with several different Concentration0 of
cyc- membranes. Recanatituted GTP-.(-S-stimUlated activity is maximal at Concentrations of at least 7 5 pg eyc- membrane pzctein/aasay.
The asaay is
linear with added G/P at low G/P Concentrations even at Buboptimal amounts of
nowever it is important touae optimal concentration0 of cyc- memcyc
branes in the &mstitution to obtain maximal aensitivity in the assay and to
ensure reproducibility of the measured actlvitof C/F. TO achieve maximal
reconstituted adenylate cyc1a.e
activity, the G ~ Fand cy=- membrane. must be
m i n at30°Crith ~econstitutionbufferpriortothe
incubated for eitleaet70
addition of the adenylate cyclase assay buffer. Under these conditions. the
a a m y is linear farfit least 20 m i n after the additionof adenylate cyclase
buffer containing Im PlATP. The conditions described above ale0 provide for
optimal extraction and rBcmBtitutian of reticulocyte G/F.
.
80
0
Adenylate Cyclase Proteins
in
1
2
3
4
Reticulocyte Maturation
5
Reticulocytes (%)
Reticulocytes (%)
P I g u t c 7. Mn2*-~tImulatcd adenylate cyclase
a~tlvlt1e. In red cell plasm.
membranes. Membranes were prepared from r a t s Injected rlth varying domes of
APR l o ) 01 fro. blood Of reticulocytotlc snlmals that
we.8 mixed I n varylng
proportlonm rlth blood from noncatlculocytotlc a n h a l a ( D).
9557