Thyroid Hormone Regulates Type I Deiodinase Messenger RNA in Rat Liver
Maria J. Berry, Anna-Lisa Kates, and P. Reed Larsen Howard Hughes Medical Institute Laboratory Department of Medicine Brigham and Women's Hospital Harvard Medical School Boston, Massachusetts 02115
Conversion of the prohormone T4 to the active hormone T3 is catalyzed by 5'-deiodinases, enzymes that have not been purified. Previous studies have shown that modulating thyroid status results in changes in type I deiodinase activity in the rat liver. We have quantitated type I deiodinase mRNA in liver by an expression assay using Xenopus laevis oocytes. We report here that changes in enzyme activity correlate closely with changes in levels of the mRNA for this enzyme, indicating that thyroid hormone regulates type I deiodinase at a pretranslational step. Using the oocyte system to express sizefractionated mRNA, we have also determined that the mRNA coding for this protein is between 1.9-2.4 kilobases in length. It has been proposed that protein disulfide isomerase (PDI) is closely related to the rat type I 5'-deiodinase. Our results indicate that this is not the case, since injection of in vitro transcribed PDI mRNA into oocytes did not result in expression of deiodinase activity, and the deiodinase mRNA could be physically separated from the 2.8-kilobase mRNA species hybridizing to rat PDI cRNA by size fractionation. (Molecular Endocrinology 4: 743-748, 1990)
and brown adipose, functions primarily to provide an intracellular source of T3 for these tissues. This enzyme exhibits a Km for T4 of -about 2 nM and is PTU resistant. Many attempts at elucidating the molecular structure of these enzymes are in progress, but these efforts have to date been unsuccessful. Regulation of both type I and type II deiodinase activities by thyroid hormone has been examined using tissue homogenates and microsomes (2-4). Activity levels of type I deiodinase in rat liver are low in the hypothyroid state and elevated in hyperthyroidism. Conversely, type II deiodinase activity levels are regulated in the opposite direction, being low in tissues from hyperthyroid animals and elevated in hypothyroidism (5). Since no specific reagents for quantitation of the enzymes have been developed, it has not been possible to confirm that these activity changes are due to alterations in the enzyme content or, if so, whether they are transcriptional or post-transcriptional. It has also been suggested that type I deiodinase, which requires reduced thiols for maximal enzyme activity, is closely related to rat protein disulfide isomerase (PDI) (6). A growing number of proteins, including intracellular enzymes and cell surface receptors, have been found to be amenable to study after expression in Xenopus oocytes. St. Germain et al. (7) recently reported expression of type I deiodinase activity in Xenopus oocytes injected with rat hepatic RNA. We have employed this system to study the effects of thyroid hormone on type I deiodinase mRNA levels and have also evaluated the putative role of PDI as a deiodinase. We observed that type I deiodinase mRNA levels correlate closely with thyroid status and with type I deiodinase activity, indicating that thyroid hormone alters the tissue concentration of the mRNA coding for the type I deiodinase.
INTRODUCTION
Virtually all of the metabolic and developmental effects of thyroid hormone are mediated by T3, which is produced from T4 by 5'-deiodination. lodothyronine 5'deiodination is catalyzed by two general classes of enzymes (1) distinguished by their tissue distribution, physiological roles, Km for substrate, and sensitivity to propylthiouracil (PTU). Type I deiodinase, present predominantly in liver and kidney, provides most of the plasma T3 in the rat. This class of enzyme exhibits a Km for T4 of ~2 HM and is sensitive to inhibition by PTU. Type II deiodinase, found in pituitary, cerebral cortex,
RESULTS
To best use the Xenopus oocyte system for our studies, an optimal source of mRNA for expressing type I deiodinase was needed. Because previous studies have
0888-8809/90/0743-0748$02.00/0 Molecular Endocrinology Copyright © 1990 by The Endocrine Society
743
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2014. at 01:45 For personal use only. No other uses without permission. . All rights reserved.
Vol 4 No. 5
MOL ENDO-1990 744
shown that type I deiodinase activity is elevated in the hyperthyroid state (2-4), rats were made hyperthyroid, and poly(A)+ RNA was isolated from liver and kidney. Deiodinase activity was produced in oocytes after the injection of mRNA from both tissues; however, the level of activity produced by liver mRNA was about 10-fold higher than the level produced by kidney mRNA. Based on this result, liver was chosen as the tissue source. Most of the activity produced in oocytes from liver mRNA was found in the supernatant after centrifugation at 10,000 x g, suggesting that the activity was not associated with mitochondria or nuclei. To ascertain whether the enzyme made in oocytes catalyzed outer ring deiodination, products of deiodination were separated by descending paper chromatography according to previously described methods (8). The rT3 degraded appeared as free iodide and 3,3'diiodothyronine in equal quantities, as expected for specific outer ring deiodination. This deiodination was completely inhibited by 1 ITIM PTU (data not shown). These data indicate that this deiodinase activity can be categorized as type I. Analysis of the time course for expression of type I deiodinase showed maximal expression 3 days after mRNA injection. Deiodinase activity was between 30-50% of this level on days 2 and 4 and lower on days 1 and 5. To further evaluate tissue sources for mRNA and, in parallel, to explore the mechanism for thyroid hormone regulation of type I deiodinase, rats were made hyperthyroid or hypothyroid, or maintained euthyroid. Hepatic microsomal proteins were prepared from these animals. Aliquots from six microsomal preparations for each thyroid state were pooled and assayed for type I deiodinase activity. Km and maximum velocity (Vmax) values were determined from the Lineweaver-Burk double reciprocal plot shown in Fig. 1. Km values for rT3 were
0.36, 0.27, and 0.44 HM, and Vmax values were 119, 421, and 759 pmol/minmg protein for hypothyroid, euthyroid, and hyperthyroid preparations, respectively. These results suggest that thyroid hormone regulates the amount of enzyme, not the activity per unit protein, in agreement with previous studies (2-4). Type I deiodinase activity was measured from the individual microsomal preparations described above. The means of activity levels from two representative assays of six microsomal preparations at each thyroid state are shown in Fig. 2A. Hyperthyroid liver microsomal proteins contained about 150% of the level of deiodinase activity found in euthyroid microsomes. The hypothyroid level of deiodinase activity was about 30% of the euthyroid level.
400-i
c o o
300-
Q. O)
E
200-
c
E
i
100
o E o.
HYPO
EUTHYROO
HYPO*
MICROSOMES
•
HYPO
•
EUTHYROID
a HYPER
-10
Fig. 1. Lineweaver-Burk Plot of Type I Deiodination Catalyzed by Liver Microsomes from Hypothyroid, Euthyroid, and Hyperthyroid Rats Liver microsomal proteins from hypothyroid, euthyroid, and hyperthyroid rats were assayed for type I deiodinase activity with rT3 concentrations varying from 60-360 nM, as described in Materials and Methods.
o o o E
100-
Q. O HYPO
EUmYROD
HYPER
RNA INJECTED Fig. 2. Effects of Thyroid Status on Type I Deiodinase Activity and mRNA A, Microsomal protein from six individual preparations each of hypothyroid, euthyroid, and hyperthyroid rat liver were assayed for type I deiodinase activity, as described in Materials and Methods. B, Pooled poly(A)+ RNA from each group of animals was injected into 18 oocytes, which were incubated for 3 days at 18 C. Each mRNA preparation was injected in at least two concentrations from 10-40 ng/oocyte in two separate injection experiments. Oocytes were homogenized, divided into triplicate aliquots of six oocytes each, and assayed for type I deiodinase activity. Deiodinase activity is expressed as counts per min of 125I released from 200,000 cpm [125l]rT3 in a 60-min incubation. Results are the mean ± SE of the two experiments.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2014. at 01:45 For personal use only. No other uses without permission. . All rights reserved.
745
Thyroid Hormone Regulation of Type I Deiodinase mRNA
Hepatic mRNA was isolated from the same animals used above and assayed for production of type I deiodinase after injection into Xenopus oocytes. Deiodinase activity was found to increase linearly with 10, 20, and 40 ng mRNA from all three thyroid states. As shown in Fig. 2B, injection of oocytes with mRNA from hyperthyroid liver produced about twice the level of deiodinase activity observed with mRNA from euthyroid liver. The level of activity produced in oocytes by injection of hypothyroid liver mRNA was 47% of the euthyroid level. These results were confirmed in a separate experiment using RNA prepared from other groups of hypothyroid, euthyroid, and hyperthyroid animals. The correlation between hepatic microsome deiodinase activity and type I deiodinase mRNA, shown in Fig. 3, was determined by plotting hepatic microsome deiodinase activity from hypothyroid, euthyroid, and hyperthyroid animals vs. deiodinase activity expressed in oocytes injected with mRNA from these animals. This correlation was highly significant. To characterize further the mRNA for type I deiodinase, we determined its mol wt. Poly(A)+ RNA was size-fractionated on low melting temperature agarose, fractions were injected into oocytes, and deiodinase activity was determined. The results are shown in Fig. 4. The highest level of deiodinase activity was encoded by the fraction containing mRNA between 1.9-2.4 kilobases (kb), indicating that the mRNA for type I deiodinase is in this size range. It has been speculated that type I deiodinase is closely related to protein disulfide isomerase (6), an enzyme involved in protein folding (9). PDI mRNA was transcribed in vitro and injected into oocytes. No type I deiodinase activity was detected. In addition, using PDI cRNA as probe, we performed Northern blots of sizefractionated mRNA from rat liver. The PDI probe hybridized to a 2.8-kb mRNA present in total poly(A)+ RNA (Fig. 5, lane T) and the 2.2- to 3.2-kb fraction (B), but
kilobases < Z
15000-1
c o
10000-
O
5000-
o o E a u
5.4
4.4
3.5
uninj
1
2
3
oocy
lOngR
200-
100-
HYPO
1.4
4
Fig. 4. Size-Fractionation of Type I Deiodinase mRNA and Expression in Xenopus Oocytes Poly(A)+ RNA from hyperthyroid rat liver was size-fractionated on low melting temperature agarose. Fractions (20 ng) were injected into groups of 12 oocytes, which were incubated for 3 days at 18 C. Oocytes were homogenized, divided into duplicate aliquots, and assayed for type I deiodinase activity. The boundaries of each size fraction, in kilobases, are indicated above the fractions and were determined from a mRNA size ladder electrophoresed in parallel lanes.
7.5
-
4.4
-
2.4
-
1.4
-
I
EUTHYROID
E
M T
CL
100
1.9
Fraction
HYPER
o
2.4
o
400-
300-
2.9
200
300
400
pmol/min/mg Protein Fig. 3. Correlation between Type I Deiodinase Activity from Injected mRNA and Liver Microsomal Fractions Type I deiodinase activity expressed in Xenopus oocytes injected with liver mRNA was plotted vs. activity in liver microsomal extracts from hypothyroid, euthyroid, and hyperthyroid rats. Linear regression analysis by the method of least squares produced a correlation coefficient of 0.948.
A B C
Fig. 5. Northern Blot of Size-Fractionated Liver mRNA with PDI Probe Poly(A)+ RNA from hyperthyroid rat liver was size-fractionated on low melting temperature agarose, and 1-^g aliquots were analyzed on a 1.1% agarose formaldehyde gel. The gel was blotted to a Duralon UV nylon membrane, probed with PDI cRNA, and autoradiographed. Lane M, RNA size markers 7.5, 4.4, 2.4, and 1.4 kb (BRL); lane T, 5 fig unfractionated mRNA from hyperthyroid rat liver; lane A, 3.2- to 4.6-kb mRNA; lane B, 2.2- to 3.2-kb mRNA; lane C, 1.7- to 2.2-kb mRNA.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2014. at 01:45 For personal use only. No other uses without permission. . All rights reserved.
Vol 4 No. 5
MOL ENDO-1990 746
not detectable in the 3.2- to 4.6-kb fraction (A) or the 1.7- to 2.2-kb fraction (C). The level of 5'-deiodinase activity expressed in oocytes injected with an identical aliquot of RNA fraction C was 476 ± 10 cpm/oocyte/ 10 ng RNA. The activities produced by injection of fractions A and B were not significantly above background. Size fractionation, oocyte injection, and Northern blotting were performed twice, with the same qualitative results. Expression of type II deiodinase mRNA in oocytes was also investigated. Type II deiodinase activity is present in brown adipose tissue, and levels of activity are dramatically increased by exposing animals to low temperature (10). RNA was isolated from cold-stimulated brown adipose tissue, injected into oocytes, and assayed for expression of deiodinase, using T4 as substrate. Type II deiodinase activity was not reproducibly detectable with several RNA preparations. To attempt to enrich for type II deiodinase mRNA, brown adipose tissue poly(A)+ RNA was size-fractionated, as described for type I RNA. Size fractions were injected, and oocytes assayed. In three separate experiments, type II deiodinase activity was not significantly different from background deiodination in uninjected oocytes.
DISCUSSION
We report here that type I deiodinase activity can be expressed in Xenopus oocytes after injection of liver mRNA, confirming the results of St. Germain et al. (7). We also found type I deiodinase mRNA in kidney. Sizefractionation of liver mRNA and expression in oocytes established a range of 1.9-2.4 kb, similar to the results of St. Germain et al. (7), who reported peak activity in a fraction centered at 18S. Information about the number of subunits comprising the type I deiodinase would have important implications for designing a cDNA cloning strategy using an oocyte expression assay. Preliminary results of affinity labeling studies suggested that type I deiodinase may be composed of two subunits (11). Our mRNA size fractionation results suggest either that type I deiodinase is encoded by a single mRNA species or, less likely, by multiple species that are very close in size. Hepatic type I deiodinase activity has previously been shown to parallel thyroid status; it is low in the hypothyroid rat and elevated in the hyperthyroid rat. The results reported here demonstrate that levels of type I deiodinase mRNA correlate very closely with the changes in deiodinase activity produced by modulating thyroid status (Fig. 3). The close agreement between the effects of thyroid hormone on type I deiodinase activity in liver microsomes and activity produced by liver mRNA injected into oocytes indicates a pretranslational effect of thyroid hormone on type I deiodinase. This effect may be due to regulation of transcription or to altered stability of existing mRNA, as the present results do not distinguish between these two possibli-
ties. This is the first evidence of thyroid hormone regulation of deiodinase mRNA, and it supports earlier speculation based on kinetic assays of crude tissue fractions (2-4). Boada ef al. (6) speculated that PDI may be type I deiodinase or a subunit of the enzyme. A putative 5'deiodinase clone identified in an expression library with antibodies against solubilized liver microsomal proteins was found by sequence analysis to be virtually identical to PDI. Type I deiodinase has been characterized as a low abundance, basic protein (12), and PDI is an abundant, highly acidic protein (13). The mRNA for PDI is ~2.8 kb (13), in contrast to the results reported here indicating a size of 1.9-2.4 kb for type I deiodinase mRNA, suggesting that the two proteins are not identical. In further support of this are our findings that oocytes injected with PDI mRNA expressed no type I deiodinase activity and that size fractionation of liver mRNA allowed the physical separation of PDI-specific mRNA from mRNA encoding type I deiodinase activity. During the course of these experiments, we became aware of the study of Schoenmakers ef al. (14), which provides additional evidence that PDI is not related to type I deiodinase. We were unable to express type II deiodinase using either total or size-fractionated mRNA, which could be due simply to the low abundance of this enzyme. While substrate turnover numbers are not known for either type I or type II deiodinase, calculations using conditions of maximal activity indicate Vmax values of about 10 pmol T 4 /hmg protein for type II deiodinase in coldstimulated brown adipose (15), vs. 1800 pmol T 4 /h-mg protein for type I deiodinase in liver (1). Thus, it is not surprising that expression of type II deiodinase in oocytes would prove difficult. The availability of a system in which to express type I deiodinase mRNA and information about the size and best sources of this mRNA have provided a starting point for designing a cloning strategy for type I deiodinase cDNA. Construction of size-selected cDNA libraries in vectors permitting in vitro transcription of RNA, which can then be screened by injection into oocytes, should result in the isolation of a cDNA clone for this enzyme.
MATERIALS AND METHODS Materials Male Sprague-Dawley rats were obtained from Charles River Laboratories (Wilmington, MA) or Zivic Miller (Allison Park, PA). Xenopus laevis frogs were from Nasco (Fort Atkinson, Wl) or Xenopus One (Ann Arbor, Ml). All chemicals were of reagent grade except reagents used in RNA preparation, which were of ultrapure grade. Polyadenylated RNA Preparation Rats were made hyperthyroid by five sc injections of T4 (12 Mg/100 g BW) over 3-5 days. Hypothyroidism was produced by giving rats 0.02% methimazole in drinking water for 3
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2014. at 01:45 For personal use only. No other uses without permission. . All rights reserved.
Thyroid Hormone Regulation of Type I Deiodinase mRNA
747
weeks. Livers from six rats were used for each RNA preparation. Liver was homogenized in 4.0 M guanidinium thiocyanate, 20 ITIM sodium acetate, 10 mM vanadyl ribonucleoside complex, and 20 mM dithiothreitol (DTT) in a Brinkmann Polytron homogenizer (Westbury, NY), followed by three passages through a 20-gauge needle to shear chromosomal DNA. The homogenate was layered onto 12-ml cushions of 5.7 M CSCI0.1 M EDTA, pH 8.0, and centrifuged at 27,000 rpm for 18 h in a Beckman SW 28 rotor (Fullerton, CA) at 15 C. RNA pellets were resuspended in 2 mM EDTA (pH 8.0)-0.1% sodium dodecyl sulfate (SDS) and extracted with an equal volume of phenol-chloroform-isoamyl alcohol (25:24:1), followed by ethanol precipitation. Precipitated RNA was resuspended in 2 mM EDTA (pH 8.0)-0.1% SDS. Polyadenylated [poly(A)+] RNA was obtained by two cycles of chromatography on oligo-(dT) cellulose (Collaborative Research, Waltham, MA). Poly(A)+ RNA was ethanol precipitated and resuspended in diethyl pyrocarbonate treated (DEPC)-H2O before injection into oocytes or agarose gel size fractionation.
cDNA insert was excised using Apa\ and Smal, which cut at the 5' and 3' ends, respectively. The cDNA was ligated into Bluescript KS (Stratagene) which had been cut with Apa\ and Smal, sites for which are located in the polylinker region. Orientation of the insert in this construct was verified by restriction mapping. Bluescript containing PDI cDNA was linearized with Not\, and mRNA was transcribed in vitro using T3 RNA polymerase. Radiolabeled transcripts of the appropriate size were verified by agarose gel electrophoresis and autoradiography. For Northern blotting, this plasmid was linearized with EcoRI, and [32P]UTP-labeled cRNA was prepared with T7 RNA polymerase, using standard techniques.
Preparation of Microsomes Microsomal proteins were prepared separately from six rats at each thyroid state. Samples of liver tissue were removed from each rat for RNA preparation, and the remaining tissue was homogenized in ice-cold hypotonic buffer containing 50 mM Tris-HCI (pH 6.8), 10 mM DTT, and 1 mM EDTA. Homogenates were made isotonic by the addition of KCI to 120 mM and centrifuged at 20,000 x g for 20 min at 4 C in a Sorvall SS34 rotor (Norwalk, CT). The 20,000 x g supematants were then centrifuged at 190,000 x g in a SW40 Ti rotor for 90 min at 4 C. Pellets were resuspended in 100 mM potassium phosphate (pH 6.9)-1 mM EDTA and stored in aliquots at - 7 0 C. Size Fractionation of RNA Poly(A)+ RNA was electrophoresed in 0.8% low melting temperature agarose (Sea Plaque, FMC) in 20 mM 3-(/V-morpholino) propane sulfonic acid (MOPS) (pH 7.0), 5 mM sodium acetate, and 1 mM EDTA with ribosomal RNA and RNA mol wt markers (Bethesda Research Laboratories, Gaithersburg, MD) run in parallel lanes. The region between 1.4 and 5.4 kb was cut into six fractions. RNA was extracted from the gel with phenol, precipitated with ethanol, resuspended in DEPC-H2O, and injected into oocytes. For Northern blots, mRNA corresponding to 1.7-2.2,2.2-3.2, and 3.2-4.6 kb was extracted from the gel and analyzed on formaldehyde gels, as described below. Northern Blots Gels for Northern blotting contained 1.1% agarose, 20 mM MOPS (pH 7.0), 5 mM sodium acetate, 1 mM EDTA, and 1.3% (wt/vol) formaldehyde. RNA samples in the same buffer containing 50% formamide and 1.3% (wt/vol) formaldehyde were heated at 70 C for 5 min., chilled on ice, and loaded on gels. After electrophoresis, gels were rinsed in 10 x SSC (1.5 M NaCI-1.6 M sodium citrate, pH 7) and blotted overnight in 20 x SSC to a Duralon-UV membrane (Stratagene, La Jolla, CA). RNA was then cross-linked to the membrane using a UV Stratalinker (Stratagene). The membrane was prehybridized with salmon sperm DNA plus E. coli tRNA, probed with PDI cRNA, and washed in 1 x SSC-0.1% SDS at 22 C. This was followed by washes in decreasing concentrations of SSC, with the final wash being in 0.1 x SSC-0.1% SDS at 65 C.
Preparation and Injection of Oocytes Xenopus laevis were anesthetized by hypothermia, and ovarian lobes were surgically removed. Oocytes were manually dissected, injected with 40 nl RNA diluted to the desired concentration in DEPC-H2O, and incubated for 3 days at 18 C in 50% Leibovitz's L-15 medium, 15 mM HEPES, 100 fig/m\ gentamycin, and 50 U/ml nystatin. Uninjected and DEPC-H2O-injected oocytes had no significant deiodinase activity. Therefore, uninjected oocytes were included as controls in each experiment. 5'-Deiodinase Assays Because type I deiodinase exhibits a 1000-fold higher Vmax/Km ratio for rT3 than for T4 (1), rT3 was used as substrate for type I deiodinase assays. T4 was used as substrate for type II deiodinase assays. RNA-injected or uninjected oocytes were homogenized in 100 mM potassium phosphate (pH 6.9)-1 mM EDTA in microcentrifuge tubes, using a Teflon pestle. Homogenates were then divided into two or three replicate assays. Reaction volumes were adjusted to 100 fi\/oocyle. Type I deiodinase reactions were initiated by the addition of 0.5 nM [125l]rT3 and 10 mM DTT. Type II reactions contained 0.25 nM [125I]T4 and were assayed at both 20 and 50 mM DTT. Reactions were incubated for 1 h at 37 C and terminated by the addition of 50 n\ horse serum and trichloroacetic acid to a 10% final concentration. Trichloroacetic acid supematants were passed over Dowex AG50 W-X2 columns. The columns were washed with 2 ml 10% acetic acid, and the eluates were counted. Microsomes were assayed in duplicate, first using 100 ng protein for all assays, then using 50 n9 microsomal protein from euthyroid and hyperthyroid animals and 200 nQ microsomal protein from hypothyroid animals. The assay buffer contained 100 mM potassium phosphate (pH 6.9), 1 mM EDTA, 10 mM DTT, 0.5 nM [125l]rT3, and 15 MM unlabeled rT3. Reactions were incubated for 30 min at 37 C and terminated as described above. Determination of Km and Vmox
Km and Vmax determinations were performed using pools of the six microsomal preparations for each thyroid state. Either 5 Mg (euthyroid and hyperthyroid) or 25 nQ (hypothyroid) of microsomal protein were incubated as described above, varying the concentration of rT3 from 60-360 nM. Reactions were incubated for 5 min at 37 C. Data were analyzed using linear regression from double reciprocal plots.
Acknowledgments
Construction of PDI Expression Vector and Transcription of PDI mRNA and cRNA
The authors are grateful to Dr. Tom Segerson, Tufts University School of Medicine, for advice on the Xenopus oocyte expression system.
The plasmid ppdi 100 (13), which contains the complete cDNA for rat PDI in pBR322, was generously provided by W. J. Rutter, University of California, San Francisco. The entire
Received November 17, 1989. Revision received February 8,1990. Accepted February 15,1990.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2014. at 01:45 For personal use only. No other uses without permission. . All rights reserved.
Vol 4 No. 5
MOL ENDO-1990 748
Address requests for reprints to: Dr. Maria J. Berry, Howard Hughes Medical Institute, Brigham and Women's Hospital, 75 Francis Street, Boston, Massachusetts 02115. This work was supported in part by NIH Grant DK-36256.
7. 8.
REFERENCES
9. 10.
1. Leonard JL, Visser TJ 1986 Biochemistry of deiodination. In: Hennemann G (ed) Thyroid Hormone Metabolism. Marcel Dekker, New York, pp 189-229 2. Kaplan MM, Utiger RD 1979 lodothyronine metabolism in liver and kidney homogenates from hyperthyroid and hypothyroid rats. Endocrinology 105:548-554 3. Kaplan MM 1979 Changes in the particulate and subcellular component of hepatic thyroxine-5'-monodeiodinase in hyperthyroid and hypothyroid rats. Endocrinology 105:548-554 4. Smallridge RC, Wartofsky L, Burman KD 1982 The effect of experimental hyperthyroidism and hypothyroidism on 5'-monodeiodinationof3,3',5'-triiodothyronineand3',5'diiodothyronine by rat liver and kidney. Endocrinology 111:2066-2069 5. Larsen PR, Silva JE, Kaplan MM 1981 Relationships between circulating and intracellular thyroid hormones: physiological and clinical implications. Endocr Rev 2:87102 6. Boada RJ, Campbell DA, Chopra IJ 1988 Nucleotide sequence of rat liver 5'-monodeiodinase (5'MD): its iden-
11.
12.
13.
14.
15.
tity with the protein disulfide isomerase. Biochem Biophys Res Commun 155:1297-1304 St Germain DL, Morganelli CM 1989 Expression of type I iodothyronine 5'-deiodinase in Xenopus laevis oocytes. J Biol Chem 264:3054-3056 Kaplan MM, Yaskoski KA 1980 Phenolic and tyrosyl ring deiodination of iodothyroninies in rat brain homogenates. JClin Invest 66:551-562 Freedman RB 1989 Protein disulfide isomerase: multiple roles in the modification of nascent secretory proteins. Cell 57:1069-1072 Silva JE, Larsen PR 1983 Adrenergic activation of triiodothyronine production in brown adipose tissue. Nature 305:712-713 Mol JA, Docter R, Kaptein E, Jansen G, Hennemann G, Visser TJ 1984 Inactivation and affinity-labeling of rat liver iodothyronine deiodinase with /V-bromoacetyl-3,3',5-triiodothyronine. Biochem Biophys Res Commun 124:475483 Mol JA, Van den Berg TP, Visser TJ 1988 Partial purification of the microsomal rat liver iodothyronine iodothyronine deiodinase I. Solubilization and ion-exchange chromatography. Mol Cell Endocrinol 55:149-157 Edman JC, Ellis L, Blacher RW, Roth RA, Rutter WJ 1985 Sequence of protein disulphide isomerase and implications of its relationship to thioredoxin. Nature 317:267270 Schoenmakers CHH, Pigmans IGAJ, Hawkins HC, Freedman RB, Visser TJ 1989 Rat liver type I iodothyronine deiodinase is not identical to protein disulfide isomerase. Biochem Biophys Res Commun 162:857-868 Silva JE, Mellen S, Larsen PR 1987 Comparison of kidney and brown adipose tissue iodothyronine 5'-deiodinases. Endocrinology 121:650-656
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2014. at 01:45 For personal use only. No other uses without permission. . All rights reserved.
Maria J. Berry, Anna-Lisa Kates, and P. Reed Larsen Howard Hughes Medical Institute Laboratory Department of Medicine Brigham and Women's Hospital Harvard Medical School Boston, Massachusetts 02115
Conversion of the prohormone T4 to the active hormone T3 is catalyzed by 5'-deiodinases, enzymes that have not been purified. Previous studies have shown that modulating thyroid status results in changes in type I deiodinase activity in the rat liver. We have quantitated type I deiodinase mRNA in liver by an expression assay using Xenopus laevis oocytes. We report here that changes in enzyme activity correlate closely with changes in levels of the mRNA for this enzyme, indicating that thyroid hormone regulates type I deiodinase at a pretranslational step. Using the oocyte system to express sizefractionated mRNA, we have also determined that the mRNA coding for this protein is between 1.9-2.4 kilobases in length. It has been proposed that protein disulfide isomerase (PDI) is closely related to the rat type I 5'-deiodinase. Our results indicate that this is not the case, since injection of in vitro transcribed PDI mRNA into oocytes did not result in expression of deiodinase activity, and the deiodinase mRNA could be physically separated from the 2.8-kilobase mRNA species hybridizing to rat PDI cRNA by size fractionation. (Molecular Endocrinology 4: 743-748, 1990)
and brown adipose, functions primarily to provide an intracellular source of T3 for these tissues. This enzyme exhibits a Km for T4 of -about 2 nM and is PTU resistant. Many attempts at elucidating the molecular structure of these enzymes are in progress, but these efforts have to date been unsuccessful. Regulation of both type I and type II deiodinase activities by thyroid hormone has been examined using tissue homogenates and microsomes (2-4). Activity levels of type I deiodinase in rat liver are low in the hypothyroid state and elevated in hyperthyroidism. Conversely, type II deiodinase activity levels are regulated in the opposite direction, being low in tissues from hyperthyroid animals and elevated in hypothyroidism (5). Since no specific reagents for quantitation of the enzymes have been developed, it has not been possible to confirm that these activity changes are due to alterations in the enzyme content or, if so, whether they are transcriptional or post-transcriptional. It has also been suggested that type I deiodinase, which requires reduced thiols for maximal enzyme activity, is closely related to rat protein disulfide isomerase (PDI) (6). A growing number of proteins, including intracellular enzymes and cell surface receptors, have been found to be amenable to study after expression in Xenopus oocytes. St. Germain et al. (7) recently reported expression of type I deiodinase activity in Xenopus oocytes injected with rat hepatic RNA. We have employed this system to study the effects of thyroid hormone on type I deiodinase mRNA levels and have also evaluated the putative role of PDI as a deiodinase. We observed that type I deiodinase mRNA levels correlate closely with thyroid status and with type I deiodinase activity, indicating that thyroid hormone alters the tissue concentration of the mRNA coding for the type I deiodinase.
INTRODUCTION
Virtually all of the metabolic and developmental effects of thyroid hormone are mediated by T3, which is produced from T4 by 5'-deiodination. lodothyronine 5'deiodination is catalyzed by two general classes of enzymes (1) distinguished by their tissue distribution, physiological roles, Km for substrate, and sensitivity to propylthiouracil (PTU). Type I deiodinase, present predominantly in liver and kidney, provides most of the plasma T3 in the rat. This class of enzyme exhibits a Km for T4 of ~2 HM and is sensitive to inhibition by PTU. Type II deiodinase, found in pituitary, cerebral cortex,
RESULTS
To best use the Xenopus oocyte system for our studies, an optimal source of mRNA for expressing type I deiodinase was needed. Because previous studies have
0888-8809/90/0743-0748$02.00/0 Molecular Endocrinology Copyright © 1990 by The Endocrine Society
743
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2014. at 01:45 For personal use only. No other uses without permission. . All rights reserved.
Vol 4 No. 5
MOL ENDO-1990 744
shown that type I deiodinase activity is elevated in the hyperthyroid state (2-4), rats were made hyperthyroid, and poly(A)+ RNA was isolated from liver and kidney. Deiodinase activity was produced in oocytes after the injection of mRNA from both tissues; however, the level of activity produced by liver mRNA was about 10-fold higher than the level produced by kidney mRNA. Based on this result, liver was chosen as the tissue source. Most of the activity produced in oocytes from liver mRNA was found in the supernatant after centrifugation at 10,000 x g, suggesting that the activity was not associated with mitochondria or nuclei. To ascertain whether the enzyme made in oocytes catalyzed outer ring deiodination, products of deiodination were separated by descending paper chromatography according to previously described methods (8). The rT3 degraded appeared as free iodide and 3,3'diiodothyronine in equal quantities, as expected for specific outer ring deiodination. This deiodination was completely inhibited by 1 ITIM PTU (data not shown). These data indicate that this deiodinase activity can be categorized as type I. Analysis of the time course for expression of type I deiodinase showed maximal expression 3 days after mRNA injection. Deiodinase activity was between 30-50% of this level on days 2 and 4 and lower on days 1 and 5. To further evaluate tissue sources for mRNA and, in parallel, to explore the mechanism for thyroid hormone regulation of type I deiodinase, rats were made hyperthyroid or hypothyroid, or maintained euthyroid. Hepatic microsomal proteins were prepared from these animals. Aliquots from six microsomal preparations for each thyroid state were pooled and assayed for type I deiodinase activity. Km and maximum velocity (Vmax) values were determined from the Lineweaver-Burk double reciprocal plot shown in Fig. 1. Km values for rT3 were
0.36, 0.27, and 0.44 HM, and Vmax values were 119, 421, and 759 pmol/minmg protein for hypothyroid, euthyroid, and hyperthyroid preparations, respectively. These results suggest that thyroid hormone regulates the amount of enzyme, not the activity per unit protein, in agreement with previous studies (2-4). Type I deiodinase activity was measured from the individual microsomal preparations described above. The means of activity levels from two representative assays of six microsomal preparations at each thyroid state are shown in Fig. 2A. Hyperthyroid liver microsomal proteins contained about 150% of the level of deiodinase activity found in euthyroid microsomes. The hypothyroid level of deiodinase activity was about 30% of the euthyroid level.
400-i
c o o
300-
Q. O)
E
200-
c
E
i
100
o E o.
HYPO
EUTHYROO
HYPO*
MICROSOMES
•
HYPO
•
EUTHYROID
a HYPER
-10
Fig. 1. Lineweaver-Burk Plot of Type I Deiodination Catalyzed by Liver Microsomes from Hypothyroid, Euthyroid, and Hyperthyroid Rats Liver microsomal proteins from hypothyroid, euthyroid, and hyperthyroid rats were assayed for type I deiodinase activity with rT3 concentrations varying from 60-360 nM, as described in Materials and Methods.
o o o E
100-
Q. O HYPO
EUmYROD
HYPER
RNA INJECTED Fig. 2. Effects of Thyroid Status on Type I Deiodinase Activity and mRNA A, Microsomal protein from six individual preparations each of hypothyroid, euthyroid, and hyperthyroid rat liver were assayed for type I deiodinase activity, as described in Materials and Methods. B, Pooled poly(A)+ RNA from each group of animals was injected into 18 oocytes, which were incubated for 3 days at 18 C. Each mRNA preparation was injected in at least two concentrations from 10-40 ng/oocyte in two separate injection experiments. Oocytes were homogenized, divided into triplicate aliquots of six oocytes each, and assayed for type I deiodinase activity. Deiodinase activity is expressed as counts per min of 125I released from 200,000 cpm [125l]rT3 in a 60-min incubation. Results are the mean ± SE of the two experiments.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2014. at 01:45 For personal use only. No other uses without permission. . All rights reserved.
745
Thyroid Hormone Regulation of Type I Deiodinase mRNA
Hepatic mRNA was isolated from the same animals used above and assayed for production of type I deiodinase after injection into Xenopus oocytes. Deiodinase activity was found to increase linearly with 10, 20, and 40 ng mRNA from all three thyroid states. As shown in Fig. 2B, injection of oocytes with mRNA from hyperthyroid liver produced about twice the level of deiodinase activity observed with mRNA from euthyroid liver. The level of activity produced in oocytes by injection of hypothyroid liver mRNA was 47% of the euthyroid level. These results were confirmed in a separate experiment using RNA prepared from other groups of hypothyroid, euthyroid, and hyperthyroid animals. The correlation between hepatic microsome deiodinase activity and type I deiodinase mRNA, shown in Fig. 3, was determined by plotting hepatic microsome deiodinase activity from hypothyroid, euthyroid, and hyperthyroid animals vs. deiodinase activity expressed in oocytes injected with mRNA from these animals. This correlation was highly significant. To characterize further the mRNA for type I deiodinase, we determined its mol wt. Poly(A)+ RNA was size-fractionated on low melting temperature agarose, fractions were injected into oocytes, and deiodinase activity was determined. The results are shown in Fig. 4. The highest level of deiodinase activity was encoded by the fraction containing mRNA between 1.9-2.4 kilobases (kb), indicating that the mRNA for type I deiodinase is in this size range. It has been speculated that type I deiodinase is closely related to protein disulfide isomerase (6), an enzyme involved in protein folding (9). PDI mRNA was transcribed in vitro and injected into oocytes. No type I deiodinase activity was detected. In addition, using PDI cRNA as probe, we performed Northern blots of sizefractionated mRNA from rat liver. The PDI probe hybridized to a 2.8-kb mRNA present in total poly(A)+ RNA (Fig. 5, lane T) and the 2.2- to 3.2-kb fraction (B), but
kilobases < Z
15000-1
c o
10000-
O
5000-
o o E a u
5.4
4.4
3.5
uninj
1
2
3
oocy
lOngR
200-
100-
HYPO
1.4
4
Fig. 4. Size-Fractionation of Type I Deiodinase mRNA and Expression in Xenopus Oocytes Poly(A)+ RNA from hyperthyroid rat liver was size-fractionated on low melting temperature agarose. Fractions (20 ng) were injected into groups of 12 oocytes, which were incubated for 3 days at 18 C. Oocytes were homogenized, divided into duplicate aliquots, and assayed for type I deiodinase activity. The boundaries of each size fraction, in kilobases, are indicated above the fractions and were determined from a mRNA size ladder electrophoresed in parallel lanes.
7.5
-
4.4
-
2.4
-
1.4
-
I
EUTHYROID
E
M T
CL
100
1.9
Fraction
HYPER
o
2.4
o
400-
300-
2.9
200
300
400
pmol/min/mg Protein Fig. 3. Correlation between Type I Deiodinase Activity from Injected mRNA and Liver Microsomal Fractions Type I deiodinase activity expressed in Xenopus oocytes injected with liver mRNA was plotted vs. activity in liver microsomal extracts from hypothyroid, euthyroid, and hyperthyroid rats. Linear regression analysis by the method of least squares produced a correlation coefficient of 0.948.
A B C
Fig. 5. Northern Blot of Size-Fractionated Liver mRNA with PDI Probe Poly(A)+ RNA from hyperthyroid rat liver was size-fractionated on low melting temperature agarose, and 1-^g aliquots were analyzed on a 1.1% agarose formaldehyde gel. The gel was blotted to a Duralon UV nylon membrane, probed with PDI cRNA, and autoradiographed. Lane M, RNA size markers 7.5, 4.4, 2.4, and 1.4 kb (BRL); lane T, 5 fig unfractionated mRNA from hyperthyroid rat liver; lane A, 3.2- to 4.6-kb mRNA; lane B, 2.2- to 3.2-kb mRNA; lane C, 1.7- to 2.2-kb mRNA.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2014. at 01:45 For personal use only. No other uses without permission. . All rights reserved.
Vol 4 No. 5
MOL ENDO-1990 746
not detectable in the 3.2- to 4.6-kb fraction (A) or the 1.7- to 2.2-kb fraction (C). The level of 5'-deiodinase activity expressed in oocytes injected with an identical aliquot of RNA fraction C was 476 ± 10 cpm/oocyte/ 10 ng RNA. The activities produced by injection of fractions A and B were not significantly above background. Size fractionation, oocyte injection, and Northern blotting were performed twice, with the same qualitative results. Expression of type II deiodinase mRNA in oocytes was also investigated. Type II deiodinase activity is present in brown adipose tissue, and levels of activity are dramatically increased by exposing animals to low temperature (10). RNA was isolated from cold-stimulated brown adipose tissue, injected into oocytes, and assayed for expression of deiodinase, using T4 as substrate. Type II deiodinase activity was not reproducibly detectable with several RNA preparations. To attempt to enrich for type II deiodinase mRNA, brown adipose tissue poly(A)+ RNA was size-fractionated, as described for type I RNA. Size fractions were injected, and oocytes assayed. In three separate experiments, type II deiodinase activity was not significantly different from background deiodination in uninjected oocytes.
DISCUSSION
We report here that type I deiodinase activity can be expressed in Xenopus oocytes after injection of liver mRNA, confirming the results of St. Germain et al. (7). We also found type I deiodinase mRNA in kidney. Sizefractionation of liver mRNA and expression in oocytes established a range of 1.9-2.4 kb, similar to the results of St. Germain et al. (7), who reported peak activity in a fraction centered at 18S. Information about the number of subunits comprising the type I deiodinase would have important implications for designing a cDNA cloning strategy using an oocyte expression assay. Preliminary results of affinity labeling studies suggested that type I deiodinase may be composed of two subunits (11). Our mRNA size fractionation results suggest either that type I deiodinase is encoded by a single mRNA species or, less likely, by multiple species that are very close in size. Hepatic type I deiodinase activity has previously been shown to parallel thyroid status; it is low in the hypothyroid rat and elevated in the hyperthyroid rat. The results reported here demonstrate that levels of type I deiodinase mRNA correlate very closely with the changes in deiodinase activity produced by modulating thyroid status (Fig. 3). The close agreement between the effects of thyroid hormone on type I deiodinase activity in liver microsomes and activity produced by liver mRNA injected into oocytes indicates a pretranslational effect of thyroid hormone on type I deiodinase. This effect may be due to regulation of transcription or to altered stability of existing mRNA, as the present results do not distinguish between these two possibli-
ties. This is the first evidence of thyroid hormone regulation of deiodinase mRNA, and it supports earlier speculation based on kinetic assays of crude tissue fractions (2-4). Boada ef al. (6) speculated that PDI may be type I deiodinase or a subunit of the enzyme. A putative 5'deiodinase clone identified in an expression library with antibodies against solubilized liver microsomal proteins was found by sequence analysis to be virtually identical to PDI. Type I deiodinase has been characterized as a low abundance, basic protein (12), and PDI is an abundant, highly acidic protein (13). The mRNA for PDI is ~2.8 kb (13), in contrast to the results reported here indicating a size of 1.9-2.4 kb for type I deiodinase mRNA, suggesting that the two proteins are not identical. In further support of this are our findings that oocytes injected with PDI mRNA expressed no type I deiodinase activity and that size fractionation of liver mRNA allowed the physical separation of PDI-specific mRNA from mRNA encoding type I deiodinase activity. During the course of these experiments, we became aware of the study of Schoenmakers ef al. (14), which provides additional evidence that PDI is not related to type I deiodinase. We were unable to express type II deiodinase using either total or size-fractionated mRNA, which could be due simply to the low abundance of this enzyme. While substrate turnover numbers are not known for either type I or type II deiodinase, calculations using conditions of maximal activity indicate Vmax values of about 10 pmol T 4 /hmg protein for type II deiodinase in coldstimulated brown adipose (15), vs. 1800 pmol T 4 /h-mg protein for type I deiodinase in liver (1). Thus, it is not surprising that expression of type II deiodinase in oocytes would prove difficult. The availability of a system in which to express type I deiodinase mRNA and information about the size and best sources of this mRNA have provided a starting point for designing a cloning strategy for type I deiodinase cDNA. Construction of size-selected cDNA libraries in vectors permitting in vitro transcription of RNA, which can then be screened by injection into oocytes, should result in the isolation of a cDNA clone for this enzyme.
MATERIALS AND METHODS Materials Male Sprague-Dawley rats were obtained from Charles River Laboratories (Wilmington, MA) or Zivic Miller (Allison Park, PA). Xenopus laevis frogs were from Nasco (Fort Atkinson, Wl) or Xenopus One (Ann Arbor, Ml). All chemicals were of reagent grade except reagents used in RNA preparation, which were of ultrapure grade. Polyadenylated RNA Preparation Rats were made hyperthyroid by five sc injections of T4 (12 Mg/100 g BW) over 3-5 days. Hypothyroidism was produced by giving rats 0.02% methimazole in drinking water for 3
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2014. at 01:45 For personal use only. No other uses without permission. . All rights reserved.
Thyroid Hormone Regulation of Type I Deiodinase mRNA
747
weeks. Livers from six rats were used for each RNA preparation. Liver was homogenized in 4.0 M guanidinium thiocyanate, 20 ITIM sodium acetate, 10 mM vanadyl ribonucleoside complex, and 20 mM dithiothreitol (DTT) in a Brinkmann Polytron homogenizer (Westbury, NY), followed by three passages through a 20-gauge needle to shear chromosomal DNA. The homogenate was layered onto 12-ml cushions of 5.7 M CSCI0.1 M EDTA, pH 8.0, and centrifuged at 27,000 rpm for 18 h in a Beckman SW 28 rotor (Fullerton, CA) at 15 C. RNA pellets were resuspended in 2 mM EDTA (pH 8.0)-0.1% sodium dodecyl sulfate (SDS) and extracted with an equal volume of phenol-chloroform-isoamyl alcohol (25:24:1), followed by ethanol precipitation. Precipitated RNA was resuspended in 2 mM EDTA (pH 8.0)-0.1% SDS. Polyadenylated [poly(A)+] RNA was obtained by two cycles of chromatography on oligo-(dT) cellulose (Collaborative Research, Waltham, MA). Poly(A)+ RNA was ethanol precipitated and resuspended in diethyl pyrocarbonate treated (DEPC)-H2O before injection into oocytes or agarose gel size fractionation.
cDNA insert was excised using Apa\ and Smal, which cut at the 5' and 3' ends, respectively. The cDNA was ligated into Bluescript KS (Stratagene) which had been cut with Apa\ and Smal, sites for which are located in the polylinker region. Orientation of the insert in this construct was verified by restriction mapping. Bluescript containing PDI cDNA was linearized with Not\, and mRNA was transcribed in vitro using T3 RNA polymerase. Radiolabeled transcripts of the appropriate size were verified by agarose gel electrophoresis and autoradiography. For Northern blotting, this plasmid was linearized with EcoRI, and [32P]UTP-labeled cRNA was prepared with T7 RNA polymerase, using standard techniques.
Preparation of Microsomes Microsomal proteins were prepared separately from six rats at each thyroid state. Samples of liver tissue were removed from each rat for RNA preparation, and the remaining tissue was homogenized in ice-cold hypotonic buffer containing 50 mM Tris-HCI (pH 6.8), 10 mM DTT, and 1 mM EDTA. Homogenates were made isotonic by the addition of KCI to 120 mM and centrifuged at 20,000 x g for 20 min at 4 C in a Sorvall SS34 rotor (Norwalk, CT). The 20,000 x g supematants were then centrifuged at 190,000 x g in a SW40 Ti rotor for 90 min at 4 C. Pellets were resuspended in 100 mM potassium phosphate (pH 6.9)-1 mM EDTA and stored in aliquots at - 7 0 C. Size Fractionation of RNA Poly(A)+ RNA was electrophoresed in 0.8% low melting temperature agarose (Sea Plaque, FMC) in 20 mM 3-(/V-morpholino) propane sulfonic acid (MOPS) (pH 7.0), 5 mM sodium acetate, and 1 mM EDTA with ribosomal RNA and RNA mol wt markers (Bethesda Research Laboratories, Gaithersburg, MD) run in parallel lanes. The region between 1.4 and 5.4 kb was cut into six fractions. RNA was extracted from the gel with phenol, precipitated with ethanol, resuspended in DEPC-H2O, and injected into oocytes. For Northern blots, mRNA corresponding to 1.7-2.2,2.2-3.2, and 3.2-4.6 kb was extracted from the gel and analyzed on formaldehyde gels, as described below. Northern Blots Gels for Northern blotting contained 1.1% agarose, 20 mM MOPS (pH 7.0), 5 mM sodium acetate, 1 mM EDTA, and 1.3% (wt/vol) formaldehyde. RNA samples in the same buffer containing 50% formamide and 1.3% (wt/vol) formaldehyde were heated at 70 C for 5 min., chilled on ice, and loaded on gels. After electrophoresis, gels were rinsed in 10 x SSC (1.5 M NaCI-1.6 M sodium citrate, pH 7) and blotted overnight in 20 x SSC to a Duralon-UV membrane (Stratagene, La Jolla, CA). RNA was then cross-linked to the membrane using a UV Stratalinker (Stratagene). The membrane was prehybridized with salmon sperm DNA plus E. coli tRNA, probed with PDI cRNA, and washed in 1 x SSC-0.1% SDS at 22 C. This was followed by washes in decreasing concentrations of SSC, with the final wash being in 0.1 x SSC-0.1% SDS at 65 C.
Preparation and Injection of Oocytes Xenopus laevis were anesthetized by hypothermia, and ovarian lobes were surgically removed. Oocytes were manually dissected, injected with 40 nl RNA diluted to the desired concentration in DEPC-H2O, and incubated for 3 days at 18 C in 50% Leibovitz's L-15 medium, 15 mM HEPES, 100 fig/m\ gentamycin, and 50 U/ml nystatin. Uninjected and DEPC-H2O-injected oocytes had no significant deiodinase activity. Therefore, uninjected oocytes were included as controls in each experiment. 5'-Deiodinase Assays Because type I deiodinase exhibits a 1000-fold higher Vmax/Km ratio for rT3 than for T4 (1), rT3 was used as substrate for type I deiodinase assays. T4 was used as substrate for type II deiodinase assays. RNA-injected or uninjected oocytes were homogenized in 100 mM potassium phosphate (pH 6.9)-1 mM EDTA in microcentrifuge tubes, using a Teflon pestle. Homogenates were then divided into two or three replicate assays. Reaction volumes were adjusted to 100 fi\/oocyle. Type I deiodinase reactions were initiated by the addition of 0.5 nM [125l]rT3 and 10 mM DTT. Type II reactions contained 0.25 nM [125I]T4 and were assayed at both 20 and 50 mM DTT. Reactions were incubated for 1 h at 37 C and terminated by the addition of 50 n\ horse serum and trichloroacetic acid to a 10% final concentration. Trichloroacetic acid supematants were passed over Dowex AG50 W-X2 columns. The columns were washed with 2 ml 10% acetic acid, and the eluates were counted. Microsomes were assayed in duplicate, first using 100 ng protein for all assays, then using 50 n9 microsomal protein from euthyroid and hyperthyroid animals and 200 nQ microsomal protein from hypothyroid animals. The assay buffer contained 100 mM potassium phosphate (pH 6.9), 1 mM EDTA, 10 mM DTT, 0.5 nM [125l]rT3, and 15 MM unlabeled rT3. Reactions were incubated for 30 min at 37 C and terminated as described above. Determination of Km and Vmox
Km and Vmax determinations were performed using pools of the six microsomal preparations for each thyroid state. Either 5 Mg (euthyroid and hyperthyroid) or 25 nQ (hypothyroid) of microsomal protein were incubated as described above, varying the concentration of rT3 from 60-360 nM. Reactions were incubated for 5 min at 37 C. Data were analyzed using linear regression from double reciprocal plots.
Acknowledgments
Construction of PDI Expression Vector and Transcription of PDI mRNA and cRNA
The authors are grateful to Dr. Tom Segerson, Tufts University School of Medicine, for advice on the Xenopus oocyte expression system.
The plasmid ppdi 100 (13), which contains the complete cDNA for rat PDI in pBR322, was generously provided by W. J. Rutter, University of California, San Francisco. The entire
Received November 17, 1989. Revision received February 8,1990. Accepted February 15,1990.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2014. at 01:45 For personal use only. No other uses without permission. . All rights reserved.
Vol 4 No. 5
MOL ENDO-1990 748
Address requests for reprints to: Dr. Maria J. Berry, Howard Hughes Medical Institute, Brigham and Women's Hospital, 75 Francis Street, Boston, Massachusetts 02115. This work was supported in part by NIH Grant DK-36256.
7. 8.
REFERENCES
9. 10.
1. Leonard JL, Visser TJ 1986 Biochemistry of deiodination. In: Hennemann G (ed) Thyroid Hormone Metabolism. Marcel Dekker, New York, pp 189-229 2. Kaplan MM, Utiger RD 1979 lodothyronine metabolism in liver and kidney homogenates from hyperthyroid and hypothyroid rats. Endocrinology 105:548-554 3. Kaplan MM 1979 Changes in the particulate and subcellular component of hepatic thyroxine-5'-monodeiodinase in hyperthyroid and hypothyroid rats. Endocrinology 105:548-554 4. Smallridge RC, Wartofsky L, Burman KD 1982 The effect of experimental hyperthyroidism and hypothyroidism on 5'-monodeiodinationof3,3',5'-triiodothyronineand3',5'diiodothyronine by rat liver and kidney. Endocrinology 111:2066-2069 5. Larsen PR, Silva JE, Kaplan MM 1981 Relationships between circulating and intracellular thyroid hormones: physiological and clinical implications. Endocr Rev 2:87102 6. Boada RJ, Campbell DA, Chopra IJ 1988 Nucleotide sequence of rat liver 5'-monodeiodinase (5'MD): its iden-
11.
12.
13.
14.
15.
tity with the protein disulfide isomerase. Biochem Biophys Res Commun 155:1297-1304 St Germain DL, Morganelli CM 1989 Expression of type I iodothyronine 5'-deiodinase in Xenopus laevis oocytes. J Biol Chem 264:3054-3056 Kaplan MM, Yaskoski KA 1980 Phenolic and tyrosyl ring deiodination of iodothyroninies in rat brain homogenates. JClin Invest 66:551-562 Freedman RB 1989 Protein disulfide isomerase: multiple roles in the modification of nascent secretory proteins. Cell 57:1069-1072 Silva JE, Larsen PR 1983 Adrenergic activation of triiodothyronine production in brown adipose tissue. Nature 305:712-713 Mol JA, Docter R, Kaptein E, Jansen G, Hennemann G, Visser TJ 1984 Inactivation and affinity-labeling of rat liver iodothyronine deiodinase with /V-bromoacetyl-3,3',5-triiodothyronine. Biochem Biophys Res Commun 124:475483 Mol JA, Van den Berg TP, Visser TJ 1988 Partial purification of the microsomal rat liver iodothyronine iodothyronine deiodinase I. Solubilization and ion-exchange chromatography. Mol Cell Endocrinol 55:149-157 Edman JC, Ellis L, Blacher RW, Roth RA, Rutter WJ 1985 Sequence of protein disulphide isomerase and implications of its relationship to thioredoxin. Nature 317:267270 Schoenmakers CHH, Pigmans IGAJ, Hawkins HC, Freedman RB, Visser TJ 1989 Rat liver type I iodothyronine deiodinase is not identical to protein disulfide isomerase. Biochem Biophys Res Commun 162:857-868 Silva JE, Mellen S, Larsen PR 1987 Comparison of kidney and brown adipose tissue iodothyronine 5'-deiodinases. Endocrinology 121:650-656
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 24 November 2014. at 01:45 For personal use only. No other uses without permission. . All rights reserved.
