Vol. 171, No. 2, 1990 September

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

14, 1990

Page5

575-580

THYROID HORMONE AND DNA BINDING PROPERTIES OF A MUTANT C-ERBAB RECEPTOR ASSOCIATED WITH GENERALIZED THYROID HORMONE RESISTANCE Stephen J. Usalal, Fredric E. Wondisford2, Tracey L. Watsonl, 1 Department

of Medicine,

2 Department 3 Molecular, Received

July

Jay B Menkel,

and Bruce D. Weintraubs

East Carolina University School of Medicine, North Carolina, 27858-4354

of Medicine,

Case Western Reserve University School of Medjcine, Cleveland, Ohio, 44106

Cellular and Nutritional Endocrinology Health, Bethesda, Maryland, 23,

Greenville,

Branch, National 20892

Institutes

of

1990

SUMMARY: We have previously reported a family, Kindred A, with autosomal dominant generalized thyroid hormone resistance in which affected members were found to have a mutation in the carboxy-terminal domain of the c-erbA8 thyroid hormone receptor. In the currentstudy, the thyroid hormone and DNA-binding properties of this mutant receptor were determined using c-erbA8 protein synthesized in vitro. Both the wild-type human placental c-erbA8 and Kindred A receptors bound [125l]-triiodothyronine, although the Kindred A receptor had decreased affinity for the hormone. The affinity for triiodothyronine was 4.5 x 109 M-1 and 2.3 x 1010 M-1 for the mutant and wild-type receptors, respectively. No abnormality of DNA-binding was detected with the Kindred A receptor using a sensitive avidin-biotin DNA-binding assay with DNA fragments containing thyroid hormone response elements. The Kindred A mutant receptor which displays abnormal triiodothyronine-binding but normal DNA-binding activities in vitro acts as a dominant negative inhibitor of thyroid hormone action in man. 0 1990Academic mess, Inc. Generalized heterogeneous

thyroid

hormone

refractoriness

resistance is a syndrome

in peripheral

in man characterized

tissues and pituitary

by

to high levels of

circulating thyroid hormones (1,2). The c-erbA8 thyroid hormone receptor on chromosome 3 (3,4) has been linked to generalized thyroid hormone resistance in multiple

kindreds with different

phenotypes

(5,6). Recently different

c-erbAB

mutations have been identified in two kindreds with GTHR. A point mutation, Mf, consisting of a guanine to cytosine replacement at nucleotide position 1318 resulted in a glycine to arginine substitution in codon 340 in one of two alleles (7). An amplified cDNA fragment containing Mf was inserted into a wild-type c-erbA8 receptor, expressed in vitro, and the mutant binding!

Mf product

appeared

to have no T3-

(7).

ABBREVIATIW

T3, L-3,5,3’ triiodothyronine. 0006-291x/90

575

$1.50

Copyrigh? 0 1990 by Academic Press, Inc. All rights of reprodtktion in any form reserved.

Vol.

171,

No.

2,

In another

BIOCHEMICAL

1990

kindred,

A, we found

a base mutation,

nucleotide

position

mutation

was tightly

linked

to the resistance

affected

members

but in none

all of seven current

study

we have

placental

p-receptor

data

the

that

to the reported the Kindred gene

Kindred

inserted

expressed

this mutant

does

bind

We have

on thyroid gene

into

that

at (6). This

A being

members.

present

in

In the

the wild-type

human

receptor

in vitro,

and present

reduced

affinity

in contrast

the DNA-binding

response

and found

P-receptor

in Kindred

T3 but with

also tested

hormone

to adenine

unaffected

A mutation

A receptor

hormone

syndrome

the Kindred

(c-erbAPl),

A receptor

of the

of ten

RESEARCH COMMUNICATIONS

cytosine

the carboxy-terminus

Mf P-receptor.

and rat growth

to affect

1643 near

AND BIOPHYSICAL

elements this mutation

properties

of the human does

of TSHP

not appear

DNA-binding.

MATERIALS AND METHODS Kindred A Receptor cDNA The cloning from fibroblast RNA of a partial Kindred A cDNA (nucleotides 13361671) with the cytosine to adenine mutation at nucleotide position 1643 has been previously reported (6). The 3’-oligomer used to amplify and clone this cDNA sequence was complementary to the 3’-untranslated portion of the c-erbAfi1 sequence (nucleotides 1672-1698) reported by Weinberger et al. (3) except it contained a thymine at nucleotide position 1680 to create an Eco RI site. This was the Eco RI site used to obtain the Bgl II (1563) - Eco RI (1680) segment and create the Kindred A receptor plasmid shown in figure 1. The remaining portion of the Kindred A receptor plasmid consisted of pGEM3 and the wild-type human placental

Wild-Type cDNA beA 101)

Khdred A cDNA ““CII. (1336-1671) A-1643 RlBgl II

RI r--J

RI

A RI

TAG-End

RI. Bgl II /

Gel Purly Bgl II

I

Kindred

A P-Receptor

Figure 1. Construction of the Kindred A P-receptor cDNA. A plasmid containing a partial c-erbAP EDNA from Kindred A with the point mutation at nucleotide position 1643, C-+A, (top, right) was used. A Bgl II - Eco RI segment of this cDNA was inserted into the identical position of the wild-type human placental P-receptor cDNA contained in pGEM3 (peA 101) asshown. The Bgl II - Eco RI contained the stop codon (TAG) of the P-receptor.

576

Vol. 171, No. 2, 1990

BIOCHEMICAL

AND BIOPHYSICAL

RESEARCH COMMUNICATIONS

B-receptor from peA 101 (3). The presence of the Kindred A mutation by sequencing the Bgl II - Eco RI segment in the construct.

was confirmed

In vitro Translation cRNAs of the Kindred A receptor cDNA, wild-type receptor cDNA in peA 101, and anti-sense Kindred A receptor were prepared according to Promega (Madison, WI) specifications using m7G (5’)ppp(5’)G from Pharmacia (Piscataway, NJ). The in vitro translation reactions were prepared using the DuPont - NEN Reticulocyte Lysate L-3%Methionine Translation Kit (Wilmington, DE). Labeled products were analyzed on 10% SDS-polyacrylamide gels. TpBinding

Studies

2 pl of each translation reaction (wild-type and Kindred A receptors, antisense reaction products and translation products without added RNA) were incubated for 18 hours at 2” in 0.5 ml of buffer (50mM NaCI/lO%v/v glycerol/2mM EDTAI5 mM 2-mercaptoethanol/ZO mM Tris-HCI, pH 7.6) as described by lnoue et al. (8). L-3,5,3’[1251]-triiodothyronine of specific activity 2200 Ci/mmol was purchased from NENDuPont and final concentrations of [125l]T3 of .77 nM to ,038 nM were used in the nitrocellulose filter-binding assay of lnoue et al. T3-binding was measured with duplicate filters, and non-specific binding was determined in the presence of a LOOfold excess of unlabeled T3. DNA-Binding

Studies

The avidin-biotin DNA-binding assay used to measure Kindred A and wild-type receptor binding to DNA fragments of the TSHP and rat growth hormone genes, and a fragment of the long terminal repeat of adenovirus 5, has been described (9). Briefly, DNA fragments containing 5’-overhands were filled in with biotin-I I-dUTP and Taq DNA polymerase. Each DNA fragment contained eleven biotinylated residues and was quantitatively precipitated by a streptavidin-agarose matrix. The biotinylated fragments were incubated with radiolabeled Kindred A receptor or wild-type B-receptor, and non-specific fragments were precipitated with a streptavidin-agarose matrix and the amount of receptor bound was quantitated using liquid scintillation counting. RESULTS The %-labeled

products of in vitro translation

from wild-type

c-erbA8 and Kindred A cRNAs were analyzed on polyacrylamide two proteins corresponding to the 52,200 and 49,100 molecular predicted

from the human placental

human placental gels and showed weight-proteins

c-erbAB cDNA (10) (data not shown).

labeled products from anti-sense Kindred A and control translation

No 3%

mixtures were

seen. TX-binding of the wild-type and Kindred A translation products were quantitated with the filter-binding assay of lnoue et al. (8). The Kindred A receptor was repeatedly shown to bind T3 with relatively high-affinity. The binding of T3 over a concentration

range of 0.04 - 0.80 nM is shown by Scatchard plot in figure 2. The

Kindred A translation product bound T3 with an affinity of approximately 4.5 X 109 M-r whereas the wild-type c-erbAB translation product bound T3 with a higher affinity of 2.3 X 1010 M-1. The anti-sense or control translation products did not bind significant

quantities

of T3. 577

Vol. 171, No. 2, 1990

BIOCHEMICAL

AND BIOPHYSICAL

0.00

RESEARCH COMMUNICATIONS

0.02

Bound (nM) Fi ure 2. Scatchard plot of TX-binding data with the Kindred A (circles) and wildproducts. The data are fit with the method of +type trrangles) in vitro translation least squares; the Kindred A and wild-type human placental c-erbAP receptor T3-affinitieswere 4.5 X 109 M-l and 2.3 X 1010 M-l, respectively.

To study the DNA-binding

properties

of the Kindred A and wild-type

the avidin-biotin

DNA binding

previously found

to be the most sensitive method

interactions

assay was employed

encompassing

(9). This assay has been for studying

and has been widely used in the characterization

elements of several genes (9,11,12).

receptors,

DNA fragments

T3-receptor-DNA of Tj-response

from the human TSHP gene

-12 to + 43 bp (11) and from the rat growth

hormone

gene

encompassing -188 to -160 bp (12) that were known to contain thyroid hormone response elements and to bind c-erbAP were used. Also, a fragment from the long terminal

repeat of adenovirus

was used for a control

5 which does not bind significant

(12). The amount

c-erbAP that bound to these fragments

AD5

of W-labeled

wild-type

amounts

are shown in figure 3. Equivalent

RAT OH

of c-erbAP

and mutant amounts

HUMAN TSHD

Figure 3. Binding of the wild-type human placental c-erbA8 and Kindred A in vitro receptors to DNA fragments containing the TSH8 segment -12 to + 43 and rat rowth hormone gene segment -188 to -160. The negative control wasa DNA 3ragment containing the adenovirus 5 long terminal repeat. The avidin-biotin DNA binding assay was employed with one picomole of the respective biotin lated DNA fragment and equivalent amounts of W-labeled wild-type and Kindre CF.A rn v&o translation products were used as quantitated by SDS-polyacrylamide gel electrophoresis.

578

Vol.

171,

No.

2,

of wild-type

1990

BIOCHEMICAL

and mutant

gel electrophoresis. bound

with

c-erbAfl

of the in vitro

were

The rat growth

comparable

BIOPHYSICAL

RESEARCH

used as quantitated

hormone

or perhaps

Kindred

AND

by SDS-polyacrylamide

and human

slightly

greater

COMMUNICATIONS

TSHP gene

avidity

fragments

to an equivalent

amount

A receptor.

DISCUSSION Two

mutations

in the c-erbA@

reported

in different

Kindred

A in codon

340 (glycine

kindreds

fibroblast

also showed

a reduced

have

been

thyroid

hormone

mutant

forms

receptors

(14,15,16),

However

such studies

(14,15).

hydrogen

bond

to form

unique

conformation

produced

in vitro

the ability

to bind

for the wild-type Kindred levels

needed

We now

biochemical

isdefective

in binding

T3 with

relatively

receptor, thyroid

The Mf mutant

receptor

is due to the differences the result

binding seven

c-erbA

cDNAs

T3-binding

including wild-type

receptor

T3-affinity

iswithin

was only

shown

of T3 relative

human

to the of the

not appear

Thyroid

hormone

human

c-erbAg

quantities

elements.

to the

which

cannot

may result

in a

transcriptional A receptor

Yet, this receptor 5-fold

when did have

less than

affected

euthyroidal

that

members

status

with

Kindred

to perturb receptors

T3 (7); it is unclear

Kindred

for T3 of the

the in vitro

structure

attention

the Kindred

A recent

the wild-type

placental

of the Mf receptor

does

the

methodologies.

of T3 concentration

The altered

human the

(17) and the

why the

not to bind

this range

function

drawn is a residue

approximately

a relatively

between

the affinities

human

M-1 and our measured

propertiesof

for high-affinity

hormone.

affinity,

was reported

of different

of the

of the a and fi

conformation

that

A

of

higher

hormones. in structure

assay compared

T3-affinity

high

and maintain

of circulating

the

is crucial

and this may in part explain

A can compensate

rather

thyroid

using

studies

had indicated

and subsequent evidence

work

such methods

of the P-receptor

for T3-binding

present

to GTHR

although regions

Proline

at the carboxy-terminus

linked

variousTs-binding about

or beta-sheet

Mf, in codon

Earlier

had not previously

at codon-448.

one in

of Kindred

studies

receptors

an alpha-helix

sequence

activation.

proline

is known

been

member

of different

mutagenesis

hormone

role of the single

proline-proline

little

In vitro

of the thyroid

important

degree, Although

the functions

of the receptors.

carboxy-terminus T3-binding.

to elucidate

mutation, A wastightly

an affected

(13).

now

resistance,

for the syndrome.

to a comparable

have

hormone

in Kindred from

and imprecise.

gene

and another

receptors

T3-affinity

employed

thyroid

responsible

nuclear

receptor

generalized

The mutation

defect

cumbersome

hormone

to histidine)

(6,7).

molecular

salt-extracted are relatively

with

448 (proline

to arginine)

and is the likely

thyroid

placental

have 579

in vitro

using

range.

A receptor

products

P-receptor

was reported that

the filter-

translational

placental

this or

(18).

to be .8-7.4X Data

of The 1010

on the of .0138

also bound

only

nM. small

p-receptor.

A receptor its ability

report

for a T3 concentration

Kindred

whether

A and Mf receptors

which

to bind been

shown

affects

to thyroid to bind

itsT3-binding hormone

response

to thyroid

hormone

In

Vol.

171, No. 2, 1990

response

elements

c-erbA@

related

may reduce

proteins

receptor

the generalized with

elucidation

via a leucine

would thyroid

the Kindred

zipper

activating

function

form which

explain

why

mutant

of gene

(20).

The Kindred

of the c-erbAB with

normal

phenotype

A mutation

DNA-binding allele

in Kindred will

by c-erbA

through

Alternatively,

of T3-regulated

c-erbAj3

TX-receptors

regulation

and other

receptor

receptor.

the activity

one mutant

resistance

A and other

of the mechanisms

inhibits only

each other

the normal

a homodimer

herein, hormone

with

RESEARCH COMMUNICATIONS

with

structure

heterodimer

may itself

as suggested

Such mechanisms

AND BIOPHYSICAL

(12, 19) and may interact

of a less active

the mutant

studies

as dimers

the transcriptional

the formation properties

BIOCHEMICAL

help

genes.

is required

for

A. Further in the

proteins.

ACKNOWLEDGMENT We wish to thank peA

Cary Weinberger

of the Scripps

Clinic,

San Diego,

California

for

101.

REFERENCES Refetoff 5. (1982) Am. J. Physiol. 243:E88-E98. :: Magner JA, Petrick P, Menezes-Ferreira M, and Weintraub BD. (1986) J. Endocrinol. Invest. 9:459-69. 3. Weinberger C, Thompson CC, Ong ES, Lebo R, Gruel DJ, Evans RM. (1986) Nature (Lond.). 324:641-46. 4. Drabkin H, Kao FT, Hartz J, Hartz I, Gazdar A, Weinberger C, Evans R, and Gerber M. (1988) Proc. Natl. Acad. SC;. USA. 85:9258-62. 5. Usala SJ, Bale AE, Gesundheit N, Weinberger C, Lash RW, Wondisford FE, McBride OW, and Weintraub, BD. (1988) Mol. Endocrinol. 2:1217-20. 6. Usala SJ, Tennyson GE, Bale AE, Lash RW, Gesundheit N, Wondisford FE, Accili D, Hauser Pand Weintraub, BD. (1990) 1. C/in. Invest. 85:93-100. 7. Sakurai A, Takeda K, Ain K, Ceccavelli P, Nakai A, Seino 5, Bell Cl, Refetoff 5, and DeGroot LJ. (1989) Proc. Nat/. Acad. Sci USA. 86:8977-8981. 8. lnoue A, Yamakawa J, Yukioka M, and Morisawa 5. (1983) Anal. Biochem. 134:176-183. C, Albert, VR, Evans, RM, and Rosenfeld, MG. 9. Glass, CK, Franc0 R, Weinberger (19871 Nature 329:738-741. 10. Sam&s HH, Forman BM, Horowitz, ZD, and Ye ZS. (1988) J. Clin. Invest. 81:957-967. 11. Wondisford, FE, Farr EA, Radovick, 5, Steinfelder, HJ, Moates, JM, McClaskey, JH, and Weintraub, BD. (1989) J. Biol. Chem. 264:14601-14604. 12. Glass, CK, Holloway, JM, Devary, OV, and Rosenfeld, MG. (1988) Cell 54:313323. 13. Weintraub, BD, Usala SJ, Bale, AE, Gesundheit N, Weinberger C, Lash RW, Wondisford, FE, McBride OW, and Menezes-Ferreira, MM. “In Progress in Endocrinology,” ed. by H. Imura, pp. 797-802, Elsevier Science Publishers, B. V. Amsterdam, 1988. 14. Munoz A, Zenke M, Gehring U, Sap J, Beug H, and Vennstrom B. (1988)

EM60 1. 7: 155-I 59. 15. Horowitz ZD, Yang C, Forman BM, Casanova J, and Samuels HH. (1989) Mol. Endocrinol. 3: 148- 156. 16. O’Donnell AL, Koenig RJ. (f990) Mol. Endocrinol. 4:715-720. 17. Creighton, TE (1989) Proteins. p. 170. W. H. Freeman & Company, San Francisco.

18. Schueler PA, Schwartz HL, Strait KA, Mariash CN, and Oppenheimer JH. (1990) Mol. Endocrinol. 4:227-234. 19. Glass CK, tipkin SM, DeVary OV, Rosenfeld MG. (1989) Cell 59:697-708. 20. Forman BM, Yang C, Au M, Casanova J, Ghysdael J, Samuels HH. (1989) Mol. Endocrinol. 3 : 16 IO- 1626.