0013-7227/00/$03.00/0
Endocrinology
Copyright © 2000 by The Endocrine Society
Vol. 141, No. 12
Printed in U.S.A.
Luteinizing Hormone Receptors Are Self-Associated in
the Plasma Membrane*
DEBORAH A. ROESS, REGINA D. HORVAT, HEIDI MUNNELLY,
B. GEORGE BARISAS
AND
Department of Physiology (D.A.R.), Cell and Molecular Biology Program (R.D.H.), and Department of
Chemistry (H.M., B.G.B.), Colorado State University, Fort Collins, Colorado 80523
ABSTRACT
We have evaluated rat LH receptor self-association and lateral
dynamics for functional and nonfunctional receptors after binding of
hormone. We demonstrate, for the first time, that grouped receptors observed in electron or light microscopy represent actual receptor
dimers or oligomers rather than simply the concentration of receptors within membrane microdomains. Fringe fluorescence photobleaching recovery methods showed that, after binding of either LH
or human CG (hCG), functional wild-type LH receptors, expressed on
293 cells (LHR-wt cells), have mobilities that are 25% lower than
those of nonfunctional LH receptors containing an arginine substitution for lysine at position 583 (LHR-K583R cells). Because lateral
diffusion coefficients in two dimensions depend only on the logarithm
T
HE NATURE OF LH receptor interactions with other
membrane proteins during signal transduction is not
well understood. Several lines of evidence suggest that LH
receptors exist in discrete complexes in the plasma membrane after hormone binding. Elegant electron microscopy
studies by Luborsky et al. (1) demonstrate that extensive
association of LH receptors into clusters containing multiple
copies of the LH receptor occurs after exposure of rat luteal
cells to high concentrations of ovine LH. Similarly, immunofluorescence studies of the LH receptor on rat granulosa
cells demonstrates the presence of large, punctate structures
on the cell membrane after ligand binding (2). In timeresolved phosphorescence anisotropy studies of receptor rotational diffusion, long rotational correlation times for the
LH receptor on bovine and ovine plasma membranes are also
consistent with the notion that LH receptors are present in
large complexes of restricted mobility (3). The physical size
of receptor-containing complexes may be indicative of the
receptor’s response to hormone binding: functional hormone-receptor complexes, i.e. those capable of activating
adenylate cyclase, exhibit significantly slower rotational dynamics than do complexes formed by hormone binding to
nonfunctional receptors or by a nonfunctional ligand binding
to a normally functioning receptor (4).
Nonetheless, the question of whether liganded LH receptors are intimately self-associated, forming receptor dimers
Received April 18, 2000.
Address all correspondence and requests for reprints to: Dr. Deborah
A. Roess, Department of Physiology, Colorado State University, Fort
Collins, Colorado 80523. E-mail: [email protected].
* This work was supported by NIH Grants HD-23236 and HD-01067
(to D.A.R.).
of the molecular size of the diffusing species, this result implies that
functional receptors exist in substantially larger membrane complexes than do nonfunctional receptors. In single-cell measurements
of fluorescence energy transfer after hormone binding, functional LH
receptors were also characterized by receptor self-aggregation. Values
for fluorescence resonant energy transfer efficiency were 13 ⫾ 2% and
17 ⫾ 3% between fluorophore-conjugated LH or hCG, respectively,
bound to receptors on LHR-wt cells. However, there was little or no
energy transfer between receptors on LHR-K583R cells. These results
suggest that receptor functionality involves receptor-receptor interactions and that the extent of such receptor self-association depends
on whether LH or hCG binds the receptor. (Endocrinology 141: 4518 –
4523, 2000)
or oligomers after binding of hormone, has not been resolved. Such interactions have been suggested by electron
microscopy studies of this receptor, which show grouping of
hormone-conjugated ferritin molecules. However, the diameter of ferritin molecules used to image LH receptors is about
24 nm (1), approximately 3-fold greater than the diameter of
the hormone itself (5). Thus, these studies fail to distinguish
actual receptor oligomerization from simple concentration of
receptors with small membrane microdomains. Similarly,
light microscopy results showing fluorescent clusters containing LH receptors (2, 6) can arise either from receptor oligomers
or from restriction of receptors to specific small domains.
Because there is evidence that at least dimeric structures
may be necessary for function of G protein-coupled receptors, such as the ␦-opioid receptor (7) and ␤2-adrenergic
receptor (8), the question of whether the functional LH receptor self-associates is of interest. We have approached this
question by first examining the differences in the lateral
dynamics of the wild-type rat LH receptor and a nonfunctional LH receptor expressed in human embryonic kidney
293 cells. The nonfunctional receptor contains a single point
mutation in lysine 583 located in the third extracellular loop,
a domain believed to be involved in signal transduction (9,
10). In cells stably expressing the mutant receptor, the cAMP
response to human CG (hCG) is either eliminated completely
(9) or reduced by over 75% (10). We have also examined
receptor-receptor interactions on individual cells, using fluorescence resonance energy transfer techniques (11) to determine whether large, slowly diffusing complexes contain
self-associated LH receptors. Fluorescence resonance energy
transfer, whether via spectroscopic methods or flow cytometric techniques (12), has proven useful in detecting mo-
4518
FUNCTIONAL LH RECEPTORS ARE SELF-ASSOCIATED
lecular associations in the plasma membrane. Because the
characteristic Förster distance Ro for the fluorescein-rhodamine pair used in these studies is approximately 56Å (13),
energy transfer between hormone-occupied LH receptors
occurs under conditions where receptors are within less than
approximately 100Å of each another (5).The results presented here suggest that functional, but not nonfunctional,
LH receptors are self-associated and present in slowly diffusing complexes.
Materials and Methods
Materials
DMEM was purchased from Irvine Scientific (Santa Ana, CA). Gentamicin and geneticin were purchased from Life Technologies, Inc. (Grand
Island, NY). HEPES was purchased from Sigma (St. Louis, MO). FBS was
purchased from Summit Biotechnology (Fort Collins, CO). Ovine LH (oLH;
NIH 28) and hCG (CR-127) were obtained from the National Hormone and
Pituitary Program, NIDDK (Baltimore, MD). Tetramethylrhodamine isothiocyanate (TrITC) and fluorescein isothiocyanate (FITC) were purchased
from Molecular Probes, Inc. (Eugene, OR).
Cell culture
Dr. Tae Ji, from the Department of Chemistry at the University of
Kentucky, kindly provided 293 cells stably transfected with the wildtype LH receptor (LHR-wt) or with an LH receptor modified in position
583 by substitution of lysine with arginine (LHR-K583R) (9). Untransfected 293 cells were maintained in DMEM containing 10% horse serum
(Summit Biotechnology), 100 U penicillin, 1000 ␮g/ml streptomycin,
and 10 mm HEPES, pH 7.4. LHR-wt cells and LHR-K583R cells were
maintained in the same medium supplemented with 400 ␮g/ml
geneticin.
4519
labeled with fluorescence donor and acceptor were incubated with 1.5
␮m TrITC- and 0.5 ␮m FITC-derivatized hormone. Cells labeled with
fluorescence acceptor were treated with 1.5 ␮m TrITC-derivatized hormone and 0.5 ␮m unlabeled hormone. The 3:1 ratio of fluorescent acceptor to donor has been shown previously to produce optimal signal
(11). After labeling for 1 h at 37 C, cells were then washed two times by
centrifugation at 300 ⫻ g for 3 min in balanced salt solution to remove
any unbound ligand. In some lateral diffusion and fluorescence energy
transfer experiments, cells were pretreated with 40 ␮g/ml cytochalasin
D for 30 min at 37 C before cell labeling.
Fringe fluorescence photobleaching recovery measurements
The optical system for performing fringe fluorescence photobleaching recovery measurements and the method used for data analysis have
been described previously (16). The microscope objective used in these
studies was a 40⫻ objective of NA 0.65 (Carl Zeiss, Inc., New York, NY).
Cells were examined under coverslip on well slides while temperature
was maintained by a thermoelectrically cooled/heated thermal stage
with a temperature range of 0 – 40 C. For fringe measurements, the region
illuminated at the sample has a 1/e2 radius of about 18 ␮m, and the
photometer acceptance region is large enough to encompass the entire
cell. The fringe spacing used in these experiments was 2.3 ␮m. Because
of the large interrogated area, 1.3 W in the bleaching pulse and 3 mW
in the probe beam were used. Unadjusted raw data were represented
directly in terms of the various parameters associated with a given
measurement, including the prebleach and immediate postbleach fluorescence levels, the extent M of fluorophore mobile on the timescale of
the experiment, and a function representing the recovery kinetics in
terms of a decay half-time. These parameters were evaluated directly by
a nonlinear least-squares procedure; and, from the measured time t1/2
at which fluorescence recovery was half-complete and from the known
optical parameters, the desired diffusion coefficient D was evaluated.
Single-cell fluorescence energy transfer
Preparation of TrITC- and FITC-derivatized hormones
Hormones were derivatized with FITC or TrITC using a modification
of methods described by Brinkley et al. (14) and described in detail
elsewhere (11). Briefly, hormones were dissolved in PBS (1.9 mm
NaH2PO4, 8.4 mm Na2HPO4, 0.15 m NaCl, PBS) containing 50 mm
sodium borate, pH 9.3. Protein concentrations were determined spectrophotometrically at 280 nm. A 5-fold molar excess of TrITC or FITC
was added to the protein solutions, and the mixtures were kept at 4 C
for 18 h in the dark. After quenching with 1 m Tris, the fluorophorederivatized hormones were separated from the unreacted free dye on a
Sephadex G-25 column. After extraction of remaining free dye with
n-butanol and extensive dialysis, the molar ratios of dye to hormone
were determined spectrophotometrically. Hormone preparations used
in these experiments had 1.0 –1.5 mol TrITC or FITC per mol oLH or
hCG. It has been previously shown that there is no effect of these
fluorophore conjugations on hormone activity (15). Before use, all fluorophore-derivatized proteins were centrifuged at 130,000 ⫻ g for 10 min
in an Airfuge (Beckman Instruments, Inc., Palo Alto, CA) to remove any
protein aggregates formed during storage at 4 C.
Labeling cells with fluorophore-derivatized hormones
Before labeling with TrITC- or FITC-derivatized hormones, cells were
incubated in balanced salt solution containing 0.1% NaN3, at 37 C for 30
min, to prevent hormone internalization (15). Typically, 2–5 ⫻ 106 cells
in 1 ml balanced salt solution were labeled with 1 ␮m TrITC-derivatized
oLH or hCG for each fringe photobleaching recovery experiment. This
hormone concentration saturates available receptors and results in maximum cAMP production by LHR-wt cells (data not shown). For fluorescence energy transfer measurements, four groups, each containing
1 ⫻ 106 cells, were labeled and examined on a given day. Thus, receptor
number per cell was comparable for cells in each group. One group of
cells was not labeled. The remaining groups were labeled with a total
hormone concentration of 2.0 ␮m. Cells labeled with fluorescent donor
alone were incubated with 1.5 ␮m unlabeled hormone (either LH or
hCG) and 0.5 ␮m of the same hormone was derivatized with FITC. Cells
Fluorescence energy transfer between FITC- and TrITC-derivatized
LH or hCG was evaluated based on the reduced rate of irreversible
photobleaching of FITC fluorophores when TrITC fluorophores were
present (11). Slower rates of fluorescence decay for cells labeled with the
FITC fluorescence donor and TrITC fluorescence acceptor, than for cells
labeled with FITC only (D), were indicative of energy transfer from fluorescence donor to acceptor and occurred only when the donor and acceptor
were separated by distances less than Ro, a characteristic of the specific
donor/acceptor pair. For FITC and TrITC, this distance is 56Å (13).
To perform these experiments, we used a fluorescence microscope
photometer based on the inverted-configuration Carl Zeiss Axiomat
microscope and associated components used for fringe fluorescence
photobleaching recovery measurements at room temperature. Fluorescence excitation was provided by an Innova 100 argon ion laser (Coherent Inc., Santa Clara, CA) operating under light control at 488 nm. The
intensity of the laser radiation focused on the cell was 15–20 mW, and
this was held constant between measurements on cells labeled with
FITC-derivatized LH or hCG only or on cells labeled with FITC- plus
TrITC-derivatized hormone. The 1/e2 Gaussian spot diameter was 18
␮m. Donor fluorescence from FITC was isolated with a standard fluorescein filter set together with a short-pass fluorescein-selective filter to
remove red tetramethylrhodamine fluorescence. This combination was
highly effective in rejecting TrITC fluorescence: TrITC-LH- or -hCGlabeled cells gave very low fluorescence signals using the fluoresceinselective filter set that were indistinguishable from those of unlabeled
cells. Signals from cells labeled with either FITC-LH or hCG only or with
FITC- and TrITC-LH or hCG were approximately 4-fold higher than
background levels. In individual experiments, cells were identified and
centered in the microscope field. At time zero, an electronically controlled shutter was opened to allow laser radiation to impinge on the cell.
Simultaneously, a computer program was activated to record the output
of the photomultiplier measuring membrane fluorescence. Data were
collected at 0.01-sec intervals for 10 sec. Typically, about 20 cells in each
sample were photobleached in this manner. The data traces were analyzed to give the energy transfer efficiency (%E), as has been described
in detail previously (11).
4520
FUNCTIONAL LH RECEPTORS ARE SELF-ASSOCIATED
Statistical analysis of data
In photobleaching recovery and fluorescence energy transfer experiments, diffusion coefficients and energy transfer efficiencies were obtained through curve fitting appropriate mathematical models to experimental data sets. These data sets contained hundreds of points, and
fitting is accomplished using the Marquardt algorithm (17). Because
each of the many observations in a single measurement provides independent information on the parameter of interest, the se of the parameter
was calculated at the same time as the fitted parameter itself. However,
because any real data set has some systematic deviation from a model
representing the parent experiment, these standard errors calculated
during the curve-fitting procedure almost certainly overestimate the
reliability of parameters. We thus present the uncertainties of a fitted
parameter x as ⬍x⬎⫾ 2s where s is the sem of a set of three to four
complete, independent determinations of x. Uncertainties in quantities,
such as percent efficiency of energy transfer, which involve parameters
obtained in at least three separate experiments, were calculated by
standard propagation of errors methods. Decisions as to whether parameters differ significantly between (18) multiple treatment groups
were made using single classification ANOVA methods (SigmaStat,
Jandel Scientific, San Rafael, CA).
Results
LH receptors are laterally mobile at 37 C after binding of
LH or hCG
We examined the lateral diffusion of the LH receptor expressed on LHR-wt and LHR-K583R cells, using fringe photobleaching recovery techniques. The entire surface of cells
was interrogated with a interferometrically-generated fringe
pattern (16). This permitted us to obtain lateral diffusion
information from a large population of LH receptors in a
single measurement. Because 293 cells exhibited low levels
of autofluorescence when excited by 514 nm light, we also
performed fringe fluorescence photobleaching recovery ex-
FIG. 1. Data traces from fringe fluorescence photobleaching recovery measurements at 37 C after binding of either TrITC-LH or TrITC-hCG to LH
receptors on LHR-wt or LHR-K583R
cells. In fringe fluorescence photobleaching recovery experiments, fluorescence for fully mobile proteins
(%M ⫽ 100%) recovers to approximately one third the prebleach level
(16). The fluorescence contribution attributable to cellular autofluorescence
has been removed from these traces, as
described in Materials and Methods.
Endo • 2000
Vol. 141 • No. 12
periments on untreated cells transfected with the appropriate form of the LH receptor, summed and averaged approximately 40 data traces from individual cells, and then
subtracted the average background signal from individual
traces obtained when LH receptors were labeled with TrITCderivatized hormones. Representative data are shown in Fig.
1. Using this protocol, lateral diffusion coefficients D for LHand hCG-occupied receptors on the various cell types ranged
from 2.1– 4.5 ⫻ 10⫺10cm2sec⫺1 at 37 C (Table 1). There was a
significant difference between the diffusion coefficients for
LH- and hCG-occupied receptors on LHR-wt cells: oLHoccupied receptors on LHR-wt cells had a diffusion coefficient of 3.2 ⫾ 1.0 ⫻ 10⫺10cm2sec⫺1, whereas that of hCGoccupied receptors was 4.5 ⫾ 1.5 ⫻ 10⫺10cm2sec⫺1. There was
no difference between the diffusion coefficients for hormoneoccupied receptors on LHR-K583R cells.
Fractions of receptors mobile at 37 C were significantly
larger for LH-occupied receptors than for hCG-occupied receptors on LHR-wt and LHR-K583R cells. After binding of
LH to receptors on LHR-wt cells, %M was 69 ⫾ 13%. This
value decreased to 43 ⫾ 3% for wild-type receptors binding
hCG. On LHR-K583R cells, %M was 94 ⫾ 11% and 71 ⫾ 10%
for receptors binding LH and hCG, respectively. After binding either LH or hCG, there were fewer receptors mobile on
LHR-wt cells than on LHR-K583R cells.
Differences in the diffusion characteristics for LH receptors on LHR-wt and LHR-K583R cells were not the result of
differences in receptor number. Before initiating each fluorescence photobleaching recovery experiment on an individual cell, we measured fluorescence counts per second
(cps) from fluorophores bound to LH receptors in the area
FUNCTIONAL LH RECEPTORS ARE SELF-ASSOCIATED
TABLE 1. Lateral diffusion coefficients (D) and mobile fraction
(% M) for LH receptors on LHR-wt and LHR-K583R cells at 37 C
Cells
LHR-wt
LHR-K583R
Ligand
oLH
oLH
hCG
hCG
oLH
oLH
hCG
hCG
Treatment
D
(10⫺10cm2sec⫺1)
%M
None
CD
None
CD
None
CD
None
CD
3.2 ⫾ 1.0
4.6 ⫾ 0.5b
4.5 ⫾ 1.5b
1.8 ⫾ 0.9c
2.1 ⫾ 0.3c
2.4 ⫾ 1.2c
2.8 ⫾ 1.0c,a
2.1 ⫾ 0.4c
69 ⫾ 13a
50 ⫾ 5b
43 ⫾ 3b
66 ⫾ 13a
94 ⫾ 11c
95 ⫾ 2c
71 ⫾ 10a
73 ⫾ 1a
a
Cells were labeled with TrITC-oLH or TrITC-hCG and subjected to
FPR measurements, as described in Materials and Methods. In addition, some HEK293 cells were pretreated with 20 ␮g/ml cytochalasin D (CD) for 1 h in some experiments before labeling with fluorescent hormones. Diffusion coefficients and mobile fractions and the
SD associated with these values were calculated from 16 – 66 measurements on individual cells. Fluorescence photobleaching recovery
data were analyzed by one-way ANOVA, and means were separated
using least-significant-differences criteria. Values for D or for % M
with different superscripts were different (P ⬍ 0.01).
illuminated by the attenuated argon ion laser. The means and
sd for counts from 20 – 40 experiments on 10 separate days
were 2934 ⫾ 1650 cps/cell on LHR-wt cells, compared with
3055 ⫾ 1726 cps/cell on LHR-K583R cells. These values do
not differ significantly.
Disruption of microfilaments increased the fraction of
mobile receptors for hCG but not LH-occupied receptors on
LHR-wt cells at 37 C
Cytoskeletal components can affect the motions of LH
receptors in some cell systems. The most pronounced effects
on protein motions have been observed with cytochalasin
d-treated ovine luteal cells, where disruption of microfilaments increased the rate of LH receptor lateral diffusion (19)
and the fraction of mobile receptors (19). On MA-10 cells,
cytochalasin D treatment resulted in faster rotational diffusion of the receptor (20). To determine whether lower values
for the mobile fraction on LHR-wt cells were caused by
restriction of receptor lateral diffusion by microfilaments,
cells were treated with cytochalasin D for 1 h before labeling
of cells for fluorescence photobleaching recovery measurements. Cytochalasin D treatment significantly affected the
measured rate of receptor lateral diffusion and the fraction
of mobile receptors on LHR-wt but not LHR-K583R cells
(Table 1). After treatment with cytochalasin D, the fractions
of mobile hCG-occupied receptors on LHR-wt cells increased
from 43 ⫾ 3% to 66 ⫾ 13%, whereas the fraction of mobile
LH-occupied receptors decreased.
Energy transfer occurs between receptors on LHR-wt cells
We then used single-cell fluorescence energy transfer
methods to evaluate whether different extents of receptor
self-association accompanied different fractions of mobile
receptors. Representative data traces showing fluorescence
energy transfer between LH- and hCG-occupied receptors
are presented in Fig. 2. As summarized in Table 2, energy
transfer efficiency was significantly higher for hCG-occupied
receptors, compared with LH occupied receptors (17% and
13%, respectively) on LHR-wt cells. On LHR-K583R cells, en-
4521
ergy transfer efficiencies between LH- and hCG-occupied receptors were 1% and 5%, respectively. These values are thus
considerably lower than those cells expressing wild-type receptor and, in fact, do not differ significantly from zero.
Disruption of microfilaments affected the ability of the
functional receptor on LHR-wt cells to form aggregated
structures. When the influence of microfilaments on receptor
organization was removed, a decrease in the fraction of mobile LH-occupied receptors on LHR-wt cells was accompanied by an increase in the extent of energy transfer between
receptors (Table 2). Significant effects of cytochalasin D treatment on fluorescence energy transfer between LH receptors
were observed only for LH-occupied receptors on LHR-wt
cells, where energy transfer efficiency between LH-occupied
receptors increased from 13 ⫾ 2% to 19 ⫾ 2%. Thus, diffusion
characteristics and energy transfer between LH-occupied receptors on LHR-wt cells, after cytochalasin D treatment, did
not differ significantly from those of untreated, hCG-occupied receptors. This suggests that there may be limits to both
the extent of energy transfer efficiency and the relative number of mobile receptors on LHR-wt cells and that microfilaments interact in a differential fashion with the complexes
formed after binding of LH and hCG. In the case of LHoccupied receptors, the cytoskeleton restricts receptor-receptor interactions. For hCG-occupied receptors, the cytoskeleton restricts the lateral diffusion of a significant fraction of the
hCG-occupied receptor population.
Discussion
We have measured comparatively large values for fluorescence energy transfer between LH or hCG bound to LH
receptors on LHR-wt cells, indicating that functional LH
receptors are self-aggregated into dimers or oligomers, and
resolving uncertainty as to whether grouping of receptors in
electron microscopy studies (1) and fluorescence microscopy
studies (2, 6) arises simply from restriction of receptors to
specific small domains or, as we have demonstrated here,
from actual receptor self-association. Values for energy transfer efficiency of 12–19% are reasonable for LH receptor
dimers or oligomers within protein clusters on the plasma
membrane (13). These results are also consistent with the
appearance of large, fluorescent clusters of wild-type rat LH
receptors on Chinese hamster ovary cells, where the receptor
is expressed as a green fluorescent protein construct (21), and
demonstrate that there are reproducible differences between
functional and nonfunctional receptors.
The lateral dynamics of wild-type LH receptors on
LHR-wt cells were typical of those measured for other membrane proteins, including, for example, the major histocompatibility complex class I antigens (22) and lymphocyte membrane glycoproteins (23). The diffusion coefficient measured
for both LH- and hCG-occupied receptors was also similar to
that measured for LH-occupied native LH receptors on ovine
luteal cells (15) and rat luteal cells (24). However, in contrast
to these other cell types, where hCG-occupied receptors
seemed laterally immobile, hCG-occupied receptors on
LHR-wt cells exhibited measurable rates for lateral diffusion.
These results for hCG-occupied receptors differed from those
previously reported in which the hCG-occupied LH receptor
4522
FUNCTIONAL LH RECEPTORS ARE SELF-ASSOCIATED
Endo • 2000
Vol. 141 • No. 12
FIG. 2. Representative data traces
from measurements of fluorescence energy transfer, at room temperature, between LH receptors on LHR-wt or LHRK583R cells. The upper trace in each
panel is the fluorescence decay from
cells labeled with fluorescent donor and
fluorescence acceptor (m). The lower
trace in each panel is from cells labeled
with fluorescence donor only (F). Calculated values for energy transfer efficiency are shown in the upper right
hand corner of each panel. On LHRK583R cells, the rates of fluorescence
decay did not differ, indicating that
there was no measurable energy transfer between LH receptors.
TABLE 2. Fluorescence energy transfer efficiency, at room
temperature, between LH receptors expressed on HEK 293 cells
Cell line
LHR-wt
LHR K583R
Treatment
None
Cytochalasin D
None
Cytochalasin D
% Energy transfer
efficiency
LH
hCG
13 ⫾ 2a
19 ⫾ 2c
1 ⫾ 1d
3 ⫾ 4d
17 ⫾ 3b,c
17 ⫾ 2b
5 ⫾ 5d
0d
As described in Materials and Methods, fluorescence energy transfer between LH receptors on HEK 293 cells was assessed on individual
cells labeled with fluorescent LH or hCG at a molar ratio of acceptor-to
donor-labeled hormone of 3:1. In some experiments, cells were preincubated with 40 ␮g/ml cytochalasin D for 30 min, at 37 C, before
addition of fluorescent hormones. The mean and SD for energy transfer efficiency was calculated from at least 60 measurements on individual cells. These values were analyzed by one-way ANOVA, and
means were separated using least-significant-differences criteria.
Values with different superscripts were different (P ⬍ 0.05).
was laterally immobile on luteal cell membranes at 37 C (15,
24), with values for fluorescence recovery, after photobleaching, of less than 20%. Nonetheless, binding of either LH or
hCG to receptors on LHR-wt cells affected both the diffusion
parameters and fluorescence energy transfer efficiency between hormone-occupied receptors. LH-occupied wild-type
receptors had larger mobile fractions and less energy transfer
between receptors than did the hCG-occupied receptors. Together with studies showing slower rotational diffusion after
binding of hCG (and thus, presumably large complexes) than
on LHR-wt cells (4), these results suggest that the receptor-
containing structures formed after binding of LH or hCG
differ structurally. Because receptor number remained constant throughout these studies, differences in receptor number do not contribute to the differential effects of hormone
binding on receptor self-association or the magnitude of the
laterally immobile fraction. It is more likely that there are
additional interactions between hCG and nearby membrane
proteins, perhaps as a result of the additional glycosylation
of the hCG molecule (25). As observed previously, binding
of chemically deglycosylated hCG to LH receptors does not
slow receptor rotational diffusion (4) or produce a large
fraction of laterally immobile receptors (26).
The components of the slowly diffusing complexes are not
known, but it is likely that these structures contain other
nonreceptor proteins. LH receptors exhibit very slow rotational motion in time-resolved phosphorescence anisotropy
studies on ovine and bovine luteal cell membranes (3), and
these slower motions are observed only when the hormonereceptor pair is functional, i.e. capable of activating adenylate
cyclase (4). LH receptors on bovine luteal cell plasma membranes are located within perhaps 100A of various nonreceptor
proteins (27), and this may be true in other species as well.
The differences between the lateral diffusion of functional
hormone receptor complexes on LHR-wt cells and nonfunctional hormone-receptor complexes on LHR-K583R cells
raise questions as to whether receptor-receptor interactions
may be necessary for signal transduction. These measurements do not, however, resolve whether receptor self-association precedes signaling. The biophysical methods applied
FUNCTIONAL LH RECEPTORS ARE SELF-ASSOCIATED
here required a finite time to label the receptor with fluorescent probes and to initiate measurements. Within that
time, receptor self-association has occurred, and the receptor
exhibits slow lateral diffusion. Nonetheless, nonfunctional
LH receptors on LHR-K583R cells were highly mobile, exhibited little or no interreceptor energy transfer, and, in rotational diffusion studies (4), had fast rotational correlation
times that were consistent with small complex sizes. Thus, in
the absence of receptor function, either as a result of receptor
mutation or binding of hormone antagonist, there is no interaction between receptors and no signaling. Functional receptors on LHR-wt cells have slow rotational correlation times
(4) and exhibit substantial energy transfer between receptors.
This suggests that microaggregation of LH receptors on
LHR-wt cells may accompany, or be required for, productive
signal transduction. In addition, receptor self-association may
persist into the times when LH receptors are nonresponsive to
hormone challenge and thus desensitized (28).
Receptor self-association has been proposed as a early
event in the function of another structurally-related hormone
receptor involved in reproductive function. Janovick and
Conn (29), using lactoperoxidase-conjugated hormones to
iodinate proximal GnRH receptors , have demonstrated that
agonist, but not antagonist, binding to the GnRH receptor
results in formation of GnRH microaggregates and have
identified nonreceptor proteins in the vicinity of the receptor.
We have directly demonstrated such microaggregates in
studies of fluorescence energy transfer between GnRH receptors (30). GnRH agonists increase energy transfer efficiency between receptors in a dose-dependent fashion, but
binding of a GnRH antagonist results in no energy transfer
between receptors. Lateral diffusion of the GnRH receptor also
depends on whether the receptor has bound GnRH agonist or
antagonist. There is a significant decrease in the fraction of
mobile receptors in response to agonist (but not antagonist)
binding. The picture that emerges from Janovick and Conn’s
work (29), as well as from our biophysical studies of GnRH
receptors (30) and LH receptors, is that hormone-responsive
receptors seem to cluster in structures that contain both selfassociated receptors and nonreceptor proteins and that are sufficiently large to exhibit reduced lateral mobility.
Acknowledgments
The authors wish to thank the National Hormone and Pituitary Program of the NIDDK for providing the ovine LH and human CG used in
these studies.
References
1. Luborsky JL, Slater WT, Behrman HR 1984 Luteinizing hormone (LH) receptor aggregation: modification of ferritin-LH binding and aggregation by
prostaglandin F2␣ and ferritin-LH. Endocrinology 115:2217–2225
2. Amsterdam A, Berkowitz A, Nimrod A, Kohen F 1980 Aggregation of luteinizing hormone receptors in granulosa cells: a possible mechanism for
desensitization to the hormone. Proc Natl Acad Sci USA 77:3440 –3445
3. Philpott CJ, Rahman NA, Kenny N, Barisas BG, Roess DA 1995 Rotational
dynamics of luteinizing hormone receptors on bovine and ovine luteal cell
plasma membranes. Biol Reprod 53:645– 650
4. Roess DA, Brady CJ, Barisas BG 2000 Biological function of the LH receptor
is associated with slow receptor rotational diffusion. Biochim Biophys Acta
1464:242–250
4523
5. Anderson W, Kang Y-H, Perotti M, Bramley T, Ryan R 1979 Interactions of
gonadotropins with corpus luteum membranes. III. Electron microscopic localization of [125I]-hCG binding to sensitive and desensitized ovaries seven
days after PMSG-hCG. Biol Reprod 20:362–376
6. Podestá EJ, Solano AR, Attar R, Sánchez ML, Molina y Vedia L 1983 Receptor
aggregation induced by antilutropin receptor antibody and biological response
in rat testis Leydig cells. Proc Natl Acad Sci USA 80:3986 –3990
7. Cvejic S, Devi LA 1997 Dimerization of the ␦ opiod receptor: implication for
a role in receptor internalization. J Biol Chem 272:26959 –26964
8. Hebert TE, Moffett S, Morello J, Loisel TP, Bichet DG, Barret C, Bouvier M
1996 A peptide derived from a ␤2-adrenergic receptor transmembrane domain
inhibits both receptor dimerization and activation. J Biol Chem
271:16384 –16392
9. Ryu K-S, Gilchrist RL, Ji I, Kim S-J, Ji TH 1996 Exoloop 3 of the luteinizing
hormone/chorionic gonadotropin receptor: Lys583 is essential and irreplaceable for human chorionic gonadotropin (hCG)-dependent receptor activation
but not high affinity hCG binding. J Biol Chem 271:7301–7304
10. Fernandez LM, Puett D 1996 Lys583 in the third extracellular loop of the
lutropin/choriogonadotropin receptor is critical for signaling. J Biol Chem
271:925–930
11. Young RM, Arnette JK, Roess DA, Barisas BG 1994 Quantitation of fluorescence energy transfer between cell surface proteins via fluorescence donor
photobleaching kinetics. Biophys J 67:881– 888
12. Trón L, Szölösi J, Damjanovich S, Helliwell SH, Ardnt-Jovin DJ, Jovin TM
1984 Flow cytometric measurement of fluorescence resonance energy transfer
on cell surfaces. Quantitative evaluation of the transfer efficiency on a cellby-cell basis. Biophys J 45:939 –946
13. Chan SS, Arndt-Jovin DJ, Jovin TM 1979 Proximity of lectin receptors on the
cell surface measured by fluorescence energy transfer in a flow system. J Histochem Cytochem 27:56 – 64
14. Brinkley MA 1992 A brief survey of methods for preparing protein conjugates
with dyes, haptens, and cross-linking reagents. Bioconjug Chem 3:2–13
15. Niswender G, Roess D, Sawyer H, Silvia W, Barisas B 1985 Differences in the
lateral mobility of receptors for luteinizing hormone (LH) in the luteal cell
plasma membrane when occupied by ovine LH versus human chorionic gonadotropin. Endocrinology 116:164 –169
16. Munnelly HM, Roess DA, Wade WF, Barisas BG 1998 Interferometric fringe
fluorescence photobleaching recovery interrogates entire cell surfaces. Biophys
J 75:1131–1138
17. Marquardt DW 1963 An algorithm for least-squares estimation of nonlinear
parameters. J Soc Indust Appl Math 11:431– 441
18. Gosset WS 1907 On the error of counting with a haemocytometer. Biometrika
5:351–360
19. Roess DA, Niswender GD, Barisas BG 1988 Cytochalasins and colchicine
increase the lateral mobility of human chorionic gonadotropin-occupied luteinizing hormone receptors on ovine luteal cells. Endocrinology 122:261–269
20. Roess DA, Jewell MA, Philpott CJ, Barisas BG 1997 The rotational diffusion
of LH receptors differs when receptors are occupied by hCG versus LH and
is increased by cytochalasin D. Biochim Biophys Acta 1357:98 –106
21. Horvat RD, Nelson S, Clay CM, Barisas BG, Roess DA 1999 Intrinsically
fluorescent luteinizing hormone receptor demonstrates hormone-driven aggregation. Biochem Biophys Res Commun 256:382–385
22. Edidin M, Zuniga M 1984 Lateral diffusion of wild-type and mutant Ld
antigens in L cells. J Cell Biol 99:2333–2335
23. Leuther M, Peacock J, Krakauer H, Barisas B 1981 Changes in lectin receptor
lateral mobilities accompany lymphocyte stimulation. J Immunol 127:893– 899
24. Roess DA, Rahman NA, Kenny N, Barisas BG 1992 Lateral and rotational
dynamics of LH receptors on rat luteal cells. Biochim Biophys Acta
1137:309 –316
25. Keutmann H, McIlroy P, Bergert E, Ryan R 1983 Chemically deglycosylated
human chorionic gonadotropin subunits: characterization and biological properties. Biochemistry 22:3067–3072
26. Roess DA, Mock E, Niswender GD, Barisas BG 1988 Cytoskeletal modulation
of hCG-LH receptor mobility in ovine luteal cells. Endocrinology 122:261–269
27. Meiklejohn BI, Rahman NA, Roess DA, Barisas BG 1997 5-Iodonapthyl-1azide labeling of specific plasma membrane proteins via energy transfer activation. Biochim Biophys Acta 1324:320 –332
28. Horvat RD, Roess DA 2000 Desensitized LH receptors are self-associated in
large, slowly-diffusing complexes. Biol Reprod [Suppl 1] 62:106
29. Janovick J, Conn P 1996 Gonadotropin-releasing hormone agonist provokes
homologous receptor microaggregation: an early event in seven-transmembrane receptor mediated signaling. Endocrinology 137:3602–3605
30. Horvat RD, Clay CM, Barisas BG, Roess DA 2000 Hormone binding causes
GnRH receptor self-association and slows receptor lateral diffusion. Biophys
J 78:391A