(Received for publication, November 12, 1996, and in revised form, January 16, 1997)
From the Division of Molecular Medicine, Department
of Medicine, Cornell University Medical College and The New York
Hospital, New York, New York 10021 and ¶ Department of Physiology
and Biophysics, Mount Sinai School of Medicine of the City University
of New York, New York, New York 10029
The roles of conserved residues in transmembrane
helices (TMs) of G protein-coupled receptors have not been well
established. A computer-generated model of the thyrotropin-releasing
hormone receptor (TRH-R) indicated that conserved Asp-71 (TM-2) could interact with conserved asparagines 316 (TM-7) and 43 (TM-1). To test
this model, we constructed mutant TRH-Rs containing polar or alanine
substitutions of these residues. The maximal activities of N43A and
N316A TRH-Rs were diminished, whereas D71A (Perlman, J. H.,
Nussenzveig, D. R., Osman, R., and Gershengorn, M. C. (1992) J. Biol. Chem. 267, 24413-24417) and N43A/N316A
TRH-Rs were inactive. Computer models of D71A and N43A/N316A TRH-Rs
show similar changes from native TRH-R in their TM bundle
conformations. The inactivity and the similarity of the computer models
of D71A and N43A/N316A TRH-Rs are consistent with the idea that Asp-71
bridges Asn-43 and Asn-316 and suggest that activity is critically
dependent on these interactions. The conservation of these residues
suggests these specific interactions involving TMs 1, 2, and 7 may be
structurally important for all members of the rhodopsin/-adrenergic
receptor subfamily of G protein-coupled receptors.
The thyrotropin-releasing hormone receptor
(TRH-R)1 (1) is a member of the
rhodopsin/-adrenergic receptor subfamily of GPCRs (2). It is thought
that all GPCRs have in common a transmembrane bundle that is composed
of seven helices. It has been proposed that residues that are highly
conserved in these helices are important in maintaining the structure
of the bundle and thereby producing a receptor that exhibits high
affinity and activity (3, 4).
Bacteriorhodopsin is the only protein that contains seven
transmembrane-spanning helices (TMs) for which a crystallographic analysis at an atomic level of resolution is available (5). However,
bacteriorhodopsin is not a GPCR and does not share sequence homology
with GPCRs. Baldwin (3) predicted a structure for GPCRs of the
-adrenergic/rhodopsin family in which the TMs are arranged in a
counterclockwise bundle (as viewed from the extracellular surface)
based on sequence homology, conservation, and polarity of residues. A
recently reported two-dimensional projection map of rhodopsin is
consistent with this structure (6). The resolution of the projection
map was not sufficiently high, however, to allow assignment of specific
helical positions and could not identify residues that form
interhelical interactions. Indirect evidence, therefore, has been used
to suggest proximity of TMs 2, 3, and 7 in rhodopsin (7), of
TMs 1, 2, 3, and 6 to TM-7 in adrenergic receptors (8, 9), of TMs
1 and 7 (10) in muscarinic receptors, and of TMs 5 and 6 of the
neurokinin NK-1 receptor (11).
Proximity of two of the three highly conserved residues, Asn in TM-1,
Asp in TM-2, and Asn in TM-7, has been suggested for several GPCRs by
demonstrations that single substitutions of Asp by Asn or of Asn by Asp
lead to diminished receptor function and double substitutions restore
function. For example, an interaction has been suggested between Asp of
TM-2 and Asn of TM-7 in the serotonin 5HT2A receptor (12).
Interestingly, the positions of Asp and Asn residues are switched in
the GnRH receptor, which contains an Asn in TM-2 and an Asp in TM-7,
and these residues were suggested to interact (13). Recently, it was
suggested that Asn in TM-1, Asp in TM-2, and Asn (and a conserved Tyr)
in TM-7 are near each other and form a polar pocket within the
1B-adrenergic receptor (14).
We have constructed a model of the complex of TRH and TRH-R (15, 16) based on a GPCR template that is different from that of bacteriorhodopsin. The model of the TRH·TRH-R complex predicts proximity of Asn 43 in TM-1, Asp 71 in TM-2, and Asn 316 in TM-7 of TRH-R. The putative binding pocket for TRH is between TMs 3, 6, and 7 and is not close to the region occupied by Asn-43, Asp-71, and Asn-316. In this study, we have examined the roles of Asn-43, Asp-71, and Asn-316 of TRH-R. We present experimental and computational evidence that Asp-71 interacts with Asn-43 and Asn-316. Our data indicate that these interactions are critical for activation of TRH-R and provide support for the proposed TM bundle topology of GPCRs.
[3H]MeTRH was obtained from DuPont. myo-[3H]Inositol was obtained from Amersham Corp. TRH was from Calbiochem and MeTRH from Sigma. Restriction endonucleases were from New England Biolabs. The cloning vector pBluescript was from Stratagene and the expression vector pCDM8 from Invitrogen. Dulbecco's modified Eagle's medium and fetal calf serum were from Collaborative Research.
MutagenesisThe full-length, mouse TRH-R cDNA in pBluescript (pBSmTRHR) (1) or pCDM8 (pCDM8mTRHR) (17) was used for mutation. Mutants were prepared by the polymerase chain reaction and were subcloned directly into pCDM8mTRHR (N43A, N43D, and D71N) or were subcloned into pBSmTRHR (N316A and N316D) and then subcloned into pCDM8mTRHR after digesting with XhoI and NotI. Double mutants were constructed in pCDM8. Construction of D71A TRH-R was described previously (18). Mutant TRH-R sequences were confirmed by the dideoxy chain termination method.
Cell Culture and TransfectionCOS-1 cells were maintained and transfected as described previously (1). In brief, cells were seeded 1 or 2 days prior to transfection at 0.7 to 1.5 × 106 cells/100-mm dish. Cells were transfected using the DEAE-dextran method as described and maintained in Dulbecco's modified Eagle's medium with 10% fetal calf serum for 1 day at which time cells were harvested and seeded into 12-well plates at 100,000 cells/well in Dulbecco's modified Eagle's medium with 5% fetal calf serum.
Receptor Binding StudiesOne day after reseeding into 12-well plates, binding experiments were carried out in buffer with cells in monolayer for 1 h at 37 °C as described elsewhere (18). Inhibitory constants (Ki) were derived from competition binding experiments for which curves were fitted by nonlinear regression analysis and drawn with the PRISM program (GraphPad Inc.).
Inositol Phosphate FormationOne day after transfection, cells in monolayer in 12-well plates were labeled with 1 µCi of myo-[3H]inositol/ml. Stimulation of IP formation was measured 1 day later for 1 h at 37 °C by methods previously described (19). None of the TRH-Rs studied stimulated IP formation in the absence of TRH. Maximal activity of a mutant TRH-R was defined as the ratio of IPs formed in response to maximally effective concentrations of TRH in cells expressing a mutant TRH-R compared with WT. The maximal activity of WT TRH-R was set at 100%. Maximal activity was assessed only for those mutant receptors which were shown to be expressed at levels that for WT TRH-Rs were sufficient for maximal activity.
Computer ModelingWe have previously constructed models of the TRH·TRH-R complex using energy minimization techniques (15) and novel mixed mode Monte Carlo/Stochastic Dynamics simulations (16). The first approach produced an initial model of the binding pocket that was later refined through the Monte Carlo/Stochastic Dynamics simulations. These models agree with experimental findings. We have now constructed a model of TRH-R that includes the ECLs. The ECLs were added to a previously constructed model of the TM domain of TRH-R (16). The disulfide bridge between Cys residues in ECL-1 and ECL-2, which has been shown to be necessary to constrain TRH-R in a high affinity conformation (20), was maintained throughout all simulations. The model with frozen helices was energy-minimized with the program CHARMM (21) and used to generate 14 energy-minimized average structures through a simulated annealing protocol (22). The details of these simulations will be presented in a separate report.2 In brief, the minimized structure was heated to 1500 K, and 14 initial structures were extracted from a trajectory at this temperature. Each of these structures was annealed to 300 K over 60 ps followed by 100 ps constant temperature simulation at 300 K. The resulting minimized average structures clustered according to their pairwise root-mean square deviation (rmsd), producing a major family of seven members at rmsd of 1.9 Å. One of these structures was retained and used to conduct a simulation of the receptor-TRH complex. TRH was manually placed in the binding pocket as described previously (16). All helical constraints were removed, and the complex was energy-minimized, heated to 300 K in 23 ps, and subjected to 1 ns of molecular dynamics simulation at 300 K. An average structure calculated from the last 600 ps of the trajectory was energy-minimized in preparation for the construction of the structures described below. To test the relationship among Asn-43, Asp-71, and Asn 316, NMR nuclear Overhauser effect distance constraints as encoded in CHARMM (Kmax = 20 kJ/mol/Å2, Kmin = 40 kJ/mol/Å2, rmin = 1.5 Å, rmax = 2.0 Å) were applied during a 2000-step minimization to induce the formation of hydrogen bonds between the three residues. Subsequently, the constraints were removed, and the structure was further minimized for 3000 steps. Only the Asn-43-Asp-71 and Asp-71-Asn-316 hydrogen bonds remained in the unconstrained structure. Several minimizations were attempted with other constraints which were later removed, to explore alternative hydrogen bonding among the three residues. The only stable hydrogen bonding pattern was Asn-43-Asp-71 and Asp-71-Asn-316. This structure was subsequently heated to 300 K and subjected to 200 ps of molecular dynamics simulation at 300 K. It was also used in preparing the four mutant receptors, i.e. N43A, D71A, N316A, and N43A/N316A TRH-Rs. Each mutant was constructed by substituting Ala for Asn or Asp and energy-minimized for 2000 steps. Subsequently, each structure was heated to 300 K and subjected to 200 ps of molecular dynamics simulation at 300 K. The energy minimized structures averaged over 100-140 ps of the corresponding simulations are presented here. In all calculations, a distant dependent dielectric function was used to approximate the effect of the environment.
Our revised model of the TRH·WT TRH-R complex indicates
proximity of conserved residues Asn-43 in TM-1, Asp-71 in TM-2, and Asn-316 in TM-7. This model does not differ significantly in the TM
bundle binding pocket from that described in our previous publication (16) but includes a newly constructed domain consisting of the three
ECLs.3 A closeup of domains including TM-1,
TM-2, and TM-7 (Fig. 1) indicates that Asp-71 forms
hydrogen bonds to both Asn-43 and Asn-316. On the other hand, no
interaction between Asn-43 and Asn-316 could be maintained in the
simulations, suggesting that these residues are involved in maintaining
receptor structure through their common bridge to Asp-71. To begin to
assess the roles of Asn-43, Asp-71, and Asn-316, substitutions were
made with Asp, Asn, and Asp, respectively. The results of binding and activation studies of WT and mutant TRH-Rs transiently expressed in
COS-1 cells are shown in Table I. The affinities of N43D
and D71N were 8- and 57-fold lower than WT TRH-R, respectively, and the
affinity of N316D TRH-R was 6-fold higher than WT TRH-R. These changes
in affinities are small compared with substitution of residues that we
have shown directly contact TRH. These data are consistent with the
model which predicts that these residues do not directly contact TRH
and that the changes in affinities are secondary to conformational
changes in the TM bundle (see below). N43D, D71N, and N316D TRH-Rs
exhibited 57, 52, and 98% of WT TRH-R maximal activity, respectively.
Thus, N43D and D71N TRH-Rs exhibited lower maximal activities than WT
TRH-R, whereas N316D was as active as WT TRH-R.
|
To further analyze the roles of these conserved residues, mutant receptors containing two substitutions of the three pairs of residues were constructed (Table I). The affinities of N43D/D71N and D71N/N316D TRH-Rs were 110- and 120-fold lower, respectively, than WT TRH-R, and no binding was detectable with N43D/N316D TRH-R. The maximal activities of N43D/D71N and D71N/N316D TRH-Rs were 7.6 and 115% that of WT TRH-R, respectively. The maximal activity of N43D/N316D could not be determined because the Bmax could not be measured. Thus, the mutant receptor containing Asn at position 71 was made to revert to WT TRH-R levels of activity by substituting Asp at position 316. This suggests that positions 71 and 316 are proximate and that an interaction between Asp-71 and Asn-316 occurs in WT TRH-R that is important for activation. We cannot determine whether the loss in activity of the double mutant N43D/D71N TRH-R was additive, which would suggest that these two residues were functionally independent, or more than additive, which would be consistent with these residues being functionally interdependent, compared with the two receptors with single mutations. Therefore, it is not clear from these data whether Asn-43 and Asp-71 are proximate (see below). As we could not determine the activity of N43D/N316D TRH-R, we could not assess whether Asn-43 and Asn-316 are functionally interdependent from these data also (see below).
To analyze our model further, alanine residues, which would minimize the possibility of introducing non-native interactions, were substituted at Asn-43, Asp-71, and Asn-316 (Table II). The affinity of N43A TRH-R was similar to WT TRH-R, and the affinities of D71A and N316A TRH-Rs were 3- and 11-fold lower, respectively, than WT TRH-R. The maximal activities of N43A and N316A TRH-Rs were 37 and 47% of WT TRH-R, respectively; D71A TRH-R was inactive. These data indicate that Asn-43, Asp-71, and Asn-316 are important for activation of TRH-R. The larger effect noted with D71A TRH-R compared with either N43A or N316A TRH-Rs is consistent with the idea that Asp-71 interacts with both Asn-43 and Asn-316. To further test this prediction, the double mutant N43A/N316A TRH-R was constructed. The affinity of N43A/N316A TRH-R was 22-fold lower than WT. More importantly, N43A/N316A TRH-R was inactive. Thus, both D71A and N43A/N316A TRH-Rs are inactive receptors. These data are consistent with our model in which Asp-71 interacts with both Asn-43 and Asn-316.
|
In our model of WT TRH-R, Asn-316 plays an important role in addition
to hydrogen bonding to Asp-71. An averaged minimized structure from the
simulation shows that the hydrogen bond to Asp-71 orients the amide
group of Asn-316 to also form a hydrogen bond with its own backbone
N-H group throughout 90% of the simulation. This introduces a major
perturbation in the helicity of TM-7 as shown schematically in Fig.
2. The helicity in this area is weakened by the presence
of Pro-317, which cannot form a hydrogen bond to Ser-313. The
interaction of Asn-316 with its own backbone prevents the formation of
a hydrogen bond to Asn-312, which is positioned above it. This local
disruption of helical hydrogen bonds unwinds the helix between Asn-312
and Asn-316. Consequently, Ser-313 forms a bifurcated hydrogen bond to
Ile-309 and Tyr-310, Ala-314 hydrogen bonds to Leu-311, leaving the
carbonyls of Asn-312 and Ala-314 without hydrogen bonding partners.
Hydrogen bonding patterns are also disrupted below Asn-316 because the
helix is kinked and unwound. Consequently, Asn-321 hydrogen bonds to
Asn-316 and Ile-319 to Ile-315 leaving Val-318 and Tyr-320 without
hydrogen bonds. A similar disruption of helicity in TM-7 was proposed
on the basis of the observation that an Asn-Pro motif (as in
Asn-316-Pro-317) is a helix breaker (23).
The effects of substituting Ala for Asp-71 or for Asn-43 and Asn-316
are illustrated in Fig. 3. Introducing an Ala
substitution in the position of Asn-316 no longer constrains TM-7 by
the hydrogen bond to Asp-71, and its helicity is restored. This occurs
both in the N316A (not shown) and the N43A/N316A mutants. Similarly, a
substitution of Asp-71 to Ala results in the loss of a hydrogen bond to
Asn-316. Unconstrained, Asn-316 turns to form a hydrogen bond with the
backbone carbonyl of Ser-313, which is left available due to the imide
nitrogen of Pro-317. Finally, the constraining effect of the hydrogen
bonding network can be noticed in the N43A/N316A double mutant receptor
in which Asp-71, no longer constrained by these residues, turns to form
a hydrogen bond with the backbone N-H of Met-51 in TM-1.
Disruption of hydrogen bonds present in WT TRH-R has major effects on
the structure of the helical bundle. As shown in Fig. 3, A
and B, mutating Asp-71 to Ala has a major effect on the
intracellular portion of the helices. A comparison to WT receptor shows
major disturbances in TM-1 (Fig. 3A) and in TM-5 and TM-6
(Fig. 3B). Despite major changes in the helicity of TM-7
(see above), its position in the helical bundle does not change much
compared with WT. The extent of the difference between the helical
bundles can be expressed by the rmsd between C atoms of WT and D71A
TRH-Rs of 1.78 Å. In contrast, the conformations of the backbones of the mutant D71A TRH-R and the double mutant N43A/N316A TRH-R are similar. As can be seen in Fig. 3, C and D, there
is excellent overlap of D71A and N43A/N316A TRH-Rs and the rmsd between
the C
atoms of the helices is only 1.01 Å. Hence, the simulations establish a proposed link between the disturbance of the helical bundle
and the similar nonresponding phenotype of the D71A and the
N43A/N316A TRH-Rs.
The major experimental finding reported herein is that the effect of substituting Asn-43 in TM-1 and Asn-316 in TM-7 with Ala in a doubly mutated TRH-R leads to the same complete loss of activity as the previously reported effect of substituting Asp-71 in TM-2 with Ala (18). These data are consistent with our model of the TRH·TRH-R complex which predicts that Asp-71 interacts with both Asn-43 and Asn-316 forming a bridge between them. Moreover, the modeling shows that the D71A and N43A/N316A TRH-Rs exhibit disruption of hydrogen bonds usually present in the native receptor that results in similar changes in the intracellular aspects of the helices of these mutant receptors. Thus, we conclude that a structural feature of TRH-R that is important for activation involves the bridging of these specific residues in TMs 1, 2, and 7.
It has been proposed that four conserved residues in TMs 1, 2, and 7 of
the 1B-adrenergic receptor, which includes the residues homologous to Asn-43, Asp-71, and Asn-316 of TRH-R, form a polar pocket
that interacts with an Arg residue of the conserved Asp-Arg-Tyr (DRY)
motif at the junction between TM-3 and intracellular loop 2 to
constrain the receptor in an inactive conformation (14). This does not
appear to be the function of these residues in TRH-R. Several
differences between TRH-R and
1B-adrenergic receptor regarding activity have been found when similar experiments were performed with homologous mutants in the two systems. 1) We have not
been able to demonstrate basal activity, that is, IP formation in the
absence of agonist, for N43A TRH-R,4 as was
shown for the corresponding N63A
1B-adrenergic receptor. It may be that a more sensitive system is needed to demonstrate constitutive activity of TRH-Rs. Alternatively, it is possible that
similar structural changes induced by the mutation of the Asn in TM-1
may result in different phenotypes of the mutant receptors. 2) N43A
TRH-R is only approximately half as active as WT TRH-R in the presence
of maximally effective concentrations of TRH, whereas N63A
1B-adrenergic receptor was reported to be as active as
WT
1B-adrenergic receptor. Thus, the role(s) of Asn in
TM-1 in basal and agonist-induced activity is different in TRH-R and
1B-adrenergic receptor. 3) R123A TRH-R retains
substantial activity in response to TRH,4 whereas the
mutant receptor with the corresponding substitution of Arg within the
Asp-Arg-Tyr motif with Ala in
1B-adrenergic receptor is
inactive. Thus, the role of this Arg is at least quantitatively different in agonist activation of TRH-R and of the
1B-adrenergic receptor. It appears, therefore, that
there are differences in the roles of highly conserved residues in
different members of the rhodopsin/
-adrenergic receptor subfamily of
GPCRs (see below).
Substitutions by alanines in mutated proteins may be the most reliable indicators of the roles of substituted native residues. Although polar substitutions for polar residues have the potential for restoring native interactions, they also may introduce strong non-native hydrogen bonding or ionic interactions. Nevertheless, the effects on activation observed with mutant TRH-Rs in which polar residues were substituted for the conserved residues support the findings observed with Ala substitutions. For example, D71N TRH-R exhibited activity that was decreased to 52% of WT TRH-R. However, when Asn-316 was mutated to Asp, the effect of mutating Asp-71 to Asn in the doubly mutated receptor, D71N/N316D TRH-R, was reversed. That is, the maximal activity of D71N/N316D TRH-R was similar to that of WT TRH-R. This restorative effect is consistent with proximity or interaction of the substituted residues (24-26). By switching residues in this way, a similar conclusion was reached with regard to these residues in the 5HT2A receptor (12). The native GnRH receptor contains Asn at the position homologous to Asp-71 of TRH-R and Asp at the position homologous to Asn-316 of TRH-R. A double mutant GnRH receptor in which these residues were switched resulted in restoration of binding though not of activation (13). Thus, these highly conserved Asn residues in TMs 1 and 7 and Asp in TM-2 are important for agonist-stimulated signaling. These data also highlight the fact that similar substitutions in different receptors may give rise to different phenotypes. Such a behavior suggests that another factor, e.g. the conformation of the intracellular loops, may be important in the appearance of a certain phenotype.
The double mutant N43D/N316D TRH-R, at a maximally effective concentration of TRH, exhibited only 15% of the WT TRH-R maximal activity. However, specific binding by this double mutant could not be measured (Table I) and, therefore, we could not ascribe the lowered IP formation to an intrinsic property of the mutant receptor. In fact, as our inability to measure specific binding was likely due to markedly lowered cell surface expression, the apparent decrease in stimulation of IPs was likely due to low receptor number. This is so because binding was detectable in several mutant receptors with lower potencies than N43D/N316D TRH-R (Tables I and II), and there is usually a good correlation between potency and affinity in TRH-R mutants (27). Although there may be several reasons for decreased expression of N43D/N316D TRH-R, it is reasonable to speculate, in conjunction with other findings described above, that it may be due to an abnormal receptor structure caused by a repulsive effect between the introduced Asp residues which would be consistent with proximity of the native residues in TRH-R.
Aside from supporting the idea that Asn-43, Asp-71, and Asn-316 are proximate and interact, the data allow conclusions to be drawn regarding the importance of the side chains of these residues in activation. Analysis of the single Ala mutants indicate that Asn-43, Asp-71, and Asn-316 are all important for activation. Asn can partially substitute for the function of Asp-71 and allow for partial activation suggesting that both oxygens of Asp-71 are important. Asp can fully substitute for Asn-316 with regard to activation which may be interpreted in at least two ways. The carbonyl group of Asn-316 (retained by Asp) may be important for activation (implying protonation of Asp-71) or the Asp introduced at position 316 may be protonated and become a potential hydrogen bond donor (like Asn). The latter mechanism has also been suggested to explain the activity of a similar mutant in the 5HT2A receptor (12).
As noted above, we have proposed that the putative binding pocket for TRH is between TMs 3, 6, and 7 in TRH-R which is not close to the region occupied by Asn-43, Asp-71, and Asn-316. Therefore, effects on affinity resulting from substitution of these residues are due to indirect effects on the binding pocket. This is consistent with the idea that conserved residues would not be involved in forming a high affinity, highly specific binding pocket. The binding affinities of N43A, D71A, and N316A TRH-Rs for MeTRH (an analog of TRH with 5-10-fold higher affinity than TRH that is used as radioligand) were little changed from WT TRH-R (Table II), which is consistent with the idea that these residues do not directly contact the ligand. In fact, the larger effects on binding of substituting Asp for Asn-43 or Asn for Asp-71 appear to be deleterious effects due to the introduced substituents rather than loss of the native side chains. Thus, these data are consistent with the idea that these conserved residues do not constitute the TRH-R binding pocket.
The model appears to support several interesting features of the
experimental data. In WT TRH-R, the hydrogen bonding network obtained
after exploring various possible patterns through the use of distance
constraints, supports the idea that Asp-71 forms hydrogen bonds with
both Asn-316 and Asn-43, while Asn-316 and Asn-43 do not interact with
each other. In the D71A and N43A/N316A mutant receptors, however, the
hydrogen bonds that maintain this network are lost. Comparison of the
structures of WT and D71A TRH-Rs shows that the helical bundle
undergoes a significant change which is localized to the intracellular
portion of the helices. The similarity of these changes for the mutant
receptors and their difference from WT TRH-R suggests that a possible
mechanism for loss of activity may be related to the rearrangement of
helices. These rearrangements will induce conformational changes in the intracellular loops which enable the interaction of the activated receptor with the G-protein. However, our simulations only provide part
of the answer since the intracellular loops are not included. Further
work will be required in constructing a model for the intracellular
loops before this question can be addressed. Nevertheless, it appears
that the phenotypic differences between similar mutations in different
receptors (e.g. N43A in TRH-R versus N63A in
1B-adrenergic receptor (14)) may result from different
consequences in the conformations of the intracellular loops caused by
similar changes in the TM portions.
The highly conserved Asp in TM-2 has been implicated in a number of
functional roles in GPCRs including activation, binding and allosteric
regulation by Na+ (28). The conserved Asn in TM-7, which is
part of a highly conserved Asn-Pro-Xaa-Xaa-Tyr (NPXXY)
motif, has been shown to be important for activation in the
5HT2A (12), angiotensin AII (29), and
2-adrenergic receptors (30). Interestingly, an extensive
review of GPCR site-specific substitutions shows that residues in TM-1
have rarely been targeted (28) (see above).
In conclusion, our experimental observations are consistent with the
idea that Asn-43 of TM-1, Asp-71 of TM-2, and Asn-316 of TM-7 are
proximate to one another in the native TRH-R and with a bifunctional
role of Asp-71 in interacting with both Asn-43 and Asn-316. Our
computational data support this idea and predict that Asp-71 forms a
bridge between Asn-43 and Asn-316 that may be important in holding
TRH-R in a conformation in which TM 1, 2, and 7 are apposed. The
conservation of these residues suggests that these interactions may be
important for the structural integrity of all members of the
rhodopsin/-adrenergic receptor GPCR subfamily.