What makes the α_1A-adrenoceptor gene express the α_1L-adrenoceptor functional phenotype?

Abstract

The α_1A-adrenoceptor is therapeutically exploited because of its prevalence in the lower urinary tract. The pharmacology shown by this lower urinary tract α_1A-adrenoceptor is different from that shown by other α_1A-adrenoceptors, which has led to it being subclassified as an α_1L-adrenoceptor. Only in the last few years was it shown that this pharmacologically distinct α_1L-adrenoceptor is a product of the α_1A-adrenoceptor gene. In this issue of the BJP, Nishimune et al. review the literature on α_1L-adrenoceptor pharmacology and discuss the possible molecular mechanisms by which the α_1A-adrenoceptor gene is able to produce two pharmacologically distinct adrenoceptor subtypes. Based primarily from their own research using cell lines transfected with α_1A-adrenoceptors, they conclude that a protein that interacts with the receptor is the most plausible explanation. The challenge remains to identify any such interacting protein and show how it is able to change the pharmacology of the receptor for different ligands.

LINKED ARTICLE

This article is a commentary on Nishimune et al., pp. 1226–1234 of this issue. To view this paper visit http://dx.doi.org/10.1111/j.1476-5381.2011.01591.x

The most effective and rapidly acting pharmacological treatments for benign prostatic hyperplasia (BPH) are the α_1A-adrenoceptor antagonists, such as tamsulosin and alfuzosin (Miano et al., 2008; receptor nomenclature follows Alexander et al., 2011). This class of BPH therapeutic agents makes over US$ 3 billion in worldwide sales (Ventura et al., 2011). Previously, non-selective α₁-adrenoceptor antagonists such as prazosin, doxazosin and terazosin were widely used, but these have now been largely superseded by tamsulosin and alfuzosin because of their greater selectivity for the α_1A-adrenoceptor subtype over the α_1B and α_1D-adrenoceptor subtypes. The proportion of the α_1A-adrenoceptor subtype expressed in the smooth muscle stroma of the prostate gland is greater than the proportion expressed in vascular smooth muscle, leading to a lower incidence of troublesome vascular side effects such as weakness, fatigue, postural hypotension and dizziness, which were commonplace with the use of the non-selective α₁-adrenoceptor antagonists.

α_1A-Adrenoceptors are abundant in the male lower urinary tract, and α_1A-adrenoceptor antagonists are very effective in relieving lower urinary tract symptoms associated with urethral obstruction caused by prostate enlargement. Despite this, prostate and other lower urinary tract tissues, from all species, do not show typical α_1A-adrenoceptor pharmacology (Nishimune et al., 2012). When used in functional isolated tissue experiments, isolated tissues from prostate gland, urethra and bladder, all exhibit a low affinity for prazosin when compared with other α₁-adrenoceptor-expressing tissues. A corresponding change in affinity is not seen with tamsulosin. This pharmacological anomaly led to the postulate that a fourth α₁-adrenoceptor existed, which was termed the α_1L-adrenoceptor.

Only recently has it been demonstrated with the use of genetically modified adrenoceptor knockout mice that the prostatic α_1L-adrenoceptor phenotype requires the expression of the α_1A-adrenoceptor gene (Gray et al., 2008; Muramatsu et al., 2008). The term α_1L-adrenoceptor is not currently recognized as an official nomenclature term. Rather, the latest edition of the Guide to Receptors and Channels states, ‘Some tissues possess α_1A-adrenoceptors that display relatively low affinity in functional and binding assays for prazosin (pK_i < 9) that might represent different receptor states (termed α_1L-adrenoceptors)’ (Alexander et al., 2011). Further, investigation of this phenomenon is critical to developing a better treatment for BPH as it would seem that men suffering from urethral obstruction resulting from BPH would benefit more from a selective α_1L-adrenoceptor antagonist rather than the selective α_1A-adrenoceptor antagonists like tamsulosin, which are currently used and show no selectivity between α_1A and α_1L-adrenoceptors. At present, there are no antagonists showing higher affinity for α_1L over α_1A-adrenoceptors.

Early attempts to explain how the α_1L-adrenoceptor phenotype could arise from the α_1A-adrenoceptor gene concentrated on whether genetic polymorphisms or splice variants of this gene could give rise to the phenotype. However, α_1A-adrenoceptors generated by known polymorphisms and splice variants in cell culture models all showed similar pharmacological characteristics to that of the α_1A-adrenoceptor (Shibata et al., 1996; Suzuki et al., 2000; Ramsay et al., 2004), providing evidence that α_1A-adrenoceptor polymorphisms and splice variants were not associated with generation of the α_1L-adrenoceptor phenotype.

Subsequently, a ‘interacting protein’ hypothesis to explain the generation of α_1L-adrenoceptors from α_1A-adrenocepors has been postulated, following observations from radioligand binding studies. The basis for this hypothesis is that radioligand binding studies of lower urinary tract tissues are almost always carried out using membrane homogenates and yield ligand affinities that fit the pharmacological profile of α_1A-adrenoceptor pharmacology. This is despite the findings that isolated intact preparations of prostate, urethra and bladder tissue display α_1L-adrenoceptor pharmacology when they have been used in functional studies. In an earlier review, Nishimune et al., (2010a) suggested that this discrepancy was caused by the homogenization process disrupting the cell membrane and thus separating α_1A-adrenoceptors from the putative ‘interacting protein’. They hypothesized that only when the α_1A-adrenoceptor is bound to the interacting protein does it display α_1L-adrenoceptor pharmacology. This idea is supported by their earlier paper that demonstrated that radioligand binding studies of tissue segments from lower urinary tract tissues produced a α_1L-adrenoceptor ligand affinity profile, while crude homogenized membrane fractions from these tissues yielded a α_1A-adrenoceptor profile (Muramatsu et al., 2005). The presumption is that the more intact tissue segments maintain a more complete membrane with little or no disruption to the α_1A-adrenoceptor – interacting protein complex.

This interacting protein theory is a plausible and logical explanation for the occurrence of α_1L-adrenoceptor pharmacology in lower urinary tract tissues expressing abundant α_1A-adrenoceptors. Indeed, the interaction of proteins also covers the possibility that a receptor heteromer may be the cause of the changed pharmacology of α_1A-adrenoceptors in the lower urinary tract. However, it is arguable whether tissue homogenization would disrupt the cell membrane sufficiently to destroy protein–protein interactions at the molecular level. Nevertheless, Nishimune et al. (2010b) identified cysteine-rich epidermal growth factor-like domain 1α (CRELD1α) as a novel down-regulating protein and therefore a protein that interacts with the α_1A-adrenoceptor. CRELD1α was identified using a yeast two-hybrid approach, with the entire open reading frame of the human α_1A-adrenoceptor gene used as bait (Nishimune et al., 2010b). Subsequent transfection of cDNA for the α_1A-adrenoceptor gene alone into CHO cells yielded cells expressing α_1A-adrenoceptors with the typical α_1A-adrenoceptor pharmacological profile, as well as a low proportion of α_1L-adrenoceptor sites. The small number of α_1L-adrenoceptors was presumably due to endogenous CRELD1α as knockdown of CRELD1α enhanced the expression of α_1A-adrenoceptors while over-expression of CRELD1α reduced α_1A-adrenoceptor expression (Nishimune et al., 2010b). Following this, they were able to produce α_1A-adrenoceptor-enhanced and α_1L-adrenoceptor-dominant cell lines that were used in ligand binding, and functional agonist and antagonist profile studies. Results for α_1A-adrenoceptor-enhanced and α_1L-adrenoceptor-dominant CHO cells were in agreement with the published profiles for α_1A and α_1L-adrenoceptor phenotypes, respectively, as seen in intact tissues.

Although the evidence presented in this paper is persuasive (Nishimune et al., 2010b), there are still questions to be answered before the story is truly convincing. For instance, CRELD1α over-expression yielded α_1L-adrenoceptor dominant cells expressing a higher proportion of α_1L-adrenoceptors; however, this was because of a reduction in α_1A-adrenoceptor binding sites rather than their conversion to α_1L-adrenoceptors as would be expected if CRELD1α were a true α_1A-adrenoceptor interacting protein. Consequently, the expression of α_1L-adrenoceptor binding sites does not appear to change regardless of CRELD1α expression. Alternatively, this observation itself could be interpreted as the CRELD protein inhibiting radioligand binding in some way, perhaps by internalization of the α_1A-adrenoceptor rather than changing its binding affinity to that of the phenotype of the α_1L-adrenoceptor. Experiments using membrane-permeable and membrane-impermeable agonists and antagonists would go a long way towards a clearer understanding of these results.

Other observations from this research throw up questions that need answers. For instance, the efficacy of agonists seen in cells with different levels of CRELD1α expression differs when their activity is compared in functional assays. This introduces the possibility of ligand-biased signalling, which needs to be addressed and could further confound progress in this area. Similar agonist efficacy differences have been observed in functional experiments with intact tissues expressing the different phenotypes. Furthermore, over-expression of CRELD1α (α_1L-dominant) also seemed to introduce an element of irreversible antagonism to prazosin at low concentrations, when compared with CRELD1α knockdown (α_1A-abundant) cells in functional assays.

The idea of a α_1A-adrenoceptor interacting protein is a logical and plausible argument to explain the transition from α_1A-adrenoceptor gene to the α_1L-adrenoceptor functional phenotype which is abundant in male lower urinary tract tissue. CRELD1α appears to go part of the way to fulfilling the criteria that one would expect of an interacting protein for this receptor, but further experimental challenges remain to strengthen its case. Such experiments might include radioligand binding of whole cells versus membrane preparations using α_1A-adrenoceptor-expressing CHO cells with and without simultaneous CRELD1α transfection. This would show whether the α_1A-adrenoceptor – CRELD1α complex can be disrupted during membrane homogenization, leading to a change in pharmacological profile. Furthermore, demonstration of a direct interaction of the α_1A-adrenoceptor and CRELD1α using resonance energy transfer techniques (FRET/BRET) in transfected CHO cells or immunoprecipitation techniques in native tissue could provide significant support for this hypothesis.

Glossary

BPH

benign prostatic hyperplasia

BRET

bioluminescence resonance energy transfer

CRELD

cysteine-rich epidermal growth factor-like domain

FRET

fluorescence resonance energy transfer

Conflicts of interest

The author states no conflict of interest.