Introduction

Top Abstract Introduction Mice with Genetically... The 2A-Subtype Mediates the... The 2B- and 2C-Subtypes:... Conclusions and Future... References

Adrenergic receptors mediate the physiological responses to the catecholamines, norepinephrine and epinephrine. They belong to the superfamily of G protein-coupled receptors containing seven putative transmembrane domains and are divided pharmacologically into ₁-, ₂-, and -adrenergic receptor types (Bylund, 1988). ₂-Adrenergic receptors are implicated in diverse physiological functions particularly of the cardiovascular system and the central nervous system. ₂-Adrenergic receptor agonists are used clinically in the treatment of hypertension, glaucoma, and attention-deficit disorder, in the suppression of opiate withdrawal, and as adjuncts to general anesthesia. ₂-Adrenergic receptors have been divided into three subtypes (_2A, _2B, and _2C) on the basis of pharmacological and molecular cloning evidence (Lomasney et al., 1991; Bylund et al., 1994; Hein and Kobilka, 1995).

Understanding the role of ₂-adrenergic receptor subtypes in these diverse functions is clearly important particularly from a pharmacological point of view. One line of evidence supporting differential functions of the subtypes is differences in their characteristics, such as their tissue distributions throughout development (Winzer-Serhan et al., 1997), and in the adult, their coupling to G proteins and regulation in response to agonist stimulation. Although in situ hybridization studies of ₂-adrenergic receptor subtype expression in mice during development and in adults (Wang et al., 1996) and rats (Nicholas et al., 1993; Rosin et al., 1993; Scheinin et al., 1994) can reveal where ₂-adrenergic receptor subtypes are expressed, these findings cannot definitively link particular subtypes to physiological functions. Furthermore, assigning the physiological functions of ₂-adrenergic receptors to specific subtypes in vivo has been difficult because of the lack of sufficiently subtype-selective agonists and antagonists. The ability to genetically manipulate ₂-subtypes provides an alternative approach to elucidating subtype-specific functions as demonstrated in recent experiments using mice with deletions, mutations, or overexpression of specific ₂-adrenergic receptor subtypes (MacDonald et al., 1997).

	Mice with Genetically Engineered 2-Adrenergic Receptor Subtypes

Top Abstract Introduction Mice with Genetically... The 2A-Subtype Mediates the... The 2B- and 2C-Subtypes:... Conclusions and Future... References

Several recent reviews have discussed the methods, advantages, and limitations of genetic engineering techniques (Wei, 1997; Rohrer and Kobilka, 1998; Yanez and Porter, 1998). There are now published reports on five mouse strains with genetic alterations of ₂-adrenergic receptor expression. Mice with a deletion of the _2A- (_2A-knockout [KO]), _2B- (_2B-KO), or _2C-gene (_2C-KO) have been generated (Link et al., 1995, 1996; Altman et al., 1999). More recently, the double knockout mice (_2AC-KO), in which both the _2A- and the _2C-genes have been deleted, have been produced (Hein et al., 1999). Mice have also been developed with a point mutation of the _2A-gene (_2A-D79N) (Macmillan et al., 1996). This mutation of the aspartate to an asparagine residue at position 79 in the second transmembrane domain of the _2A-adrenergic receptor selectively uncouples the receptor from the activation of K⁺ channels in vitro, although coupling to Ca²⁺ channels and adenylyl cyclase activity is maintained (Surprenant et al., 1992). It was expected that the expression of this mutation in the intact animal would provide insight into the signal transduction mechanisms mediating the effects of _2A-adrenergic receptor stimulation. However, _2A-D79N mice showed an approximately 80% reduction in _2A-adrenergic receptor binding despite normal mRNA levels. The receptors that were expressed showed the expected pharmacological characteristics but were unable to couple to K⁺ or Ca²⁺ channels (Lakhlani et al., 1997). Thus, the _2A-D79N receptor expressed in vivo exhibits distinct characteristics compared with its expression in vitro, and this has served as a functional knockout. All four of the mouse strains described above are viable and fertile and appear grossly normal. Apparently, none of the ₂-adrenergic receptor subtypes are absolutely required for embryonic development or adult survival, although one or more of the subtypes may play a role in normal development. In addition to knockout strategies, transgenic techniques have also been applied to ₂-adrenergic receptors, and a strain of mice has been generated in Kobilka's laboratory with approximately 3-fold overexpression (OE) of the _2C-gene (_2C-OE) under the control of its homologous promoter (Sallinen et al., 1997).

Results from experiments using mice with genetic alterations of ₂-adrenergic receptor expression are summarized in Tables 1 and 2. Several complicating factors should be kept in mind when interpreting the results from these experiments. Compensatory changes, such as the up- or down-regulation of another component of a signaling pathway, could offset the loss of a functional receptor in a genetically engineered mouse. These compensatory changes could also be the cause of a phenotype. A phenotype could result from developmental changes rather than from altered expression of a receptor in the adult, or the altered receptor expression could be a distant cause in a complex chain of physiological events. Some of the data obtained with particular animals, however, argue against compensatory changes occurring at least after manipulation of the _2A-adrenergic receptor subtype (Janumpalli et al., 1998). There can also be remarkable differences in inbred mouse strains, necessitating the use of appropriate wild-type strains in experiments with KO and transgenic mice. Altered expression of a receptor could cause different phenotypes in young and old mice, males and females, different genetic backgrounds, or different environments. Crabbe and coworkers (1999) recently reported that different behavioral phenotypes were found by different laboratories using the same mouse strains, even different phenotypes in the same laboratory at different times, indicating that behavioral experiments in genetically altered mice are particularly vulnerable to variability. Thus, reproducibility is crucial for one to have confidence in the results from modifications of gene expression.

View this table: [in this window] [in a new window]   TABLE 1 Physiological effects of altering 2-adrenergic receptor gene expression in mice The 2A-subtype mediates most of the classical effects of 2-adrenergic receptor agonists.

View this table: [in this window] [in a new window]   TABLE 2 Effects of altered 2C-adrenergic receptor gene expression on behavior

	The 2A-Subtype Mediates the Classical Effects of 2-Adrenergic Receptor Agonists

Top Abstract Introduction Mice with Genetically... The 2A-Subtype Mediates the... The 2B- and 2C-Subtypes:... Conclusions and Future... References

Through experiments with the _2A-KO and _2A-D79N mice, most of the classical effects of ₂-adrenergic receptor agonists can be attributed to the _2A-subtype. Mice with a mutated or deleted _2A-subtype do not exhibit the hypotensive, sedative, antinociceptive, anesthetic-sparing, or hypothermic effects in response to ₂-adrenergic agonists.

Hypotensive Effects. ₂-Adrenergic agonists activate ₂-receptors in the rostral ventrolateral medulla, decreasing sympathetic outflow, which causes a reduction in arterial blood pressure and heart rate (Guyenet, 1997). In addition to these centrally mediated responses, there is a transient hypertensive response caused by ₂-adrenergic receptor-mediated vasoconstriction of vascular smooth muscle. The hypothesis of _2A-adrenergic receptor involvement in the centrally mediated cardiovascular responses was based on _2A-adrenergic receptor expression in the rostral ventrolateral medulla (Nicholas et al., 1996), and _2A-adrenergic receptor involvement was confirmed in _2A-D79N mice. The hypotensive response to administration of ₂-adrenergic receptor agonists was abolished, demonstrating that the _2A-subtype plays a principal role in this response (Macmillan et al., 1996). The bradycardic response to agonist also was blunted in _2A-D79N mice (Macmillan et al., 1996). These results have been confirmed in both _2A-D79N and _2A-KO mice (Altman et al., 1999; Zhu et al., 1999). Furthermore, the hypertensive response was abolished in _2B-KO mice, and the hypotensive effect was immediate and accentuated. The bradycardic response in _2B-KO mice was normal, and _2C-KO mice showed no differences from wild-type strains in their hypertensive, hypotensive, and bradycardic effects (Link et al., 1996). The _2A-subtype appears to play a role in vasoconstriction at least in some vascular compartments because the hypertensive response in _2A-D79N mice was absent when the agonist was administered through the femoral artery (Macmillan et al., 1996). These results demonstrate that the _2A-adrenergic receptor mediates the hypotensive and bradycardic effects of ₂-adrenergic agonists. In contrast, the _2B-adrenergic receptor appears to be the main mediator of the pressor response that results from ₂-adrenergic agonist administration.

Sedative Effects. The sedative effects of dexmedetomidine were examined in _2A-D79N mice by Rotarod, loss of righting reflex (Lakhlani et al., 1997), and spontaneous locomotor activity tests (Hunter et al., 1997). In all cases, _2A-D79N mice showed no sedation in response to dexmedetomidine, indicating that the _2A-adrenergic receptor mediates the sedative effects of ₂-agonist administration. In contrast, both the _2B-KO and _2C-KO mice showed dose-dependent reductions in locomotor activity in response to dexmedetomidine that were indistinguishable from wild-type mice (Hunter et al., 1997). ₂-Adrenergic agonists appear to induce sedation by activating autoreceptors in the locus coeruleus, reducing its spontaneous rate of firing (Nacif-Coelho et al., 1994). Several lines of evidence have implicated the _2A-subtype in this action, including the prominent expression of _2A-receptor mRNA and protein in the locus coeruleus seen with in situ hybridization and immunohistochemical studies (Nicholas et al., 1993; Rosin et al., 1993; Wang et al., 1993; Scheinin et al., 1994). In _2A-D79N mice, ₂-adrenergic receptor agonists were unable to alter the spontaneous firing rate of locus coeruleus neurons, confirming the role of the _2A-subtype (Lakhlani et al., 1997).

Antinociceptive Effects. Another therapeutic use of ₂-adrenergic receptor agonists is analgesia (Eisenach et al., 1996). The antinociceptive effect of dexmedetomidine has been studied in the ramped hot-plate test as well as in hot-water immersion and intense light tail-flick latency tests. In all of these tests, _2A-D79N mice showed no antinociceptive response to dexmedetomidine (Hunter et al., 1997; Lakhlani et al., 1997). In contrast, dexmedetomidine induced normal dose-dependent antinociception in _2B-KO and _2C-KO mice in the tail immersion test (Hunter et al., 1997). Spinal analgesia was examined in _2A-D79N mice using tail-flick latency tests and the Substance P behavioral test, which uses inhibition of Substance P-induced behaviors as an indirect measure of antinociception. In the tail-flick latency test, both intrathecal brimonidine and clonidine induced dose-dependent antinociception in wild-type but not _2A-D79N mice (Stone et al., 1997; Fairbanks and Wilcox, 1999). In the Substance P behavioral test, the antinociceptive effect of intrathecal ₂-adrenergic agonists was blunted in _2A-D79N compared with wild-type mice. Presumably, the remaining antinociceptive effect in _2A-D79N mice is due to residual _2A-adrenergic receptor activity, although a small effect due to another subtype cannot be ruled out. Thus, the _2A-subtype is the predominant subtype involved in the analgesic effects of ₂-adrenergic receptor agonists.

Other Effects. Presynaptic inhibition of norepinephrine release is a classic ₂-adrenergic function. Dexmedetomine potently inhibited neurotransmitter release in the vasa deferentia of _2A-D79N, _2B-KO, and _2C-KO mice. This inhibitory effect, however, was greatly attenuated in _2A-KO mice, and the stimulatory effect of the ₂-adrenergic antagonist yohimbine was attenuated as well (Altman et al., 1999). Similar results have been found in the brain (hippocampus and occipito-parietal cortex) and the heart (atrium) of _2A-KO mice (Trendelenburg et al., 1999). These data indicated that the _2A-subtype is the most important in mediating presynaptic ₂-adrenergic receptor inhibition of neurotransmitter release, although a role for at least one other subtype seemed probable. Recent studies on the sympathetic nerves in the heart of _2A-KO and _2C-KO mice as well as in mice lacking both the _2A- and the _2C-subtypes (double knockout; _2AC-KO) have confirmed and extended these conclusions. In the _2A-KO but not the _2C-KO mouse, the maximal inhibitory effect of brimonidine on norepinephrine release was significantly reduced but not eliminated as compared with the wild type. In the _2AC-KO mouse, however, the inhibitory effect of brimonidine was completely abolished (Hein et al., 1999). Further experiments in these mice indicate that the _2A-receptor inhibits transmitter release at high stimulation frequencies, whereas the _2C-subtype regulates release at lower levels. The regulation at both high and low frequencies appears to be physiologically important (Hein et al., 1999).

	The 2B- and 2C-Subtypes: Fewer Defined Functions

Top Abstract Introduction Mice with Genetically... The 2A-Subtype Mediates the... The 2B- and 2C-Subtypes:... Conclusions and Future... References

_2B-Subtype. In comparison with the _2A-subtype, relatively less has been discovered about the functions of the _2B- and _2C-subtypes through knockout experiments. As noted above, the _2B-subtype appears to have a dominant role in eliciting the vasoconstrictor response to ₂-adrenergic agonists because this response is lacking in _2B-KO mice (Link et al., 1996). The _2B-adrenergic receptor has also been implicated in salt-induced hypertension. When subjected to subtotal nephrectomy followed by dietary salt loading, the increase in blood pressure was much greater in _2C-KO and wild-type mice as compared to _2B-KO mice (Makaritsis et al., 1999a). The significance of this effect is enhanced by the fact that it was obtained with heterozygous _2B-KO mice (due to the difficulty in breeding homozygous mice because their survival is limited), and thus the authors conclude that a full complement of _2B-receptor genes is necessary to raise blood pressure in response to dietary salt loading. Although the role, if any, of the _2A-subtype cannot be determined from these studies, the data imply that the _2B- but not the _2C-subtype is prominently involved in the development of salt-induced hypertension.

_2C-Subtype. Unlike its counterparts, the _2C-subtype does not appear to play a major role in cardiovascular regulation or the other classical effects of ₂-adrenergic receptors. The cardiovascular and sedative effects of dexmedetomidine were normal in _2C-KO mice. Sallinen and coworkers (1997) reported small, but opposite, changes in the hypothermic effect of dexmedetomidine in _2C-KO and _2C-OE mice, indicating that the _2C-subtype may play a role in this effect secondary to the prominent role of the _2A-subtype. In both _2C-KO and _2C-OE mice, dexmedetomidine induced dose-dependent reductions in monoamine turnover indistinguishable from those in wild-type animals. However, _2C-OE mice showed slightly increased basal levels of dopamine and its metabolite homovanillic acid, whereas _2C-KO mice showed slightly decreased levels of metabolites of dopamine (homovanillic acid), norepinephrine (3-methoxy-4-hydroxyphenylglycol), and serotonin (5-hydroxyindoleacetic acid) (Sallinen et al., 1997). The opposite findings for homovanillic acid in _2C-KO and _2C-OE mice point to a possible role for _2C-adrenergic receptors in the regulation of dopamine systems in the brain.

	Conclusions and Future Directions

Top Abstract Introduction Mice with Genetically... The 2A-Subtype Mediates the... The 2B- and 2C-Subtypes:... Conclusions and Future... References

The subtype involved in many ₂-adrenergic receptor-mediated physiological functions is now known (at least in the mouse), but there are still many unanswered questions concerning their functional significance. For example, what is the role of each of the subtypes in development? Because the _2A-subtype mediates most of the classical effects of ₂-adrenergic agonists, it is doubtful that an _2A-selective agonist would have a substantially better clinical profile than the currently available agents. On the other hand, because the _2A-subtype has not yet been shown to be important in cognitive functions, whereas the _2C-subtype does appear to play a role in these functions, it may turn out that selective _2A-agents may have fewer central nervous system side effects than nonselective agents. Drugs acting at _2B- or _2C-adrenergic receptors are likely to have fewer of the classical ₂-adrenergic side effects than _2A-specific agents. However, because the functions of these subtypes are not as clear as those of the _2A-subtype, the therapeutic value of _2B- and _2C-selective drugs is also unclear. It would appear likely, however, that _2C-selective agents may be useful in at least some central nervous system disorders.

A comparison of the studies published to date using mice with altered expression of ₂-adrenergic receptors reveals some inconsistencies such as the role, if any, of the _2B- or _2C-subtypes in ₂-adrenergic-mediated spinal analgesia. The recent development of _2AC-KO "double-knockout" mice may help answer these questions (Hein et al., 1999). _2AC-KO mice (as well as _2AB-KO mice if they can be produced), when tested against _2A-KO mice, may show the involvement of the _2C- (and _2B-) subtype(s) in various functions that heretofore have been masked by the dominance of the _2A-subtype.

Generation of mice with inducible gene knockouts would more closely resemble acute blockade of specific ₂-adrenergic receptor subtypes and avoid any compensatory adaptations that might occur during development. Thus, inducible KO mice might reveal physiological roles of ₂-adrenergic receptors that have been masked by compensatory changes in current ₂-adrenergic receptor KO mice. In addition, inducible knockout mice might allow such compensatory modifications to be studied as they develop. However, for some responses, such as suppression of epileptogenesis, this time-intensive and expensive experimental strategy may not be warranted. This is because the responses in _2A-D79N mice are indistinguishable from wild-type mice treated with idazoxan, and idazoxan administration has no further effect in _2A-D79N mice evaluated in the kindling paradigm (Janumpalli et al., 1998). Furthermore, transgenic studies using a limited promoter region of the _2A-adrenergic receptor subtype indicate that we do not yet know how to achieve faithful reproduction of the expression profile of this subtype (Wang et al., 1996).

KO and transgenic mice are likely to be important tools in drug development for determining the physiological site of action for newly developed pharmacological agents. Such an approach has already been used to determine that the hypotensive effects of two putative imidazoline-1 receptor agonists, moxonidine and rilmenidine, are mediated predominantly by _2A-adrenergic receptors in the mouse (Fairbanks and Wilcox, 1999; Zhu et al., 1999).

The ability to probe subtype-specific functions in mice by altering the same ₂-adrenergic receptor (_2A-KO and _2A-D79N mice; _2C-KO and _2C-OE) and the general consistency of the results strengthens the conclusions drawn from these studies. Despite their acknowledged limitations, these genetic approaches have provided, and are expected to continue to provide, considerable insight into the functions of ₂-adrenergic receptor subtypes.