F. Mayor Jr, Departamento de Biología Molecular, Centro de Biología Molecular ‘Severo Ochoa’ (CSIC-UAM), Universidad Autónoma, 28049 Madrid, Spain. Fax: + 34 91 397 47 99, Tel.: + 34 91 397 48 65, E-mail: email@example.com
G protein-coupled receptor kinase 2 (GRK2) and β-arrestin 1 are key regulatory proteins that modulate the desensitization and resensitization of a wide variety of G protein-coupled receptors (GPCRs) involved in brain functions. In this report, we describe the postnatal developmental profile of the mRNA and protein levels of GRK2 and β-arrestin 1 in rat brain. The expression levels of GRK2 and β-arrestin 1 display a marked increase at the second and third week after birth, respectively, consistent with an involvement of these proteins in brain maturation processes. However, the expression attained at birth and during the first postnatal week with respect to adult values (45–70% for GRK2, ≈ 30% for β-arrestin 1) is relatively high compared to that reported for several GPCRs, indicating the existence of changes in the ratio of receptors to their regulatory proteins during brain development. On the other hand, we report that experimental hypothyroidism results in changes in the patterns of expression of GRK2 and β-arrestin 1 in cerebral cortex, leading to a 25–30% reduction in GRK2 levels at several stages of development. Such changes could help to explain the alterations in GPCR signaling that occur during this pathophysiological condition.
G protein-coupled receptor kinases (GRKs) comprise a family of serine-threonine kinases [1,2] that specifically phosphorylate the agonist-occupied form of G protein-coupled receptors (GPCRs). This is followed by binding to the phosphorylated receptor of members of a second family of proteins, termed β-arrestins, leading to receptor uncoupling from heterotrimeric G proteins and signal shut off. This process is generally known as desensitization, a general feature of GPCR signalling that involves a loss of receptor responsiveness after acute or sustained activation. GRK2 (also referred to as β-adrenergic receptor kinase 1) and β-arrestin 1 are the more abundant GPCR regulatory proteins and display a ubiquitous expression pattern [2,3]. Notably, these proteins show high levels of expression in brain and are widely present throughout the nervous system [4–6]. The distribution patterns of GRK2 and β-arrestin 1 proteins in the adult brain do not match exactly, but both proteins are mainly localized in neurons at postsynaptic densities. Therefore, GRK2 and β-arrestin 1 appear to be appropriately located to regulate a variety of G protein-coupled neurotransmitter receptors. In this regard, these regulatory proteins have been reported to desensitize several GPCRs present in the nervous system, such as α-adrenergic and β-adrenergic, muscarinic, dopaminergic, and angiotensin receptors, among others [1,2,7]. Although much knowledge about the onset of neurotransmitter responses and the developmental profiles of several receptors implicated in neurotransmission exists, the ontogenic pattern of expression of their putative regulatory proteins GRK2 and β-arrestin 1 has not been described.
On the other hand, information about the regulatory factors that govern GRK2 or β-arrestin 1 expression in the nervous system is very limited to date. It has been extensively reported that thyroid hormone (T3) plays an important role in normal brain development, and that its deficiency results in severe morphological and biochemical alterations in the brain [8,9]. Despite such important effects, only the expression of a few genes in the brain has been demonstrated to be directly regulated by thyroid hormones ([10–12] and references therein). Furthermore, hypothyroidism does alter the expression levels of several synaptic G protein-coupled receptors and also modifies their functional state [13–16]. Given that GRK2 and β-arrestin 1 are involved in the regulation of GPCR responses, the possibility that these proteins are targets for the action of thyroid hormones deserved to be explored.
In this context, we report for the first time the protein and mRNA expression levels of GRK2 and β-arrestin 1 in whole rat brain during postnatal development. We also report the effect of experimental hypothyroidism on GRK2 and β-arrestin 1 mRNA and protein expression levels during the postnatal period in the rat cerebral cortex. These data may contribute to a better understanding of the role(s) of these proteins in brain maturation, as well as help to establish a possible correlation with the developmental patterns of synaptic G protein coupled-receptors.
Materials and methods
Wistar rats raised in our animal facilities were used and the rules of the European Community Council (Directive of 24 November 1986, 86/609/ECC) concerning maintenance and handling of animals were followed. Hypothyroidism was induced by a combination of chemical and surgical thyroidectomies, as described . To induce fetal and neonatal hypothyroidism, dams were given 0.02% methyl-mercapto-imidazol (MMI; Sigma Chemical Co., St Louis, MO) in their drinking water from the 9th day after conception. Additionally, on postnatal day 5 pups were surgically thyroidectomized. Hypothyroidism in adult rats was induced by thyroidectomy at postnatal day 40 followed by 20 days of MMI treatment. This procedure results in profound hypothyroidism, as shown by very low thyroid hormone concentrations in the brain, and by obvious physiological landmarks . For brain analysis at the moment of birth, term fetuses were delivered with intact placentas by rapid hysterectomy of cervically dislocated dams at 22 days of gestation and immediately sacrified. Animals were killed at different postnatal ages by decapitation. Cerebral cortices or total brains were removed and frozen at −70 °C until used.
Ribonuclease protection assays
Total RNA was extracted from cerebral cortices or whole brain at different developmental stages by the method of Chomczynski and Sacchi  and specific mRNA expression was analyzed by RNase protection assays (RPA-II; Ambion). Specific probes for rat GRK2 and β-arrestin 1 transcripts were generated by PCR using bovine GRK2 and β-arrestin 1 cDNAs (generously provided by J. L. Benovic, Thomas Jefferson University, Philadelphia, PA, USA) as templates, by using oligonucleotides bearing EcoRI and HindIII restriction sites (underlined) to amplify the desired regions. For GRK2 (nucleotides (nt) 1795–1897), the forward primer 5′-CCACGAGGAATTCTACGCCCTGGGTAAGGACTGC-3′, and the reverse primer 5′-AAGCTTGGGGAACAGGTAGAAGTATCGCCGCTGCC-3′ were used; and for β-arrestin 1 (nt 731–849), the forward primer 5′-CACGGAGAATTCATCAGCGTCAATGTCCATGTCACC-3′ and the reverse primer 5′-GGGAAGCTTGTACTGAGCTGTGTTGAAGAGACAGATGTCTGC-3′ were used. Amplified fragments were subcloned in Bluescript vectors (Stratagene). The vectors were linearized and single-stranded antisense transcripts generated by T7 polymerase in the presence of [α-32P]CTP (400 Ci·mmol−1; Amersham Corp.). Control sense transcripts were generated in a similar way by using T3 polymerase. Total RNA obtained from rat brain (30 µg) as described above was hybridized with antisense riboprobes at 45 °C overnight in 20 µL hybridization buffer (80 mm formamide, 100 mm sodium citrate pH 6.4, 300 mm sodium acetate pH 6.4, and 1 mm EDTA). Subsequently, reaction samples were digested with a mixture of RNases A and T1 for 40 min at 37 °C and protected fragments were resolved in a 6% polyacrilamide/8 m urea gel, visualized by autoradiography, and quantified by laser densitometry. A control probe for rat GAPDH was obtained from Ambion and all data were normalized according to GAPDH expression. Control rat GRK2 and β-arrestin 1 cDNAs were kindly provided by R. J. Lefkowitz (Duke University, Durham, NC, USA). Full-length mRNA was obtained from these constructs by using T3 and T7 polymerase, respectively.
Determination of GRK2 and β-arrestin 1 protein levels
Whole rat brain or cerebral cortex was homogenized using a Polytron device in 4 vol. of 20 mm Tris/HCl, pH 7.5, 5 mm EDTA, 5 mm EGTA and protein inhibitors (2 µg·mL−1 aprotinin, 30 µg·mL−1 bacitracin, 30 µg·mL−1 trypsin inhibitor, 1 mm benzamidine, 1 mm phenylmethanesulfonyl fluoride). The homogenate was centrifuged (1500 g, 5 min, 4 °C) to obtain a crude postnuclear supernatant. Aliquots of these lysates containing 100 µg of protein were resolved in 10% SDS/PAGE and transferred to nitrocellulose membranes for 60–75 min in 10 mm NaHCO3, 3 mm Na2CO3 pH 10, plus 20% methanol using a Trans-Blot cell from Bio-Rad. The filters were blocked with 10 mm Tris/HCl pH 7.5, 150 mm NaCl (NaCl/Tris) and 5% fat-free dried milk. GRK2 protein was detected with AbFP1, a polyclonal antibody raised against a fusion protein containing aminoacids 50–145 of bovine GRK2 , whereas β-arrestin 1 was detected with Ab186, a polyclonal antibody raised in our laboratory (P. Penela, unpublished results) against amino acids 172–286 of bovine β-arrestin 1. The specifity of these antibodies has been validated previously [6,21,22]. Blots were developed using a chemiluminiscent method (ECL, Amersham) after incubation with a goat antirabbit antibody conjugated to peroxidase.
mRNA levels of GRK2 and β-arrestin 1 in rat brain during postnatal development
The levels of GRK2 and β-arrestin 1 mRNAs were quantified by a sensitive RNase protection assay in total RNA isolated from whole brain of rats 0–21 days old. The first three weeks after birth represent the most active phase of brain growth and comprise the time of most vulnerability to thyroid hormone deficiency . Both GRK2 and β-arrestin mRNAs were specifically detected with cRNA probes generated by PCR using GRK2 and β-arrestin 1 cDNAs as templates (Fig. 1). A RNA fragment of the expected size was protected by the GAPDH probe incubated with brain RNA preparations, whereas β-arrestin 1 and GRK2 probes yielded two bands in close proximity to the size corresponding to the theoretical hybrids to be formed (Fig. 1, upper panel). This is not a consequence of some intrinsic feature of the probes, given that a single band is protected when they are incubated with their corresponding control sense RNA sequences (Fig. 1, lower panel). This could suggest the existence of different transcripts for β-arrestin 1 and GRK2. However, a similar pattern of bands was obtained in the presence of full-length sense RNA synthesized in vitro from plasmids bearing rat GRK2 or β-arrestin 1 cDNAs (Fig. 1, lower panel), indicating that the observed heterogeneity is probably due to the presence in the bovine probes of several mismatches with the rat mRNA transcripts. Both bands were taken into consideration when analyzing the developmental expression patterns of GRK2 and β-arrestin 1.
Neonatal brain displays relatively high levels of mRNA expression for both GRK2 and β-arrestin 1 (38% and 52%, respectively, compared to the adult period; Fig. 2). These levels are maintained during the first days of postnatal development, except for a modest transient increase in GRK2 mRNA levels detected around day 5 after birth (59% versus 38% at birth, P < 0.05). β-arrestin 1 expression is markedly augmented at day 9, attaining levels slightly higher than adult values, while GRK2 mRNA undergoes a delayed rise compared to β-arrestin 1, attaining almost adult levels at postnatal day 20. These data are different with respect to the postnatal development of several synaptic GPCRs, for which expression levels are generally low at birth and sharply rise during the first two weeks of life [23–25].
Levels of GRK2 and β-arrestin 1 in the developing rat brain
The analysis of GRK2 and β-arrestin 1 protein levels, by using specific antibodies in total brain samples obtained at different stages after birth, reveals a developmental profile (Fig. 3) that does not exactly match the pattern of mRNA expression described above. The amount of GRK2 in the first days after birth is maintained at levels equivalent to those observed in neonatal brain (43% of adult values). This is followed by a clear increase from the end of the first postnatal week to day 20. Likewise, β-arrestin 1 levels also remained without significant changes throughout the first 9 days of life, with values fluctuating around 25–37% of adult expression, whereas a clear increase is detected in the third postnatal week. Therefore, although it appears that the GRK2 and β-arrestin 1 expression levels attained during brain development are mainly driven by transcriptional mechanisms, postranscriptional events must also be operating to explain why mRNA and protein levels are not parallel at some postnatal stages. This fact is exemplified by comparing the relative expression of protein and mRNA for GRK2 on postnatal day 9 and immediatly after birth. Thus, although mRNA values are similar at both time periods (38% and 43% of adult level at postnatal days 0 and 9, respectively), protein expression has already attained adult levels at day 9, while at birth kinase levels are markedly lower (43% of adult levels).
Expression of mRNA for GRK2 and β-arrestin 1 during postnatal development in the cerebral cortex of hypothyroid rats
We assessed the effect of hypothyroidism on the developmental pattern of GRK2 and β-arrestin 1 mRNA by performing RNase protection assays in total RNA samples from cerebral cortex. In euthyroid animals, the relatively high expression levels of β-arrestin 1 mRNA at day 5 increase slightly at day 20, then noticeably decrease after this age (66% of reduction between days 20–60; Fig. 4). This biphasic pattern is similar to that obtained with whole brain (Fig. 2), although higher changes in mRNA levels between day 5 and day 20 are noted in the latter. In the period considered, hypothyroidism induces even more elevated mRNA levels at day 5, which remain high afterwards, thus blunting the developmentally programmed decrease in β-arrestin 1 mRNA that takes place in normal animals after postnatal day 20, and leading to a constant level of expression throughout the period analyzed. On the other hand, hypothyroidism did not significantly affect the normal pattern of GRK2 mRNA expression, which increases nearly two-fold from day 5 to day 20 (when maximal levels are attained) and is sustained afterwards.
Effects of hypothyroidism on GRK2 and β-arrestin 1 levels
In order to test whether the effect of hypothyroidism on β-arrestin 1 mRNA abundance was also reflected by changes in protein levels, we perfomed immunoblots using extracts from cerebral cortices of normal and hypothyroid animals obtained at different postnatal ages (Fig. 5). Euthyroid cortex displays low amounts of β-arrestin 1 at day 5 (17% of levles at day 60), in clear contrast with the near-maximal mRNA levels observed at this stage (see Fig. 4). This discrepancy suggests the existence of additional postranscriptional regulatory mechanisms that take place during development. β-arrestin 1 levels subsequently increase up to day 20 and remain unchanged at day 60 in normal animals. Hormone deficiencies induce significantly higher protein levels of β-arrestin 1 at day 5 with respect to control rats, in good agreement with the mRNA data, although the expression pattern thereafter is very similar to that observed during normal conditions. Interestingly, GRK2 protein levels underwent significant variations as a consequence of hormone deficiency, even if mRNA levels were not affected (Fig. 4). Thus, hypothyroidism led to a 25–30% decrease in GRK2 levels at day 5 and 60, although the overall developmental profile of GRK2 protein was not modified. These results suggest that GRK2 expression in cerebral cortex is mainly affected by hypothyroidism at a postranscriptional level, whereas changes in transcription and/or transcript stability are observed for β-arrestin 1.
GRK2 and β-arrestin 1 are the most abundant members of their protein families in the adult brain, and have been reported to participate in the modulation and signaling of a variety of GPCRs. Since the cellular complement of these proteins is important to determine the efficacy of GPCR signal transduction [1,2,26,27], we have explored their expression pattern during postnatal development.
The analysis of the developmental profiles of GRK2 and β-arrestin 1 in rat brain raises several issues. First, the overall developmental pattern of both GRK2 and β-arrestin 1 is consistent with a functional involvement of these proteins in brain maturation processes. The mRNA and protein levels of β-arrestin 1 and GRK2 in whole rat brain display a marked increase at the second or third week of life, respectively, which correlates with the period of maximal synaptic connections, attained between postnatal days 12 and 24. The developmental profiles of several receptors involved in neurotransmission, which are targets for the action of GRK2 and β-arrestin 1, such as dopaminergic , adrenergic , or muscarinic receptors , also show noticeable increases starting on the first postnatal week. However, it should be noted that the relative levels of GRK2 and β-arrestin 1 at days 0–5 compared to adult values are, in general, much higher than those of the above mentioned GPCR. This suggests that the main role of GRK2 and β-arrestin 1 during this early period would be to modulate other GPCRs that are more actively expressed at this stage, and also indicates the existence of changes in the ratio of a given receptor to its regulatory proteins during brain development. Such variations in the receptor : GRK2-β-arrestin 1 ratio may represent an important physiological mechanism for modulating GPCR signalling and for determining the specificity of regulation of certain GPCRs [3,27].
Second, GRK2 and β-arrestin 1 profiles are not completely parallel through postnatal brain development. This is consistent with the observation that both proteins are not always expressed simultaneously in the same brain regions [4,5], indicating some degree of functional specifity for the combination of GRK and β-arrestin isoforms. Finally, it is also apparent that both transcriptional and post-transcriptional mechanisms operate to modulate the expression levels of GRK2 and β-arrestin 1 during brain development, and our data suggest that the contribution of such mechanisms to the overall expression of these proteins varies during development.
The expression pattern of several GPCRs is altered by T3 hormone deficiencies during brain development [15,28,29]. The adrenergic transduction system is particularly affected by the thyroid state [16,29–31], and it contributes to the brain disorders induced by hypothyroidism. We report here that hypothyroidism slightly but significantly decreases cerebral cortex GRK2 protein levels at day 5 and 60 of postnatal development without affecting kinase mRNA levels. This is consistent with the fact that no thyroid-response elements have been identified in the human GRK2 promoter  and suggests a postranscriptional mechanism. In this regard, we have recently reported that GRK2 is rapidly degraded by the proteasome-dependent pathway . An altered rate of protein degradation is one of the several modifications observed in some hypothyroid tissues as a result of an impairment in ubiquitin-dependent proteasome activity . It is tempting to suggest that alterations in proteasome function elicited by hypothyroidism lead to a decrease in GRK2 protein stability in such conditions.
It is worth noting that the extent of change in GRK2 levels that we report here, although modest, is in the same range of that observed in other physiopathological situations, such as in cerebral cortex or locus coruleus after opiate treatment , in lymphocytes of hypertensive patients  or in other experimental models , thus suggesting that such changes might be physiologically relevant. The observed decrease in GRK2 levels in hypothyroid cortex from 60-day-old rats adds to other reported alterations in GPCR signalling componentes induced by T3 deficit, including adenylyl cyclase , some Gαs isoforms , and several GPCRs, such as β-adrenoceptors [13,37]. Since changes in the cellular complement of GRK2 can alter receptor responsiveness [1,2,27] a decrease in GRK2 levels may contribute to the reported alterations of GPCR signaling in hypothyroid conditions.
The authors thank Drs J. L. Benovic and R. J. Lefkowitz for experimental tools and A. Morales for skillful secretarial assistance. This work is supported by grants from Ministerio de Educación y Cultura (PM95-0033 and PM98-0020) and the European Union (BMH4–3596) to F. M. Jr and from Plan Nacional (SAF98–0060) to A. M. The Centro de Biología Molecular ‘Severo Ochoa’ holds an Institutional grant from Fundación Ramón Areces.