Secondary Logo

Journal Logo

Clinical Aspects

A Tri-Nucleotide Pattern in a 3’ UTR Segment Affects The Activity of a Human Glucocorticoid Receptor Isoform

Green, Tajia L.; Leventhal, Stacey M.; Lim, Debora; Cho, Kiho; Greenhalgh, David G.

Author Information
doi: 10.1097/SHK.0000000000000750

Abstract

INTRODUCTION

Gene expression is highly regulated by epigenetic phenomena. The 3’ untranslated regions (UTRs) of genes contribute significantly to that regulation by affecting mRNA stability, localization, and translation (1–4). There are several mechanisms through which the 3’UTR influences mRNA activity. AU-rich elements (ARE) are known to affect mRNA stability and microRNAs (miRNAs) bind to the mRNA and generally decrease expression (3, 5). The length of the 3’UTR, often regulated by alternative polyA sites, can also modulate mRNA expression (6). PolyA sites proximal to the stop codon tend to be preferred sites of regulation. They create shorter 3’ UTR regions and usually result in greater protein expression (2, 7, 8). The increase in translation is possibly due to decreased availability of regulatory positions. Additionally, the 3’ UTR secondary structure can also significantly influence mRNA translation (3).

Glucocorticoids, through binding to the glucocorticoid receptor, are major regulators of homeostasis. They are also used in the treatment of a wide variety of inflammatory conditions including septic shock. As such, securely controlling glucocorticoid receptor expression and activities are essential. The human glucocorticoid receptor (hGR) has an extensive 3’ UTR region with multiple putative polyA signals (9, 10). During the initial identification of the hGR coding sequence, Hollenberg et al. (10) found that the proximal polyA signal at position 3101 (X03225) (corresponding to position 3963 of the current hGRα reference mRNA sequence NM_001018077) is the preferred site in a human placental cDNA library. It has also been found that an additional regulatory element, the class I ARE motif AUUUA, is overrepresented in the hGR 3’ UTR (9, 11). There are 10 AUUUA sites in the 3’ UTR of the biologically active splice variant, hGRα, and 4 sites in hGRβ, the major alternative splice variant. Three of the AUUUA motifs in hGRβ have been shown to be involved in hGRβ destabilization and for the AUUUA motif at position 3669, a naturally occurring A to G polymorphism has been linked to rheumatoid arthritis (12, 13). Schaaf and Cidlowski (13) also found that altering all 10 AUUUA sites in hGRα increased mRNA stability. Other important components in regulating hGR expression are the miRNA binding sites. The hGR 3’ UTR, in general, has been predicted to have numerous miRNA sites (11). Two miRNAs, miR-18 and miR-124a, have been found to decrease protein expression and there is evidence that miR-124a may be involved in tissue-specific expression (14, 15). In a separate investigation in mice, miR-96, miR-101a, miR-142-3p, and miR-433 were individually able to decrease glucocorticoid receptor expression by up to 40% (11, 16).

An interesting hGR isoform has been identified in our laboratory, hGR-S1(-349A) (Fig. 1). This splice variant retains intron H between exons eight and nine and also has a deletion at position 349, resulting in a frame shift and early termination which generates a novel 3’ UTR region (17). The putative protein has only 118 amino acids and lacks the transactivation, DNA binding, and ligand binding domains, which are the major functional domains. hGR-S1(-349A), however, has a hyperactive response to high doses of exogenous steroids in comparison to the reference hGRα. This hyperactivity is lost when the 3’ UTR region is removed. In this study, we delved into the structure and function of the 3’ UTR region to identify the component(s) contributing the hyperactivity of hGR-S1(-349A) and report an intriguing phenomenon relating to the 3’ UTR length which results in an autonomous hyperactive function of hGR-S1(-349A).

Fig. 1
Fig. 1:
Diagram of hGRα and hGR-S1(-349A).hGR consists of nine exons of which exons two through nine are translated. The biologically active isoform, hGRα, is 777 amino acids long and has a transactivation domain, DNA binding domain (DBD), hinge region (H), and ligand binding domain (LBD). hGR-S1(-349A) is a previously identified hyperactive isoform (17). hGR-S1(-349A) retains intron H between exons eight and nine, and has a single base pair deletion at position 349 of the coding sequence resulting in a frame shift and early termination. Its putative protein length is 118 amino acids. Arrows indicate primers used to amplify both isoforms. The stop codons and the -349A deletion are marked with vertical lines.

PATIENTS AND METHODS

Identification, nomenclature, and construction of hGR recombinant isoforms

The discovery and identification of hGR-S1(-349A) (JN797823) was described earlier (17). The unique hGR isoform was isolated from one of 97 volunteers who were surveyed for variability in their receptors. The nomenclature of the isoforms was also discussed in the previous publication and was based on the published coding sequence of the most common hGR (hGRα). The reference hGRα mRNA sequence was obtained from the National Center for Biotechnology Informatics (NCBI) (NM_001018077). The term “hGR-S” signifies an alternative “splice” variant and the 1 signifies the first variant found in our laboratory. The term “-349A” signifies loss of an adenine (A) at site 349 of the coding sequence.

Polymerase chain reaction (PCR) was used to create the various isoforms. For the 3’ UTR deletions, the hGR-S1(-349A) expression construct was used as the template and a forward primer upstream of the start codon (pc4A-1A) and a reverse primer specifically designed for the desired 3’ UTR length was used. These isoforms were named hGR-S1(-349A) followed by their 3’ UTR length. To create the isoforms lacking the H intron, PCR was used to amplify the first half from an isoform that contained the -349A deletion, and a back section from a different isoform without intron H. A second PCR reaction was then performed with the two sequences to concatenate the pieces into the desired single isoform. The complete list of primers used can be found in the Table, Supplementary Digital Content 1, http://links.lww.com/SHK/A480. The conjoined isoforms, as well as the other deletion isoforms that were created, were checked for accuracy by sequencing (MC Laboratories, South San Francisco, Calif or UCDNA Sequencing Facility, Davis, Calif). The hGR isoforms were cloned into the pcDNA4/HisMax vector (Life Technologies, Grand Island, NY) for functional analysis.

Measurement of transactivation potentials of hGR isoforms

An HEK293 cell subclone, tsA201, was obtained from Dr. Daniel Feldman at Shriners Hospitals for Children Northern California (2003) and was used for transfection with plasmids containing individual hGR isoforms. The cells were validated by Short Tandem Repeat profiling (ATCC, Manassas, Va). For each experiment, 10,000 to 12,000 tsA201 cells, suspended in 100 μL of antibiotic-free Dulbecco's modified Eagle medium (Life Technologies) and supplemented with 10% fetal bovine serum (Atlanta Biologicals, Lawrenceville, Ga), were seeded on day zero in a 96-well plate and then incubated at 37°C with 5% CO2. On day one, the cells were transfected with the hGR isoforms and a glucocorticoid response element (GRE)-luciferase reporter plasmid (PathDetect GRE Cis-Reporter Plasmid; Agilent Technologies, La Jolla, Calif) using Fugene 6 (Promega, Madison, Wis) per the manufacturer's protocol. On day two, the cells were treated with graded concentrations of hydrocortisone (10–2 μM to 1 μM) or vehicle (0.9% saline) for 24 h before measurement of activity. The hydrocortisone concentrations are based on previous experiments (17–19). Pharmaceutical grade hydrocortisone sodium succinate (Pfizer, New York, NY; clinical anti-inflammatory adult dosage, 15–240 mg; half-life, 8–12 h) was used for these assays. 1 μM mifepristone (RU486; Sigma, St. Louis, Mo) was also used as a glucocorticoid antagonist. 50 mM ethanol was used as the vehicle control. Transactivation potential was determined with a Luciferase Assay Kit (Agilent). A Perkin-Elmer MicroBeta Trilux (Perkin-Elmer, Waltham, Mass) was used to measure luminescence.

Western blot analysis of hGR isoforms

tsA201 cells transfected with recombinant hGR isoforms were lysed in ice-cold lysis buffer (Agilent) supplemented with cOmplete Protease Inhibitor Cocktail (Roche, Indianapolis, Ind) and the supernatants harvested. Extracted protein was run on a 4% to 20% Criterion gel (BioRad, Hercules, Calif). Separated proteins were transferred to a polyvinylidene difluoride Hybond-P membrane (GE Healthcare, Piscataway, NJ) and then blocked with 5% nonfat dry milk, washed, and incubated overnight with the HisG-HRP antibody (Life Technologies) in 5% milk in phosphate-buffered saline with 0.05% Tween-20. Protein visualization was performed via chemiluminescence using the ECL Plus Western Blot Detection System (GE Healthcare).

PCR quantitation

Total RNA was isolated from tsA201 cells transfected with recombinant hGR isoforms with an RNeasy Mini Kit (Qiagen, Valencia, Calif) followed by cDNA synthesis with QuantiTech Reverse Transcription Kit (Qiagen) with an oligo(dT) primer (pr49.3: GGCCACGCGTCGACTAGTACTTTTTTTTTTTTT). Real-time RT-PCR was then used to quantify the amount of recombinant hGR mRNA expression with Brilliant III Ultra-Fast SYBR QPCR (Agilent). This was performed using specific DNA oligonucleotide primers, pd4A-1D (GGATCCAGTGTGGTGGAATTCTGCA) and hGR 2–2I (CTGCTGTGGGAATCCCAGGTCA). A second PCR reaction with B2M was performed as an internal control (B2M-1A: CAGGTTTACTCACGTCATCCAG; B2M-2A: CAAAGTCACATGGTTCACACG).

Statistical analysis

For luciferase assays, all experiment samples were run in triplicate, except for the vehicle treatments that were run in duplicate, and each experiment repeated at least two times. After multiple experiments confirmed the data patterns, the data from multiple experiments were combined and normalized for presentation of figures and statistics. Values are expressed as means with error bars representing standard error of the mean (SEM). SEM values for the 3D graph are available in Table, Supplementary Digital Content 2, http://links.lww.com/SHK/A481. The results were compared by one-way ANOVA and significance was confirmed with a Tukey's post hoc test.

RESULTS

Retention of intron H and reference hGRα 3’ UTR do not contribute to hyperactivity of hGR-S1(-349A)

In a previous work, we found that the deletion, frame shift, and early termination in hGR-S1(-349A) creates a unique new 2805 bp 3’ UTR region. In addition, the newly formed 3’ UTR of hGR-S1(-349A) is necessary for its hyperactivity (17). Within the 3’ UTR are two unique components, the retained intron H and 303 base pairs (bp) of reference hGRα 3’ UTR after the reference stop codon. To investigate whether or not these two additions were responsible for the unexpected augmentation in activity of hGR-S1(-349A), either the 303 bp of 3’ UTR, intron H, or both were removed (Fig. 2A). The transactivation potential of these new constructs was tested after stimulation with graded doses of hydrocortisone. All three new constructs [without reference 3’ UTR (-UTR), without intron H (-H), and without reference 3’ UTR/intron H (-H/-UTR)] had an augmented, dose-dependent response to hydrocortisone (Fig. 2B). At 0.01 μM, there is minimal response; however, the activity of hGR-S1(-349A) without intron H (-H) is greater than the other constructs (P <0.01). At 0.1 μM of hydrocortisone, all constructs have significant activity although both constructs without intron H (-H and -H/UTR) had greater activity than either hGR-S1(-349A) or hGR-S1(-349A) without the 303 bp of 3’ UTR (-UTR) (P <0.05). The greatest activity was at the highest concentration, 1 μM, and there was no significant difference in the transactivation potential between any of the altered constructs and the original hGR-S1(-349A).

Fig. 2
Fig. 2:
Effect of intron H and reference hGRα 3’ UTR on the transactivation potential of hGR-S1(-349A).A, Line diagrams of the new hGR-S1(-349A) variant isoforms in comparison to reference hGRα. For reference, the stop codon (dashed gray vertical line) and primers used for amplification (arrows) are identified. Due to the primer position there is an additional 303 bp of hGRα 3’ UTR after the reference stop codon. The sequences for hGR-S1(-349A) and its variants, -UTR (removal of the 303 bp of reference 3’ UTR), -H (the removal of intron H), and -H/UTR (the removal of 303 bp of reference 3’UTR and intron H) follow. The position of the -349A deletion and new stop codon (vertical lines) are indicated. For comparison, the position of the reference hGRα stop is shown (dashed gray vertical line). B, When stimulated with graded doses of hydrocortisone, all constructs have a hyperactive response at higher doses (0.1 μM and 1 μM). The -H construct has a significantly greater hyperactive response at 0.01 μM (**: P <0.01), and at 0.1 μM the responses of -H and -H/-UTR are both greater than hGR-S1(-349A) and -UTR (*: P <0.05). There is no significant difference in transactivation potential between hGR-S1(-349A) and its variants at 1 μM hydrocortisone. H indicates intron H; Hyd, hydrocortisone; UTR, 303 bp of reference 3’ UTR. Data is a combination of seven experiments and is presented as means ± SEM.

1303 bp of 3’ UTR in hGR-S1(-349A) contributes to an autonomous hyperactive response

Due to the deletion, frame shift, and early termination, hGR-S1(-349A) has a novel 2805 bp 3’ UTR. To determine if there is an optimal 3’ UTR length or a critical region necessary for the hyperactivity of hGR-S1(-349A), the 3’ UTR of hGR-S1(-349A) was truncated in approximate 300 bp increments from the 3’ end of the reference hGRα. The transactivation potential of the deletion constructs was then tested with graded doses of hydrocortisone (0.01 μM, 0.1 μM, and 1 μM) (Fig. 3 and see Table, Supplementary Digital Content 2, http://links.lww.com/SHK/A481, which lists SEM values). The response of hGRα and hGR-S1(-349A) confirmed previous results where both isoforms have dose-dependent responses to hydrocortisone with hGRα activity peaking at 0.1 μM and hGR-S1(-349A) displaying a hyperactive response that peaks at 1 μM (17). Removing just the first 300 bp of 3’ UTR resulted in an almost complete loss of activity and this was the case with most of the other deletion constructs. Unexpectedly, the construct retaining 1303 bp of 3’ UTR resulted in a transactivation potential even greater than hGR-S1(-349A). Even more interesting was the finding that the hyperactivity did not require hydrocortisone stimulation. Therefore, the hyperactivity was autonomous of steroid stimulation. The hGR-S1(-349A)(1303) isoform is more than 4.5 times greater than reference hGRα peaks at 0.1 μM of hydrocortisone. The hGR-S1(-349A)(1303) also has more than double the activity of hGR-S1(-349A) when it peaks at 1 μM of hydrocortisone.

Fig. 3
Fig. 3:
Transactivation potentials of ∼300 bp serial 3’ UTR deletions of hGR-S1(-349A).The 3’UTR of hGR-S1(-349A) was truncated by ∼300 bp increments and stimulated with graded doses of hydrocortisone (selected isoforms shown). Line drawings of each isoform are shown under the graph. As expected, there is a dose-dependent response by hGRα which peaks at 0.1 μM of hydrocortisone, and a dose-dependent response by hGR-S1(-349A) which peaks at 1 μM hydrocortisone. The transactivation potential of hGR-S1(-349A)(1303) is significantly greater than all other isoforms with saline vehicle or with hydrocortisone at all concentrations. The activities of all other isoforms are negligible, as are hGRα and hGR-S1(-349A) with the saline vehicle. Hyd indicates hydrocortisone. Data is a combination of two experiments and is presented as means. (See Table, Supplementary Digital Content 2, http://links.lww.com/SHK/A481, which lists SEM values).

Subsequently, to determine if the classical hGR pathway plays a role in signaling for these two hyperactive hGR isoforms, hGR-S1(-349A) and hGR-S1(-349A)(1303) were treated with 1 μM of the glucocorticoid antagonist, mifepristone (RU486), before being stimulated with 1 μM of hydrocortisone. Mifepristone is able to effectively and significantly decrease the transactivation potentials of both constructs, indicating that the two isoforms interact with exogenous hydrocortisone for their activity (Fig. 4).

Fig. 4
Fig. 4:
Transactivation potentials of hGR-S1(-349A) and hGR-S1(-349A)(1303) with mifepristone treatment.Before being stimulated with 1 μM of hydrocortisone, hGR-S1(-349A) and hGR-S1(-349A)(1303) were treated with 1 μM of mifepristone or ethanol vehicle. The transactivation potentials of both isoforms were effectively blocked by mifepristone treatment. Data is a combination of two experiments and is presented as means ± SEM.

Incremental deletions of the 3’ UTR surrounding hGR-S1(-349A)(1303) reveals a peak in activity with 1303 bp of 3’ UTR

To further dissect and possibly pinpoint a critical region of the 3’ UTR region involved in the autonomous hyperactivity seen in the hGR-S1(-349A)(1303) isoform, the 3’ UTR region was dissected to look at an area 50 bp up- and down-stream of 1303 bp by adding or removing 3’ UTR in ∼10 bp increments, creating constructs with 1342 bp to 1253 bp of 3’ UTR (Fig. 5A). When stimulated with graded doses of hydrocortisone in the luciferase assay, we found that the 3’ UTR regions surrounding hGR-S1(-349A)(1303) also have autonomous hyperactivity that is not dependent on hydrocortisone stimulation (Fig. 5, B and C). The amount of activation is directly dependent on 3’ UTR length and activity peaks with 1303 bp of 3’ UTR. The pattern was the same with all concentrations of hydrocortisone examined. Interestingly, an anomaly occurs with 1293 bp and 1263 bp of 3’ UTR in that there is virtually a complete loss of activity for both isoforms with vehicle and with hydrocortisone.

Fig. 5
Fig. 5:
Transactivation potentials of ∼10 bp serial deletions of 3’ UTR surrounding hGR-S1(-349A)(1303).A, Line diagrams depict the constructs altering the 3’ UTR in ∼10 bp increments surrounding hGR-S1(-349A)(1303), resulting in UTR lengths from 1342 bp to 1253 bp. The 3’ UTR of all new isoforms terminate in what would correspond to Exon 5 of hGRα. B, Transactivation potential of hGR-S1(-349A)(1342) through hGR-S1(-349A)(1253) without hydrocortisone (only 1 μM saline shown) revealed the transactivation potential is directly related to length of 3’ UTR with a peak in activity at hGR-S1(-349A)(1303). C, Transactivation potentials of the isoforms with hydrocortisone (only 1 μM shown) revealed that, similar to the vehicle controls, transactivation potentials are directly related to 3’ UTR length with activity peaking at hGR-S1(-349A)(1303). There is an unexpected loss of activity at 1293 bp and 1263 bp either with or without hydrocortisone stimulation. Data is a combination of four experiments and is presented as means ± SEM.

Deletion of the 3’ UTR in 1 bp increments surrounding hGR-S1(-349A)(1293) reveals a 3 bp cyclical loss in activity

To further investigate the inexplicable loss of activity of the 1293 bp 3’ UTR construct, an additional set of 3’ UTR deletion constructs was created in the 20 bp region surrounding hGR-S1(-349A)(1293) in increments of 1 bp (from 1302 bp to 1284 bp) (Fig. 6A). Luciferase assay results reveal a very interesting pattern. At all concentrations of vehicle and hydrocortisone in this deletion series, there is a loss of activity when the 3’ UTR ends at every third nucleotide while all other sites maintain the autonomous hyperactive activity (Fig. 6, B and C). This pattern of loss in activity continued until the 1263 bp position where the second drop in activity was seen. Additionally, investigation of the 3’ UTR region upstream of hGR-S1(-349A)(1303), from 1303 to 1313 showed that the pattern continued (Fig. 7).

Fig. 6
Fig. 6:
Transactivation potentials of 1 bp serial deletions of 3’ UTR surrounding hGR-S1(-349A)(1293).A, Line diagrams depicting the constructs altering the 3’UTR of hGR-S1(-349A)(1293) in 1 bp increments (1303 bp to 1283 bp of 3’ UTR). B, Transactivation potential of hGR-S1(-349A)(1303) through hGR-S1(-349A)(1283) without hydrocortisone (only 1 μM saline shown) shows a cyclic loss of activity every third base pair. C, Similar to vehicle controls, the transactivation potential of the constructs after treatment with graded doses of hydrocortisone (only 1 μM shown) has a cyclic loss of activity every third base pair. Data is a combination of four experiments and is presented as means ± SEM.
Fig. 7
Fig. 7:
Transactivation potentials of 1 bp serial deletions of 3’ UTR upstream of hGR-S1(-349A)(1303).A, Line diagrams depicting the constructs altering the 3’UTR upstream of hGR-S1(-349A)(1303) in 1 bp increments (1313 bp to 1303 bp of 3’ UTR). B, Transactivation potential of hGR-S1(-349A)(1313) through hGR-S1(-349A)(1303) without hydrocortisone (only 1 μM saline shown) shows the continued cyclic loss of activity at every third base pair. C, The transactivation potential of the isoforms after treatment with graded doses of hydrocortisone (only 1 μM shown) also has a cyclic loss of activity every third base pair. Data is a combination of four experiments and is presented as means ± SEM.

Subsequently, the transcript and protein expression of selected hGR-S1(-349A) deletion constructs was examined to determine if a difference in expression led to the decrease in activity. The constructs from 1302 to 1300 and 1287 to 1285 were transfected into tsA201 cells alongside hGR-S1(-349A)(1303). The expression of all isoforms was compared by Western blot. The luciferase assay confirmed a loss of activity and the Western blot showed that the 1302 and 1287 isoforms expressed proteins that were similar in size and amount in comparison to all the other deletion isoforms (Fig. 8, A and B). Additionally, real-time RT-PCR was used to contrast the expression of hGR-S1(-349A)(1303) and hGR-S1(-349A)(1293), an active and nonactive isoform, respectively (Fig. 8C). The real-time RT-PCR showed that there was also no difference in mRNA expression between the active and inactive isoforms.

Fig. 8
Fig. 8:
No difference in expression of inactive isoforms.A, Western blot for hGR-S1(-349A)(1303) with two sets of 3’ UTR deletion isoforms that have loss of activity, 1302 to 1300 UTR and 1287 to 1285 UTR. All isoforms express the same ∼18 kd putative protein and there is no difference in expression between the autonomous hyperactive isoforms and the inactive isoforms. B, Corresponding luciferase data for same isoforms treated with 1 μM hydrocortisone confirms no activity at 1302 UTR and 1287 UTR. Experiment was performed once as a confirmation and is presented as means ± SEM. C. Real-time RT-PCR for hGR-S1(-349A)(1303), an active isoform, and hGR-S1(-349A)(1293), an inactive isoform showed no difference in mRNA expression. Data is a combination of two experiments and is presented as means ± SEM.

Screening for homology of the active 1303 3’ UTR sequence in other genes

Since there was such a profound influence of the 1303 bp 3’ UTR of the hGR-S1(-349A) on activity, a 250 bp region overlapping the 3’ end of hGR-S1(-349A)(1303) was aligned against the other members of the nuclear hormone receptor family. It was found that this region occurs in approximately the same area in all of the receptors with ∼50% identity (Fig. 9A). This region was BLAST analyzed against the NCBI database for its presence in other genes and matched to the androgen, mineralocorticoid, and progesterone receptors in humans and other species (Fig. 9B).

Fig. 9
Fig. 9:
Homology of hGR-S1(-349A)(1303) active region UTR to other nuclear receptors.A, Alignment results of a 250 bp region of 3’ UTR (shaded box) surrounding the end of hGR-S1(-349A)(1303) with the nuclear hormone receptors: androgen receptor (AR), estrogen receptor-α (ERα), mineralocorticoid receptor (MR), and progesterone receptor (PR). The 250 bp region aligned in the same area in all the receptors from the end of the hinge to the start of the ligand binging domain with coverage of 252 bp to 268 bp and approximately 50% identity (gray shaded region). D indicates DNA binding domain; H, hinge region; LBD, ligand binding domain; TD, transactivation domain. B, The 250 bp region was also subjected to a BLASTn analysis. The results for each of the nuclear receptors are shown.

DISCUSSION

Many studies of genes focus on polymorphisms within the coding region or promoters in the variable 5’ UTR (20–22). The role of the 3’ UTR in gene activity, especially in mutations that result in truncated proteins, has often been underappreciated (23). In the hGR, there have been reports of both activation and repression of activity in truncated versions of the receptor, which can be dependent on cell type (19, 23, 24).

Usually, single-nucleotide alterations that result in early termination lead to a decrease in activity, which is a logical expectation for a truncated protein. In our case, we found that hGR-S1(-349A) has an unexpected hyperactive effect with steroid treatment even though the truncated protein lacks many of the known functional components (ligand binding, DNA binding, and a significant portion of the transactivation domains). The 3’UTR is also required since activity is almost nonexistent when using an isoform without the 3’ UTR region (17). Furthermore, the hyperactivity of this truncated isoform is blocked by mifepristone, indicating that the classical glucocorticoid receptor pathway is still somehow involved in its function. This opens the door to possible alternative methods of action that have yet to be explored. It is conceivable that the shortened protein acts as a cofactor to augment the activity of the full-sized hGR. It is possible that during periods of profound stress, such as after a major burn injury, a splice variant, like the hGR-S1(-349A), is produced and results in an increased sensitivity of hGRα to endogenous corticosteroids. This cofactor, in essence, “supercharges” the patient's stress response to allow him or her to handle such a severe insult. We are currently evaluating hGR expression in patients with severe burns to determine whether this variant is present. To date, we have found the -349A deletion in four patients and all subjects express an isoform retaining intron H.

Furthermore, the hGR-S1(-349A)(1303) isoform, which has a 3’UTR length of 1303 bp and is hyperactive with or without the addition of steroids, may conceivably act like an oncogene in that it has lost any negative feedback and is constantly “turned on” (8). It is unlikely, however, to expect that a variant such as this would be found in nature under normal physiologic conditions since overexpression of the glucocorticoid response would lead to a severe case of Cushing syndrome that would most likely be fatal in utero. Cushing syndrome results in the overproduction of cortisol and has profound effects on the untreated patient. This hypothesis could be tested in knockout studies but it is unlikely that continuous expression of the GR without steroid binding would lead to a viable organism.

There have also been reports that many genes express alternative UTRs (25). In studies that have examined the effects of the 3’ UTR in the reference hGR, three areas of interest have become the focus as possible regulatory regions: AU- or GU-rich elements, miRNAs, and polyadenylation sites (3). The 1303 bp 3’ UTR of hGR-S1(-349A)(1303) was examined for these elements (RegRNA 2, http://regrna2.mbc.nctu.edu.tw/) (26). This analysis identified no AU- or GU-rich elements or polyadenylation sites; however, three putative miRNA sites (distance from the hGRα stop codon: 233 bp, hsa-miR-4667-5p; 377 bp, hsa-miR-4669; 907 bp, hsa-miR-4793-5p) were identified. These three miRNA target sites were initially found in a screening study but not much is known about their activities (27).

Most miRNAs act by reducing the stability of the mRNA or by repression of protein production. Western blot and real-time PCR data, however, indicated that there is no difference in mRNA or protein expression among the hGR-S1(-349A) isoforms with variant 3’ UTR (Fig. 8). Therefore, there is a small possibility these miRNAs sites might be involved in the hyperactivity, but they have not been examined.

Analysis of the deletion isoforms near the 1303 3’ UTR region revealed an unusual pattern in which there was a drop-off in activity at every third nucleotide (Fig. 10). The reason for the drop-off in activity remains unclear. In the three-nucleotide codon, the first two positions are important in determining coding specificity. Thus, it could be speculated that this type of nucleotide hierarchy partitioning may factor in the observed trinucleotide pattern of activity loss.

Fig. 10
Fig. 10:
Illustration of loss of function at every third nucleotide of the 3’ UTR for hGR-S1(-349A)(1303) to hGR-S1(-349A)(1253).Illustration of nucleotide and relative activity at each 3’ UTR position, 1303 to 1253, in grayscale, with greatest activity in black to no activity in white for 1 μM. Saline is on the top half and hydrocortisone is on the bottom half. Loss of activity is continued at every third position throughout the series. Hyd indicates hydrocortisone; Sal, saline.

The hGR-S1(-349A)(1303) 3’ UTR was surveyed for open reading frames (ORFs). A significant ORF was found that begins at the same position as the hGR-D alternative translation isoforms and terminates using a stop codon in the vector backbone (28). Due to the truncation and use of the vector sequence, the 3’ end of this transcript would be drastically different from the previously reported hGR-D isoforms. A preliminary study indicated that a putative protein may be produced from this start codon, and it will be important to confirm the expression and analyze possible functions.

Since the hGR is a member of the family of closely related nuclear receptors, this phenomenon may also be reflected in the other steroid hormone receptors (progesterone, mineralocorticoid, and androgen). Since the ∼1303 bp region of the hGR-S1(-349A) UTR would correspond to the reference hGRα transactivation site τ2, this would be a well-conserved area. Therefore, the findings from this study may have direct applicability to the other members of this receptor family.

While determining how a truncated hGR isoform, hGR-S1(-349A), becomes hyperactive in response to steroids, we identified a region of the 3’UTR that confers autonomous and profound hyperactivity. Even more interesting is the finding that this 3’UTR region is present in other nuclear receptors. This section of the nuclear receptors may be a major regulator of many nuclear receptor activities. It is well known that there are variable responses to steroid treatments for many diseases. Some patients respond to normal doses of steroids while other patients are “non-responders” who either do not respond or need higher doses for a response. It was clearly illustrated in the sepsis studies by Annane et al. (29) that some patients are “responders” and others are “non-responders” when tested for their response to adrenocorticotropic hormone. Our hypothesis is that variations in the hGR may explain these differential responses. Therefore, elucidating the mechanism behind this ordered anomaly may prove valuable in understanding the complex layers of gene regulation, as well as variations in patient response to glucocorticoid treatment.

REFERENCES

1. Berkovits BD, Mayr C. Alternative 3’ UTRs act as scaffolds to regulate membrane protein localization. Nature 2015; 522 7556:363–367.
2. Ngondo RP, Carbon P. ZNF143 is regulated through alternative 3’UTR isoforms. Biochimie 2014; 104:137–146.
3. Barrett LW, Fletcher S, Wilton SD. Regulation of eukaryotic gene expression by the untranslated gene regions and other non-coding elements. Cell Mol Life Sci 2012; 69 21:3613–3634.
4. Hughes TA. Regulation of gene expression by alternative untranslated regions. Trends Genet 2006; 22 3:119–122.
5. Bartel DP. MicroRNAs: target recognition and regulatory functions. Cell 2009; 136 2:215–233.
6. Elkon R, Ugalde AP, Agami R. Alternative cleavage and polyadenylation: extent, regulation and function. Nat Rev Genet 2013; 14 7:496–506.
7. Tian B, Manley JL. Alternative cleavage and polyadenylation: the long and short of it. Trends Biochem Sci 2013; 38 6:312–320.
8. Mayr C, Bartel DP. Widespread shortening of 3’UTRs by alternative cleavage and polyadenylation activates oncogenes in cancer cells. Cell 2009; 138 4:673–684.
9. Vedeckis WV, Ali M, Allen HR. Regulation of glucocorticoid receptor protein and mRNA levels. Cancer Res 1989; 49 (8 suppl):2295s–2302s.
10. Hollenberg SM, Weinberger C, Ong ES, Cerelli G, Oro A, Lebo R, Thompson EB, Rosenfeld MG, Evans RM. Primary structure and expression of a functional human glucocorticoid receptor cDNA. Nature 1985; 318 6047:635–641.
11. Vandevyver S, Dejager L, Libert C. Comprehensive overview of the structure and regulation of the glucocorticoid receptor. Endocr Rev 2014; 35 4:671–693.
12. Derijk RH, Schaaf MJ, Turner G, Datson NA, Vreugdenhil E, Cidlowski J, de Kloet ER, Emery P, Sternberg EM, Detera-Wadleigh SD. A human glucocorticoid receptor gene variant that increases the stability of the glucocorticoid receptor beta-isoform mRNA is associated with rheumatoid arthritis. J Rheumatol 2001; 28 11:2383–2388.
13. Schaaf MJ, Cidlowski JA. AUUUA motifs in the 3’UTR of human glucocorticoid receptor alpha and beta mRNA destabilize mRNA and decrease receptor protein expression. Steroids 2002; 67 7:627–636.
14. Turner JD, Alt SR, Cao L, Vernocchi S, Trifonova S, Battello N, Muller CP. Transcriptional control of the glucocorticoid receptor: CpG islands, epigenetics and more. Biochem Pharmacol 2010; 80 12:1860–1868.
15. Vreugdenhil E, Verissimo CS, Mariman R, Kamphorst JT, Barbosa JS, Zweers T, Champagne DL, Schouten T, Meijer OC, de Kloet ER, et al. MicroRNA 18 and 124a down-regulate the glucocorticoid receptor: implications for glucocorticoid responsiveness in the brain. Endocrinology 2009; 150 5:2220–2228.
16. Riester A, Issler O, Spyroglou A, Rodrig SH, Chen A, Beuschlein F. ACTH-dependent regulation of microRNA as endogenous modulators of glucocorticoid receptor expression in the adrenal gland. Endocrinology 2012; 153 1:212–222.
17. Baker AC, Green TL, Chew VW, Tung K, Amini A, Lim D, Cho K, Greenhalgh DG. Enhanced steroid response of a human glucocorticoid receptor splice variant. Shock 2012; 38 1:11–17.
18. Lasker MV, Leventhal SM, Lim D, Green TL, Tung K, Cho K, Greenhalgh DG. Hyperactive human glucocorticoid receptor isoforms and their implications for the stress response. Shock 2015; 43 3:228–232.
19. Tung K, Baker AC, Amini A, Green TL, Chew VW, Lim D, Nguyen ST, Yee KS, Cho K, Greenhalgh DG. Novel hyperactive glucocorticoid receptor isoform identified within a human population. Shock 2011; 36 4:339–344.
20. Breslin MB, Geng CD, Vedeckis WV. Multiple promoters exist in the human GR gene, one of which is activated by glucocorticoids. Mol Endocrinol 2001; 15 8:1381–1395.
21. Turner JD, Muller CP. Structure of the glucocorticoid receptor (NR3C1) gene 5’ untranslated region: identification, and tissue distribution of multiple new human exon 1. J Mol Endocrinol 2005; 35 2:283–292.
22. Turner JD, Vernocchi S, Schmitz S, Muller CP. Role of the 5’-untranslated regions in post-transcriptional regulation of the human glucocorticoid receptor. Biochim Biophys Acta 2014; 1839 11:1051–1061.
23. Geng CD, Pedersen KB, Nunez BS, Vedeckis WV. Human glucocorticoid receptor alpha transcript splice variants with exon 2 deletions: evidence for tissue- and cell type-specific functions. Biochemistry 2005; 44 20:7395–7405.
24. de Lange P, Segeren CM, Koper JW, Wiemer E, Sonneveld P, Brinkmann AO, White A, Brogan IJ, de Jong FH, Lamberts SW. Expression in hematological malignancies of a glucocorticoid receptor splice variant that augments glucocorticoid receptor-mediated effects in transfected cells. Cancer Res 2001; 61 10:3937–3941.
25. Nagasaki H, Arita M, Nishizawa T, Suwa M, Gotoh O. Species-specific variation of alternative splicing and transcriptional initiation in six eukaryotes. Gene 2005; 364:53–62.
26. Huang HY, Chien CH, Jen KH, Huang HD. RegRNA: an integrated web server for identifying regulatory RNA motifs and elements. Nucleic Acids Res 2006; 34 (web server issue):W429–W434.
27. Persson H, Kvist A, Rego N, Staaf J, Vallon-Christersson J, Luts L, Loman N, Jonsson G, Naya H, Hoglund M, et al. Identification of new microRNAs in paired normal and tumor breast tissue suggests a dual role for the ERBB2/Her2 gene. Cancer Res 2011; 71 1:78–86.
28. Lu NZ, Cidlowski JA. Translational regulatory mechanisms generate N-terminal glucocorticoid receptor isoforms with unique transcriptional target genes. Mol Cell 2005; 18 3:331–342.
29. Annane D, Sebille V, Charpentier C, Bollaert PE, Francois B, Korach JM, Capellier G, Cohen Y, Azoulay E, Troche G, et al. Effect of treatment with low doses of hydrocortisone and fludrocortisone on mortality in patients with septic shock. JAMA 2002; 288 7:862–871.
Keywords:

3’ untranslated region (UTR); cyclic activity; glucocorticoid receptor; hyperactivity; truncated isoform

Supplemental Digital Content

© 2017 by the Shock Society