Norepinephrine transporter variant A457P knock-in mice display key features of human postural orthostatic tachycardia syndrome

Abstract

Postural orthostatic tachycardia syndrome (POTS) is a common autonomic disorder of largely unknown etiology that presents with sustained tachycardia on standing, syncope and elevated norepinephrine spillover. Some individuals with POTS experience anxiety, depression and cognitive dysfunction. Previously, we identified a mutation, A457P, in the norepinephrine (NE; also known as noradrenaline) transporter (NET; encoded by SLC6A2) in POTS patients. NET is expressed at presynaptic sites in NE neurons and plays a crucial role in regulating NE signaling and homeostasis through NE reuptake into noradrenergic nerve terminals. Our in vitro studies demonstrate that A457P reduces both NET surface trafficking and NE transport and exerts a dominant-negative impact on wild-type NET proteins. Here we report the generation and characterization of NET A457P mice, demonstrating the ability of A457P to drive the POTS phenotype and behaviors that are consistent with reported comorbidities. Mice carrying one A457P allele (NET(+/P)) exhibited reduced brain and sympathetic NE transport levels compared with wild-type (NET(+/+)) mice, whereas transport activity in mice carrying two A457P alleles (NET(P/P)) was nearly abolished. NET(+/P) and NET(P/P) mice exhibited elevations in plasma and urine NE levels, reduced 3,4-dihydroxyphenylglycol (DHPG), and reduced DHPG:NE ratios, consistent with a decrease in sympathetic nerve terminal NE reuptake. Radiotelemetry in unanesthetized mice revealed tachycardia in NET(+/P) mice without a change in blood pressure or baroreceptor sensitivity, consistent with studies of human NET A457P carriers. NET(+/P) mice also demonstrated behavioral changes consistent with CNS NET dysfunction. Our findings support that NET dysfunction is sufficient to produce a POTS phenotype and introduces the first genetic model suitable for more detailed mechanistic studies of the disorder and its comorbidities.

Fig. 1.

Generation of NET A457P knock-in mice. (A) Schematic of the wild-type (WT) SLC6A2 (NET) locus, targeting vector and NET locus following recombination. ‘WT NET’ shows the location of the targeting arms, and BglII sites and 3′ probe used in Southern confirmation. White arrows depict the PCR primers used to genotype A457P mice. ‘A457P Construct’ depicts the position of the floxed Neo/CRE cassette and the PGK cassette. Neo is under the control of the RNA polymerase II promoter and CRE is driven by the testis-specific angiotensin converting enzyme promoter, producing excision of the floxed region upon germline targeting. A457P in exon 9 in the 5′ arm is depicted. ‘Targeted A457P Allele’ illustrates the construct targeted to the NET locus. The BglII sites and 3′ probe location illustrate the shift in size of the BglII fragment, produced by insertion of the construct, to 11.6 kb (from 7.8 kb in WT NET). Arrows indicate PCR primers used to confirm correct insertion at both the 5′ (white arrows) and 3′ (black arrows) end of NET. The gray arrow shows a forward primer used with the 5′ arm reverse primer (white arrow) to genotype the A457P mutation through PCR and dideoxy sequencing. ‘Targeted A457P Allele Floxed Region Excised’ shows the final status of the NET allele, with the Neo/CRE cassette excised and one loxP site remaining in intron 11. White arrows depict the PCR primers used to genotype A457P mice, as shown above in WT NET. (B) Southern analysis of BglII-digested DNA from NET^+/+ and NET^+/P mice. (C) PCR genotype analysis of NET^+/+, NET^+/P and NET^P/P mouse DNA to detect the residual loxP site.

Fig. 2.

[³ H]NE transport and expression in A457P knock-in mice. Data are expressed as mean percent ± s.e.m. of NET^+/+ mice (A–C,E,F) or as raw values ± s.e.m. (D). *P<0.05, **P<0.01 and ***P<0.001 compared with NET^+/+ mice (two-tailed t-test). (A) [³H]NE transport in the SNS, n=5–6. Vent., ventricle; VD, vas deferens. (B,C) Single point [³H]NE transport in brain, n=6–7. (D) Saturation kinetics of [³H]NE transport in brain, n=8. Cortex NET^+/+ (black squares), V_MAX=6.6±0.40 moles×10⁻¹⁶/μg/minute, K_M=87.8±20.9 nM; cortex NET^+/P (white squares), V_MAX=4.6±0.50 moles×10⁻¹⁶/μg/minute, K_M=96.4±38.1 nM; hippocampus NET^+/+ (black triangles), V_MAX=8.2±0.77 moles×10⁻¹⁶/μg/minute, K_M=118.3±40.6 nM; hippocampus NET^+/P (white triangles), V_MAX=5.5±0.54 moles×10⁻¹⁶/μg/minute, K_M=136.8±46.3 nM. (E,F) NET expression in total and surface fractions of cortex (CTX) and hippocampus (HPC) synaptosomes in NET^+/P (n=7; E) and NET^P/P (n=7; F) mice compared with NET^+/+ mice.

Fig. 3.

Radiotelemetry in NET^+/+ versus NET^+/P mice. (A) 3-day collection of HR measures; data were analyzed and graphed in 30-minute increments. Traces shown start at lights off in the colony room. (B) HR during 3 days of data collection, expressed as means ± s.e.m. averaged over the 24-hour period and during 12-hour day and night periods. Significance was determined by twoway ANOVA. Genotype: F(1,27)=15.59, P<0.001. Time of day: F(2,27)=18.26, P<0.0001. *P<0.05, post-hoc t-test, n=5. (C) Mean arterial BP measured during the same 3-day period as in A and B (n=5).

Fig. 4.

Baroreflex sensitivity in NET^+/+ versus NET^+/P mice. (A,B) Dose response of systolic BP (SBP) to PHE (A) and NTP (B). Two-way ANOVA revealed effects of dose for both PHE and NTP [F(4,50)=61.78, P<0.01, n=3–6 for NTP and F(5,60)=23.53, P<0.01, n=3–6 for PHE]. (C,D) Baroreceptor sensitivities [the ratio of the change in HR to the changes in SBP at 20 μg/kg PHE (C) and 30 μg/kg NTP (D)] were calculated. BRS, baroreflex sensitivity; PHE, phenylephrine; NTP, nitroprusside.

Fig. 5.

Catecholamines and metabolites in A457P knock-in mice. Plasma and urinary NE and DHPG (A,C), and DA and EPI (B,D) in male (A,B) and female (C,D) A457P mice. Data are expressed as the mean percent ± s.e.m. compared with NET^+/+ mice. Significance was determined by two-way ANOVA. Male plasma genotype, F(2,135)=50.77, P<0.001; male urine genotype, F(2,215)=10.82; P<0.001; female plasma genotype, F(2,110)=20.21, P<0.001; female urine interaction, F(8,85)=2.88, P<0.01. *P<0.05, **P<0.01 and ***P<0.001 compared with NET^+/+. n=6–12 (male plasma), 9–20 (male urine), 8–9 (female plasma) and 6–7 (male urine).

Fig. 6.

NET^+/+ versus NET^+/P mice in behavioral tests. Data are expressed as mean ± s.e.m. (A–C) Behavior in the EPM. For time spent in the different zones (A), significance was determined by two-way ANOVA. Interaction: F(2,96)=13.74, P<0.0001. n=16–18. Freezing behavior (B) and head dips (C) were also assessed. n=16–18. (D,E) Open field exploration. Time spent in the center versus surrounding zones (D) and distance travelled (E) were measured. n=16–18. (F) Time immobile in the last 4 minutes of the TST. n=7–9. *P<0.05, **P<0.01, post-hoc t-test.