Aging reveals a role for nigral tyrosine hydroxylase ser31 phosphorylation in locomotor activity generation

Abstract

Background: Tyrosine hydroxylase (TH) regulates dopamine (DA) bioavailability. Its product, L-DOPA, is an established treatment for Parkinson's disease (PD), suggesting that TH regulation influences locomotion. Site-specific phosphorylation of TH at ser31 and ser40 regulates activity. No direct evidence shows that ser40 phosphorylation is the dominating mechanism of regulating TH activity in vivo, and physiologically-relevant stimuli increase L-DOPA biosynthesis independent of ser40 phosphorylation. Significant loss of locomotor activity occurs in aging as in PD, despite less loss of striatal DA or TH in aging compared to the loss associated with symptomatic PD. However, in the substantia nigra (SN), there is equivalent loss of DA or TH in aging and at the onset of PD symptoms. Growth factors increase locomotor activity in both PD and aging models and increase DA bioavailability and ser31 TH phosphorylation in SN, suggesting that ser31 TH phosphorylation status in the SN, not striatum, regulates DA bioavailability necessary for locomotor activity.

Methodology and principal findings: We longitudinally characterized locomotor activity in young and older Brown-Norway Fischer 344 F(1) hybrid rats (18 months apart in age) at two time periods, eight months apart. The aged group served as an intact and pharmacologically-naïve source of deficient locomotor activity. Following locomotor testing, we analyzed DA tissue content, TH protein, and TH phosphorylation in striatum, SN, nucleus accumbens, and VTA. Levels of TH protein combined with ser31 phosphorylation alone reflected inherent differences in DA levels among the four regions. Measures strictly pertaining to locomotor activity initiation significantly correlated to DA content only in the SN. Nigral TH protein and ser31 phosphorylation together significantly correlated to test subject's maximum movement number, horizontal activity, and duration.

Conclusions/significance: Together, these results show ser31 TH phosphorylation regulates DA bioavailability in intact neuropil, its status in the SN may regulate locomotor activity generation, and it may represent an accurate target for treating locomotor deficiency. They also show that neurotransmitter regulation in cell body regions can mediate behavioral outcomes and that ser31 TH phosphorylation plays a role in behaviors dependent upon catecholamines, such as dopamine.

Figure 1. Dopamine, tyrosine hydroxylase protein, and phosphorylation stoichiometry (ps) comparisons in nigrostriatal and mesoaccumbens pathways.

Analyses of striatum (str), substantia nigra (SN), nucleus accumbens (NAc), and ventral tegmental area (VTA) from all test subjects (n = 14, seven pairs of 12 and 30 month old rats from locomotor study). a. total DA recovery The mean relative total recovery of DA (in ng) was Str = 214±16; SN = 8.4±0.8; NAc =  59.6±5.8; VTA = 6.0±1.0. b. total TH protein recovery The mean TH recovery (in ng) was Str = 473±49; SN = 59±9; NAc = 68±9; VTA = 26±5. c. TH-Normalized DA tissue content Str 0.56±0.11, SN 0.18±0.03, NAc 0.77±0.03, VTA 0.22±0.03. There was a significant difference in DA recovered per TH among the four regions (ANOVA (p<0.0001) and between Str and SN (p<0.001), Str and VTA (p<0.01), SN and NAc (p<0.001), and VTA and NAc (p<0.001). d. ser31 phosphorylation stoichiometry (PS) Str 0.33±0.02, SN 0.09±0.01, NAc 0.36±0.01, VTA 0.10±0.01. There was a significant difference in ser31 phosphorylation among the four regions (ANOVA (p<0.0001) and between Str and SN (p<0.001), Str and VTA (p<0.001), SN and NAc (p<0.001), and VTA and NAc (p<0.001). e. ser40 PS Str 0.022±0.001, SN 0.045±0.004, NAc 0.049±0.004, VTA 0.035±0.002. There was a significant difference in ser40 phosphorylation among the four regions (ANOVA (p<0.0001) and between Str and SN (p<0.001), Str and NAc (p<0.001), Str and VTA (p<0.05), and NAc and VTA (p<0.05). f. ser19 PS Str 0.06±0.01, SN 0.13±0.01, NAc 0.12±0.01, VTA 0.25±0.01. There was a significant difference in ser19 phosphorylation among the four regions (ANOVA (p<0.0001) and between Str and SN (p<0.001), Str and NAc (p<0.01), Str and VTA (p<0.001), SN and VTA (p<0.001) and NAc and VTA (p<0.001).

Figure 2. Tyrosine hydroxylase effective stoichiometry in nigrostriatal and mesoaccumbens pathways.

Values are obtained by multiplying total inherent TH protein content by phosphorylation stoichiometry in each sample. a. Ser31 es. Stoichiometry determined by anti-phospho ser31 TH against calibrated standards for ser31 and divided by total TH protein. b. Ser40 es c. ser19 es Values were calculated based upon each subject's ps value and multiplied by the mean total TH. Region-to-region difference for ser31 es was similar to the difference in total DA (as shown in Fig. 1a ) recovered from the longitudinally-characterized inherent locomotor activity patterns for the test subjects.

Figure 3. Dopamine bioavailability and correlation to locomotor activity generation.

Total DA in striatum ((str) a, c, e) versus substantia nigra ((SN) b, d, f) and correlation to mean lifetime locomotor activity measures in all 14 subjects, ages 12 and 30 months (n = 7 each age group). The following are the Spearman correlation values and p-values; Movement number, a. str, p = 0.65, r = 0.13 b. SN, **p = 0.008, r = 0.68, Horizontal activity, c. str, p = 0.58, r = 0.16 d. SN, **p = 0.008, r = 0.72; Time spent moving, e. str, p = 0.42, r = 0.24 e. SN, *p = 0.01, r = 0.64. Additional statistical correlations are presented in Table 1.

Figure 4. Longitudinal-characterization of locomotor activity during aging.

Mean for the age tested represented by horizontal line and the locomotor activities of test subjects represented by points. The young group was tested at 4 and 12 months and the old group tested at 22 and 30 months. a. movement number (ANOVA, group p<0.0001; post-hoc 12 vs. 30 mos, p<0.05): aging difference within group, 4 vs. 12 mos., p<0.05; 22 vs. 30 mos., p<0.05 (Student's paired t-test used for all within group comparisons). b. horizontal activity (ANOVA, group p<0.0001; post-hoc 12 vs. 30 mos., p<0.01): aging difference within group, 4 vs. 12 mos., p<0.05; 22 vs. 30 mos., p<0.05. c. time spent moving (ANOVA, group p<0.0001; post-hoc 12 vs. 30 mos., p<0.05): aging difference within group, 4 vs. 12 mos., p<0.05; 22 vs. 30 mos., p<0.05. Data not shown: total distance (ANOVA, group p<0.0001; post-hoc 12 vs. 30 mos, p<0.05): age difference within group, 4 vs. 12 mos., p<0.05; 22 vs. 30 mos., p>0.05, and movement velocity (ANOVA, group p<0.005; post-hoc 12 vs. 30 mos., p>0.05): aging difference within group, 4 vs. 12 mos., p<0.05; 22 vs. 30 mos., p<0.05.

Figure 5. Individual locomotor activities with respect to group throughout the life span.

Young adult and old rats were tested at 4 and 22 months old, and again at 12 and 30 months old. There was a significant correlation in locomotor activity between the mean performance for each test subject at the first testing period (ten trial average) and second testing period (seven trial average) of locomotor activity tested eight months later: a. movement number, p<0.0001, r = 0.865; b. horizontal activity (HAC), p = 0.0001, r = 0.846; c. time spent moving (as seconds per hour), p<0.0001, r = 0.868. Data not shown are movement speed (cm/sec), p = 0.0001, r = 0.848, and distance traveled (cm/hr), p<0.0001, r = 0.881.

Figure 6. Aging and DA tissue content.

Values are ng DA/ mg protein in 12- and 30-month old subjects in the nigrostriatal (a, b) and mesoaccumbens (c, d) dopaminergic pathways. a. striatum, ns, 12 mo.,178±17, 30 mo., 152±19, b. substantia nigra, **p<0.01, 12 mo., 7.6±0.5, 30 mo. 4.9±0.6, c. nucleus accumbens, ns, 12 mo., 83±6.9, 30 mo. 74±2.8, d. ventral tegmental area, ns, 12 mo., 9.0±1.5, 30 mo., 9.4±2.5. These statistical relationships (Student's t-test, unpaired) also held in the comparison of total DA recovery (not normalized to protein) in each region (Fig S4).

Figure 7. Nigral TH protein loss in aging.

a. Representative blot showing the ∼40% decrease in total TH (tTH) protein in the SN in a 30-month old BNF rat compared to the tTH content in the 12-month old. All TH protein assays are conducted against a standard curve with quantified TH protein. A specific quantity of total protein is loaded from the sample to determine the ng tTH per µg protein and the quantity from the sample is interpolated from the standard curve. b. nigral total TH protein content between the 12- and 30-mo old subjects, 0.053±0.010 and 0.028±0.005 ng per µg protein, respectively. p<0.05 (Student's t-test, unpaired).

Figure 8. Nigral TH phosphorylation stoichiometry in aging.

a. Representative blot showing the 30% decrease in TH phosphorylation at ser31 in the 30-month old BNF rat. All TH phosphorylation assays are conducted against a standard curve with quantified site-specific phosphorylation levels so actual quantities of phosphorylated TH at each phosphorylation site are loaded in the standard curve range and the quantity of phosphorylated TH from the sample is interpolated from the standard curve. Because actual quantities of total TH (tTH) vary from sample-to-sample, it is not always possible to load equal amounts of tTH. Total protein load among samples is kept within 25 µg if at all possible to keep any carrier protein effects upon the signal to an absolute minimum. In this case, total protein load differed by 10 µg (80 µg for the 12 mo sample and 90 µg for the 30 mo sample). The standard curve source is from a calibrated PC12 cell extract with a phosphorylation stoichiometry at ser31 at 0.09 ng p31 per ng tTH. b. Site-specific phosphorylation of TH in the SN in test subjects. The 30-month group had significantly reduced phosphorylation at ser 31 (p = 0.029, unpaired Student's t-test). No significant change was observed at ser40 or ser19.

Figure 9. Effective stoichiometry at ser31 and locomotor activity.

Effective stoichiometry of ser31 (the product of ser31 phosphorylation stoichiometry and TH protein quantity) correlates with the highest ever recorded locomotor activity measures (movement number, horizontal activity, and time spent moving) of the 17 trials for each test subject. Spearman correlation stats are as follows; movement number (a), r = 0.59, p = 0.03, horizontal activity (b), r = 0.65, p = .01, time spent moving (c), r = 0.57, p = 0.03.