Hyperactivity, dopaminergic abnormalities, iron deficiency and anemia in an in vivo opioid receptors knockout mouse: Implications for the Restless Legs Syndrome

Abstract

Previous studies have uncovered a potential role of the opioid system in iron hemostasis and dopamine metabolism. Abnormalities in both of these systems have been noted in human RLS. Autopsy studies of human RLS have shown an endogenous opioid deficiency in the thalamus. Opioids, particularly prolonged-release oxycodone/naloxone, have been approved in Europe to be a second-line therapy for severe restless legs syndrome (RLS). To study the role of opioid receptors in the pathogenesis of RLS, we used a triple knockout (KO) mouse strain that lack mu, delta, and kappa opioid receptors and explored the behavioral and biochemical parameters relevant to RLS. The triple KO mice showed hyperactivity and a trend of increased probability of waking during the rest period (day) akin to that in human RLS (night). Surprisingly, triple KO mice also exhibit decreased serum iron concentration, evidence of anemia, a significant dysfunction in dopamine metabolism akin to that noted in human RLS, as well as an increased latency in response to thermal stimuli. To our knowledge, this is the first study to demonstrate that the endogenous opioid system may play a role in iron metabolism and subsequently in the pathogenesis of anemia. It is also the first study showing that opioid receptors are involved in the production of motor restlessness with a circadian predominance. Our findings support the role of endogenous opioids in the pathogenesis of RLS, and the triple KO mice can be used to understand the relationship between iron deficiency, anemia, dopaminergic dysfunction, and RLS.

1. Introduction

The Restless legs syndrome (RLS) is a common neurological movement disorder, which affects up to 10% of the population across the world with about 2.5% being severely enough affected to desire medical attention [1]. The prevalence of the disease is much higher in a group with low iron levels [1]. RLS patients typically have a strong urge to move the legs, with or without abnormal sensations. The symptoms usually become worse during inactivity such as lying or sitting and are more prominent at night. There is at least a partial and temporary relief through activity [1].

Dysfunctional dopaminergic system and iron deficiency have been considered as possible mechanisms of the development of RLS [2, 3]. One of the primary treatments for RLS is dopaminergic agonists [1]. Iron-deficiency anemia is a well-established trigger for human RLS [1]. Either oral or intravenous iron can be used to treat RLS in certain cases [1]. It is generally believed that brain iron deficiency (BID) leads to a hyperdopaminergic and hyperglutamatergic states [4]. In addition, it has been pointed out that functional changes of the ascending arousal systems and deficient GABA-mediated inhibitory control may be responsible for the enhanced arousal mechanisms critical to RLS [5]. Recent experimental and clinical studies suggest that a BID-induced hypoadenosinergic state may also contribute to sensorimotor signs of RLS and the enhanced arousal state [6]. Ligands of the auxiliary alpha-2-delta subunit of voltage-dependent calcium channels such as pregabalin or gabapentin are used to treat RLS [1]. In another study of human RLS employing neuroimaging, the severity of RLS increased with a decrease in the availability of serotonin transporters in the pons and medulla [7]. Therefore, RLS is a disease caused by dysfunctions in multiple neurotransmitter systems.

Our previous autopsy studies of human RLS have shown an endogenous opioid deficiency of both beta-endorphin and met-enkephalin in the thalamus of RLS patients compared to controls [8]. Administration of opioids improve symptoms of RLS and the combination of oxycodone/naloxone has been approved in Europe to treat the disease [9]. In our previous in vitro studies exploring the interaction among the endogenous opioid, dopaminergic and iron regulatory systems, we induced an iron deficiency in the substantia nigra cells of rats by chelating iron with deferoxamine. This destroyed primarily dopaminergic cells. If the endogenous opioid analog delta-opioid peptide (p-ALA2, p-Leu5) enkephalin (DADLE) was first added to the system before iron chelation, the destruction of dopaminergic systems was largely prevented [10]. This suggested a mechanism using the iron regulatory system, the dopaminergic system, and the endogenous opioid system. In addition, an early study indicates that peripheral administration of beta-endorphin significantly elevated the pain threshold only in the iron-deficient rats, but not the wild types [11]. The group also found that the function of morphine and haloperidol, a dopamine D2 receptor antagonist, in elevation of the pain threshold was stronger in iron-deficient rats than in the wild types [11]. These findings, in combination with ours, imply that the iron regulatory system, the dopaminergic system, and the endogenous opioid system might potentially interact in RLS.

The current investigation extends our work to an in vivo animal model where behavioral and biochemical parameters were explored relevant to human RLS. It is difficult to capture the human correlate of RLS in an animal model [12]. First of all, the diagnosis of human RLS heavily depends on patients’ reports of the symptom of the urge to move limbs, which cannot be obtained from animals. Second, essential features that can be used as surrogates of or show analogy with human RLS and encompass a valid animal model have not been agreed upon. However, increased activity at rest time, altered sleep-wake cycles, hyperactivity, sensory abnormalities, iron deficiency, and dysfunctional dopaminergic system are some features commonly assessed. Here, we generated a strain of mice with triple knockout of all three subtypes of opioid receptor (mu, delta, and kappa). We did not use it as a complete model of RLS; instead, the animal model was used to validate the relevance of the opioid system for behavioral alterations as they can be observed in human RLS and its potential involvement in the dopaminergic system, iron metabolism and anemia. We hypothesized that complete knockout of all three subtypes of the opioid receptors could cause disrupted iron homeostasis, dysfunctional dopamine metabolism, and RLS-related behavioral phenotypes.

2. Material and methods

2.1. Animals

The triple knockout (KO) mice were generated from mu opioid receptor, delta opioid receptor and kappa opioid receptor KO mice imported from the Jackson Laboratory (Stock Nos: 007559, 007557 and 007558, respectively). Wild type (WT) mice were age-matched non-littermates in the same genetic background. Mice were housed in standard mouse cages at 72°F under 12-hour light, 12-hour dark (12 LD) cycle condition within a specific-pathogen-free (SPF) facility. The colony had a quarterly monitoring program of common rodent pathogens conducted by the Animal Care Service of the University of Florida.

2.2. Behavioral studies

Eight triple KO and 4 WT male mice with an average age of 7 months were used in the continuous open field analysis as previously described [13]. Briefly, each mouse was placed in the center of a VersaMax Legacy open field apparatus with enough corncob bedding, food, and water under 12 LD conditions. The apparatus contained infrared sensors along the walls to detect any breaks in the beams, which were then decoded by VERSDATA version 2.70–127E (AccuScan Instruments INC.) into behavioral patterns. The data were recorded every 15 min throughout the experiment. Data from the last four days were separated into light-on and light-off phases, and the total distances during each phase were added up and coded as day 4, night 4, day 5, night 5, day 6, night 6, day 7 and night 7 respectively. The analysis was conducted based on all four periods in each phase. Separately, the total distance for each 15 min of the last four days was recoded. If the total distance traveled in 15 min is 0, the mouse was considered as sleeping, and the data were coded as 0; otherwise, the mouse was considered as awake, and the data were coded as 1. Data were separated into a light-on and light-off phase. Each phase had four periods with 48 data points each.

In a separate experiment, 12 triple KO and 17 WT male mice with an average age of 4 months were tested for open field activity lasting only 30 min during light on or daytime period [14]. Bright illumination (approximately 1 k lux at the center by a 60 W white bulb) was focused on the center of each field.

Thirteen triple KO and 9 WT male mice with an average age of 4 months were tested for the perception of warm stimuli. Each mouse was placed in an acrylic restrainer with the distal end of its tail protruding on a metal surface maintained at 55°C. The timer was turned on once the tail touched the surface and immediately stopped when the mouse flicked its tail away from the heat.

2.3. Colorimetric assay for serum iron, ferritin, and transferrin

Blood was collected by cardiac puncture method from 7 triple KO (4 males, 3 females) and 8 WT mice (5 males, 3 females) with an average age of 10 months. The blood was allowed to clot and then separated by centrifugation at 1,500 g for 10 min. The serum was removed and centrifuged again at 1,500 g for 10 min for further purification. The iron concentration and transferrin level were quantified using QuantiChrom Iron Assay Kit (BioAssay Systems Inc.) and Mouse Transferrin ELISA Kit (Immunology Consultants Laboratory Inc.), respectively. Serum was collected with the same method from 5 male triple KO and 5 WT mice. The ferritin level was measured by Mouse Ferritin ELISA Kit (Immunology Consultants Laboratory Inc.), according to the manufacture’s instruction.

2.4. Mineral measurement in striatal tissues

Eight triple KO and 8 WT male mice with an average age of 4 months were sacrificed and perfused with ice-cold saline. Their striata were dissected out and quickly frozen with liquid nitrogen. The samples were shipped to Veterinary Diagnostic Laboratory of Michigan State University and measured by atomic absorption spectroscopy.

2.5. Complete blood count

Blood was collected by retro-orbital blood collection method from 14 triple KO and 9 WT male mice with an average age of 3 months. The blood was analyzed with a HemaVet machine (The Americas Drew Scientific Inc.) by the ACS Clinical Diagnostics Laboratory of the University of Florida.

2.6. Striatal monoamine analysis

Fourteen triple KO (7 males, 7 females) and 12 WT mice (6 females, 6 males) with an average age of 3 months were sacrificed, and their striata were dissected out and quickly frozen with liquid nitrogen. The samples were shipped to Neurochemical Core at Vanderbilt University for analysis. Dopamine (DA) and serotonin (5-HT), as well as their respective metabolites 3,4-dihyroxyphenylacetic acid (DOPAC), homovanillic acid (HVA) and 5-hydroxyindoleacetic acid (5-HIAA), were measured as described previously [15].

2.7. Quantification of food and water intake

The experiment was conducted essentially according to others [16]. Twelve triple KO and 8 WT male mice were acclimated to individual cages for six days. Food and water intakes were measured daily for four days. During the test, deionized water was available from a 25 ml plastic serological pipette with 0.2 ml gradations inserted into the cage. The top of the pipette was closed with a rubber stopper. The drinking tubes were placed to the mouse’s right of the food hopper. The sprouts were 25 mm above the cage bottom. Water intake was calculated from the scale on the pipette, and food intake was derived from the weight of the whole cage. The body weights of the mice were measured at the beginning and the end of the 4-day test.

2.8. Western blot

Western blot was done as previously described [17]. The striata were dissected from 6 triple KO and 8 WT brains with an average age of 4 months and homogenized in 200 μl of ice-cold lysis buffer (Tris-HCL 50mM, pH=7.4; NaCl 175mM; EDTA 5mM, pH=8.0) containing protease inhibitor cocktail (Roche); 22 μl of ice-cold 10% Triton X-100 was added in the homogenate. The mixtures were incubated for 30 minutes on ice and centrifuged at 10,000 × g for 15 minutes at 4°C. The supernatant was used as protein samples for Western blot. The protein concentration of the supernatant was measured by protein assay reagent (Bio-Rad). An aliquot of the supernatant corresponding to 30 μg of protein was mixed with 2 × loading buffer containing 2-mercaptoethanol and boiled for 5 minutes, chilled on ice and spun down. The proteins were separated on a 10% SDS-PAGE gel and transferred to Millipore Immobilon –FL transfer membranes (PVDF). The PVDF membranes were washed in 0.1M PBS for 5 min and blocked with LI-COR Odyssey blocking buffer for 1 hour. The membranes were incubated overnight at 4°C with goat polyclonal D1R antibody (Santa Cruz, sc-31478) at 1:200 dilution, mouse monoclonal D2R antibody (Santa Cruz, sc-5303) at 1:500 dilution, rabbit polyclonal SERT antibody (Millipore Sigma, AB9706) at 1:500 dilution, or goat glyceraldehyde-3-phosphate dehydrogenase (GAPDH) antibody (Santa Cruz, sc-20357) at 1:2000 dilution in the blocking buffer. The membranes were washed with 0.1M PBS containing 0.1% Tween 20 4 times, for 5 min each, then treated for 1 hour with LI-COR IRDye 800CW donkey anti-goat IgG (H+L), LI-COR IRDye 800CW donkey anti-mouse IgG (H+L), or LI-COR IRDye 680RD donkey anti-rabbit IgG (H+L) at 1:15,556 dilution. After being washed four times with 0.1M PBS containing 0.1% Tween 20 for 5 min each and 0.1M PBS 3 times for 5 min each, the membranes were dried, and the signals were detected and quantified by a LI-COR Odyssey imaging system.

2.9. Data processing and statistical analysis

SAS statistical package was used. Total distance traveled in the continuous open field and the tail flick data were analyzed by logistic regression with the GEE model. The results of the GEE model were presented as relative changes between WT and triple KO, with WT normalized to 0 without the error bar. Probability of waking was analyzed by binomial logistic regression. Tail flick and metabolism data were analyzed by a mixed model with repeated measurement. Data obtained from 30 min open field test, colorimetric assays, striatal mineral concentration, complete blood test, and high-performance liquid chromatography (HPLC) were processed with mixed model ANOVA. Western blot data were analyzed with Student’s t-test. The p values in 30 min open field and monoamine analysis have been adjusted for multiple comparisons using the Benjamini-Hochberg-Yekutieli false discovery rate [FDR (p < 0.05)].

3. Results

3.1. Hyperactivity in the triple KO mice

To assess the diurnal total activity levels of the triple KO mice, we housed mice in open field activity chambers for consecutive seven days under normal 12 LD conditions. We found that the triple KO mice exhibited an increased total distance traveled compared with the WT mice only during the light-on phase, when the mice would normally be resting or sleeping (Fig. 1A, p<0.005), but not during the light-off phase (Fig. 1A, p=0.80). We also found that the mutant mice had a trend of increase in the probability of waking in the light-on phase (Figure 1B, left panel, p=0.19). The results indicate circadian rhythm-dependent hyperactivity in the triple KO that is compatible with RLS patients, whose symptoms mostly occur or worsen in the evening or at night [18]. We also measured the food and water consumption of these mice to determine their metabolic functions. There were no significant changes in food or water intake of the triple KO mice compared with the WT mice (Supplementary figure 1).

Figure 1.

Activity level and sensory tests. (A) Continuous open field showed that the total distance traveled by the triple KO mice was increased during the light-on phase, but not in the light-off phase. (B) The triple KO mice showed a trend of increased probability of waking in the continuous open field study. (C) The 30-min open field test indicated that the triple KO mice showed a significant increase in horizontal activity level. (D) The triple KO mice had a significant increase in vertical activity level in the 30-min open field test. (E) The stereotypic activity was increased in the triple KO mice in the 30-min open field test. (F) The triple KO mice showed no change in clockwise circling but did show an increase in counterclockwise circling in the 30-min open field test. (G) The triple KO mice had a decreased anxiety level in the 30-min open field test. ***p<0.005. (H) Tail flick test to determine sensory perception to warm stimuli. The triple KO mouse showed decreased pain sensitivity compared to the WT. Data in figures A, B, and H were processed with logistic regression with GEE model. The results were presented as relative differences, and WT group was normalized to 1 without error bar. *p<0.05.

Next, we assessed activity levels of the triple KO and the WT mice in a short-term open field test (30 min) during the daytime. We found that the triple KO mice had significant increases in both horizontal and vertical activities (Fig. 1C, D, p<0.005; Table 1). Furthermore, the triple KO mice showed an increased level of stereotypical behavior and a decreased level of anxiety (Fig. 1E, G, p<0.005; Table 1). It has been found that opioid receptors are involved in modulating neural processes essential to depressive-like and anxiety-related behaviors in different ways. For example, mu opioid receptor single KO mice show a decreased anxiety level while delta or kappa opioid receptor single KO mice have an increased anxiety level [19]. Finally, the triple KO mice had a significant increase in the counterclockwise (CCW) circling (Fig. 1F, left panel, p<0.005; Table 1), but not in the clockwise (CW) circling compared with the WT mice (Fig. 1F, right panel, p=0.053; Table 1), which resembles what we found in the Btbd9 KO model of RLS [15]. We will present the relevant data below. These data, taken together, indicate that there is an increase in activity levels of the triple KO mice only during the daytime, which is their rest period.

Table 1.

Open field activity parameters during 30-min short term test.

Parameters	WT	Triple KO	Adjusted p values
Horizontal activity	5,234 ± 469	8,609 ± 587	0.0003***
Total distance	3,381 ± 458	5,605 ± 574	0.005**
NO. of movement	306 ± 18	447 ± 22	0.0005***
Movement time	347 ± 37	547 ± 46	0.008**
Resting time	1,453 ± 37	1,253 ± 46	0.003***
Vertical activity	198 ± 36	609 ± 45	0.002***
NO. of vertical movement	91 ± 15	245 ± 19	0.0003***
Vertical time	88 ± 14	237 ± 18	0.0004***
Stereotypy activity	2,035 ± 286	3,968 ± 359	0.0004***
NO. of stereotypy bout	234 ± 10	292 ± 13	0.003***
Stereotypy time	236 ± 27	386 ± 34	0.002***
Clockwise revolution	13 ± 2	20 ± 3	0.053
Counterclockwise revolution	11 ± 3	24 ± 3	0.003***
Center time	200 ± 41	540 ± 51	0.0008***
CTRDIST:TOTDIST	0.18 ± 0.02	0.37 ± 0.03	0.0002***

Activity levels and total distance are presented in mean centimeters ± standard error of the mean (SEM). Times are presented as mean seconds ± SEM. NO., number of beam break count; CTRDIST, central distance; TOTDIST, total distance. WT, wild-type mice. The p values have been adjusted for multiple comparisons using the Benjamini-Hochberg-Yekutieli false discovery rate [FDR (p < 0.05)].

^**p<0.01

^***p<0.005.

3.2. Thermal sensory alterations in the triple KO mice

Patients with idiopathic and secondary human RLS show sensory abnormalities to pinprick and temperature-based modalities [20–22]. We assessed the triple KO mice for abnormalities in the sensory system using the tail-flick test. Interestingly, we found the triple KO mice had an increased latency to response to warm stimuli (Fig. 1H, p<0.05), which is opposite to what we found in the Btbd9 mutant mice, but consistent with thermal hypoaestheisa as observed in secondary RLS patients [20]. Opioid receptors are involved in the regulation of diverse nociceptive processes with different effects. For example, decreased tail withdrawal latencies in kappa opioid receptor KO females but increased latencies in the triple KO females have been reported [23].

3.3. Altered iron metabolism and blood profile in the triple KO mice

Iron deficiency is thought to be a potential reason for the development of RLS, with evidence showing that a lower iron level is correlated with symptoms in RLS [3]. Ferritin is an iron-storage protein [24]. Transferrin binds to iron and helps iron transportation [24]. Both are important proteins involved in the iron homeostasis [24]. We measured the iron levels in both the serum and the striatum to test whether the triple KO mice had an alteration in the iron homeostasis. We found that the triple KO mice had a significant decrease in the serum iron level (Figure 2A, p<0.05) and a trend of decrease in the serum ferritin level (Fig. 2B, p=0.07), but no significant changes in the transferrin level compared to the WT mice (Fig. 2C, p=0.65). With a complete blood count test, we further found significantly decreased levels of hemoglobin (HB) and hematocrit (HCT; Table 2) in the triple KO. Low HB and HCT may indicate a state of anemia. The result is consistent with the finding that people with iron-deficiency anemia are at higher risk of developing RLS [25]. However, we failed to find any mineral changes in the striatum (Table 3). These results suggest that there is a peripheral iron deficiency in the triple KO mice.

Figure 2.

Iron concentration in serum. (A) The triple KO mice showed a decreased level of iron in serum. (B) The triple KO mice had a trend of decrease in the serum ferritin level (p=0.07) but a normal level of transferrin level. *p<0.05.

Table 2.

Complete blood count (CBC) test.

CBC test Results	WT		Triple KO		p values
	Mean	SEM	Mean	SEM
RBC M/uL	10.68	±0.17	10.30	±0.14	0.10
HB g/dL	15.43	±0.23	14.62	±0.18	0.01*
HCT %	51.50	±0.82	49.11	±0.65	0.03*
PLT %	1,187.72	±863.54	2,045.18	±689.85	0.45
MCV fL	48.23	±0.44	47.70	±0.35	0.36
MCH pg	14.46	±0.10	14.20	±0.084	0.07
MCHC g/dL	29.98	±0.25	29.79	±0.20	0.56
RDW %	16.83	±0.19	17.22	±0.15	0.13
MPV fL	4.28	±0.077	4.41	±0.061	0.22
PDW %	22.23	±0.62	21.81	±0.50	0.61

The values of each parameter are presented in mean ± SEM. RBC, red blood cells; HB, hemoglobin; HCT, hematocrit; PLT, platelet count; MCV, mean corpuscular volume; MCH, mean corpuscular hemoglobin; MCHC, mean corpuscular hemoglobin concentration; RDW, red blood cell distribution width; MPV, mean platelet volume; PDW, platelet distribution width. M: million; K: thousand; WT, wild-type mice.

^*p<0.05.

Table 3.

Levels of minerals in the striatum.

Minerals	WT		Triple KO		Adjusted p values
	Mean	SEM	Mean	SEM
Iron	68	±4	74	±4	1.0
Cobalt	0.06	±0.003	0.07	±0.003	0.7
Copper	19	±1	20	±1	0.9
Manganese	1.70	±0.30	2.95	±0.33	0.2
Molybdenum	0.20	±0.01	0.22	±0.01	0.8
Selenium	1.06	±0.04	1.07	±0.04	0.8
Zinc	65	±3	68	±3	0.7

The values of neurochemicals represent means ± SEM in ug/g of dry tissue. The p values have been adjusted for multiple comparisons using the Benjamini-Hochberg-Yekutieli false discovery rate [FDR (p < 0.05)].

3.4. Altered dopaminergic metabolism in the triple KO mice

Levodopa and dopamine D₂/D₃ receptor (D₂/D₃R) agonists can be used as a treatment for RLS [2]. We analyzed the striatum of the triple KO mice using western blot for changes in dopamine receptors (D₁R, D₂R) and serotonin transporters (SERT), as well as HPLC for changes in dopamine (DA); serotonin (5-HT) and their metabolites 3,4-dihyroxyphenylacetic acid (DOPAC), homovanillic acid (HVA) and 5-hydroxyindoleacetic acid (5-HIAA). There were no significant changes in levels of D1R, D2R, SERT, DA, 5-HT, and their metabolites in the triple KO mice (Supplementary figure 2; Table 4). Then we calculated the DOPAC to DA ratio for each mouse. With mixed model ANOVA, we observed a decrease in DOPAC to DA ratio (Table 3, p<0.005). It is likely due to the trend of decreased DOPAC (p=0.23) and relatively increased DA (p=0.67) in the mutant mice. This suggests that while the gross levels of DA and DOPAC are not altered, DA metabolism is significantly decreased in the striatum of the triple KO mice.

Table 4.

Levels of dopamine, serotonin and their metabolites in the striatum.

Neurotransmitters and metabolites	WT		Triple KO		Adjusted p values
	Mean	SEM	Mean	SEM
NA	4.6	±3.0	8.9	±3.0	0.4
DOPAC	10.0	±0.6	9.0	±0.6	0.3
DA	140	±8	145	±7	0.7
5-HIAA	5.1	±0.3	6.2	±0.3	0.1
HVA	18	±0.9	18	±0.9	0.9
5-HT	15.7	±0.7	17.7	±0.7	0.3
3-MT	12.7	±0.9	14.6	±0.8	0.4
DOPAC/DA	0.071	±0.002	0.062	±0.001	0.003***
HVA/DA	0.130	±0.003	0.130	±0.003	0.4
5-HIAA/5-HT	0.32	±0.01	0.35	±0.01	0.2
3-MT/DA	0.092	±0.004	0.100	±0.004	0.3

The values of neurochemicals represent means ± SEM in ng/ul of tissue. The turnover of metabolites is shown as ratios of neurochemicals. NA, Norepinephrine; DOPAC, 3,4-Dihydroxyphenylacetic acid; DA, Dopamine; 5-HIAA, 5-Hydroxyindoleacetic acid; HVA, Homovanillic acid; 5-HT, Serotonin; 3-MT, 3-Methoxytyramine. WT, wild-type mice. The p values have been adjusted for multiple comparisons using the Benjamini-Hochberg-Yekutieli false discovery rate [FDR (p < 0.05)].

^***p<0.005.

4. Discussion

In this in vivo mouse model, a total knockout of all opioid receptor subtypes (mu, delta, kappa) resulted in increased activity at rest period, abnormalities in dopamine metabolism, serum iron deficiency and mild anemia analogous to that seen with human RLS [1]. It should be noticed that the triple KO mice may not be a complete animal model for RLS. Our previous autopsy studies of human RLS and in vitro cell culture study have shown a role of the endogenous opioid system in RLS and administration of opioids results in clinical improvement of RLS, leading to the registration of oxycodone/naloxone as the 2nd line treatment of RLS. Therefore, triple KO mice are the first valuable animal model of any kind in demonstrating that the endogenous opioid system may play a role in iron metabolism and thereby may contribute to the pathogenesis of anemia. The animal model revealed a potential mechanism for the clinical effectiveness of opioids, a non-dopaminergic agent, in treating RLS [26].

As this is a global knockout of all three opioid receptors, it is not possible to differentiate between central and peripheral mechanisms involved in the symptoms. It is, however, remarkable that the triple knockout model showed a peripheral iron deficiency as evidenced by low serum iron in both male and female triple knockout animals and no change in serum transferrin. This is in line with the common view that iron deficiency is thought to be the best-established chemical abnormality in RLS [3]. RLS patients show an iron deficiency in serum [27, 28] and cerebrospinal fluid (CSF) [29, 30]. Despite the significant and relevant changes in serum iron levels, triple KO mice showed no decrease in iron levels in the striatum. This appears to contrast reports showing that iron is lower in multiple brain regions like the substantia nigra (SN) [31–33] and was also described to be proportional to the severity of the disease [27, 31, 33]. Subsequently, brain iron deficiency was suggested to be a better hallmark for the diagnosis of RLS, which is not reflected in the current triple KO mice. However, none of the available RLS animal models showed brain iron deficiency except for the iron-deprived rodent [34]. Similarly, the 12 MRI imaging studies on brain iron measurement, one did not show any change, and the other showed an increase [35]. A recent case study reported iron accumulation in several brain regions, including the SN [36]. It is notable that iron-sensitive MRI methods have their limitations. For instance, R2 relaxation is affected by water content changes not related to iron, and R2*/R2′ has possible contaminations from the background field gradient [37]. Phase imaging or quantitative susceptibility mapping (QSM) provides more reliable iron quantification. Initial analysis of 15 RLS patients using phase imaging showed reduced iron content in SN, thalamus, putamen, and pallidum [38]. The results were not replicated with a large cohort of 39 RLS patients using QSM, and the researchers did not find iron changes in SN, putamen, and globus pallidus [39]. Considering the variability of iron content in the brain regions with the different methods, the decrease in iron levels may be a less stringent criterion for RLS or may indicate that there are different subsets of RLS patients.

Further evidence for RLS patient heterogeneity was also seen following iron supplementation. Administration of iron orally or intravenously significantly improves RLS symptoms in a subset of patients with low serum ferritin levels [40], but only about half of the patients responded to the iron supplementation [41]. The prevalence of clinically significant RLS is 23.9% in iron-deficient anemia (IDA) population, which is nine times higher than the general population [25]. On the other hand, the data show that three out of four IDA patients do not have RLS. Therefore, it is likely that RLS is a heterogeneous disorder with diverse etiology, usually but not always accompanied by iron deficiency.

A recent study showed that RLS patients with normal plasma iron levels revealed a mitochondrial iron deficiency in peripheral blood mononuclear cells (PBMCs) of RLS patients [42]. The reduced mitochondrial respiratory capacity and the intracellular iron content could be improved in vitro by dopaminergic treatment of PBMCs [42]. Since opioid receptors, especially the delta opioid receptor, participate in the regulation of mitochondrial structure and function [43, 44], it is tempting to speculate that the absence of opioid receptors may alter the function of mitochondria causing the iron deficiency and the activation of the cellular hypoxia-inducible factor (HIF) pathway [45, 46]. This is consistent with the finding that the endogenous opioid analog delta-opioid peptide prevents the destruction of dopaminergic systems caused by iron chelation in vitro [10].

Further evidence for a role of the opioid system in iron metabolism and storage comes from cell culture studies where the use of opioid receptor agonists DAMGO and morphine increased ferritin heavy chain levels in the cortical neurons [47]. Ferritin in animals is composed of 24 ferritin light chain (FLC) and ferritin heavy chain (FHC) in ratios that vary in different cell types, and their functions are not interchangeable [48]. The application of opioid agonists interferes the balance between FLC and FHC, therefore may induce functional alterations of ferritin. Additionally, the same study also found an increase in the FHC protein level in the cortex of morphine-treated animals [47]. A preliminary study in postmortem human brain tissue showed an elevation in the numbers of FHC-positive cells in the frontal cortex of opiate drug abusers [49]. Furthermore, male opiate drug abusers have higher levels of ferritin and a higher hematological index value in their blood samples compared with the control group [50]. A separate study indicates that patients with non-insulin-dependent diabetes (NIDDM) who are addicted to opium show higher serum iron levels compared to NIDDM patients who do not have an addition to the drug [51]. Heroin and opium dependent people also have increased HCT, MCV and MCHC levels [52]. These results are consistent with our observation that mice lacking opioid receptors showed significantly decreased HB and HCT.

In addition to the role of opioid receptors and iron handling and storage, there are also interactions of the dopamine and opioid system that may be relevant for RLS-like phenotype in the triple knockout mice. Although the levels of DA or DA metabolites did not significantly change in the striatum of the triple KO mice, the ratio of DOPAC to DA was significantly decreased, which suggests a decreased striatal DA metabolism. Opioid receptors are extensively expressed in the striatum and DA terminals [53]. Previous studies show that regional administration of DAMGO leads to a reduction of DA release in the rostral and caudal dorsal striatum (DS), but an enhancement in the medial DS [54]. Furthermore, activation of mu-opioid receptors (MORs) on striatal cholinergic interneurons robustly suppresses its spontaneous firing [55], which in turn influences DA release through nicotinic receptor-mediated facilitation [56]. Also, activation of kappa opioid receptors (KORs) on the nigrostriatal afferents from the SN to the striatum inhibits DA release and boosts the activity of DAT [57, 58]. In the absence of opioid receptors, the triple KO mice lost the regulation of endogenous opioids in the striatal dopaminergic circuit, therefore might have activated compensatory mechanisms that turn down the DA release and metabolism.

Dopamine agonists are one of the primary treatments for RLS patients. It has been found that RLS patients exhibit abnormal dopaminergic profiles in the putamen and the SN; in addition, changed levels of dopamine metabolites have been found in the CSF [59–61]. Moreover, significantly decreased D2 receptor (D2R) expression in putamen [61] and D2R striatal binding potential have been observed [62, 63]. Finally, RLS patients show an altered level of dopamine transporter (DAT) in the striatum compared to controls [64, 65]. These data, taken together, indicate a dysfunction in the dopaminergic system, caused by the “hyperdopaminergic” presynaptic state, the “hypodopaminergic” postsynaptic state, or both [2].

The dopaminergic system interacts with glutamatergic and adenosine systems in the striatum [6]. In addition to changes in the dopaminergic system, clinical studies also indicate a presynaptic hyperglutamatergic state in RLS [66]. Mammalian cell line exposed to an iron chelator has increased level of adenosine A2A receptor (A2AR) [67]. Rats and rodents with severe BID showed a consistent upregulation of striatal A2AR [67–69], and a pronounced downregulation of A1R both in the striatum and in the cortex [69]. A1R forms heteromer with D1R, while A2AR largely coexpresses with D2Rs and enkephalin mRNA in the striatum where it modulates dopaminergic activity [70]. In addition, MORs and delta opioid receptors (DORs) inhibit specific striatal inputs from the thalamus and cerebral cortex, respectively [71]. Therefore, the knockout of opioid receptors may lead to increased striatal glutamatergic inputs. Moreover, it has been found that morphine causes a significant decrease of adenosine levels in some brain regions known to regulate states of arousal [72]. Hence, the lack of opioid receptors may result in enhanced adenosine activity. Dysfunctional glutamatergic and adenosine systems may in turn, influence the dopaminergic system [4].

RLS patients have a characteristic circadian rhythm-involved symptom. They usually have a strong urge to move the legs, especially at night and at rest, which can be relieved by movement [18]. The “urge to move” is difficult to model in animals. Previous models of RLS, therefore, mainly focused on testing the motor restlessness or alterations in sleep efficiency. For example, A11 dopaminergic nucleus lesioned rats demonstrated alterations in the standing episodes and the total standing time [73], as well as increased frequencies of limb movement during both the light and the dark periods [74] compared with the sham ones. Heterozygous Meis1 mutant mice showed hyperlocomotion in both a 30 min open field test and an open field test lasting for 7 days [13, 75]. Homozygous Btbd9 KO mice were hyperactive in a 30 min open field test, with a trend of increase in wheel running (WR) counts during the rest phase and a significant increase in activity level when placed in constant darkness (DD) [15]. The BXD 40 females with iron deficiency showed increased activity in the last part of the active phase and the first part of the rest phase compared with other BXD strains [12]. The D3KO mice displayed significantly increased daily locomotor activities than their WT controls [13]. There lacks evidence showing that the hyperactivity observed in these animal models is restricted to the rest or inactive time only. The triple KO mice, however, showed clear diurnal hyperactivity, with the increased total distance traveled appearing specifically during the rest phase, but not the active phase.

Lastly, RLS patients show hypersensitivity to a pinprick as well as tactile hypoesthesia and dysesthesia to non-noxious cold stimuli (paradoxical heat sensation) [21, 22]. Also, RLS patients show hyperalgesia to blunt pressure and hyperaesthesia to vibration [20]. These somatosensory abnormalities of RLS patients are thought to be related to an altered central sensitization and likely caused by a maladaptive sensory-motor cortical plasticity [76]. With repetitive low-frequency transcranial magnetic stimulation on the sensory-motor network, RLS patients reported subjective improvement of sensory discomforts [77]. Our triple KO mice showed sensory changes to heat in that they showed an increased latency to warm stimuli, which provide a potential animal model for further testing of the sensory-motor network.

There are limitations in the current study. We did not have periodic limb movement assessment. The link between the peripheral iron deficiency and central alterations of DA metabolism is not clear. Lack of all three opioid receptors does not allow allocation to specific receptor subtype. Furthermore, substitution studies with opioids to analyze possible restoration of the defects in iron metabolism, DA release, and DA metabolism will need to await similar studies on models where endogenous opioid peptides are lacking.

5. Conclusions

Mice lacking mu, delta, and kappa opioid receptors exhibit hyperactivity specifically during the rest phase akin to that seen in human RLS. In addition, these mice had iron-deficiency anemia and dopaminergic dysfunction which is also seen in human RLS. Sensory abnormalities were noted similarly akin to those in human RLS. The mouse model will be useful to study the relationship among endogenous opioid system, iron deficiency, anemia, dopaminergic dysfunction, and the pathogenesis of RLS. Additionally, because opioid receptors KO mice, including triple KOs, usually are used in opioid tolerance and addiction studies, these results may be of interest, especially for behavioral studies in various diseases beyond RLS.

Highlights.

• Triple opioid receptor KO mice showed peripheral iron deficiency and anemia.

• Triple opioid receptor KO mice showed diurnal hyperactivity.

• Triple opioid receptor KO mice had decreased thermal sensitivity.

• Triple opioid receptor KO mice showed decreased dopamine metabolism.

Abbreviations

3-MT

3-methoxytyramine

5-HIAA

5-hydroxyindoleacetic acid

5-HT

serotonin

12 LD

12-hour light, 12-hour dark cycle condition

BID

brain iron deficiency

CBC

complete blood count

CCW

counterclockwise circling

CW

clockwise circling

D₁R

D1 dopamine receptors

D₂R

D2 dopamine receptors

DA

dopamine

DOPAC

3,4-dihyroxyphenylacetic acid

HB

hemoglobin

HCT

hematocrit

HPLC

High-performance liquid chromatography

HVA

homovanillic acid

MCH

mean corpuscular hemoglobin

MCHC

mean corpuscular hemoglobin concentration

MCV

mean corpuscular volume

MOR

mu opioid receptor

MPV

mean platelet volume

NA

norepinephrine

PDW

platelet distribution width

PLT

platelet count

RBC

red blood cell

RDW

red blood cell distribution width

RLS

restless legs syndrome

SERT

serotonin transporter

SN

substantia nigra

Triple KO

mu, delta, and kappa opioid receptor knockout mouse

WT

wildtype mice