**2.3 Osmotic minipump infusions**

278 Neuroscience – Dealing with Frontiers

intraventricular administration of a GABAA receptor agonist (Soriano *et al.* 1997, Yamada *et al.* 1996); SNc TH mRNA is elevated followed systemic administration of nicotine or α-7 nicotinic acetylcholine receptor agonists (Serova & Sabban 2002); and we recently reported that the number of SNc TH+ cells in adult mice can be increased or decreased by direct brain infusions of drugs targeting SNc neuronal activity (Aumann *et al.* 2011, Aumann *et al.* 2008). These data are consistent with the possibility that the DA phenotype of adult SNc neurons

Understanding how the DA phenotype of neurons in the adult SNc is regulated will highlight strategies and molecular (drug) targets to facilitate DA phenotype acquisition and maintenance by either transplanted or endogenous cells. The aims of the experiments described in this report were to determine whether infusions of drugs that increase the number of SNc DA (TH+) cells in normal mice: (1) increase the number of these cells in a 6-

All experimental procedures on animals were approved by the Florey Neuroscience Institutes animal ethics committee and conform to the National Health and Medical

Male C57Bl/6J mice and male Sprague-Dawley rats were used for this study. Throughout the experiments all animals were housed in a climate-controlled (22oC) room on a 12/12hour light/dark cycle with ad libitum access to food (standard rodent chow) and

Unilateral 6-OHDA lesions were made in the mice and rats when they were 8-weeks old. We aimed for partial (40-60%) loss of SNc DA neurons because another form of compensatory plasticity (i.e. spouting of surviving nigrostriatal projections) in the nigrostriatal system is hampered with lesions >75% (Stanic et al. 2003). Mice were anesthetized with an i.p. injection of 5% chloral-hydrate in sterile phosphate buffered saline, whereas rats were anesthetized with 1-2% isofluorane in air. Their heads were secured in a stereotaxic frame and a midline incision made over the skull. A small (2-3mm diameter) hole was drilled through the skull overlying the left (mice) or right (rats) SNc with a dental burr. In mice, 1.5µg/µl 6-OHDA (Sigma-Aldrich) in distilled H2O with 0.2mg/ml ascorbic acid was prepared on the morning of the lesions and kept on ice until injected to minimize oxidation. 1.6µl of the 6-OHDA solution was injected slowly (1.0µl/min.) through a 26gauge sterile needle at stereotaxic coordinates: 3.0mm posterior to Bregma, 1.5mm lateral to the midline, 4.0mm deep. In rats, two 1.0µl injections of 2.0µg/µl 6-OHDA in distilled H2O with 0.2mg/ml ascorbic acid were made through a glass micropipette at stereotaxic coordinates: (1) 3.7mm anterior to Lambda, 1.7mm lateral to the midline, 8.1mm deep; and (2) 3.7mm anterior to Lambda, 2.1mm lateral to the midline, 7.5mm deep. At the completion of each injection the injection needle was left in situ for 2mins. to allow toxin diffusion then the needle was slowly withdrawn to minimize toxin backtracking up the needle track. The skin

can be gained or lost in an activity-dependent way.

OHDA rodent model of PD; and (2) restore motor deficits.

Research Council of Australia's published code of practice.

**2. Methods 2.1 Animals** 

water.

**2.2 6-OHDA lesion** 

Some weeks following the 6-OHDA lesion, when SNc TH+ cell loss was maximal, a cannula [ALZET® (Cupertino, CA, USA) or PlasticsOne® (Roanoke, VA, USA)] was implanted into the midbrain or striatum on the same side as the lesion, through which a drug (or vehicle) was infused continuously via an osmotic pump (ALZET® model #1002 in mice & model #2006 in rats) for a period of time (see Results section for specific timings). To prime the pumps prior to implantation they were filled with drug (or vehicle) and immersed in 37oC sterile saline overnight. The next day animals were anaesthetized and prepared as described above. The cannula was implanted at the following stereotaxic coordinates: (1) For SNc infusions in mice 3.0mm posterior to Bregma, 1.5mm lateral to the midline, 4.0mm deep; (2) For SNc infusions in rats 3.7mm anterior to Lambda, 1.9mm lateral to the midline, 7.8mm deep; (3) For striatal infusions in rats 0.5mm anterior to Bregma, 3.0mm lateral, 3.5mm deep. The cannula was glued to the skull with dental cement and the attached pump was placed in a subcutaneous 'pocket' created in the interscapular region.

In rat experiments the period of drug delivery was extended beyond the lifetime (6 weeks) of the first implanted pump by replacing it with a new filled pump. This was done by anaesthetizing the rat with 1-2% isofluorane in air, making a small incision in the skin overlying the old pump, clamping the vinyl tube closed, removing the old pump (and flow moderator), replacing it with a new primed pump (and new flow moderator), releasing the clamp and suturing the skin. The implanted cannula remained undisturbed throughout this procedure.

#### **2.4 Behavior**

The behavioral studies described below were performed on cohorts of animals typically comprising 12 animals, 6 treated with drug and 6 with vehicle. Each behavioral test was performed at the same time of day within each cohort and each animal was studied concurrently. If availability of equipment prevented this, animals were studied consecutively in a different pseudo-randomized order and equipment was thoroughly cleaned with 80% ethanol after each animal to remove any distracting odors.

#### **2.4.1 Rotational response to amphetamine**

Animals were injected intraperitonealy with 5mg/kg amphetamine and immediately placed into a cylindrical chamber (17cm diameter by 17cm high for mice & 31cm diameter by 31cm high for rats). In some rat experiments rotational behavior was also assessed for 20mins. prior to amphetamine. The behavior of each animal in its chamber was recorded on videotape for at least 1hour following amphetamine. On days where amphetamine was administered, no other behavioral tests were performed.

Rotational behavior was measured off-line using Ethovision® XT animal tracking software (Noldus Information Technology, Wageningen, Netherlands). One rotation was defined as a cumulative 360o change in heading direction of the animal's centre of mass, with no time

Activity-Dependent Regulation of the Dopamine

until immunohistochemical processing.

polystyrene mounting medium.

**2.6 Stereology** 

mouse (1:600, Chemicon). No counterstain was performed.

stereologist without prior knowledge of the treatment received.

Phenotype in the Adult Substantia Nigra: Prospects for Treating Parkinson's Disease 281

slides coated with 0.1% chrome alum and 1% gelatin in distilled H2O and stored at -80oC

For TH immunohistochemistry, mounted sections were post-fixed in 10% neutral buffered formalin (5min.), incubated for 10min. in blocking solution (0.1M PBS, 0.3% Triton X-100 & 3% normal goat serum), then for 72hours at 4oC in rabbit anti-TH primary antibody (1:1500, Chemicon, Temecula, CA, USA) with 0.3% Triton X-100 and 1% normal goat serum in PBS. This was followed by 2hours incubation at room temperature in a biotinylated secondary antibody (1:1000, goat anti-rabbit IgG, Dako, Denmark). Next, sections were incubated for 1hour in 0.02% avidin peroxidase with 0.75% Triton X-100 in PBS at room temperature, then in cobalt- and nickel-intensified diaminobenzidine (DAB) for 20mins., then 3% hydrogen peroxide was added to the DAB solution for a further 2mins. to complete the chromagen reaction. Rinses (3 x 5min. each) in PBS were performed between each step. Sections were counterstained with 1% neutral red for 8min., washed in H2O, dehydrated in a series of graded ethanol solutions (50-100%) and cleared in X3B before being coverslipped with a

For DAT immunohistochemistry the protocol was the same as for TH immunohistochemistry with the following exceptions. Prior to incubation in blocking solution an antigen retrieval step was performed in which the sections were heated in 0.2% citrate buffer (pH 6) and allowed to cool at room temperature for 30min. Sections were incubated in a blocking solution (5% normal goat serum, 0.3% Triton X-100 in 0.1M PBS) for 10min. The primary antibody was mouse anti-DAT (1:3000, Chemicon) and the biotinylated secondary antibody was sheep anti-

To estimate the numbers of TH+ and TH- neurons in the SNc following 6-OHDA lesion and drug treatment, a fractionator sampling design was used, as previously described in detail (Parish *et al.* 2001), using a stereology program (Stereo Investigator, MicroBrightField, VT, USA) attached to a microscope. The SNc was delineated at its anatomical boundaries with reference to rodent brain atlases (Franklin & Paxinos 2008, Paxinos & Watson 2007) and based on high cell-packing density, cell morphology, and TH immunoreactivity (Nelson *et al.* 1996). Counts of TH+ and TH- SNc neurons (glia were excluded on the basis of soma diameter <5µm) within a counting frame (45µm x 35µm) were made at regular pre-determined intervals (x=145µm, y=145µm) using a 60x oilimmersion objective lens. The cell nucleus was the counting unit. TH+ cells were immunoreactive against TH and TH- cells were not immunoreactive against TH but counterstained. Every 5th section of the series was analyzed (i.e. 80µm apart) and the volume of the SNc was estimated according to Cavalieri's method (Gundersen *et al.* 1988), which estimates on the basis of area, section thickness and distance between sections. Cells on the treated and contralateral (internal control) sides of each brain were always counted consecutively to ensure consistency of cell classifications. Cell count in the lesioned and treated SNc was expressed as a proportion of cell count in the contralateral internal control SNc in each animal. Cell counts were performed by an experienced SNc

limit. If an animal interrupted a turn in one direction with a turn in the other of more than 90o, the cumulative change in the first direction was reset to zero.

#### **2.4.2 Locomotor cells**

Locomotion was measured using photo-optic locomotor cells (Truscan Photobeam; Coulbourn Instruments, Allentown, PA, USA), which automatically detect horizontal- (locomotor) and vertical-plane (rearing) movements. Cell dimensions were 42cm wide x 42cm deep x 38cm high. Each rat was placed into a locomotor cell (6 rats at a time) for 30mins. in a quiet room with low-level light.

#### **2.4.3 Cylinder test**

Lateralized forelimb movement deficits were assessed using the cylinder test. This test was always performed in conjunction with the locomotor cells – half the cohort was run in the locomotor cells while the other half was run in the cylinder test, then vice verse (in a different pseudo-randomized order each time). Each rat was placed in a clear plastic cylinder (19cm diameter and 26cm high) standing vertically on a bench (1 rat at a time). A video camera recorded behavior for 10mins. in a quiet room with low-level light. Whilst in the cylinder the rats reared up to explore the only route of escape, supporting themselves against the cylinder wall with their left and/or right forelimbs. Two mirrors were positioned behind the cylinder to observe forelimb movements obscured by the rat's body. The number of times each forelimb touched the wall of the cylinder over the 10mins. was counted by an observer without prior knowledge of the treatment received.

#### **2.4.4 Corridor test**

Lateralized motor or attention deficits were assessed using the corridor test. No other test was performed on the day of the corridor test. Three days prior to the corridor test, rats were food-deprived by rationing chow at the rate of 2.5g/100g body weight. Two rats were run concurrently (in a different pseudo-randomized order each time). Each rat was placed in a corridor measuring 240cm long by 7cm wide by 21cm high, enclosed at each end but open at the top. Food (chocolate rice-puff cereal) was available *ad libitum* at multiple, evenly spaced points (13cm apart) along the left and right sides of the corridor. A video camera recorded their behavior for 10mins. in a quiet room with low-level light. Whilst in the corridor the rats explored freely along its length including sniffing and eating the cereal at their leisure. The number of times each rat sniffed at or ate cereal on their left and right sides was recorded by an observer without prior knowledge of the treatment received.

#### **2.5 Immunohistochemistry**

At the end of the experiment animals were killed with an overdose of anesthetic (sodium pentobarbitone, 100mg/kg, i.p.). Before their heart stopped beating the animals were perfused transcardially with warm (37oC) heparinized (0.1%) 0.1M phosphate buffered saline (PBS) followed by cold (4oC) 4% paraformaldehyde plus 0.2% picric acid in 0.1M phosphate buffer (PB). The brains were removed and placed at 4oC in PBS with 20% sucrose. Once equilibrated with 20% sucrose in PBS, the brains were frozen and 16µm thick coronal sections were cut through the striatum and SNc. Sections were collected directly onto glass slides coated with 0.1% chrome alum and 1% gelatin in distilled H2O and stored at -80oC until immunohistochemical processing.

For TH immunohistochemistry, mounted sections were post-fixed in 10% neutral buffered formalin (5min.), incubated for 10min. in blocking solution (0.1M PBS, 0.3% Triton X-100 & 3% normal goat serum), then for 72hours at 4oC in rabbit anti-TH primary antibody (1:1500, Chemicon, Temecula, CA, USA) with 0.3% Triton X-100 and 1% normal goat serum in PBS. This was followed by 2hours incubation at room temperature in a biotinylated secondary antibody (1:1000, goat anti-rabbit IgG, Dako, Denmark). Next, sections were incubated for 1hour in 0.02% avidin peroxidase with 0.75% Triton X-100 in PBS at room temperature, then in cobalt- and nickel-intensified diaminobenzidine (DAB) for 20mins., then 3% hydrogen peroxide was added to the DAB solution for a further 2mins. to complete the chromagen reaction. Rinses (3 x 5min. each) in PBS were performed between each step. Sections were counterstained with 1% neutral red for 8min., washed in H2O, dehydrated in a series of graded ethanol solutions (50-100%) and cleared in X3B before being coverslipped with a polystyrene mounting medium.

For DAT immunohistochemistry the protocol was the same as for TH immunohistochemistry with the following exceptions. Prior to incubation in blocking solution an antigen retrieval step was performed in which the sections were heated in 0.2% citrate buffer (pH 6) and allowed to cool at room temperature for 30min. Sections were incubated in a blocking solution (5% normal goat serum, 0.3% Triton X-100 in 0.1M PBS) for 10min. The primary antibody was mouse anti-DAT (1:3000, Chemicon) and the biotinylated secondary antibody was sheep antimouse (1:600, Chemicon). No counterstain was performed.

#### **2.6 Stereology**

280 Neuroscience – Dealing with Frontiers

limit. If an animal interrupted a turn in one direction with a turn in the other of more than

Locomotion was measured using photo-optic locomotor cells (Truscan Photobeam; Coulbourn Instruments, Allentown, PA, USA), which automatically detect horizontal- (locomotor) and vertical-plane (rearing) movements. Cell dimensions were 42cm wide x 42cm deep x 38cm high. Each rat was placed into a locomotor cell (6 rats at a time) for

Lateralized forelimb movement deficits were assessed using the cylinder test. This test was always performed in conjunction with the locomotor cells – half the cohort was run in the locomotor cells while the other half was run in the cylinder test, then vice verse (in a different pseudo-randomized order each time). Each rat was placed in a clear plastic cylinder (19cm diameter and 26cm high) standing vertically on a bench (1 rat at a time). A video camera recorded behavior for 10mins. in a quiet room with low-level light. Whilst in the cylinder the rats reared up to explore the only route of escape, supporting themselves against the cylinder wall with their left and/or right forelimbs. Two mirrors were positioned behind the cylinder to observe forelimb movements obscured by the rat's body. The number of times each forelimb touched the wall of the cylinder over the 10mins. was

Lateralized motor or attention deficits were assessed using the corridor test. No other test was performed on the day of the corridor test. Three days prior to the corridor test, rats were food-deprived by rationing chow at the rate of 2.5g/100g body weight. Two rats were run concurrently (in a different pseudo-randomized order each time). Each rat was placed in a corridor measuring 240cm long by 7cm wide by 21cm high, enclosed at each end but open at the top. Food (chocolate rice-puff cereal) was available *ad libitum* at multiple, evenly spaced points (13cm apart) along the left and right sides of the corridor. A video camera recorded their behavior for 10mins. in a quiet room with low-level light. Whilst in the corridor the rats explored freely along its length including sniffing and eating the cereal at their leisure. The number of times each rat sniffed at or ate cereal on their left and right sides

At the end of the experiment animals were killed with an overdose of anesthetic (sodium pentobarbitone, 100mg/kg, i.p.). Before their heart stopped beating the animals were perfused transcardially with warm (37oC) heparinized (0.1%) 0.1M phosphate buffered saline (PBS) followed by cold (4oC) 4% paraformaldehyde plus 0.2% picric acid in 0.1M phosphate buffer (PB). The brains were removed and placed at 4oC in PBS with 20% sucrose. Once equilibrated with 20% sucrose in PBS, the brains were frozen and 16µm thick coronal sections were cut through the striatum and SNc. Sections were collected directly onto glass

counted by an observer without prior knowledge of the treatment received.

was recorded by an observer without prior knowledge of the treatment received.

90o, the cumulative change in the first direction was reset to zero.

**2.4.2 Locomotor cells** 

**2.4.3 Cylinder test** 

**2.4.4 Corridor test** 

**2.5 Immunohistochemistry** 

30mins. in a quiet room with low-level light.

To estimate the numbers of TH+ and TH- neurons in the SNc following 6-OHDA lesion and drug treatment, a fractionator sampling design was used, as previously described in detail (Parish *et al.* 2001), using a stereology program (Stereo Investigator, MicroBrightField, VT, USA) attached to a microscope. The SNc was delineated at its anatomical boundaries with reference to rodent brain atlases (Franklin & Paxinos 2008, Paxinos & Watson 2007) and based on high cell-packing density, cell morphology, and TH immunoreactivity (Nelson *et al.* 1996). Counts of TH+ and TH- SNc neurons (glia were excluded on the basis of soma diameter <5µm) within a counting frame (45µm x 35µm) were made at regular pre-determined intervals (x=145µm, y=145µm) using a 60x oilimmersion objective lens. The cell nucleus was the counting unit. TH+ cells were immunoreactive against TH and TH- cells were not immunoreactive against TH but counterstained. Every 5th section of the series was analyzed (i.e. 80µm apart) and the volume of the SNc was estimated according to Cavalieri's method (Gundersen *et al.* 1988), which estimates on the basis of area, section thickness and distance between sections. Cells on the treated and contralateral (internal control) sides of each brain were always counted consecutively to ensure consistency of cell classifications. Cell count in the lesioned and treated SNc was expressed as a proportion of cell count in the contralateral internal control SNc in each animal. Cell counts were performed by an experienced SNc stereologist without prior knowledge of the treatment received.

Activity-Dependent Regulation of the Dopamine

Phenotype in the Adult Substantia Nigra: Prospects for Treating Parkinson's Disease 283

Fig. 1. Effect of the SK channel agonist 1-EBIO on the number of TH+ SNc cells in 6-OHDAlesioned adult mice. 6-OHDA was injected into the left SNc of 8-week old male C57Bl/6J mice to reduce the number of SNc TH+ cells to ~50% of normal (see vehicle-treated mice, black symbols). Two weeks later a cannula was implanted into the left SNc, through which 1-EBIO or vehicle was infused (from an attached minipump) continuously for a further 2 weeks. The number of TH+ cells in the left and right SNc of each mouse was estimated using unbiased stereology by an observer who was unaware of the treatment received. The data for each mouse (filled squares) are expressed as the ratio of the number of TH+ SNc cells on the lesioned and treated (left) side relative to the internal control (right side). The mean of each treatment group is represented by a horizontal line. Lesion and vehicle-treated mice (n=15, black symbols) have a ~50% reduction, on average, in the number of TH+ SNc cells. Treatment with 100μM 1-EBIO (n=9, pale red symbols) appears to have had a

beneficial effect (see text) in some mice, but there was no change on average. Treatment with 200μM 1-EBIO (n=10, red symbols) appears to have had an even stronger beneficial effect in some mice, resulting in an average increase in SNc TH+ cells to ~80% of normal. Note, however, that this increase was not statistically significant (p=0.068, one-way ANOVA).

We tested the neuro-restorative effects of orally administered riluzole on motor symptoms and TH+ SNc cell number in a mouse PD model. Three weeks after 6-OHDA SNc lesion, mice were administered riluzole in their drinking water for 2 weeks. Both 3μM and 30μM riluzole doses were tested. Note these were the projected daily systemic concentrations based on the average daily volume of drinking water consumed and body-weight of each mouse. Every 3 days following the lesion and throughout the treatment period, the rotational behavior of the mice in response to amphetamine (5mg/kg i.p.) was measured (figure 2A & A'). Following 6-OHDA lesion and before treatment began (figure 2A), the relative number of ipsiversive rotations (rotations toward the side of the lesion) increased progressively to reach a maximum by 12 days [p=0.035 (time), two-way ANOVA]. Presumably this reflects gradual depletion of DA in the left striatum caused by progressive degeneration of DA neurons in the left SNc. The relative number of contraversive rotations
