Developing a Rat Model for Bipolar Disorder

Summary

This article presents a protocol for the induction of a unique rat model of bipolar disorder that captures both mania-like and depressive-like behavior.

Abstract

Bipolar disorder is a mental health condition characterized by extreme mood swings, including periods of emotional highs (mania) and lows (depression). While the exact underlying neurobiology is not yet fully understood, imbalances in neurotransmitter systems, particularly dopamine, appear to play a central role. For this reason, manipulations of dopaminergic pathways have been used to model mania or depression in rodents. However, models that accurately represent the typical switch between these two episodes are rare, limiting face validity. In a unique model, modern techniques are used to temporarily increase dopamine D1 receptor expression, which has been implicated in the pathology of bipolar disorder. A tetracycline-inducible lentiviral construct that expresses the dopamine D1 receptor under the control of the calmodulin kinase II alpha promoter is stereotactically injected into the medial prefrontal cortex of adult rats. Dopamine D1 receptor overexpression is achieved by adding the tetracycline analog doxycycline to the animals' drinking water, leading to an increase in reward-related, impulsive, and risk-taking behaviors and a decrease in anxiety. These behaviors resemble a mania-like phenotype. By removing doxycycline from the drinking water, a depressive-like phenotype, characterized by increased helplessness and anhedonia, can be induced within the same animal. This article provides a step-by-step protocol for performing the surgery, as well as procedures for inducing the bipolar disorder-like phenotype. Additionally, considerations for assessing behavioral changes associated with mania-like and depressive-like behavior are described. This promising model, which demonstrates good construct and face validity, offers a valuable tool for further investigating the pathophysiological mechanisms of bipolar disorder.

Introduction

Bipolar disorder (BD) is a severe mood disorder affecting around 1% of the world´s population¹. It is characterized by episodes of extreme mood, depression and mania, alongside euthymic states. Symptoms of depressive episodes in BD resemble those of unipolar depression. Patients show reduced interest and pleasure in activities and feelings of sadness, hopelessness, and worthlessness. Additionally, changes in appetite, sleeping behavior, as well as cognitive impairments can often be observed². Manic episodes are characterized by abnormally heightened mood, decreased need for sleep, social disinhibition, increased self-esteem, and feelings of grandiosity, as well as heightened risk-taking and irritability².

The disease etiology of BD seems to be a complex interplay of genetic and developmental factors³, but the exact mechanisms involved in its pathophysiology are still not fully understood. Symptoms are thought to arise from imbalances in neurotransmitter systems⁴ and, particularly, studies focusing on the dopamine system have been influential⁵. For example, Berk et al.⁶ postulated the dopamine hypothesis, assuming that a hyperdopaminergic state underlies mania, while depression arises from hypodopaminergia. Since then, evidence from animal models, as well as pharmacological and imaging studies, have gathered strong support for an association between manic symptoms and hyperdopaminergia. Also, a connection between decreased dopaminergic signaling and depressed episodes could be found, although to a lesser extent⁷. In addition, results from genetic investigations have strengthened the idea of a dopamine hypothesis of BD⁸.

To further shed light on the role of the dopamine system in BD, animal models can be used to investigate the neurobiological mechanisms that underlie symptoms. Applications and limitations of disease models are often assessed based on three validation criteria, originally proposed by Willner⁹. These include face-, construct-, and predictive validity. Face validity describes the ability of the model to mimic the behavioral characteristics of the disorder. Construct validity is reached when the pathophysiology and etiology of the disorder are the basis of the model, while predictive validity implies that pharmacological treatment of the disorder can be reproduced within the model.

So far, various rodent models have contributed to an understanding of BD¹⁰ and include a wide range of genetic modifications, pharmaceutical interventions, as well as environmental manipulations¹¹.

Experimental manipulations of the Clock gene, e.g., have been shown to induce a mania-like phenotype in mice. The transcription factor CLOCK plays an important role in regulating circadian rhythms, and genetically altered mice, expressing a protein that cannot activate Clock transcription, are characterized by hyperactivity and increased reward responses¹². The resulting phenotype seems to be mediated by differentially regulated genes for dopaminergic signaling in the brain's ventral tegmental area¹³.

Directly influencing dopamine signaling via the administration of dopamine-increasing drugs, such as the psychostimulant amphetamine, has been shown to induce hyperlocomotion, and subsequent withdrawal has been linked to depressive-like symptoms, including anhedonia¹⁴. Pharmacological challenges with ketamine or the dopamine D2/D3 receptor agonist quinpirole have also been shown to induce behavior relevant to BD^15,16.

In addition to pharmacological intervention, manipulations of the environment, such as sleep deprivation, can be employed to induce behavioral phenotypes relevant to BD¹⁷. Sleep-deprived animals show a mania-like phenotype characterized by increased locomotion and emission of ultrasonic vocalizations that are associated with changes in dopamine signaling¹⁸.

There are numerous other rodent models to study depressive¹⁹ or mania-like²⁰ behavior. However, while all these models have strongly contributed to an understanding of BD pathology, they are limited by studying only one episode at a time or short-term effects. In contrast, modeling the characteristic switch between affective states has been difficult to achieve.

Here, a protocol for a unique rat model for BD is presented. It demonstrates increased face validity by inducing both episodes in one animal using a single targeted manipulation of the dopamine system, i.e., conditionally overexpressing the dopamine D1 receptor (DRD1) in the medial prefrontal cortex (mPFC) from a tetracycline-inducible lentiviral construct. By driving gene transcription under the control of the calmodulin kinase II alpha (CamKIIa) promoter, DRD1 is mainly expressed in glutamatergic neurons, thus increasing the specificity of the genetic manipulation.

The original lentivirus backbone pRRL.cPPT.WPRE.Sin was provided by Dr. Didier Trono (Ecole Polytechnique Fédérale de Lausanne, Switzerland)²¹ and modified by replacing the GFP minigene with a polylinker site (lentivirus vector PL13). PL13 was then used to produce PL13.pTRE2.DRD1.CamKIIa.rtTA3 or PL13.pTRE2.dsRedExpress.CamKIIa.rtTA3. The cDNA of rat DRD1 was obtained from Dr. David Sibley (NINDS/NIH)²², and the reverse tetracycline-controlled activator 3 (rtTA3) cDNA from Drs. Atze Das and Ben Berkhout (Academic Medical Center, University of Amsterdam)²³. The CamKIIa promoter DNA was provided by Dr. Karl Deisseroth (Stanford University, CA), and the dsRedExpress and the tetracycline response element 2 (pTRE2) sequences were subcloned from inhouse plasmids pcDNA3.1-dsRedExpress and pcDNA3.1-pTRE2, respectively. Viral vectors were generated by subcloning PCR-amplified DNA sequences flanked by restriction sites.

The model using this viral vector has demonstrated that overexpression of DRD1 in mPFC CamKIIa-positive neurons leads to a mania-like phenotype^24,25, while the subsequent downregulation of gene expression induces depressive-like behavior²⁶. Since the disease-like phenotype can be repeatedly induced in one animal²⁷, the model reflects a high level of face validity. In addition, manipulations of the dopamine system hold strong construct validity for animal models of BD⁷, as changes in DRD1 levels^28,29 or DRD1 polymorphisms have been associated with BD pathology^30,31,32.

Other animal studies have also led to an increased understanding of the functions of prefrontal DRD1. For example, a decrease in DRD1 has been a consistent finding in models of depression^33,34, while optogenetic stimulation of DRD1 in mPFC glutamatergic neurons reduces anxiety and induces antidepressive effects³⁵. In a recent publication by Wu et al.³⁶, the role of mPFC DRD1 in affective state transitions has been demonstrated. This study highlights that these receptors are crucial for underlying changes in excitatory synapse plasticity.

Altogether, employing a rat model of BD that consists of targeted and conditional manipulation of DRD1 in CamKIIa-positive neurons of the mPFC constitutes a model system with high construct and face validity and, thus, exhibits strong potential for translational research on BD.

In the following, surgical procedures for model generation are described. Additionally, methodological considerations for model induction and behavioral assessments will be presented alongside representative results of the resulting disease-like phenotype. Possible obstacles and influencing factors in model generation and behavioral assessment are discussed, and an outlook on future directions is given.

Protocol

The protocol for stereotactic injection described here has been approved by the LANUV (Landesamt für Natur, Umwelt und Verbraucherschutz, Northrhine-Westfalia, Germany). Adult male Sprague Dawley rats (350-650 g body weight) were used. The reagents and equipment used in this study are listed in the Table of Materials.

1. The lentiviral constructs

NOTE: A third-generation lentiviral system is used for the conditional expression of DRD1 or red fluorescent protein (dsRed) as a control condition^25,27.

1. Produce the lentivirus based on the protocol by Stewart et al.³⁷ with packaging plasmids 8454 and 8455 from the Addgene repository.
   NOTE: If virus production is not planned independently, many core facilities provide high-titer lentivirus, such as Charité Berlin, Germany.
2. Titer concentrated viruses and store them at -80 °C.
3. Prepare 2 × 10⁷ transducing units (TU) per µL for injections.
4. Transport viruses to the surgical suite on dry ice.

2. Animals

NOTE: The rat model for BD has been established in adult male Sprague Dawley rats (350-650 g body weight). For investigation of female rats or earlier developmental time points, it is crucial to consider that expression of DRD1 in the mPFC changes during development and can be influenced by the estrous cycle^38,39,40.

1. Pair-house rats with animals of the same condition with food and water ad libitum under constant temperature and humidity conditions (45%-65 % relative humidity, temperature 22 °C ± 2 °C).
2. Keep rats under an inverse 12 h light-dark cycle (lights off at 11 a.m.), as behavioral investigation should be performed during the animals' active phase in the dark.
3. Give animals at least seven days to acclimate to the facility and handling by the experimenters prior to the start of any experiments.

3. Stereotactic injection of the viral construct

NOTE: Perform surgery under a safety hood (precaution for working with lentivirus) and aseptic conditions.

1. Preparation
   1. Ensure all necessary materials are available and functional (Table of Materials).
   2. Set up the stereotactic frame with the syringe holder attached to the stereotactic arm. Connect the syringe holder to the syringe pump.
   3. Set up the dental drill and mount a 0.9 mm burr.
   4. Lay the heating pad and set it to 37 °C. Elevate the heating pad to an appropriate height for easier positioning of the rat.
   5. Cover the heating pad with an absorbent drape.
   6. Prepare autoclaved surgical instruments on a sterile surface.
   7. Mount a 10 µL stereotaxic injection syringe with a 33 G injection needle to the syringe holder.
   8. Withdraw 2.3 µL of viral suspension for bilateral injections of 1 µl each. Visually confirm the successful withdrawal. Make sure to perform this step as the last preparatory step to minimize the time it takes for the lentivirus to be at room temperature.
2. Analgesia and anesthesia induction
   1. On the morning of surgery, administer meloxicam (1 mg/kg body weight, p.o.).
   2. Twenty min before the start of surgical procedures for analgesia, inject the rat with buprenorphine (0.5 mg/kg body weight, s.c.).
   3. Turn on the anesthesia machine with an oxygen flow of 0.8-1 L/min.
   4. Flood the induction chamber with 4 % Isoflurane and place the rat in the induction chamber.
   5. After successful induction of anesthesia, visible by slowed breathing and loss of consciousness, remove the rat from the induction chamber and move it to the stereotactic frame.
3. Positioning of the rat
   1. Make sure anesthesia flow is switched to the nose mask.
   2. Transfer the rat from the induction chamber to the stereotactic frame, placing its front teeth in the holder.
   3. Properly place the anesthesia mask over the nose and turn the isoflurane to 1.5-2.3% for maintenance.
   4. Protect the eyes using sterile eye cream.
   5. At least 10 min before making any incision, locally inject the rat with lidocaine (10 mg/kg body weight, s.c.) directly below the planned incision site.
   6. Secure the rat in the stereotactic frame using ear bars. Ensure the ear bars are even and at a level head position.
   7. Trim fur around the incision site using scissors. Remove fur pieces using a cellulose pad dampened with a skin antiseptic.
   8. Disinfect the surgical field using a skin antiseptic.
4. Craniotomy and injection of viral construct
   1. Ensure proper anesthesia by checking for the absence of toe reflex.
   2. Disinfect hands and switch to sterile gloves before touching any equipment.
   3. Make a small medial incision (~1.5 cm) using a scalpel blade.
   4. Secure access to the surgical field by pushing skin to the sides with bulldog clamps.
   5. Clean the surgical field from blood and remaining tissue using sterile swaps. Ensure proper vision of the bregma and sufficient anterior space.
   6. Set A/P and M/L coordinate to zero based on bregma.
   7. Move the stereotactic arm to coordinates A/P + 2.7 and M/L ± 0.4 and visualize using a disinfected pencil.
   8. Drill a hole of ~1 mm diameter, covering injection sides for both hemispheres.
   9. Remove any blood using a sterile swap.
   10. Set D/V coordinates to zero at the surface of the brain and slowly lower the injection needle to -2.8 to inject into the prelimbic area of the mPFC.
   11. Wait for 5 min to allow relaxation of the tissue.
   12. Inject 1 µL of viral suspension with a rate of 0.1 µL/min.
   13. Allow 5 min for absorption before slowly removing the needle.
   14. Repeat the injection in the other hemisphere.
5. Closure and post-operative care
   1. Remove the needle and close the opening in the skull using bone wax.
   2. Remove bulldog clamps and suture the skin (3-0 surgical suture).
   3. Inject the rat with meloxicam (1 mg/kg, s.c.) for postoperative analgesia.
   4. Turn off the anesthesia, remove the animal from the stereotactic frame, and place it into its home cage. Ensure that the rat fully wakes up.
   5. Rinse the syringe with 100% ethanol to deactivate the remaining lentivirus, followed by distilled H₂O in preparation for the next injection.
   6. Perform postoperative analgesia with meloxicam (1 mg/kg, p.o.) every 24 h over 3 days and score the animals' health status for 1 week.
   7. Single house animals for the first 24 h after surgery to prevent others from tampering with the sutures. Place them back with their cage mates afterward.

4. Doxycycline treatment for model induction

NOTE: Start model induction as soon as 24 h after the injection. One can also wait longer time periods between injection and induction of up to several months, e.g., to test larger cohorts of animals at the same time. This has been shown to not influence the functionality of the viral construct.

1. Induction of a mania-like episode
   1. To induce a mania-like phenotype, give animals 0.5 g/L doxycycline hyclate by adding it to the drinking water. This induces viral transcription and the overexpression of additional DRD1.
   2. Prepare the doxycycline-containing water freshly every 48 h to 72 h, a timespan where the stability of doxycycline is not influenced, even in non-opaque water bottles⁴¹.
      NOTE: After doxycycline treatment for seven days, virus-mediated overexpression will have reached its maximum, and one can perform a behavioral investigation during the mania-like episode.
2. Induction of a depressive-like episode
   1. Switch the animals back to normal drinking water, to induce a depressive-like episode.
   2. Wait for 4 days until viral transcription has stopped, then perform a behavioral assessment of the depressive-like episode.
   3. Perform subsequent episode inductions following the same pattern.

5. Behavioral assessment

NOTE: Following model induction, one can assess bipolar-like behavior. Different frameworks of translation from clinical symptoms to behavioral patterns observable in rodents have been proposed. One of the most influential is research domain criteria⁴², where changes in domains of functioning and behavior, possibly affected in psychiatric disorders, are examined. It is, however, important to note that due to the species barrier, some symptoms, e.g., suicidality, cannot be investigated in rodents⁴³. Because of their advanced cognitive and emotional capabilities, rat models have especially strong potential for translational symptom assessment⁴⁴, allowing for more elaborate testing procedures. Considerations for behavioral assessment are described in Table 1.

1. Plan behavioral investigation as a battery of behavioral tests⁴⁵, to provide a comprehensive picture of the resulting phenotype.
2. Pay attention to performing more invasive testing last.
3. Consider changes in behavior potentially resulting from previous testing experience when trying to test one animal in both disease-like episodes.
   NOTE: Depending on the question at hand, it can be beneficial to test different groups of rats either during the mania- or the depressive-like episode, which would also allow for tissue collection during the respective episode. Experience has shown that testing behaviorally naïve animals can result in a more prominent phenotype in certain behavioral tests.
4. Also, take into account other factors such as housing conditions⁴⁶ or sex of the experimenter⁴⁷.
5. Adopt all possible measures to reduce unintended stress to the animals, not only for animal welfare, but also to exclude possible interactions of psychiatric phenotypes with stress⁴⁸.
6. Pay special attention to the circadian rhythm, as disruptions of the circadian rhythm are a symptom of BD¹⁷. Since rats are most active during dusk and dawn, one should test under dim red light with animals housed under an inverse light-dark cycle⁴⁹.
   NOTE: Most of the representative results presented here have been collected following this approach. A disease-like phenotype is, however, still observable if behavioral assessment is performed without switching the animals' day-night cycle²⁷.
7. Always preregister the experiment and conduct and describe following PREPARE⁵⁰ and ARRIVE⁵¹ guidelines.

النتائج

When doxycycline is added to the animals' drinking water, additional DRD1 will be expressed, and after 7 days, there will be sufficient overexpression to test the animal for mania-like behavior. So far, an increase in reward-related behaviors has been demonstrated. Mania-like animals drink more sucrose solution in relation to water in a two-bottle choice test when compared to controls²⁵. When put in an observation box with a receptive female and observed for 25 min, mania-like animals show more sexual mounts compared to controls²⁷ (Figure 1A). In a cocaine self-administration paradigm, they administer more cocaine under a fixed-ratio schedule and show a higher break point in a progressive-ratio schedule. Their dose-response curve is shifted towards higher sensitivity to low doses²⁵. This shift in sensitivity is also observed in increased motivational salience in several place conditioning paradigms. Mania-like animals spent more time in the conditioned sides for nicotine, alcohol, and cocaine in comparison to controls²⁵. Increased novelty seeking and more impulsive choices in a T-maze-based test on delayed discounting were also found²⁵. In an operant rat version of the Iowa Gambling Task, mania-like animals decide more often for the disadvantageous (high risk, high gain) choices compared to controls²⁴ (Figure 1B). Anxiety in mania-like animals is reduced, as indicated by more time spent on the open arms in the elevated plus maze²⁵.

A depressive-like phenotype can be induced by terminating DRD1 over-expression. In the depressive-like episode, an increase in helplessness could be observed. In a triadic paradigm of helplessness, the group that was for the first time presented with an electric shock (Figure 1C), as well as the group that had learned to control the shock, were more helpless with increased escape latencies when compared to their respective controls²⁷. The groups in which helplessness was induced, did not show any differences between the experimental and control animal. Anhedonia was found in the two-bottle choice test for sucrose²⁷ and in sexual behavior (unpublished data). In the marble burying test, depressive-like animals were also more anxious²⁶ (Figure 1D).

The described animal model not only offers the possibility to investigate mania- or depressive-like behavior, but it also provides a unique opportunity to observe a switch in behavior when terminating the DRD1 overexpression, resembling the switch from mania to depression in patients. Here, it is important to keep habituation to certain behaviors in mind and choose tests with minimal habituation. For example, increased sexual behavior in the mania-like episode and in the depressive-like episode, a reduction of such behavior to levels like seen in control animals was shown. In this experiment, three mania-/depressive-like cycles were induced within the same animal²⁷. For sucrose drinking, the preference for the sucrose solution in the mania-like state was not only reduced to control levels when switched to the depressive-like state but decreased²⁷. In the rat version of the Iowa gambling task, the number of disadvantageous choices was increased in mania-like animals but not significantly different from controls when the animals were in the depressive-like state. In the latter state, the number of overall earned pellets, was reduced compared to control animals²⁴.

Overall, animals show a robust bipolar-like phenotype, observable in different behavioral domains during both episodes. The switch between episodes in this model contributes to enhanced face validity. An overview of affected behavioral domains is given in Figure 2.

Figure 1: Behavioral changes in mania- and depression-like states following viral DRD1 overexpression. During the viral DRD1 overexpression, in the mania-like state, animals show more sexual mounts (A) and an increase in risky choices in the Iowa Gambling Task (B) compared to controls. After the termination of the overexpression, animals switch to a depressive-like state. They show an increase in helplessness (C) and anxiety (D). *p < 0.05; **p < 0.01; error bars indicate standard error of the mean. Please click here to view a larger version of this figure.

Figure 2: Behavioral phenotype of the model. In the mania-like episode, there is an increase in reward-related behaviors (e.g., sexual behavior), impulsivity, and risk-taking. Anxiety was reduced in the elevated plus maze test. During the depressive-like episode, anxiety was increased in the marble burying test, sexual behavior was reduced, and animals showed more helplessness. The image of the rat in the figure was taken from Servier Medical Art and is licensed under CC BY 4.0. Please click here to view a larger version of this figure.

Table 1: Considerations for behavioral assessment. The table highlights important considerations for the main experimental steps during behavioral assessment. Please click here to download this Table.

Discussion

Here, a novel rat model for BD with increased face validity is presented. A targeted manipulation of DRD1 in the mPFC allows for the induction of a mania- and depressive-like phenotype in the same animal. Representative results highlight an observable disease-like phenotype in both episodes. The model is relatively easy to apply. Two inducible lentiviral vectors expressing either DRD1 or dsRed as control are needed. For the production and the use of lentiviral systems in animals, certain safety levels are required, which need to be in place. If the necessary equipment for virus production is not available, experience working with core facilities has been positive.

The most crucial step for generating the model is the stereotactic injection of the lentiviral system. Stereotactic surgeries are well-established procedures in neuroscience, and success rates among trained investigators are high. There are two main possible sources of error. Problems with anesthesia can lead to fatalities during the surgical procedure. Here, using isoflurane-inhalation anesthesia, as described in the protocol, has proven the best approach, as easily adjustable drug levels constitute a clear benefit compared to injection anesthesia. Since isoflurane does not provide any analgesic effects and meningeal nociceptors are sensitive to stimulation⁵², it is recommended to use an opioid for intraoperative analgesia. Combined with appropriate postoperative medication, as described in the protocol, there are no observable signs of postoperative pain. However, possible influences on research questions, e.g., with regard to interactions of dopamine with the opioid system, should always be considered, and an appropriate medication regime should be chosen accordingly⁵³. If the surgery is performed under aseptic conditions, the occurrence of infections or impairment of wound healing is rare. If problems during surgery or recovery occur, troubleshooting should focus on the correct execution of the described protocol. Ensuring aseptic working conditions and precise dosing of medication are essential. Administration of fluid or glucose solutions can additionally aid recovery. If infections occur, treatment should not include any tetracyclines as these will interact with transcription of the viral systems. First line treatment for postoperative wound infections would be enrofloxacin, possibly combined with carprofen.

Another possible source of error during surgery is the placement of the injection outside the target area. This, however, rarely occurs when the protocol is followed correctly and proper positioning of the animal's head is ensured. Successful placement should always be verified. While placement of the dsRed-expressing virus is easily detectable in control animals, placement verification of DRD1-expressing viruses requires extra steps. Performing antibody staining against different parts of the viral construct did not produce satisfactory results. It is recommended to verify virus placement by dissecting the mPFC and performing a PCR to detect rtTA3 transcripts as described in Beyer et al.²⁴. It is also important to note that the injection of the virus should be bilateral with equal amounts of virus. Cerebral and behavioral lateralization has been shown to differ in patients with bipolar disorder^54,55, and unilateral viral injections might not induce the desired behavioral phenotype.

Induction of viral DRD1 expression and the mania-like episode by adding doxycycline to the drinking water works very well. Substituting normal drinking water with doxycycline has been proven to not cause marked changes in drinking behavior. However, fluid consumption should be monitored. Modulations are possible, if other substances, e.g., medications, are intended to be administered via drinking water. Doxycycline administration could also be performed via food pellets. But this has not been validated yet.

For behavioral investigation, several considerations are listed in Table 1. Especially model-specific requirements need to be evaluated when planning an experiment. For example, it should be decided if two groups of animals will be tested, or if one animal will undergo behavioral assessment during both episodes, which may require retesting. If the bipolar-like phenotype is not detectable during the behavioral investigation, although placement could be verified, troubleshooting can focus on various factors possibly influencing the behavioral outcome. Changes in experimenters or the circadian rhythm should be critically evaluated during the process, as stressful conditions in the environment can influence behavioral outcomes.

While the model demonstrates good construct and face validity, predictive validity still needs to be evaluated. Chronic lithium administration, as the first-line treatment of BD⁵⁶, should be successful in preventing model-induced changes in behavior. Responses to other medications used in BD, such as antipsychotic or anticonvulsive drugs, might be investigated to fully test the predictive validity of the model.

Additionally, a current limitation is that model validity in female animals still needs to be assessed in future studies. While there is a trend to include female animals in preclinical research, this is often still neglected. For the presented model, interactions of the dopamine system with the estrous cycle are to be expected. It is, however, unclear to what extent they will occur. It is also important to bear the general limitations of psychiatric animal models in mind. While the possibility of inducing both disease-like episodes in one rat provides increased face validity, externally induced changes still differ from spontaneous occurrence and cycling of disease episodes in patients with BD. Since the model is solely based on targeted manipulation of the dopamine system, major effects will be caused by alterations in dopamine transmission and related secondary effects. Contributions of other systems to the symptomology of BD are therefore not accounted for.

In conclusion, the presented model has strong potential for investigating BD, as both disease episodes can be studied in a single animal. This presents unique possibilities for the investigation of transitions between episodes compared to most established models. The presented protocol requires equipment and technical skills which are available in most preclinical research labs, making it widely applicable. So far, the resulting behavioral phenotype has been robust across different behaviors. Other domains like social behavior⁵⁷ or cognitive functions are yet to be explored. While the presented protocol focused on behavioral outcomes, there are various possibilities for future applications to investigate molecular mechanisms further. Expanding investigations to understand underlying mechanisms in BD pathogenesis, especially regarding the transition between episodes, may lead to the identification of therapeutic targets that could eventually be translated into future clinical applications.

Materials

Name	Company	Catalog Number	Comments
0.9 mm burr	FST	19007-09	Burr for craniotomy
10 µl Neuros Syringe	Hamilton	65460-06	Mounted to syringe pump for injection
1ml single use Syringes	Braum	9166017V	Administration of medication
33 G Needles	Hamilton	65461-02	Replacement needles for neuros syringe
4-way valve	UNO	180000259	For simultaneous connection of induction chamber and face mask
Absorbent Drape	Sabanindas	1834014	Covering equipment before placing the animal
Anaesthetic Gas Filter	UNO	180000140	Anesthesia fume collection
Anasthesia mask for stereotactic	Hugo Sachs Electronic	73-4922	Administering anesthesia during surgery
Anesthesia vaporiser	UNO	180000002	Provide and adjust levels of vaporised isoflurane
Bone wax	SMI	Z046	Closing the hole in the skull
Bulldog clamps	FST	18038-45	To retain skin and allow access to the surgical field
Buprenorphine	Elanco	18760711	Interoperative analgesia
Cannula	Tegler	T138339	Administration of medication
Cellulose swabs	Meditrade	1177	Cleaning Skin
Connector	UNO	180000005	Connecting anesthesia tubing to face mask
Control Unit for heating pad	UNO	180000122	Controlling heating pad
Dental Dril	Saeyang	SMT K-38	Dental drill for craniotomy; equipable with fine dental burrs
Desktop digital stereotaxic instrument	RWD	E03135-002	Fully equipped stereotactic frame with digital manipulator
Destilled H2O	-	-	Rinsing the syringe
Doxycycline hyclate	Sigma aldrich	D9891	For model induction
Dry ice	-	-	Transporting viral suspension
Earbars	RWD	68302	Head fixation in the stereotactic frame
Ethanol	-	-	Rinsing the syringe and deactivating virus
Flowmeter	UNO	CM2	Verify and adjust flow rate
Forceps - anatomical	FST	11000-12	Holding skin
Forceps - surgical	FST	11027-12	Holding skin
Heating pad	UNO	180000028	Heating pad for keeping the animal warm during surgery
Induction chamber	UNO	180000233	Chamber for initial induction of anesthesia
Isoflurane	CP Pharma	V7005232.00.00	Anesthesia
Lentiviral suspension	-	-	Lentiviral construct coding for DRD1 or dsRed for model induction
Lidocaine	Combustin	8780701	Local analgesia
Meloxicam	Boehringer Ingelheim	7578423	Pre- and postoperative analgesia
Needle holder	FST	91201-13	Sutering
Oxygen concentrator	UNO	180000399	Providing oxygen for anesthesia
PE Tubing	-	-	Connecting components of the anesthesia machine to induction chamber & face mask
Pencil	-	-	Marking the correct side for craniotomy
Scalpel blade holder	FST	10003-12	To hold scalpel blade
Scapel blades	FST	10011-00	Fine surgical blade for incision
Scavenger Unit	UNO	180000260	Controlling capacity of fume collector
Skin disinfectant	Bode	975042	Disinfacting skin before incision
Sterile cotton swabs	Boettger	1102241	Cleaning surgical field
Sterile eye cream	Bayer	1578675	Protect eyes during surgery
Surgical Scissors	FST	14000-12	Trimming fur and cutting suture material
Suture 3-0 polyglycolic acid	SMI	11201519	Suturing skin
Syringe pump	KdScientific	788130	Syring pump with connectable holder