Method of effecting neuroprotection using soluble neuregulin isoforms

Abstract

The present invention refers to soluble Neuregulin-1 isoforms representing Posttranslational Neuregulin-1 modifications as medication in cognition-related neurological disorders, in particular schizophrenia, Alzheimer's and Parkinson's diseases.

Claims

1. A method of treating a subject suffering from traumatic brain injury, comprising systemically administering to said subject a recombinant soluble Type I Neuregulin-1 isoform which is an about 15 to about 35 kD N-terminal fragment of Type I Neuregulin-1 , wherein said recombinant soluble Type I Neuregulin-1 isoform is the first 245, 246, 247, 248, 249, or 250 amino acids of Type I Neuregulin-1 .

2. The method of claim 1, wherein said recombinant soluble Type I Neuregulin-1 isoform is the first 245 amino acids of Type I Neuregulin-1 .

3. The method of claim 1, wherein said recombinant soluble Type I Neuregulin-1 isoform crosses the blood-brain barrier.

4. The method of claim 1, wherein said recombinant soluble Type I Neuregulin-1 isoform reduces dopaminergic cell death in said subject compared to a subject that is not treated with said recombinant soluble Type I Neuregulin-1 isoform.

5. The method of claim 1, wherein the Type I Neuregulin-1 isoform has an isoelectric point (pl) of about 4 to about 10.

6. The method of claim 1, wherein the Type I Neuregulin-1 isoform is a modified polypeptide, wherein the modifications are selected from phosphorylation, glycosylation, methylation, myristoylation, oxidation and any combination thereof.

7. The method of claim 1, wherein said recombinant soluble Type I Neuregulin-1 isoform is administered in an amount of 3 to 600 ng/kg body weight of the subject to be treated.

8. The method of claim 1, wherein said recombinant soluble Type I Neuregulin-1 isoform is administered in an amount of 150 to 300 ng/kg body weight of the subject to be treated.

9. The method of claim 1, wherein said recombinant soluble Type I Neuregulin-1 isoform is administered in an amount of about 3, 5, 20, 50, 150, 200, 300 or 600 ng/kg body weight of the subject to be treated.

10. The method of claim 1, wherein said recombinant soluble Type I Neuregulin-1 isoform is administered in an amount of about 20 ng/kg body weight of the subject to be treated.

11. The method of claim 1, wherein the administration is by intravenous or intraperitoneal injection.

Description

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS

(1) FIG. 1: Various reviews and numerous research articles on Neuregulin 1 show the key functional position of NRG 1 as an upstream regulatory principle of mechanisms thought to be pivotal in neurodegenerative diseases, neurological disorders, as well as physiological function:

(2) NRG are key parts of functional complexes, consisting at least of neuregulins (NRG), receptor tyrosine kinases (ErbB receptors), heparansulfate proteoglycans (HSPG) and NMDA receptors (NMDAR), which are transiently and activity-dependent assembled together in cholesterol (CHO)-rich membrane microdomains. In particular the shaping of calcium signals is important for the interaction with subsynaptic scaffolding proteins by posttranslational modifications (PSD-95, by interaction with certain phosphorylated domains, like PDZ- or SH-domains on partner proteins). The PSD-95 complex directly regulates pro-inflammatory enzymes like nitric oxide synthase (NOS, iNOS is inducible, nNOS is neuronal) and Cox-2 (cyclooxygenase-2), which promote their effects in a complex relationship with related, but not necessarily downstream mechanisms, involving NAD.sup.+-dependent enzymes like PARP-1 (poly-ADP-ribose polymerase-) and Sir-2 (sirtuin-2); PARG is poly(ADP-ribose) glycohydrolase the complementary and antagonistic enzyme to PARP-1, HDAC are histone deacetylases, the general class of enzymes which includes Sir-2. MPTP stands for the mitochondrial permeability transition pore. DRP-2 is dihydropyrimidinase-related protein 2. Also other important membrane proteins, like certain nicotinic acetylcholine receptors (nAChRa7), GABAA receptors (GABAAR) amyloid precursor protein (APP) and proteases (PS) are transiently organized in lipid rafts and acquire different functional properties outside the usual phospholipid (PL) environment, details in (Schrattenholz and Soskic 2006).

(3) FIG. 2: Summary of learning experiments in Morris water maze: animals treated with a daily dose of 3 ng/kg (i.v.) of the soluble extracellular domain of neuregulins-1 beta (NRG-1 beta-ECD) were significantly better in learning than vehicle treated animals; IAE: inner area entry; IAEF inner area entry frequency; TS: time spent in inner area; DT: distance travelled in inner area.

(4) FIG. 3: Reduction of Amphetamine-induced hyperactivity by NRG-1 beta-ECD, a widely accepted model for schizophrenia. Concentrations ranged from 15 to 600 ng/kg (i.v. injection 15 minutes prior to amphetamine application). A positive control of 0.125 mg/kg Haloperidol was included. Whereas Haloperidol like other non-typical and typical antipsychotics usually reduce activity below control level (indicated here by dotted lines labelled veh/veh, in blue for crossings and in magenta for rears), NRG-1 beta-ECD reduction asymptotically approaches control levels of activity, but does not cause further reduction. The low effective concentrations of NRG-1 beta-ECD and the absence of negative effects (reduction of activity below vehicle control levels) are the outstanding properties in this model. The effects are significant with p<0.05;

(5) FIG. 4: Summary of learning experiments with APPPS mouse model of cerebral amyloidosis and Alzheimer's disease in a Morris water maze: animals treated with a daily dose of 200 ng/kg i.p. NRG-1 beta-ECD were significantly better in learning than vehicle treated animals; IAE: inner area entry; IAEF inner area entry frequency; TS: time spent in inner area; DT: distance travelled in inner area.

(6) FIG. 5: HPLC quantification of dopamine and its metabolites: The columns labelled with asterisks are highly significant.

(7) TABLE-US-00001 Legend S Saline (control) aM acute MPTP aMN acute MPTP and NRG-1 beta-ECD aN acute NRG-1 beta-ECD cM chronic MPTP cMN chronic MPTP and NRG-1 beta-ECD cN chronic NRG-1 beta-ECD

(8) FIG. 6: Metabolism of dopamine by MAO-B and COMT.

(9) FIG. 7: MPTP exposition leads to a significant loss of dopaminergic neurons in the substantia nigra (aMPTP, p=0.0005; and cMPTP, p=0.0075). The ip application of 20 ng/kg of NRG-1 beta-ECD leads to a reversal (aNR-MPTP, p=0.57, i.e. not different from vehicle control) or clear and significant improvement of the MPTP lesion (cNR-MPTP; p=0.0097); in the chronic model (5 days daily ip application of 20 ng/kg of NRG-1 beta-ECD) there is also a significant effect number of dopaminergic neurons (cNR; p=0.0002);

(10) TABLE-US-00002 Legend NaCl Saline (control) aMPTP acute MPTP aNR- acute MPTP and NRG-1 beta-ECD MPTP aNR acute NRG-1 beta-ECD cMPTP chronic MPTP cNR- chronic MPTP and NRG-1 beta-ECD MPTP cNR chronic NRG-1 beta-ECD

(11) FIG. 8: Two representative images of 2D-Western blots of brain proteins of APPPS mice stained for Neuregulin-1 are shown of each, a treated and good learning animal (top) and non-treated animal with inferior learning performance (below).

(12) The numbers in the upper part are pI values of the 2D gel.

(13) FIG. 9: A Western blot experiment compairs the abundance of the NRG-1 ECD-fragment in post mortem cortical material from Alzheimer patients and controls.

(14) FIG. 10: 2D-PAGE shows, that the acidic isoform of NRG-1 -ECD, with a pI of approx. 5-5.5 and a molecular weight of approx 25-32 kD in these experiments is clearly diminished in Alzheimer's patients brains.

(15) FIG. 11: is a schematic of the pool used for the Morris Water Maze assessment.

(16) FIG. 12: is a schematic of the pool used for the Morris Water Maze assessment showing the zones for computing the animals' track records.

DETAILED DESCRIPTION OF THE INVENTION

Examples

(17) General:

(18) In all of the following experiments fragments of Neuregulin-1 beta have been used, comprising only the extracellular domain (ECD) of the entire transcript of the human nrg-1 gene. They had a molecular weight of approx 25-32 kD and isoelectric points between approx 5 and 9.5, depending upon phosphorylation and/or glycosylation status.

(19) The physiologically active form of Neuregulin-1 isoform has a pI of approx. 5.5. The physiologically active form has an pI of approx. 5.5 (most of the experiments were carried through with a commercially available isoform produced in E. coli, with a molecular weight of 26 kD and an pI of approx. 9.0)

(20) This isoform is a recombinant soluble human NRG-1 beta fragment consisting of the first 245 amino acids of NRG-1, purchased from R & D Systems, Inc. (Catalog No. 377-HB-CF). It will be named NRG-1 beta-ECD in the following. This active isoform has a pI of approx. 9.0

(21) We also tested a corresponding fragment of NRG-1 with 8 kD, only comprising the EGF domain, purchased from R&D Systems (Catalog No. 396-HB). This fragment appears to be neuroprotective as well in vitro and in vivo, but was not investigated in depth because of much higher proliferative properties, which raised concerns about cancerogenity.

Example 1

(22) Initial Toxicology Data Indicate that NRG1 (ECD) has No Adverse Effects in Acute Toxicology and In Vitro Mutagenicity Tests. There was no acute intravenous toxicity in rats: All animals survived until the end of the study period. No clinical signs were observed during the course of the study. The body weight of the animals was within the range commonly recorded for this strain and age. No macroscopic findings were recorded at necropsy. The median lethal dose of NRG1 (ECD) after single intravenous administration to female rats, observed over a period of 14 days is: LD.sub.50 (female rat): greater than 5000 ng/kg body weight. Daily intravenous administration of Neuregulin over a period of seven days at dose levels of 50, 200 and 600 ng/kg body weight/d did not result in any premature death. No clinical signs were recorded. The treatment did not affect the food consumption and body weight development. The no observed effect level (NOEL) was established at 600 ng/kg body weight/d. In the mouse lymphoma thymidine kinase locus assay using the cell line L5178Y according to the OECD Guideline for the Testing of Chemicals, No. 476 In vitro Mammalian Cell Gene Mutation Test, NRG1 (ECD) was non-mutagenic. In the chromosome aberration test in Chinese hamster V79 cells according to the OECD Guideline for the Testing of Chemicals, No. 473, NRG1 (ECD) did not induce structural chromosome aberrations.

(23) Moreover, in none of the animal experiments carried out with regard to efficacy (some of them going on for several months with daily iv applications) did we ever observe adverse effects of NRG1 (ECD).

(24) Application of NRG1 (ECD) in the various animal models described below, was either by intravenous (iv) or intraperitoneal (ip) injections; concentrations were ranging from 3-600 ng/kg.

Example 2

(25) Learning and Memory: Spatial Learning with and without NRG-1 Beta-ECD Application

(26) Methods:

(27) The Morris Water Maze assesses spatial learning. It requires animals to swim in a water-filled pool and to find a rescue platform submerged just below the surface. It is obligatory that the platform is placed away from the walls of the maze and that animals have reference points visible from the water surface that permit estimation of location, but are not close enough to the target to permit associative learning. The animals are trained that rescue only comes via the platform meaning that all animals which do not find the platform, are guided to the platform and allowed to rest before being removed from the set-up. Therefore, one of the most important reference points for the mouse is the human operator.

(28) The experiment aims at determining two key parameters associated with murine spatial recall: the rate at which the mice learn to relocate the platform the ability to retain the information in the short term (within a training period or overnight)
Animals

(29) The study is performed with two groups of APP/PS mice (Meyer-Luehmann et al. 2006; Radde et al. 2006), one of which is treated by a daily dose of NRG-1 beta-ECD and the other one is sham treated as a control. Each group consists of 8 males which are nine weeks old at the beginning of their first series of experiments.

(30) The first series of experiments started with two subgroups of 8 treated and 8 untreated mice on week 42 and will last for 15 days. Further series of identical experiments will be performed 6, 12, etc. weeks later.

(31) For a second pair of subgroups (8 treated and 8 untreated nine weeks old males) the same series of experiments started on week 48, so the experiments of these subgroups lag exactly 6 weeks behind the ones of the first subgroups.

(32) Apparatus

(33) The learning aptitude of the treated and untreated APP/PS mice is assessed using a circular Morris water maze which should be large enough to provide searching space without exhausting the mouse. Utmost care needs to be taken to keep each detail of the experimental setup as invariable as possible throughout all experiments.

(34) In the current study, a pool of 120 cm diameter is used which is placed at an exactly reproducible position in the lab with always identical orientation. At fixed positions in the pool, a white, translucent, circular platform of 15, 10, or 5 cm diameter is placed that extends to just below the water surface (so it is invisible to the mice) and that the animals can climb onwhich is the only means to rest out of the water. To assist climbing, the platform is coated with a gauze grip surface (see FIG. 11).

(35) In order to perform the rescue procedure in probe trials as detailed below, the platform is equipped with a mechanism that allows for automatically raising and lowering it without direct operator intervention. Thus, depending on its height the platform is accessible to the swimming mice or not On-demand platform (Buresova et al. 1985).

(36) Platform locations are always situated in a ring shaped, concentric region of the pool with inner and outer diameters 40 cm and 80 cm, respectively. Four quadrants are defined such that the platform occupies the central region of one of them (the target quadrant). For further details on platform sizes and positions see below.

(37) In order to make sure that platform position is exactly the same throughout an entire series of experiments; a socket will be firmly affixed to the floor of the pool on which the platform can be mounted with a minimum of spatial tolerance. On top of the platform, in its center, there is another mounting for a (proximal) cue sticking out of the water which is well visible on the video recording as well as to the mice swimming in the pool. For a check of platform position, a brief video recording will be taken without an animal but with the cue plugged into the platform whenever the platform or the video camera have been manipulated with in any way.

(38) The water is made opaque using low fat milk powder. The water temperature should be cold enough to encourage searching for an exit but not so cold that the animals suffer or are exhausted. As a fair compromise, water temperature is monitored at the start of each experiment and modulated with either warm water or ice flakes to 18 C. Between individual trials, temperature is readjusted as needed.

(39) Four distal cues (of different simple geometric shapes and different colors, height 20 cm) are attached 20 cm above the sides of the pool, one in each quadrant. Care is taken to place each cue in exactly the same location throughout all experiments. The entire pool is enclosed by a white translucent curtain. Lighting is dimmed and diffuse.

(40) A video camera is firmly mounted at an exactly vertical position above the center of the pool, such that the pool completely fills the video image. Video recordings are taken at PAL resolution (720576 pixels, 25 frames per second), at the least. The videos are evaluated by an automatic tracking system that allows for flawless detection of the animals movements with time.

(41) Mice are placed into the water using a special devise that is mounted on a stick, so they can be watered at exactly defined spots along the rim of the pool without the operator entering the cabin made up of the translucent curtain.

(42) Experimental Design

(43) In each session, mice are placed into the pool at predefined sites and are allowed to swim for 60 s. Animals' motion tracks are recorded by a video tracking system, and parameters are computed from which conclusions regarding the animals' learning aptitude can be drawn (most notably the period until the mouse hits the platform for the first time=escape latency; further details see below). If a mouse succeeds in finding the platform it is left resting there for a short period of time (15 s). Otherwise, after 60 s of swimming, the mouse is guided to the platform by the operator and allowed to rest for 15 s. Afterwards, it is picked up by the operator, dried gently and returned to its housing or prepared for the next swim.

(44) On each day of experiments, one trial per mouse is performed in the early morning. Each trial consists of two consecutive swims originating from two different quadrants, but never from the target quadrant. Exact watering sites (and platform positions whenever applicable) are assigned randomly for each swim of each day, but do not differ between the individual mice during that day.

(45) If mice turn out to learn extremely slowly the number of swims per trial or trials per day may be increased (and vice versa). Moreover, in many mouse strains younger animals learn very quickly, so after four or five days of training, escape latencies remain constant at a few seconds only which is equally true in treated as in untreated animals. However, for the statistical evaluation, it is of advantage if the curve of escape latencies over training days does not saturate but rather decreases monotonically. Therefore, an experimental design is used in which the problem to solve becomes more difficult with training progress: On predefined days the platform is replaced with a smaller one while the platform's center coordinates remain the same. If and when platforms are swapped may be determined independently for each series of experiments and ought to depend on the outcomes of the preceding series.

(46) In each series of experiments mice are subjected to three different kinds of tasks: Cued place navigation. The platform is marked with a cue, and the mouse is allowed to swim until it finds the platform. This procedure tests associative learning and serves for dividing mice into two experimental groups the learning aptitudes of which are as similar as possible. Moreover, in the second and further series of experiments, cued place navigation supports blanking recollection of the position of the platform in preceding series. Hidden platform acquisition training. The platform is invisible to the mouse and located at the same position as during the preceding swim. This task allows for monitoring the mouse's progress in recalling the exact location of a hidden platform (spatial learning). Probe trial testing. In this task, the on-demand platform is maximally lowered underneath the surface and the mouse is allowed to swim freely searching for it. Probe trial testing assesses the animals' absolute recall which, in this context, can be also interpreted as conviction, persistence or certainty regarding the platform location. The conventional approach to interpreting the experiment is that animals that have firmly fixed the location of the platform will more persistently search in a limited location and thus spend more time in the zone next to the platform. In probe trial testing, there is a risk that the inability to find the platform may reduce the incentive to swim to the platform zone. In order to keep these irritations as small as possible, the modalities of human rescue ought to remain the same so there is some spatial constancy despite the absence of the platform. Therefore, after 60 s of swimming the platform is lifted to just beneath the surface, the mouse is guided there by the operator and allowed to rest for 15 s before being taken out of the apparatus. On all days of probe trial testing, only one swim is performed.

(47) Approximately 60 minutes prior to each trial, mice are treated daily with either 5 ng/kg NRG-1 beta-ECD (suspended in black 6 mouse serum and provided i.v. in a volume of 20 L per mouse) or with 20 L of vehicle i.v., respectively.

(48) On day 1 of the first series of experiments, all mice in the study receive sham treatment only. Thereafter, mice are assigned to the Neuregulin and the control groups such that the distributions of escape latencies match in both groups.

(49) In each series of experiments the following chronology is adhered to: day 1. Cued platform search with platform of size 10 cm and position changing for each swim. day 2. Cued platform search with platform of size 10 cm and position changing for each swim. day 3. Cued platform search with platform of size 10 cm and position changing for each swim. day 4. Cued platform search with platform of size 10 cm and same position as the last one on day 3. day 5. Hidden platform search with platform of size 15 cm and same position. day 6. Hidden platform search with platform of size 15 cm and same position. day 7. Hidden platform search with platform of size 15 cm and same position. day 8. Hidden platform search with platform of size 10 cm and same position. day 9. Probe trial testing. day 10. Hidden platform search with platform of size 10 cm and same position. day 11. Hidden platform search with platform of size 10 cm and same position. day 12. Hidden platform search with platform of size 5 cm and same position. day 13. Hidden platform search with platform of size 5 cm and same position. day 14. Hidden platform search with platform of size 5 cm and same position. day 15. Probe trial testing.

(50) It may be necessary to aid unlearning of the platform position from a preceding set of experiments by allowing the mice to freely swim for a few days without platform present.

(51) The rate of learning is assessed by monitoring each training/test session and noting the success of the animals in finding the platform as well as the evolution of the search strategy from skirting the sides of the pool to moving away from the sides to search in the near to central area where the platform lies.

(52) Measured PARAMETERS

(53) From the animals' video recordings, each mouse's motion track is extracted and exported as a series of x, y, and time coordinates for further processing. Care needs to be taken to reliably identify each track's staring point and to avoid tracking errors. Simultaneously, a number of parameters are computed from which conclusions regarding the animals' learning aptitude can be drawn (see below). Parameter recordings are halted after 60 s or if the mouse has found the platform (whichever happens earlier).

(54) For the definition of parameters to be computed from the animals' track records the following zones are defined (see FIG. 12):

(55) In order to keep evaluations as flexible as possible, four concentric target zones (centered about the platform) of 5.5 to 30 cm diameter are employed.

(56) Parameters computed from the animals' track records include: Total distance travelled Overall average speed Number of entries to the pool center Time in the pool center Latency to first entry to the pool center Distance travelled to first entry to the pool center Number of entries to the inner area Time in the inner area Distance travelled in the inner area Latency to first entry to the inner area Distance travelled to first entry to the inner area and for each target zone 1 to 4 and the target quadrant Number of entries to the zone Time in the zone Distance travelled in the zone Latency to first entry to the zone Distance travelled to first entry to the zone Distance from beginning of track to nearest point of zone Average distance from the zone when outside the zone Minimum distance from the zone when outside the zone Time to minimum distance from the zone when outside the zone Time getting closer to the zone Time getting further away from the zone Time moving towards the zone Time moving away from the zone Number of head entries to the zone Time of head in the zone Distance of head travelled in the zone Latency to first entry of head to the zone Average distance of head from the zone when outside the zone Minimum distance of head from the zone when outside the zone Initial heading error Average heading error Number of exits from the zone

(57) For each day of the experiment, the readings of parameters of learning progress in the treated and untreated groups are compared to each other statistically.

(58) On examining the track records of individual mice, a human observer is able to come to a fairly realistic perception of the animals' assertiveness in locating the platform which is not fully reflected in the measured parameter values. Therefore, track records are also manually inspected and the animals' recall of the platform position is rated.

(59) Results:

(60) Those animals treated with a daily dose of 3 ng/kg NRG-1 beta-ECD i.v. 30 min prior to training were significantly better in learning-related parameters than the vehicle treated group.

(61) Neuregulin not only improved learning, but treated animals had also developed more advanced search strategies: More treated animals entered the inner area of the pool (11 vs. 7, p=0.019), entries to the inner area occurred more often (2.17 vs. 0.92 times, p=0.02), time spent and distance travelled in the inner zone was longer (6.51 s vs. 2.13 s, p=0.09 and 0.64 m vs. 0.25 m, p=0.031, respectively).

(62) The results of the learning experiments in a Morris water maze are summarized in FIG. 2.

Example 3

(63) Schizophrenia: Amphetamine-Induced Hyperactivity in the Rat

(64) Methods:

(65) The method, which detects antipsychotic and anti-Parkinson activity, follows that described by Costall et al. 1978 and uses an activity meter similar to that described by Boissier and Simon 1966.

(66) Amphetamine induces hyperactivity in this test situation. Hyperactivity is antagonized by classical and atypical antipsychotics acting on dopaminergic systems at the limbic level, and is potentiated by anti-Parkinson drugs.

(67) Rats are injected with d-amphetamine (3 mg/kg i.p.) and are immediately placed in the activity meter.

(68) The activity meter consists of 12 covered Plexiglass cages (402525 cm) contained within a darkened cabinet. Each cage is equipped with two photocell assemblies at each end of the cage, 3 cm above the floor, in order to measure the number of movements by each animal (one per cage) from one end of the cage to the other. Two additional photocell assemblies are placed at 20 cm above the floor to record rearing. The scores for activity and rearing are recorded by computer over 10-minute intervals and cumulated over a 30-minute period.

(69) 15 rats were studied per group. The test was performed blind. The test substance was evaluated at 8 doses, administered i.v. 15 minutes before amphetamine, and compared with a vehicle control group. The experiment also included a control group not treated with amphetamine.

(70) Haloperidol (0.125 mg/kg i.v.), administered under the same experimental conditions, was used as reference substance.

(71) The experiment therefore included 16 groups.

(72) Data were analyzed by comparing treated groups with appropriate control using unpaired Student's t tests.

(73) Results:

(74) As shown in FIG. 3, NRG-1 beta-ECD in a dose-dependent manner inhibits the amphetamine-induced hyperactivity in an animal model for schizophrenia. Conspicuously, the experiments reveal outstanding properties of NRG-1 beta-ECD: The effects shown in FIG. 3 are strongest in the second half of the experiment (minutes 20-40). In the first 20 minutes only a smaller effect can be found, this delayed effect points to further processing of the protein. The effective concentrations of NRG-1 beta-ECD used here are about 200-1000 times lower than those used for typical control neuroleptica like Haloperidol (125 g/kg). In contrast to Haloperidol, Clozapine, Olanzapine etc. there are no negative effects observed in that NRG-1 beta-ECD does not reduce activity of test animals below vehicle control levels.

Example 4

(75) Schizophrenia: Prepulse Inhibition

(76) Rodents with NRG1 knock-out show significantly impaired prepulse inhibition (PPI) linking NRG1 to schizophrenia. A widely used surrogate measure of psychosis in animal models, PPI is considered a schizophrenia endophenotype. It was reported that there are neurophysiological effects of missense mutations of a nonsynonymous single nucleotide polymorphism located on NRG1 (rs3924999) on PPI after extensive genotyping, in both schizophrenia and healthy control populations (Hong et al. 2007). We tested the effect of NRG-1 beta-ECD on PPI. The results so far may be summarized as follows:

(77) At 105 dB, NRG-1 beta-ECD showed a general trend towards re-establishment of PPI (+26%, +23% and +36%, at 150, 300 and 600 ng/kg respectively), although the effect did not reach statistical significance and was not observed at 115 dB. It had no effects on spontaneous movements in the absence of stimulus at 150 or 300 ng/kg but significantly decreased spontaneous movements in the absence of stimulus at 600 ng/kg (20% and 29%, on average and peak intensities respectively, p<0.05, this is similar to aripiprazole). NRG-1 beta-ECD had no effects on the reaction to the pre-pulse alone.

(78) The results so far suggest the absence of significant effects on apomorphine-induced PPI deficits for Propsy100 over the dose-range 150-300 ng/kg and a decrease of spontaneous movements as well as a trend towards re-establishment of PPI at 600 ng/kg i.v. in the Pre-pulse Inhibition (PPI) Test in the rat (deficits induced by apomorphine).

(79) In this series of experiments, the reference substance, aripiprazole, had weak but significant activity at 3 mg/kg i.p., but not at 10 mg/kg i.p., in the same test.

(80) All together and under conditions used, NRG-1 beta-ECD appears to affect PPI at higher concentrations around 600 ng/kg. These results surprisingly open a novel understanding of recent neurobiological research implying (NRG1) as one of the leading candidate genes in schizophrenia.

Example 5

(81) Learning and Memory in an Animal Model for Alzheimer's Disease (APPPS dt Mice)

(82) The animal experiments testing learning and memory with or without application of the soluble extracellular domain of Neuregulin 1 (NRG-1 beta-ECD) in a Morris water maze set up described above for normal mice, have been repeated double transgenic mouse model for cerebral amyloidosis (APPPS mice(Meyer-Luehmann et al. 2006; Radde et al. 2006)).

(83) Here again those animals which were treated with a daily dose of NRG-1 beta-ECD (here 200 ng/kg i.p. were applied) 30 min prior to training were significantly better in learning-related parameters than the vehicle treated group.

(84) Neuregulin not only improved learning, but treated animals had also developed more advanced search strategies: More treated animals entered the inner area of the pool (12 vs. 7, p=0.009), entries to the inner area occurred more often (2.0 vs. 0.7 times, p=0.03), time spent and distance travelled in the inner zone was longer (5.3 s vs. 2.1 s, p=0.09 and 0.7 m vs. 0.3 m, p=0.025, respectively).

(85) The results of the learning experiments with APPPS mouse model of ceretral amyloidosis and Alzheimer's disease in a Morris water maze are summarized in FIG. 4.

Example 6

(86) Neuregulin 1-Beta MPTP Mouse Model of Parkinson's Disease

(87) Methods:

(88) Male C57Bl/6 mice of 10 weeks were used in the MPTP (1-Methyl-4-phenyl-1, 2, 3, 6-tetrahydropyridine) model for Parkinson's disease.

(89) Brain tissue is dissected (Substantia Nigra, Striatum, Cortex) of 10 weeks old male C57Bl6 mice (N=10 per group) at different times after treatment (0, 1, 3, 7, 21 days) with NaCl (controls) or MPTP (acute and subchronic models). The methods follow published procedures (Hglinger et al. 2007; Hglinger et al. 2004).

(90) TABLE-US-00003 Time after injection of MPTP Treatment 0 days 1 days 3 days 7 days 21 days MPTP acute N = 10 N = 10 N = 10 N = 10 N = 10 MPTP chronic N = 10 N = 10 N = 10 N = 10 N = 10 NaCl N = 10 Total: N = 110

(91) MPTP is dissolved as a powder in 0.9% NaCl and is injected intraperitoneal (acute application: 420 mg/kg, each at 2 hour intervals; chronic application: 530 mg/kg, each at 24 hour intervals). These injections take approx. 10 seconds, animals were sacrificed at defined time points (see table) by cervical dislocation. The procedures follow published protocols (Hglinger et al. 2007; Hglinger et al. 2004; Liberatore et al. 1999; Przedborski and Vila 2003, Vila and Przedborski 2003).

(92) 0 days after last MPTP-administration: Loss of striatal dopaminergic nerves

(93) 1 day after last MPTP-administration: Beginning microglia-activation

(94) 3 days after last MPTP-administration: Maximum of microglia-activation

(95) 7 days after last MPTP-administration: Maximum of astrocyte-activation

(96) 21 days after last MPTP-Administration: Maximum of cell death

(97) 21 days after intracerebral infusion of NRG-1 beta-ECD and a control peptide via Alzet Mini pumps, followed by MPTP treatment (acute vs. chronic), a histological quantification of dopaminergic neurons of the middle brain was performed according to stereological principles. Also, a biochemical quantification of dopamine and its metabolites in the striatum is performed by HPLC. Procedures are performed according to published protocols (Hglinger er et al. 2007; Hglinger et al. 2004).

(98) TABLE-US-00004 Treatment Infusion N MPTP acute NRG-1 beta-ECD N = 10 Control acute NRG-1 beta-ECD N = 10 MPTP chronic NRG-1 beta-ECD N = 10 Control chronic NRG-1 beta-ECD N = 10 MPTP acute Control peptide N = 10 Control acute Control peptide N = 10 MPTP chronic Control peptide N = 10 Control chronic Control peptide N = 10 Total: N = 80
Results:

(99) As shown in FIG. 5, the results of the HPLC measurements of dopamine and its metabolites reveal a clear effect of administration of NRG-1 beta-ECD during MPTP insult, in this model for Parkinson's disease.

(100) The effects are non-classical: whereas there is no significant effect upon dopamine levels, neither during MPTP-insult nor in acute or chronic controls of NRG-1 beta-ECD administration, there are pronounced and clear effects on the concentrations of DOPAC and HVA. The chronic administration of NRG-1 beta-ECD results in a clear and significant reduction of this metabolite in the absence of MPTP insult, whereas in the acute regimen only a slight decrease is observed. During the chronic condition of the MPTP insult, NRG-1 beta-ECD causes a significant increase of homovanillic acid (HVA), an effect, which is even more pronounced in the absence of MPTP-insult.

(101) These results are can be interpreted by a down regulation of MAO-B during the chronic NRG-1 beta-ECD administration and/or COMT up regulation. Under the conditions applied a huge and significant positive effect on survival of dopaminergic neurons was observed. NRG-1 beta-ECD is also highly neuroprotective in this model. Given the ip injection during this series of experiments, the clear efficacy also proves again that NRG-1 beta-ECD is highly efficient in passing the blood brain barrier.

(102) FIG. 6 shows the metabolic scheme which appears to be affected by NRG-1 beta-ECD administration: dopamine is converted by MAO-B to DOPAC and by COMT to 3-MT; homovanillic acid is subsequently generated form both metabolites by COMT from DOPAC and by MAO-B from 3-MT; NRG-1 beta-ECD administration is obviously regulating activities of both enzymes.

(103) Even more important and as shown in FIG. 7, in the MPTP model of Parkinson's disease there is a clear and significant neuroprotective effect of NRG-1 beta-ECD becoming apparent by histological quantification of dopaminergic neurons of the middle brain. The stereological method has been described elsewhere (Liberatore et al., 1999, Przedborski & Vila, 2003; Vila & Przedborski, 2003; Hglinger et al., 2004; Hglinger et al., 2007).

(104) Taken together, there is a surprisingly clear and beneficial neuroprotective effect in the MPTP animal model of Parkinson's disease: The effects prove again that the intraperitoneal administration of very low concentrations of NRG-1 beta-ECD (e.g. 20 ng/kg) is sufficient to achieve efficacy and thus that NRG-1 beta-ECD passes the blood brain barrier. Also the complex influence on dopamine metabolites (HPLC results; FIG. 5) points to regulation of MAO-B and COMT by NRG-1 and NRG-1 beta-ECD.

Example 7

(105) Identification of an Acidic Posttranslational Isoform of NRG-1 Beta-ECD as the Active Principle

(106) We have published evidence, that in learning and memory a particular posttranslational acidic isoform of NRG-1 beta-ECD is the active form (Schillo et al. 2005a). Here we show that similar patterns are observed in animal models of Alzheimer's disease and post-mortem brain tissue from Alzheimer's and Parkinson's disease patients. We conclude that this acidic isoform is the active principle.

(107) Methods:

(108) For staining Western blots we used the following antibodies: anti-NRG1-ECD, rabbit polyclonal (sc-28916 Lot: I 2905 Santa Cruz; H-210) Neuregulin-1 (H-210) is a rabbit polyclonal antibody raised against amino acids 21-230 mapping within an N-terminal extracellular domain of Neuregulin-1 isoform HRG- of human origin. Neuregulin-1 (H-210) is recommended for detection of Neuregulin-1 isoforms HRG-, HRG-1A, HRG-2B, HRG-3, HRG-1, HRG-2, HRG-3 (GGF), GGF2 and SMDF of mouse, rat and human origin by Western Blotting (starting dilution 1:200, dilution range 1:100-1:1000), immunoprecipitation [1-2 g per 100-500 g of total protein (1 ml of cell lysate)] and immunofluorescence (starting dilution 1:50, dilution range 1:50-1:500).

(109) Secondary antibodies were:

(110) Anti goat, HRP

(111) sc-2922 Lot: C1405 Santa Cruz

(112) Anti rabbit, HRP

(113) Sc-2054 Lot: G 2005 Santa Cruz

(114) Next to immunostaining we performed MALDI-TOF and Q-TOF mass spectrometry to confirm NRG-1 beta-ECD.

(115) Now we find a very similar pattern in APPPS mouse model of cerebral amyloidogenesis and Alzheimer's disease as shown in FIG. 8. The concentration of this particular acidic isoform of NRG-1 beta-ECD at roughly a pI of 5.0 is considerably higher in treated APPPS mice which are at the same time better learners.

(116) In FIG. 8 two representative images are shown of each, a treated good learning animal (top) and non-treated animal with inferior learning performance. (below).

(117) FIG. 9 shows the results of a Western Blot experiment using post mortem cortical material from each 9 Alzheimer patients and age-matched controls. It clearly reveals, that the NRG-1 -ECD fragment is significantly less abundant in the Alzheimer cases. As an internal control the abundance of NRG-12 was measured, which appears not to be affected by the memory-loss associated with the disease.

(118) A further investigation of the this specific Alzheimer- and memory-associated isoform of NRG-1 -ECD by Western blots of 2-dimensional gels (2D-PAGE) of the same post mortem human brain material used for FIG. 9, reveals as shown by representative examples in FIG. 10, that it is indeed the acidic isoform of NRG-1 -ECD which is diminished in the Alzheimer condition.

CONCLUSIONS

(119) We present here for the first time functional evidence of in vivo effects of posttranslational modifications of the transcript of nrg-1 gene, in particular a truncated form generated by proteolytic cleavage, comprising the extracellular domain of NRG-1 beta with MW 15-35, pI 4-10; more specifically we found an antipsychotic activity in animal models for schizophrenia, probably based on regulation of MAO-B and COMT, at concentrations of 5-600 ng/kg (i.v.). In contrast to control neuroleptics which are used at concentrations which are 100-1000-fold higher, there was no negative effect observed.

(120) Moreover we found a neuroprotective effect in MPTP model of Parkinson's disease at concentrations of 3-300 ng/kg (i.v.)

(121) Moreover we found a positive effect on memory- and learning in respective animal models (Morris water maze) for learning and cerebral amyloidosis and Alzheimer's disease

(122) Given the adverse effects of many atypical antipsychotics, currently in use (Haddad and Sharma 2007) we conclude that soluble NRG-1-ECD fragment with EGF domains of SMDF, NRG-1 alpha, but in particular NRG-1 beta might be useful as a stand-alone or co-medication for the treatment of schizophrenia, bipolar disorder and depression.

(123) It might also be used in the same sense in other diseases of the central nervous system, like neurodegenerative disorders like Alzheimer's and Parkinson's disease, amyotrophic lateral sclerosis, multiple sclerosis, stroke, traumatic brain and spinal chord injuries.

(124) Soluble NRG-1-ECD proteins have these very broad effects due to a central role in neuronal signal transduction, in particular mediating glutamate signalling and excitotoxicity, which plays a central role in all indications mentioned above (Schrattenholz and Soskic 2006).

REFERENCES

(125) 1. Arnold S E, Talbot K, Hahn C G (2005) Neurodevelopment, neuroplasticity, and new genes for schizophrenia. Prog Brain Res 147:319-345 2. Bao J, Lin H, Ouyang Y, Lei D, Osman A, Kim T W, Mei L, Dai P, Ohlemiller K K, Ambron R T (2004) Activity-dependent transcription regulation of PSD-95 by neuregulin-1 and Eos. Nat Neurosci 7:1250-1258 3. Benzel I, Bansal A, Browning B L, Galwey N W, Maycox P R, McGinnis R, Smart D, St C D, Yates P, Purvis I (2007) Interactions among genes in the ErbB-Neuregulin signalling network are associated with increased susceptibility to schizophrenia. Behav Brain Funct 3:31 4. Bertram I, Bernstein H G, Lendeckel U, Bukowska A, Dobrowolny H, Keilhoff G, Kanakis D, Mawrin C, Bielau H, Falkai P, Bogerts B (2007) Immunohistochemical evidence for impaired neuregulin-1 signaling in the prefrontal cortex in schizophrenia and in unipolar depression. Ann N Y Acad Sci 1096:147-156 5. Blackwood D H, Pickard B J, Thomson P A, Evans K L, Porteous D J, Muir W J (2007) Are some genetic risk factors common to schizophrenia, bipolar disorder and depression? Evidence from DISCI, GRIK4 and NRG1. Neurotox Res 11:73-83 6. Boissier J R, Simon P (1966) [On the potentiation of DOPA effects by monoamine oxidase inhibitors]. Psychopharmacologia 8:428-436 7. Britsch S (2007) The neuregulin-I/ErbB signaling system in development and disease. Adv Anat Embryol Cell Biol 190:1-65 8. Buonanno A, Fischbach G D (2001) Neuregulin and ErbB receptor signaling pathways in the nervous system. Curr Opin Neurobiol 11:287-296 9. Buresova O, Krekule I, Zahalka A, Bures J (1985) On-demand platform improves accuracy of the Morris water maze procedure. J Neurosci Methods 15:63-72 10. Costall B, Hui S C, Naylor R J (1978) Modulation of amphetamine hyperactivity by DPI injected into rat nucleus accumbens [proceedings]. Br J Pharmacol 64:461P 11. Eilam R, Pinkas-Kramarski R, Ratzkin B J, Segal M, Yarden Y (1998) Activity-dependent regulation of Neu differentiation factor/neuregulin expression in rat brain. Proc Natl Acad Sci USA 95:1888-1893 12. Esper R M, Pankonin M S, Loeb J A (2006) Neuregulins: versatile growth and differentiation factors in nervous system development and human disease. Brain Res Brain Res Rev 51:161-175 13. Falquet L, Pagni M, Bucher P, Hulo N, Sigrist C J, Hofmann K, Bairoch A (2002) The PROSITE database, its status in 2002. Nucleic Acids Res 30:235-238 14. Farmer A, Elkin A, McGuffin P (2007) The genetics of bipolar affective disorder. Curr Opin Psychiatry 20:8-12 15. Fischbach G D (2007) NRG1 and synaptic function in the CNS. Neuron 54:495-497 16. Geuna S, Nicolino S, Raimondo S, Gambarotta G, Battiston B, Tos P, Perroteau I (2007) Nerve regeneration along bioengineered scaffolds. Microsurgery 27:429-438 17. Glabe C (2006) Biomedicine. Avoiding collateral damage in Alzheimer's disease treatment. Science 314:602-603 18. Go R C, Perry R T, Wiener H, Bassett S S, Blacker D, Devlin B, Sweet R A (2005) Neuregulin-1 polymorphism in late onset Alzheimer's disease families with psychoses. Am J Med Genet B Neuropsychiatr Genet 139:28-32 19. Golub M S, Germann S L, Lloyd K C (2004) Behavioral characteristics of a nervous system-specific erbB4 knock-out mouse. Behav Brain Res 153:159-170 20. Guo W P, Wang J, Li R X, Peng Y W (2006) Neuroprotective effects of neuregulin-1 in rat models of focal cerebral ischemia. Brain Res 1087:180-185 21. Haddad P M, Sharma S G (2007) Adverse effects of atypical antipsychotics: differential risk and clinical implications. CNS Drugs 21:911-936 22. Hahn C G, Wang H Y, Cho D S, Talbot K, Gur R E, Berrettini W H, Bakshi K, Kamins J, Borgmann-Winter K E, Siegel S J, Gallop R J, Arnold S E (2006) Altered neuregulin 1-erbB4 signaling contributes to NMDA receptor hypofunction in schizophrenia. Nat Med 12:824-828 23. Hanninen K, Katila H, Saarela M, Rontu R, Mattila K M, Fan M, Hurme M, Lehtimaki T (2007) Interleukin-1 beta gene polymorphism and its interactions with neuregulin-1 gene polymorphism are associated with schizophrenia. Eur Arch Psychiatry Clin Neurosci 24. Hglinger G U, Breunig J J, Depboylu C, Rouaux C, Michel P P, varez-Fischer D, Boutillier A L, Degregori J, Oertel W H, Rakic P, Hirsch E C, Hunot S (2007) The pRb/E2F cell-cycle pathway mediates cell death in Parkinson's disease. Proc Natl Acad Sci USA 104:3585-3590 25. Hglinger G U, Rizk P, Muriel M P, Duyckaerts C, Oertel W H, Caille I, Hirsch E C (2004) Dopamine depletion impairs precursor cell proliferation in Parkinson disease. Nat Neurosci 7:726-735 26. Holbro T, Hynes N E (2004) ErbB receptors: directing key signaling networks throughout life. Annu Rev Pharmacol Toxicol 44:195-217 27. Hong L E, Wonodi I, Stine O C, Mitchell B D, Thaker G K (2007) Evidence of Missense Mutations on the Neuregulin 1 Gene Affecting Function of Prepulse Inhibition. Biol Psychiatry 28. Hu X, Hicks C W, He W, Wong P, Macklin W B, Trapp B D, Yan R (2006) Bace1 modulates myelination in the central and peripheral nervous system. Nat Neurosci 9:1520-1525 29. Karoutzou G, Emrich H M, Dietrich D E (2007) The myelin-pathogenesis puzzle in schizophrenia: a literature review. Mol Psychiatry 30. Kwon O B, Longart M, Vullhorst D, Hoffman D A, Buonanno A (2005) Neuregulin-1 reverses long-term potentiation at CA1 hippocampal synapses. J Neurosci 25:9378-9383 31. Li B, Woo R S, Mei L, Malinow R (2007) The neuregulin-1 receptor erbB4 controls glutamatergic synapse maturation and plasticity. Neuron 54:583-597 32. Liberatore G T, Jackson-Lewis V, Vukosavic S, Mandir A S, Vila M, McAuliffe W G, Dawson V L, Dawson T M, Przedborski S (1999) Inducible nitric oxide synthase stimulates dopaminergic neurodegeneration in the MPTP model of Parkinson disease. Nat Med 5:1403-1409 33. Mathew S V, Law A J, Lipska B K, vila-Garcia M I, Zamora E D, Mitkus S N, Vakkalanka R, Straub R E, Weinberger D R, Kleinman J E, Hyde T M (2007) {alpha}7 nicotinic acetylcholine receptor mRNA expression and binding in postmortem human brain are associated with genetic variation in Neuregulin 1. Hum Mol Genet 34. McIntosh A M, Moorhead T W, Job D, Lymer G K, Munoz M S, McKirdy J, Sussmann J E, Baig B J, Bastin M E, Porteous D, Evans K L, Johnstone E C, Lawrie S M, Hall J (2007) The effects of a neuregulin 1 variant on white matter density and integrity. Mol Psychiatry 35. Meeks T W, Ropacki S A, Jeste D V (2006) The neurobiology of neuropsychiatric syndromes in dementia. Curr Opin Psychiatry 19:581-586 36. Meyer-Luehmann M, Coomaraswamy J, Bolmont T, Kaeser S, Schaefer C, Kilger E, Neuenschwander A, Abramowski D, Frey P, Jaton A L, Vigouret J M, Paganetti P, Walsh D M, Mathews P M, Ghiso J, Staufenbiel M, Walker L C, Jucker M (2006) Exogenous induction of cerebral beta-amyloidogenesis is governed by agent and host. Science 313:1781-1784 37. Muller N, Schwarz M (2006) Schizophrenia as an inflammation-mediated dysbalance of glutamatergic neurotransmission. Neurotox Res 10:131-148 38. Nadri C, Belmaker R H, Agam G (2007) Oxygen restriction of neonate rats elevates neuregulin-1alpha isoform levels: Possible relationship to schizophrenia. Neurochem Int 51:447-450 39. Owen M J, Craddock N, Jablensky A (2007) The genetic deconstruction of psychosis. Schizophr Bull 33:905-911 40. Ozaki M, Itoh K, Miyakawa Y, Kishida H, Hashikawa T (2004) Protein processing and releases of neuregulin-1 are regulated in an activity-dependent manner. J Neurochem 91:176-188 41. Ozaki M, Sasner M, Yano R, Lu H S, Buonanno A (1997) Neuregulin-beta induces expression of an NMDA-receptor subunit. Nature 390:691-694 42. Pertusa M, Morenilla-Palao C, Carteron C, Viana F, Cabedo H (2007) Transciptional control of cholesterol biosynthesis in Schwann cells by axonal neuregulin 1. J Biol Chem 43. Przedborski S, Vila M (2003) The 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine mouse model: a tool to explore the pathogenesis of Parkinson's disease. Ann N Y Acad Sci 991:189-198 44. Radde R, Bolmont T, Kaeser S A, Coomaraswamy J, Lindau D, Stoltze L, Calhoun M E, Jaggi F, Wolburg H, Gengler S, Haass C, Ghetti B, Czech C, Holscher C, Mathews P M, Jucker M (2006) Abeta42-driven cerebral amyloidosis in transgenic mice reveals early and robust pathology. EMBO Rep 7:940-946 45. Rimer M, Barrett D W, Maldonado M A, Vock V M, Gonzalez-Lima F (2005) Neuregulin-1 immunoglobulin-like domain mutant mice: clozapine sensitivity and impaired latent inhibition. Neuroreport 16:271-275 46. Ross C A, Margolis R L, Reading S A, Pletnikov M, Coyle J T (2006) Neurobiology of schizophrenia. Neuron 52:139-153 47. Schillo S, Pejovic V, Hunzinger C, Hansen T, Poznanovic S, Kriegsmann J, Schmidt W J, Schrattenholz A (2005b) Integrative proteomics: functional and molecular characterization of a particular glutamate-related neuregulin isoform. J Proteome Res 4:900-908 48. Schillo S, Pejovic V, Hunzinger C, Hansen T, Poznanovic S, Kriegsmann J, Schmidt W J, Schrattenholz A (2005a) Integrative proteomics: functional and molecular characterization of a particular glutamate-related neuregulin isoform. J Proteome Res 4:900-908 49. Schrattenholz A, Soskic V (2006) NMDA receptors are not alone: dynamic regulation of NMDA receptor structure and function by neuregulins and transient cholesterol-rich membrane domains leads to disease-specific nuances of glutamate-signalling. Curr Top Med Chem 6:663-686 50. Schubert C (2006) Alzheimer disease: BACE1 branches out. Nat Med 12:1123 51. Scolnick E M, Petryshen T, Sklar P (2006) Schizophrenia: do the genetics and neurobiology of neuregulin provide a pathogenesis model? Hary Rev Psychiatry 14:64-77 52. Thomson P A, Christoforou A, Morris S W, Adie E, Pickard B S, Porteous D J, Muir W J, Blackwood D H, Evans K L (2007) Association of Neuregulin 1 with schizophrenia and bipolar disorder in a second cohort from the Scottish population. Mol Psychiatry 12:94-104 53. Vila M, Przedborski S (2003) Targeting programmed cell death in neurodegenerative diseases. Nat Rev Neurosci 4:365-375 54. Willem M, Garratt A N, Novak B, Citron M, Kaufmann S, Rittger A, Destrooper B, Saftig P, Birchmeier C, Haass C (2006) Control of peripheral nerve myelination by the beta-secretase BACE1. Science 314:664-666 55. Woo R S, Li X M, Tao Y, Carpenter-Hyland E, Huang Y Z, Weber J, Neiswender H, Dong X P, Wu J, Gassmann M, Lai C, Xiong W C, Gao T M, Mei L (2007) Neuregulin-1 enhances depolarization-induced GABA release. Neuron 54:599-610 56. Xie F, Padival M, Siegel R E (2006) Association of PSD-95 with ErbB4 facilitates neuregulin signaling in cerebellar granule neurons in culture. J Neurochem 57. Xu Z, Croslan D R, Harris A E, Ford G D, Ford B D (2005a) Extended therapeutic window and functional recovery after intraarterial administration of neuregulin-1 after focal ischemic stroke. J Cereb Blood Flow Metab 58. Xu Z, Croslan D R, Harris A E, Ford G D, Ford B D (2006) Extended therapeutic window and functional recovery after intraarterial administration of neuregulin-1 after focal ischemic stroke. J Cereb Blood Flow Metab 26:527-535 59. Xu Z, Ford G D, Croslan D R, Jiang J, Gates A, Allen R, Ford B D (2005b) Neuroprotection by neuregulin-1 following focal stroke is associated with the attenuation of ischemia-induced pro-inflammatory and stress gene expression. Neurobiol Dis 19:461-470 60. Xu Z, Jiang J, Ford G, Ford B D (2004) Neuregulin-1 is neuroprotective and attenuates inflammatory responses induced by ischemic stroke. Biochem Biophys Res Commun 322:440-446 61. Yang X L, Huang Y Z, Xiong W C, Mei L (2005) Neuregulin-induced expression of the acetylcholine receptor requires endocytosis of ErbB receptors. Mol Cell Neurosci 28:335-346