WO2016210431A1 - Method of treating sepsis-induced myopathy - Google Patents

Abstract

A method for treating muscle-impaired myopathy and paralysis in a patient experiencing sepsis has the step of administering a 5-HT receptor agonist to the patient. The 5-HT receptor agonist is administered orally, intravenously, intramuscularly, via soft tissue or subcutaneous injection, rectally, vaginally, intranasally, intraophthalmically, intrathecally or topically. Preferably the 5-HT receptor agonist is specific for the 5-HT2C receptor.

Description

METHOD OF TREATING SEPSIS-INDUCED MYOPATHY

TECHNICAL FIELD

[0001] This application relates generally to pharmacologic therapy and more specifically, to the treatment of sepsis-induced weakness and paralysis with a 5-HT₂ receptor agonist.

BACKGROUND

[0002] While not well publicized, sepsis is a large public health problem, filling many hospital beds and debilitating many patients. Long a mysterious, complex systemic condition with manifold manifestations, sepsis is still not well treated: It lasts a long time and has debilitating sequelae.

[0003] In the US alone, more than 750,000 people per year have sepsis with over half requiring ICU care and mechanical ventilation. As many as one third to one half of ventilated patients go on to develop global, persistent weakness. This group is estimated at 100,000 to 150,000 patients per year, which is ten times the occurrence of spinal cord injury. The number of individuals diagnosed with global, persistent weakness is believed to be underestimated because most patients are not evaluated for mild to moderate weakness following critical illness.

[0004] The severe weakness aspect of sepsis is what forces many patients into the ICU, particularly for mechanical ventilation. Debilitated patients in the ICU are at high risk for bed sores, pressure induced nerve palsies such as peroneal neuropathy, infections, particularly from central lines, IVs and feeding tubes. The longer patients spend in the ICU, the more debilitated they become, with higher chances of myopathy and neuropathy, both of which contribute to persistent weakness and increased mortality. Recovery and return to productivity are delayed as the patient slowly recuperates. Some individuals never return to full energy.

[0005] From the above problems, it can be seen that costs of ignoring persistent weakness are huge. The high incidence of prolonged weakness (just from sepsis) is huge. Assuming only 100,000 patients per year and $100,000/patient stay (extrapolated from an appropriate hospital stay in 1996), we arrive at $10,000,000,000. And that does not take into account 1) the higher end of the estimated population of global, persistent weakness; 2) undiagnosed patients; and 3) lost productivity.

1

INCORPORATED BY REFERENCE (RULE 20.6) SUMMARY

[0006] In one embodiment there is a method for treating muscle-impaired myopathy and paralysis in a patient experiencing sepsis. The method has the step of administering a 5-HT receptor agonist to the patient. The 5-HT receptor agonist is administered orally, intravenously, intramuscularly, via soft tissue or subcutaneous injection, rectally, vaginally, intranasally, intraophthalmically, intrathecally or topically. Preferably the 5-HT receptor agonist is specific for the 5-HT2C receptor.

BRIEF DESCRIPTION OF THE FIGURES

[0007] Referring to the drawings:

[0008] FIGs. 1A, IB and 1C are graphs illustrating the inability of septic rats to sustain MN firing;

[0009] FIGs. 2A, 2B and 2C are graphs showing that the inability of septic rats to sustain MN persists after resolution of sepsis;

[0010] FIGs. 3 A through 3L are graphs showing the buildup of inactivation that can trigger MN inexcitability;

[0011] FIGs. 4A, 4B and 4C are graphs showing cessation of firing in a septic MN with minimal slow Na inactivation;

[0012] FIG. 5 is a series of graphs showing subthreshold oscillations in septic MN compared to an MN model;

[0013] FIGs. 6A and 6B are graphs showing the control and septic MNs firing on descending ramps after injected current;

[0014] FIG. 7 is a graph showing that blocking PIC eliminates repetitive firing;

[0015] FIGs. 8 A, 8B, 8C and 8D are graphs showing that increasing PIC with Quipazine restored the ability to sustain firing of a MN in a septic rat; and

[0016] FIGs. 9A, 9B and 9C are graphs showing that increasing PIC with lorcaserin normalized MN firing in a rat eight days after induction of sepsis.

DETAILED DESCRIPTION

[0017] All prior use of 5-HT2 receptor agonists has focused on central actions in a host of conditions such as addiction, appetite, anxiety, cognition, memory, mood, sexual dysfunction,

INCORPORATED BY REFERENCE (RULE 20.6) sleep and vasoconstriction. No use has been reported to alleviate weakness and paralysis at the muscle in the septic ICU patient or the patient who suffers from peripheral neuropathy.

[0018] In the discovery process, we queried how we could normalize debilitating lack of muscular excitability, particularly in sepsis. We first considered the implications of our recent report: Inability to sustain firing (termed stuttering) had been described in striatal neurons (S and W). Stuttering was dependent on the balance between inward current mediated by persistent inward currents (PICs) and outward current mediated by subthreshold voltage- activated K channels (S&W), decreased PIC or increased subthreshold K current causing stuttering. Because of the availability of pharmacological agents to affect PIC, we initially focused on manipulation of PIC rather than manipulation of subthreshold K current.

[0019] To determine the role of PICs in repetitive firing of control MNs, we blocked sodium PICs by injecting riluzole (a 5-HT₂ receptor blocker specific for the 5-HT₂c receptor) intraperitoneally into two control rats. The first rat received 10 mg/kg and the second only 4 mg/kg. Both doses eliminated repetitive firing such that only one or two action potentials were generated during sustained current injection. These data confirmed the central role of PICs in the ability of MNs to fire repetitively.

[0020] We have also preliminarily tested whether increasing PIC could improve firing of septic MNs. First, we induced sepsis in three rats and a day later gave each 3 mg/kg of serotonergic agonist quipazine IP. We picked quipazine because it activates 5-HT₂c receptors as well as additional subclasses of 5-HT₂ receptors and 5-HT3 receptors. Moreover, quipazine has been reported to increase PICs in MNs. In each of the rats, 1 MN was studied prior to quipazine injection. Prior to injection, none of the MNs were able to sustain firing. However, after quipazine administration, we found that two-thirds of the studied MNs sustained firing without pausing for 5 sec. From this we concluded that quipazine, already available for human use, would be a candidate for treatment of this previously unappreciated disorder. The following examples provide the thought processes and rationales for our claimed invention.

Example 1

[0021] Reducing MN excitability within the CNS contributes to both acute and chronic weakness in septic rats. Our lab has reported proficiency in acquiring electrophysiological measures from MNs and their motor units. Our experimental approach enables in vivo examination of the two possible mechanisms by which the central nervous system (CNS) can

3

INCORPORATED BY REFERENCE (RULE 20.6) contribute to feeble motor-unit activation: (1) reduction in the intrinsic excitability of MNs (see below) and/or (2) reduction in net synaptic excitation of MNs.

[0022] Contributions of the peripheral nervous system to weakness in sepsis are well established. Peripheral neuropathy is well documented and myosin loss leading to myopathy has been reported (Williams et al., 1999). In addition, we have discovered inexcitability of both muscle and peripheral nerve (Rich et al., 1996; Rich et al., 1997; Novak et al., 2009). In this example, we will examine the relative contribution of each of these mechanisms including reduced intraspinal MN excitability to weakness during various stages of sepsis and recovery from sepsis.

[0023] Our recently published study demonstrated that during sepsis in the rat, there was a reduction in the intrinsic excitability of spinal MNs supplying the medial gastrocnemius muscle (Nardelli et al., 2013). Excitability was measured in terminal studies of anesthetized adult rats that were either healthy or made septic by cecal ligation and puncture a day earlier. Statistical comparison of MNs sampled from septic and healthy rats revealed a striking and consistent difference in repetitive firing evoked by intracellular current injection. Repetitive firing of individual MNs in septic rats was slower and more erratic such that septic MNs fired one third as many spikes during a five second current injection (Nardelli et al., 2013). None of the previously described reductions in muscle and nerve excitability displayed this dramatic and rapid in onset and thus reduced muscle and nerve excitability had not previously been considered in a rationale for function or treatment. We have extended control studies to include sham operated rats and found that while sham surgery induced defects in repetitive firing in 40% of MNs a day after surgery, the defect was milder than in septic rats where 100% of MNs fired abnormally (see Fig. 2 below, n = 4 sham surgery rats). We concluded the stress of surgery itself or mild infection from sham surgery was also a contributing factor to reduced MN excitability at early times after surgery.

[0024] To determine the relative importance and time course of all contributors to weakness that have been identified in septic rats to date, we proceeded to determine the relative contribution of myopathy, neuropathy and reduced MN excitability by measuring muscle force in response to electrical stimulation in three distinct locations in the motor unit: 1) muscle, 2) nerve and 3) the MN. We began with direct muscle stimulation for 5 seconds at 50 Hz to assess weakness due to myopathy. Direct muscle stimulation bypasses the MN, nerve and neuromuscular junction (NMJ), so any observed weakness is attributable to myopathy. In problems of excitation-contraction coupling, atrophy or death of muscle fibers,

4

INCORPORATED BY REFERENCE (RULE 20.6) force generation in response to direct muscle stimulation should be reduced. To date we have measured whole muscle force in three control rats and five rats a day after induction of sepsis. Mean force normalized to body weight was 12.1 ± 4.0 grams in controls and 5.8 + 1.0 grams in septic rats. The 50% reduction in force in septic rats suggests myopathy is a significant contributor to weakness soon after induction of sepsis. We had found a similar early development of myopathy in patients with severe sepsis (Khan et al., 2006). Therefore, likely contributors to sepsis-induced myopathy include loss of excitability, atrophy and loss of muscle myosin, which are consistent with earlier research.

[0025] We next measured force generation following 50 Hz of nerve stimulation.

Stimulation of nerve rather than muscle requires normal function of axons and NMJs to generate normal force and thus picks up defects due to neuropathy or failure of

neuromuscular transmission. We have reported experience with using repetitive stimulation to study of failure of neuromuscular transmission in a canine model of motor neuron disease. We measured generated force at the beginning of the force record during nerve stimulation to examine reduction in force due to neuropathy and at the end of the record to include the effect of failure of neuromuscular transmission. We have yet to see any decrement in force generation during nerve stimulation. Although our data were limited, it was consistent with findings in patients with no failure of neuromuscular transmission. We have not detected neuropathy in our current data set of five septic rats. Our results and those of others suggest we may detect neuropathy in a larger sample of septic rats.

[0026] After examining the contribution of myopathy and neuropathy to weakness, we next looked for a contribution of reduced MN excitability. We measured two parameters of force generation in response to current injection into a single MN: 1) the motor unit force generated by a single MN action potential and 2) the mean sustained motor unit force in response to five seconds of current injection into the MN soma. In septic (n = 4) and control (n = 4) rats, the MNs studied to date showed no reduction in force of septic motor units relative to control (Fig 1C twitch force). However, given the wide range of motor unit forces seen in control rats (reportedly 10-20 fold), it was not surprising to see no change, unlike the 50% force reduction secondary to myopathy that was seen in response to whole muscle stimulation. What was notable was that despite having similar twitch tensions, motor units in septic rats generated only 20% as much force during sustained current injection into the MN soma (Fig 1).

5

INCORPORATED BY REFERENCE (RULE 20.6) [0027] Shown in Figure 1 are firing and force generation produced by a motor unit in a control rat and a motor unit from a septic rat. The traces were selected from two MNs that both had a rheobase current of 9 nA. In both sets of traces, the current injected was 14 nA (5nA above rheobase) to allow for direct comparison of firing. During the five second current injection into the MN of the septic rat there was a dramatic failure of repetitive firing, such that average motor unit force generated is reduced by more than 80% (bottom trace A vs B). It is important to note that in addition to the reduction in average force over the five second current injection, there was an inability of motor units from septic rats to provide smooth force (bottom trace Fig IB) due to the marked variation in instantaneous firing rate (top plot, Fig IB). This was in marked contrast to the smooth force generated by motor units from control rats (Fig 1A, bottom trace) that normally resulted from steady instantaneous firing rate (top trace Fig 1A). Thus, in addition to causing weakness, the reduction in MN excitability resulted in inability to produce movements requiring precision (dexterity) or steady force (posture). This finding that rat motor units could not generate smooth force helps explain a clinical finding: We have noted weak patients who are recovering from sepsis are very shaky and cannot generate smooth movement. These clinical observations and our work in rats thus spark a hypothesis of a new underlying mechanism (unsteady MN firing).

Reduced excitability of CNS neurons provides the first potential mechanistic explanation for the CNS dysfunction of the sort that we necessarily expect in patients experiencing their encephalopathy and coma.

[0028] From data to date, we estimated the largest contributors to weakness at the one-day time point are myopathy (-50% reduction in maximal force), and reduced MN excitability (- 80% reduction in force). The relative contribution of myopathy and neuropathy to weakness in patients has not been determined, but both have been reported (Khan et al., 2006).

[0029] We have also studied the time course of reduced MN excitability (see FIGs. 2A, 2B and 2C). For these experiments rats were allowed to recover from sepsis. We injected current into MNs at days 3 and 7-8 after sepsis induction. We found problems in MN firing persisted even at the 8-day time point. The absence of significant recovery of MN excitability within the first week after sepsis induction also supports our hypothesis of reduced MN excitability as a contributor to chronic weakness and disability following sepsis.

[0030] For future studies, we will measure the relative contributions of myopathy, neuropathy and reduced MN excitability in electrophysiologic studies of rats in both the acute and chronic recovery phases of sepsis.

6

INCORPORATED BY REFERENCE (RULE 20.6) [0031] Following up on our preliminary data suggests MN excitability was not normalized 8 days following induction of sepsis (a time when sepsis has resolved), we will extend our study to 1 month and 4 months following induction of sepsis. At each time point, we will also determine the contribution of myopathy and neuropathy to weakness.

[0032] After identifying drugs that improve MN excitability (see below), we will test the effects of those drugs on force generation at various key time points identified

electrophysiologically to determine their drug impact on both acute and chronic weakness.

[0033] For this and other experiments, the effects of sepsis are studied by comparison of treated vs. control groups. Data sampled from rats in each group are pooled and tested for group differences using nested ANOVA, which identifies treatment effects as well as dependence on individual rats.

[0034] The preliminary indication is that reduced MN excitability begins early and persists. We initially expected that MN excitability would recover within a week; whereas, neuropathy and/or myopathy might persist and cause weakness at the one- and four-month time points. However, preliminary data suggest reduced MN excitability may extend for longer periods and contribute significantly to chronic weakness. If indeed reduced MN excitability persists, there would be benefit in treating it to reverse chronic weakness in patients (see preliminary data below). This hypothesis is very exciting because it opens the potential of treatment with 5-HT₂ receptor agonists, unlike other weakness causes of atrophy, myosin loss and axonal death for which there is no current therapy.

[0035] Preliminary results indicate that myopathy contributes to both acute and chronic weakness: This conclusion was supported by finding reduced force generation in response to direct stimulation of muscle at both immediate and later time points after sepsis induction. If we continue to obtain this outcome we will perform Western blots of myosin to look for loss of muscle myosin as a contributor to myopathy. Loss of myosin from muscle has previously been reported with sepsis. We have reported extensive experience with Western blots of myosin in critical illness myopathy. We have also performed immunostaining of muscle to look for atrophy and disorganization of sarcomeres in a rat model of critical illness myopathy.

[0036] Neuropathy may also contribute to chronic weakness: If we find that reduction in the force generated by a single nerve stimulus is greater than the reduction following direct muscle stimulation at long time points, we will conclude neuropathy contributes to chronic

7

INCORPORATED BY REFERENCE (RULE 20.6) weakness. In this case we will perform histologic analysis of nerve as we have done previously in septic rats (Novak et al., 2009).

[0037] The etiology of weakness may evolve over time: The three outcomes listed above are not exclusive. Preliminary data supports both reduced excitability and myopathy while published studies in rat support outcome neuropathy (Cankayali et al., 2007; Novak et al., 2009). Our study will specify how the relative contributions to weakness of myopathy, neuropathy and MN excitability evolve over time.

Example 2

[0038] Here we will further test the hypothesis that reduction in the ratio of persistent inward current (PIC) to subthreshold K current underlies reduced excitability of the CNS portion of MNs in sepsis.

[0039] During recording of firing from septic MNs, we consistently noted a "stuttering" pattern of firing that was very rarely observed in -50 MNs examined by us in control rats for other studies (Nardelli et al., 2013). Identifying the mechanism of this abnormal sensitivity to repetitive firing is key to understanding how sepsis causes weakness. Evidence outlined below suggests a novel mechanism.

[0040] The goal of this example aim is to determine the changes in active membrane properties that underlie inability of MNs to sustain firing. We will take a three-pronged approach to generate and test hypotheses: 1) Record from control and septic MNs using square pulses and ramp depolarizations to determine changes in membrane properties that are associated with reduced excitability. 2) Use a compartmental MN model to determine theoretical constraints on potential mechanisms underlying firing-induced reduction of excitability. 3) Use the dynamic clamp technique to counteract abnormal conductance that may underlie inexcitability. Our preliminary data using these approaches points to a defect in slowly gating, subthreshold, voltage-gated conductance as the cause of inability to sustain firing in septic rats.

[0041] We have begun a systematic analysis of the most likely candidate mechanisms for inability to sustain firing. We began by determining if we could induce a failure to sustain firing in control MNs. A control MN was stimulated by injecting current into the soma for 2 minutes (Fig 3 A-C). Action potential (AP) amplitude progressively decreased and the MN stopped firing after 95 seconds (arrow on black trace). In 3C the rate of AP rise (dV/dt), an

8

INCORPORATED BY REFERENCE (RULE 20.6) index of the amount of sodium current available, is plotted vs time to allow estimation of slow inactivation. Once slow inactivation has reduced the amount of sodium current to 20% its original value (90 vs 450), the MN can no longer fire.

[0042] We modeled build-up of slow inactivation to determine whether we could mimic the findings from our intracellular recording using a compartmental MN model (Powers et al., 2012) that was modified to represent the properties of rat MNs. The model included persistent inward current (PIC) and transient Na currents, a delayed rectifier K current and a Ca-activated K current to produce the post-spike after-hyperpolarization (AHP). Figs. 3D-F show the response of a model MN with slow inactivation present to a prolonged 10 nA current pulse. We speeded up simulation speed and the effects of slow inactivation using a one-compartment model, and the rate of slow inactivation was increased 10-fold, reducing the time scale by 10-fold. Because Na current had to charge a large capacitance in the one- compartment model, the rate of rise of the AP was slower than in the neuron in which the Na current was just charging the soma and proximal dendrite membrane. However, the model reproduced the finding that discharge fails at nine seconds, after a large decrease in spike height and rate of rise. If slow inactivation was removed (Fig. 3G-I), the model MN became able to sustain firing throughout the current injection.

[0043] We next tested whether reduction of slow inactivation prevented failure to sustain firing of control MNs in vivo. Dynamic clamp was applied as needed to increase the AHP to prevent buildup of sodium channel slow inactivation. This application of dynamic clamp made the MN (that was unable to sustain firing in 3A) able to sustain firing throughout stimulation (Fig 3J). AP amplitude and rate of rise were preserved, suggesting that the reason the MN was able to keep firing was that sodium channels had not been slow inactivated (Fig 3K, 3L). Thus both modeling and dynamic clamp identified slow inactivation as a candidate for cessation of firing following prolonged current injection into control MNs.

[0044] Is buildup of slow inactivation the mechanism underlying failure to sustain firing in septic MNs? In septic MNs there are two differences that argue against slow inactivation as the sole mechanism underlying reduced excitability. 1) The time course of inability to sustain firing in septic MNs is much faster than in control MNs (Fig 4A, 1-3 seconds vs ~ 2 minutes, Fig 3A). 2) Looking at the characteristics of individual APs in septic MNs, we see that failure to sustain firing was associated with a much milder reduction in AP amplitude and rate of rise than with slow inactivation in control MNs (Fig 4B vs Fig 3B). In septic MNs dV/dt was still above 200 mV/ms when firing fails. In the control, dV/dt was approximately

9

INCORPORATED BY REFERENCE (RULE 20.6) 90 mV/ms at point of spike failure (Fig 4C vs Fig 3C). We were unable to model failure of spiking due to slow inactivation alone without large reduction in AP amplitude and rate of rise. We tested the role of slow inactivation in pausing of septic MNs using dynamic clamp to increase the AHP and were unable to prevent pausing (data not shown). Thus, recording of spikes, modeling of data and dynamic clamp all suggest slow inactivation does not have the characteristics to account for the failure to sustain MN firing in sepsis. This is a clear example of the power of testing hypotheses in an animal model which saved us from pursuing drugs that decrease slow inactivation of sodium channels.

[0045] Next we analyzed the role of subthreshold conductances in regulation of MN excitability. If failure to sustain firing in septic MNs was not due to buildup of slow Na inactivation, what other mechanism caused the defect in firing? Clues were obtained by examining unique features of MN firing in septic rats. In septic MNs there are subthreshold oscillations in membrane potential between spiking (Fig 5- left panel). The oscillations in septic MNs are similar to oscillations reported in mouse MNs in a low excitability state (Iglesias et al, 2011). Iglesias et. al. suggested oscillations are due to a low ratio of PICs to subthreshold voltage-dependent K currents. Using dynamic clamp to either increase PIC or reduce subthreshold K conductance eliminated the oscillations (Iglesias et al., 2011). In confirmation of their findings, we could not model subthreshold oscillations using either PIC or subthreshold K conductance alone. Only when we included both conductances were we able to reproduce oscillations similar to those in septic MNs (Fig 5-right panel).

[0046] Activation of subthreshold currents has been studied previously in rat MNs using slowly rising and falling triangular injected current waveforms. Ramp depolarization in rat MNs turns on both persistent inward currents determines the rate of firing and the current at which firing turns on and off (Hamm et al., 2010). When inward currents predominate, their activation allows MNs to continue firing at a lower level of current on the descending limb of the command than was required to initially recruit firing. They may additionally fire at higher rates during the descending limb (Turkin et al. 2010). Figure 6A shows an example of a frequency-current (F-I) plot in a normal rat MN in which Δ 1= 3.5 nA. Δ I = recruitment I - de-recruitment I, and is a positive number when PIC is activated. In contrast, Fig. 6B shows a septic MN in which Δ I = -2.2 nA indicating predominance of subthreshold K current over PIC. Note the septic MN firing rate is much more variable on both the rising and falling phase than in the control. To date we have studied the response to current ramps of 20 MNs (6 naive control, 4 sham surgery control, and 10 septic rats). In septic rats mean Δ I = -0.5 nA

10

INCORPORATED BY REFERENCE (RULE 20.6) ± 2.6 nA (SD); whereas, mean Δ I in control MNs = +1.1 + 1.3 nA. Data from the four sham surgery MNs showed a mixed phenotype. More sham surgery rats will need to be studied to make a conclusion about its effect. Septic MN responses are quite variable. The range of Δ I responses in septic MNs was -10 to 2.5 nA vs a range of -1.8 to 3.5 nA for control MNs. In the proposed experiments, we will systematically compare the frequency of occurrence of different patterns of F-I behavior and values of Δ I between control, septic and sham surgery MNs.

[0047] In other tests, the above sequence of recording, followed by modeling to generate hypotheses, which are then tested using dynamic clamp, is the approach we will take in this example to characterize subthreshold currents.

[0048] Analysis of subthreshold currents from recordings: We use the bridge balance mode to measure instantaneous firing frequency and hysteresis in control and septic MNs (Hamm et al., 2010). We also use the discontinuous current clamp mode to look for slow changes in baseline potential during pauses in firing of septic MNs to determine whether subthreshold K conductance is increased or reduced during pauses. If we see slow changes in baseline potential, we intersperse brief hyperpolarizing pulses during stimulation to measure changes in input resistance that would be predicted by slow activation of subthreshold voltage activated K conductance.

[0049] Modeling of MN excitability: We use data from our recordings in Experiment 1 as a guide to determine whether our understanding is sufficient to accurately model behavior of both control and septic MNs. Specifically, we determine the amplitude and activation kinetics of PICs and subthreshold K conductances necessary to model pauses in firing that last up to 2 seconds.

[0050] Dynamic clamp to test hypotheses generated by Experiments 1 and 2: The detailed modeling from Experiment 2 informs us as to the key parameters that could induce stuttering in firing and reduce excitability of septic MNs. Based on our current preliminary data, we predict key parameters will be reduction in PIC and an increase in subthreshold K

conductance. Proceeding along that line, 1) we use dynamic clamp to increase PIC and reduce subthreshold K conductances in septic MNs to determine whether we can restore the ability of MNs to sustain firing; and 2) we reduce PIC and increase subthreshold K in control MNs to determine whether we can trigger pauses in firing that mimic those in septic MNs.

11

INCORPORATED BY REFERENCE (RULE 20.6) [0051] While testing with the dynamic clamp can yield useful information, it is important and difficult to properly position the claim, particularly in dendrites. However, this concern is lessened by the fact that we are studying differences in excitability in response to current injection in the soma of MNs. Thus, the currents underlying the defect in excitability that we see must be near the MN soma and can thus be reasonably simulated by dynamic clamp.

[0052] One possibility is that PIC is reduced relative to subthreshold K conductance. Four pieces of evidence support this hypothesis: 1) A more negative value of Δ I in septic MNs in response to ramp depolarization. 2) Modeling showing that reduction in PIC or an increase in K conductance can trigger stuttering. 3) Dynamic clamp data showing that either increasing PICs or reducing subthreshold K currents improves firing in septic MNs. 4) Dynamic clamp showing that reducing PICs or increasing subthreshold K currents in control MNs mimics the failure to sustain firing in septic MNs. These outcomes strongly support the approach (see below) to develop therapy for weakness.

[0053] Modeling the defect in MN excitability provides strong support for the possibility that we have identified the mechanism underlying inability of MNs to sustain firing. This provides the first ever demonstration of a cellular defect of excitability in the CNS secondary to sepsis. Furthermore, reduction in the ratio of PIC to K conductance explains why patients have such severe trouble recruiting motor units. PICs are thought to function to amplify synaptic currents to help MNs reach threshold. In rats, we bypass dendrites by injection of current into the soma so we are underestimating the effects of reduced PIC on recruitment of MNs.

[0054] The most important alternative outcome would be finding that neither PIC nor subthreshold K currents are involved in the inability to sustain firing as this would change our approach to developing therapy. While this seems unlikely given preliminary data provided below, in this case we would consider the possibility that a hyperpolarized shift in the voltage dependence of sodium channel inactivation contributes to the reduction in MN excitability. In previous studies where we identified an increase in sodium channel inactivation as the mechanism underlying reduced excitability, we found abnormalities of single action potentials fired from rest. The lack of abnormality in single action potentials suggests the mechanism underlying reduced MN excitability is distinct from the hyperpolarized shift in sodium channel inactivation. However, two recent computational studies suggest a hyperpolarized (left) shift in activation and inactivation of sodium channels, when coupled with reduced efficacy of the Na-K ATPase, can lead to stuttering in firing of neurons with

12

INCORPORATED BY REFERENCE (RULE 20.6) pauses that last for seconds. These studies raise the possibility that reduced excitability in MNs involves multiple mechanisms, but shares the shift in sodium channel inactivation seen in peripheral nerve and muscle. We would probe this possibility by adding in Na-K ATPase to the model and varying its efficacy to determine whether we can recreate single action potentials that are normal, but induce stuttering during repetitive firing.

[0055] This outcome would raise the possibility that the same change in sodium channel behavior that we identified in skeletal muscle, occurs in MNs. In this case, we would consider treating with drugs such as L-NAME that block nitric oxide synthase as our work in skeletal muscle shows denervated muscle excitability is improved in the NOS-1 knockout mouse and thus suggests blocking nNOS will improve excitability.

Example 3

[0056] Develop therapy that normalizes MN excitability. Inability to sustain firing (termed stuttering) has been previously described in striatal neurons. The presence of stuttering was dependent on the balance between inward current mediated by PICs and outward current mediated by subthreshold voltage- activated K channels (Sciamanna and Wilson, 2011). Either decreasing PIC or increasing subthreshold K current could convert a continuously firing striatal neuron into one that stuttered. Although PIC is thought to be carried by both Na and Ca channels, pharmacological agents that increase and decrease PIC are available.

Because of the ease of pharmacologically manipulating PIC we are initially focusing in this Aim on manipulation of PIC rather than manipulation of subthreshold K current.

[0057] Manipulation of PICs: To determine the role of PICs in repetitive firing of control MNs we blocked sodium PICs by injecting riluzole (Harvey et al., 2006b; Kuo et al., 2006) intraperitoneally into two control rats. In the first rat we injected 10 mg/kg and in the second we lowered the dose to 4 mg/kg. In both rats repetitive firing was eliminated such that only 1 or 2 APs were generated during sustained current injection (Fig 7, n = 6 MNs). These data confirm the central role of PICs in ability of MNs to fire repetitively (Harvey et al., 2006a; Harvey et al., 2006b; Kuo et al., 2006).

[0058] We have also preliminarily tested whether increasing PIC improves firing of septic MNs. One day after induction of sepsis we gave three rats 3mg/kg of the serotonergic agonist quipazine IP. Quipazine activates 5-HT₂c receptors as well as additional subclasses of 5-HT₂ receptors (Smith et al., 1995) and 5-HT receptors (Cappelli et al., 1998). Quipazine has been reported to increase PICs in MNs (Smith et al., 1995; Cappelli et al., 1998; Kim et al., 1999;

13

INCORPORATED BY REFERENCE (RULE 20.6) Brumley and Robinson, 2005). In each of the 3 rats, 1 MN was studied prior to injection of quipazine and consistent with data shown in Fig 2, 100% of the MNs were unable to sustain firing. After quipazine we were able to record from one MN in each rat. Two thirds of MNs sustained firing after injection of quipazine. In one rat we recorded from a single MN throughout the period before and after injection of quipazine. Shown in Fig 8 A-C is the response of this MN to a five second current injection (6 nA above rheobase for all time points). Before quipazine the MN did not sustain firing. After quipazine the MN fired without pausing for five seconds. Shown in Fig 8D is the response of this MN to ramp current injection before and after quipazine. Prior to quipazine the MN was barely able to fire on the descending ramp and had a Δ I of -6 nA. After quipazine, firing on the descending phase was greatly improved and Δ I increased to -2nA. These data strongly suggest that treatment with quipazine improves repetitive firing of MNs from septic rats. Note that the variability of firing rate on both the ascending and descending ramp was also reduced following quipazine.

[0059] We have also treated one rat 8 days after induction of sepsis with 3 mg kg of the selective 5-HT₂c agonist lorcaserin (Thomsen et al., 2008; Higgins et al., 2012). Lorcaserin (trade name BELVIQ) is now FDA approved for weight loss (Witkamp, 2011). 5-HT₂c receptors appear to be particularly important neuromodulatory receptors involved in regulation of PIC as activation of 5-HT₂c receptors greatly amplifies reflex strength

(Machacek et al., 2001) and has been proposed to underlie increased MN excitability that underlies spasticity following spinal cord injury (Murray et al., 2010). We held a single MN impalement for 2 hours and recorded the firing pattern before and at various times after injection of lorcaserin (Figure 9). In full recognition that this is a sample of 1 and needs further validation, we observe that the similarity of the responses to lorcaserin and quipazine is encouraging. Prior to injection of lorcaserin several abnormalities were present: 1) The MN was unable to sustain steady firing, 2) There was trial to trial variability in response to the same current injection (compare traces in A and B), 3) There was marked variability in instantaneous firing rate (IFR, top traces of A, B and C). 1.5 hours after injection of lorcaserin, recording from the same MN demonstrates a number of improvements in firing (Fig 9C): 1) the MN sustained firing throughout the 5s current injection, 2) trial-to-trial variability was greatly reduced (traces not shown), 3) variability in IFR was greatly reduced (top trace) and 4) IFR was increased.

[0060] Next steps call for testing drugs to modulate PIC: Our preliminary data suggests reduction in PICs may be an important contributor to reduced excitability of septic MNs. We

14

INCORPORATED BY REFERENCE (RULE 20.6) will identify a dose of riluzole in control rats that recreates the stuttering firing pattern and subthreshold oscillations present in MNs in septic rats. When we find a dose of riluzole that induces stuttering in MNs from control rats, this indicates reduction of PIC is sufficient to account for the defect in MNs from septic rats. We will also determine whether increasing PICs with quipazine and lorcaserin can correct firing patterns of septic MNs.

[0061] Further plans are laid for expected results: 1) Lorcaserin restores normal MN excitability. If lorcaserin normalizes MN excitability in septic rats and improves motor unit force generation as outlined above, we move to identify and treat patients recovering from critical illness. The team has a neuromuscular specialist who has performed a number of clinical studies in patients with ICU acquired weakness and collaborates with a number of clinical leaders in the field. The team is well positioned to start a multi-center clinical trial.

[0062] 2) If quipazine, but not lorcaserin restores normal MN excitability. While this suggests increasing PICs by activating 5-HT receptors is a good therapeutic strategy, it also suggests the effect of quipazine is through a class of 5-HT receptors other than 5-HT₂c. In this case we look for and test FDA approved drugs that activate the subclasses of 5-HT receptors activated by quipazine.

[0063] 3) If neither lorcaserin nor quipazine restore normal MN excitability. This suggests serotonergic agonists are not likely to be effective in treating weakness. There are two reasons serotonergic agonists might have little effect on MN excitability: 1) there could be down-regulation of the serotonergic receptor subtypes that trigger increases in PIC or 2) our hypothesis of the mechanism underlying reduced MN excitability is incorrect. If we find 5-HT agonists have little impact on MN excitability we re-examine our results from dynamic clamp and modeling. If they clearly suggest increasing PIC should normalize firing, we conclude that 5-HT receptors that modulate PIC are down-regulated on MNs and next try norepinephrine agonists to increase PICs (Heckman et al., 2003; Machacek and Hochman, 2006; Heckman et al., 2008a).

[0064] Alternative approach to normalize MN excitability if increasing PIC alone is insufficient: If attempts to increase PIC do not normalize firing of septic MNs, we turn our attention to blocking subthreshold K currents. We can block the Kvl K channel family as they mediate stuttering (Sciamanna and Wilson, 2011). Use of dendrotoxin to block Kvl channels (Southan and Robertson, 2000) requires infusion into the lumbosacral spinal canal. 4-aminopyridine has been given to systemically to block Kvl channels (Baker et al., 2011),

15

INCORPORATED BY REFERENCE (RULE 20.6) but results need to be interpreted in light of 4-AP's additional block of fast gating K currents (Johnston et al., 2010). We could also target the KCNQ family of channels (Kv7 channels) as these channels can mediate a switch between phasic and tonic firing (Brown and

Passmore, 2009). Kv 7.2 and Kv7.3 can be blocked by systemic administration in vivo of selective blockers such as linopirdine, XE991 or DMP-543 (Miceli et al., 2008; Brown and Passmore, 2009; Ipavec et al., 2011).

Example 4

[0065] While the above examples utilized available 5-HT₂ receptors agonist quipazine and lorcaserin, we have not determined if those are the optimal existing compounds for treating sepsis-induced motor weakness. In this example, we screen the available drugs and their analogs against our animal models to establish the most likely candidates. While we start the screening with FDA-approved drugs to help patients as soon as possible, we also screen other as yet unapproved analogs for improved activity. We also develop a human cell model for screening or obtain such from experts in the 5-HT₂ receptor agonist pharmacology.

[0066] Reference throughout this specification to an "embodiment," an "example" or similar language means that a particular feature, structure, characteristic, or combinations thereof described in connection with the embodiment is included in at least one embodiment of the present invention. Thus appearances of the phrases an "embodiment," and "example," and similar language throughout this specification may, but do not necessarily, all refer to the same embodiment, to different embodiments, or to one or more of the figures. Additional, reference to the words "embodiment", "example" or the like for two or more features, elements, etc., does not mean that the features are necessarily related, dissimilar, the same, etc.

[0067] Each statement of an embodiment or example is to be considered independent of any other statement of an embodiment despite any use of similar or identical language characterizing each embodiment. Therefore, where on embodiment is identified as "another embodiment," the identified embodiment is independent of any other embodiments characterized by the language "another embodiment." The features, functions and the like described herein are considered to be able to be combined in whole or in part one with another as the claims and/or art may direct, either directly or indirectly, implicitly or explicitly.

16

INCORPORATED BY REFERENCE (RULE 20.6) [0068] As used herein, "comprising," "including," "containing," "is," "are," "characterized by," and grammatical equivalents thereof are inclusive or open-ended terms that do not exclude additional un-recited elements or method steps. "Comprising" is to be interpreted broadly and including the more restrictive terms "consisting of and "consisting essentially of."

[0069] Reference throughout this specification to features, advantages, or similar language does not imply that all of features and advantages that may be realized with the present invention should be or are in any single embodiment of the invention. Rather, language referring to the features and advantages is understood to mean that a specific feature, advantage or characteristic described in connection with an embodiment is included in at least one embodiment of the present invention. Thus, discussion of the features and advantages, and similar language, throughout this specification may, but does not necessarily, refer to the same embodiment.

[0070] Furthermore, the described features, advantages, and characteristics of the invention may be combined in any suitable manner in one or more embodiments. One skilled in the relevant art will recognize that the invention can be practiced without one or more of the specific features or advantages of a particular embodiment. In other instances, additional features and advantages may be recognized in certain embodiments that may not be present in all embodiments of the invention.

17

INCORPORATED BY REFERENCE (RULE 20.6)

Claims

1. A method for treating muscle-impaired myopathy and paralysis in a patient

experiencing sepsis, the method comprising administering a 5-HT receptor agonist to the patient.

2. The method of claim 1 wherein the 5-HT receptor agonist is administered orally.

3. The method of claim 1 wherein the 5-HT receptor agonist is administered

intravenously.

4. The method of claim 1 wherein the 5-HT receptor agonist is administered

intramuscularly or via soft tissue or subcutaneous injection.

5. The method of claim 1 wherein the 5-HT receptor agonist is administered rectally.

6. The method of claim 1 wherein the 5-HT receptor agonist is administered

vaginally.

7. The method of claim 1 wherein the 5-HT receptor agonist is administered

intranasal.

8. The method of claim 1 wherein the 5-HT receptor agonist is administered

ophthalmically.

9. The method of claim 1 wherein the 5-HT receptor agonist is administered

intrathecally.

10. The method of claim 1 wherein the 5-HT receptor agonist is administered

topically.

11. The method of claim 1 wherein the 5-HT receptor agonist comprises a 5-HT₂c receptor agonist.

12. A method for treating muscle impaired myopathy and paralysis in a patient who suffers from peripheral neuropathy, the method comprising administering a 5-HT receptor agonist to the patient.

13. The method of claim 12 wherein the 5-HT receptor agonist is administered orally.

14. The method of claim 12 wherein the 5-HT receptor agonist is administered

intravenously.

15. The method of claim 12 wherein the 5-HT receptor agonist is administered

intramuscularly or via soft tissue or subcutaneous injection.

16. The method of claim 12 wherein the 5-HT receptor agonist is administered

rectally.

18

INCORPORATED BY REFERENCE (RULE 20.6)

17. The method of claim 12 wherein the 5-HT receptor agonist is administered vaginally.

18. The method of claim 12 wherein the 5-HT receptor agonist is administered intranasally.

19. The method of claim 12 wherein the 5-HT receptor agonist is administered ophthalmically.

20. The method of claim 12 wherein the 5-HT receptor agonist is administered intrathecally.

21. The method of claim 12 wherein the 5-HT receptor agonist is administered topically.

22. The method of claim 12 wherein the 5-HT receptor agonist comprises a 5-HT₂c receptor agonist.

23. The method of claim 1 wherein the 5-HT receptor agonist comprises a 5-HT₂c receptor agonist.

19

INCORPORATED BY REFERENCE (RULE 20.6)