Complement inhibition abrogates nerve terminal injury in Miller Fisher syndrome.код для вставкиСкачать
Complement Inhibition Abrogates Nerve Terminal Injury in Miller Fisher Syndrome Susan K. Halstead, PhD,1 Peter D. Humphreys, MSc,1 John A. Goodfellow, BSc,1 Eric R. Wagner, BSc,1 Richard A. G. Smith, PhD,2 and Hugh J. Willison, FRCP1 A large body of clinical and experimental data indicate that complement activation is an important mechanism for neuronal and glial injury in Guillain–Barré syndromes. Inhibition of complement activation therefore might be expected to limit the progression of the disease. Using in vitro and in vivo models of the Guillain–Barré syndrome variant, Miller Fisher syndrome, we have shown previously that anti-GQ1b ganglioside antibodies target the presynaptic motor nerve terminal axon and surrounding perisynaptic Schwann cells, thereby mediating destructive injury through deposition of membrane attack complex. Here, we have used this model to investigate the effects of a novel therapeutic inhibitor of complement activation, APT070 (Mirococept), both in vitro and in vivo. In these models, APT070 completely prevents membrane attack complex formation, and thereby has a major neuroprotective effect at the nerve terminal, as assessed by immunohistology of perisynaptic Schwann cell and axonal integrity. These data provide a rationale for considering clinical trials of APT070 in Guillain–Barré syndrome, its variant forms, and other complement dependent neuromuscular disorders. Ann Neurol 2005;58:203–210 Serological, immunological, and pathological studies all implicate a role for complement-mediated damage to neuronal and Schwann cell membranes in Guillain– Barré syndromes (GBS).1– 8 Modeling of the GBS variant Miller Fisher syndrome (MFS) in the mouse has used a range of complement fixing anti-GQ1b antibodies.9,10 In one model, the pore-forming membrane attack complex (MAC) of complement is critical in mediating neuronal and perisynaptic Schwann cell (pSC) injury at the presynaptic neuromuscular junction (NMJ).11 Inhibition of complement might be expected to attenuate the disease process. Complement activation is negatively regulated by soluble and membrane-bound molecules including CR1, CD59, and decay accelerating factor (CD55).12 Studies on experimental allergic neuritis have shown that complement inhibition or depletion with soluble CR1 or cobra venom factor provide protection against disease advancement and tissue injury.13,14 APT070 is a novel type of complement regulator derived from a bacterially expressed region of CR1 combined with a membrane-localizing peptide that confers unique cell- and tissue-binding properties.15 APT070 protects complement-mediated tissue injury in animal models of vascular shock,15 rheumatoid arthritis,16 and renal transplant reperfusion injury.17 It is well tolerated when given systemically to healthy volunteers and by local administration to patients with rheumatoid arthritis and patients undergoing renal transplantation (Inflazyme Pharmaceuticals, British Columbia, Canada). In this study, we investigated the complement inhibitory effects of APT070 in mouse models of MFS to establish whether this agent might be a suitable candidate for treatment trials of antibody-mediated, complement-dependent forms of GBS in humans. From the 1Division of Clinical Neurosciences, Institute of Neurological Sciences, Southern General Hospital, Glasgow, Scotland; and 2 Inflazyme Pharmaceuticals, Suite 425, 5600 Parkwood Way, Richmond, British Columbia, Canada. Published online Jul 27, 2005 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/ana.20546 Received Dec 5, 2004, and in revised form May 16, 2005. Accepted for publication May 17, 2005. Material and Methods Proteins APT070 was prepared by posttranslational linkage of an N-myristoylated 17-mer peptide to a C-terminus cysteine added to the first three short consensus repeat (CCP) domains of human CR1 expressed in Escherichia coli as described previously.15–17 Bovine serum albumin (BSA) was obtained from Sigma (Poole, United Kingdom). Address correspondence to Dr Willison, University of Glasgow, Department of Neurology, Ground Floor Neurology Block, Southern General Hospital, Glasgow, G51 4TF, Scotland. E-mail: email@example.com © 2005 American Neurological Association Published by Wiley-Liss, Inc., through Wiley Subscription Services 203 Mice, Antibodies, and Sera Male Balb/c mice were obtained from Harlan (Bicester, Oxor, United Kingdom) and used at 4 to 6 weeks old. The IgM anti-ganglioside monoclonal antibody CGM3 was derived from mice inoculated with a GT1a-bearing Campylobacter jejuni lipooligosaccharide. CGM3 reacts with gangliosides GQ1b, GD3, and GT1a.18,19 This antibody has similar specificity and pathogenic effects to human MFS sera.10,20 CGM3 concentration was measured using quantitative enzyme-linked immunosorbent assay (Bethyl Laboratories, Montgomery, TX). An MFS serum positive for antiGQ1b IgG antibodies (titre ⬎1/12,500) was studied with institutional ethical approval according to local guidelines. Normal human serum (NHS) from a single donor stock was fresh frozen and stored in aliquots at ⫺70°C to preserve complement activity for use as a source of heterologous complement. CGM3, MFS serum, and NHS were dialyzed against Ringers’ solution (116mM NaCl, 4.5mM KCl, 1mM MgCl2, 2mM CaCl2, 1mM NaH2PO4, 23mM NaHCO3, 11mM glucose, pH 7.4) for 24 hours at 4°C. Immunohistology Unfixed hemidiaphragm cryostat sections (8 –20m) were air-dried, then stored at ⫺20°C. To localize NMJs, we used Texas Red and dipyrromethene boron difluoride (BODIPY)labeled ␣-bungarotoxin (␣-BTx; diluted 1/750 to 1.3g/ml; Molecular Probes, Eugene, OR) that binds to postsynaptic nicotine acetylcholine receptors (nAChr). Complement component C4 was detected with rabbit anti-C4 (1:200; Sigma), followed by fluorescein isothiocyanate (FITC)–conjugated goat anti–rabbit IgG (1:300; Southern Biotechnology Associates, Birmingham, AL). C3c was detected with FITClabeled rabbit anti-C3c (1/300; Dako, Ely, United Kingdom). MAC was detected using mouse anti–human C5b-9 (1/50; Dako), followed by FITC-conjugated goat anti– mouse IgG (1/300). Mouse IgM deposits were identified with FITC-labeled anti–mouse IgM antibodies (1/300; Southern Biotechnology Associates). All antibody incubations were for 1 hour at 4°C. S100 staining was used to label pSCs in whole-mount triangularis sterni muscle preparations (see later). Tissue was fixed with 4% formaldehyde in phosphate-buffered saline (PBS) for 20 minutes, then rinsed with PBS for 10 minutes. Muscle samples then were incubated with 0.1M glycine for 10 minutes, rinsed in PBS for 10 minutes, and incubated with rabbit anti-S100 (1/200; Dako) in blocking solution (1% goat serum, 0.4M L-lysine, 0.5% Triton X-100 [Sigma Labs, St. Louis, MO] in PBS) overnight at 4°C. Samples were rinsed three times in blocking solution for 5 minutes, incubated in goat anti–rabbit IgG–conjugated CY5 (1:200; Southern Biotechnology Associates) for 3 hours at 4°C, followed by three 5-minute rinses in blocking solution. For neurofilament (NF) staining, sections of unfixed tissue were preincubated for 1 hour at room temperature (RT) with Texas Red–conjugated ␣-BTx, rinsed, immersed in ethanol at ⫺20°C for 20 minutes, then incubated overnight at RT with the rabbit polyclonal serum 1211 (1/750; reactive with phosphorylated NF; Affiniti Research Products, Exeter, United Kingdom), followed by goat anti–rabbit IgG (1/300; Southern Biotechnology Associates). 204 Annals of Neurology Vol 58 No 2 August 2005 pSC viability was assessed using ethidium homodimer 1 (EthD-1; 2M, a membrane impermeant dye that labels with red fluorescence the nucleic acids of membranepermeable cells; Molecular Probes) as described previously.11 NMJs were identified in 15m cryostat sections by staining with bodipy-conjugated ␣-BTx, and the percentage of NMJs with EthD-1–positive nuclei at end-plates was calculated. In Vitro Nerve–Muscle Preparations Mice were killed by CO2 inhalation, subject to UK Home Office guidelines. Hemidiaphragms were dissected and prepared for in vitro studies in pregassed (95% O2/5% CO2) Ringers’ solution at RT (20 –22°C).21 The APT070 dose– response curve was established using four mouse hemidiaphragms on three separate occasions. Untreated control sections of diaphragm were removed from each muscle preparation for baseline immunohistological analysis of NF, activated complement proteins, and IgM levels. Muscles were preincubated with 48, 24, 12, or 0g/ml APT070 in Ringers’ solution (1 hour at 32°C), rinsed in Ringers’ solution, followed by CGM3 (50g/ml) plus APT070 (48, 24,12, or 0g/ml), normalized to a protein concentration of 48g/ml with BSA for 2 hours at 32°C, then 30 minutes at 4°C, then equilibrated for 10 minutes at RT. Muscle preparations then were rinsed in Ringers’ solution and exposed to 40% NHS plus APT070 (48, 24, 12, or 0g/ml) for 1 hour at RT. In subsequent bioassays, hemidiaphragm preparations were incubated with APT070 or BSA as a placebo control (48g/ml) in Ringers’ solution for 1 hour at 32°C. Hemidiaphragms were rinsed in Ringers’ solution, incubated as described earlier with CGM3 (50g/ml) or MFS serum (diluted 1:1 in Ringers’ solution) plus APT070 or BSA (48g/ ml), then exposed to 40% NHS plus APT070/BSA (48g/ ml) for 1 hour at RT. Tissue was assessed for levels of mouse IgM, C4 or C3c, MAC, NF, and pSC viability using EthD-1. For illustrations of nerve terminals containing whole NMJs with innervating nerve fibers, we removed and treated triangularis sterni muscles with CGM3, APT070 (or BSA as placebo), and NHS as described earlier. Wholemount preparations then were stained for MAC, NF, S100, and EthD-1. Passive Immunization To demonstrate APT070 regulatory effects in vivo, we conducted passive immunization of live mice as described previously.11 Mice were injected intraperitoneally (IP) with 1.5mg CGM3, followed 15 hours later by an intravenous (IV) injection of APT070 (580g), or an equivalent dose of BSA. Two hours later, mice were given 0.5ml NHS by IP injection. In a further series of experiments, a comparison was made between prophylactic treatment with APT070 (580g, followed 1 hour later by NHS) and simultaneous administration of APT070 (580g) with NHS (ie, treatment at onset of neurological injury). Animals were observed for 3 hours after dosing with NHS; then they were killed by CO2 asphyxiation before tissue removal for immunohistology. Diaphragms were assessed for levels of IgM, C3c, MAC, NF, and EthD-1 as described earlier. Image Acquisition, Quantitation, and Statistical Analysis Digital images were captured using both Zeiss Pascal (Carl Zeiss, Welayn Garden, UK) and Bio-Rad MRC1024 (BioRad, Hemel Hempstead, UK) confocal laser-scanning microscopes. Image-analysis measurements were made using Scion Image (Scion Corporation, Frederick, MD) analysis software. For quantitative analysis of IgM, C3c, MAC, and NF, three staining runs of each marker were performed on tissue from at least three individual hemidiaphragm preparations or passively immunized mice, and they were quantified as previously described.21 All studies were observer blinded. For immunohistological analysis of nonparametric data, statistical comparisons were made using the Mann–Whitney U test at a 5% level of significance. For comparison of EthD-1–positive pSCs at the NMJ, the 2 test was used at a 1% level of significance. Results Protective Effects of APT070 In Vitro To assess the dose range of APT070 that might be required for neuronal and pSC protection in the neuropathy model in vitro, we first conducted a dose finding study. APT070 was assessed across the concentration range of 48, 24, 12, and 0g/ml and administered in all incubation steps to maximize the possibility of finding a beneficial therapeutic effect. Across this APT070 dose range, there was a gradual reduction in the deposition of the intermediate complement protein, C3c, at the NMJ (Fig 1). MAC deposits were dramatically reduced to undetectable levels at all three concentrations of APT070 (see Figs 2A, B). Thus, APT070 had a more pronounced inhibitory effect on MAC deposition than C3c deposition across this concentration range. In the absence of APT070, NF immunoreactivity was abolished over nerve terminals (see Figs 1, 2C, and 2D), indicating that a severe level of MAC-mediated neuronal injury has occurred.11 In the presence of all concentrations of APT070, NF levels were protected, with a minor dose-dependent effect. Thus, tissue was protected (to levels equal to untreated control tissue) when incubated with 48g/ml APT070 ( p ⫽ 0.488). NF signal was also significantly different to untreated controls in the samples incubated with 12 ( p ⫽ 0.017) and 24g/ml ( p ⫽ 0.001). pSC death was monitored Fig 1. APT070 dose–response curve. Mouse hemidiaphragm preparations incubated with APT070 (0, 12, 24, and 48g/ml) at all stages of the in vitro preparation compared with untreated control tissue from the same mouse. An increasing concentration of APT070 results in dose-dependent reduction of the intermediate complement protein C3c. Membrane attack complex (MAC) deposits were greatly reduced at all concentrations of APT070 examined, indicating that APT070 has a stronger inhibitory effect on the C5 convertase compared with the C3 convertase. Neurofilament (NF) signal (a measure of axonal integrity) demonstrates NF preservation to control levels (p ⫽ 0.488) at 48g/ml APT070. nAChr ⫽ nicotinic acetylcholine receptor. Mann–Whitney U test: asterisk denotes significantly different from control, and number sign indicates significantly different from 0g/ml APT070 (maximal lesion). Halstead et al: Complement Inhibition in MFS 205 Fig 2. Illustrative images of whole-mount muscles demonstrating neuronal and perisynaptic Schwann cell (pSC) protection at the neuromuscular junctions (NMJs) treated with APT070. Mouse triangularis sterni in vitro preparations incubated with APT070 or bovine serum albumin (BSA) at 48g/ml in the presence of CGM3 and normal human serum (NHS). (A) Membrane attack complex (MAC; green) deposits overlying two BSA-treated NMJs (red). Severe MAC-mediated injury to the pSCs has resulted in loss of S100 labeling (blue) directly over the NMJ, although S100 labeling is still present in myelinating Schwann cells. (B) No MAC (green) is present at NMJs (red) treated with the complement inhibitor APT070, and pSC integrity is indicated by the presence of S100 labeling (blue) over the NMJ. (C, D) Neurofilament signal (green) at the NMJs (red). BSA-treated NMJs (C) show loss of neurofilament (NF) signal over the terminal suggesting axonal injury, compared with APT070-protected NMJs (D). (E, F) Ethidium homodimer (EthD-1; green) was used to assess pSC membrane integrity. (E) BSA-treated NMJs have EthD-1– positive pSC nuclei over the NMJ (red), which are protected in the presence of APT070 (F). Asterisks (E, F) indicate pSC nuclei. Scale bar ⫽ 20m. by EthD-1 uptake and was found to be extensive, as expected, in tissue exposed to CGM3 plus complement in the absence of APT070.11 Thus, in the APT070untreated control sample, 1,480 of 2,313 (64%) NMJs had 1 or more EthD-1–positive nuclei compared with 206 Annals of Neurology Vol 58 No 2 August 2005 16% (n ⫽ 2,281; 2 test, p ⬍ 0.01) at 12g/ml APT070, 3% (n ⫽ 1,879; 2 test, p ⬍ 0.01) at 24g/ml APT070, and 1% (n ⫽ 2,824; 2 test, p ⬍ 0.01) at 48g/ml APT070 (see Figs 2E, F). The concentration of CGM3 added to the organ bath was constant in all experiments, and abundant IgM deposits over nerve terminals were observed under all conditions studied (data not shown). The exploratory dose finding study indicated that APT070 administered at 48g/ml appeared amply sufficient to preserve both neuronal and pSC integrity in vitro. To validate these neuroprotective effects using an irrelevant placebo control protein (BSA) and to test for C4 inhibition (as an indicator of classic pathway activation/inhibition), we conducted in vitro hemidiaphragm preparations at 48g/ml APT070 (or BSA as placebo) in three independent experiments. In this set of experiments, levels of C4-derived products were reduced ( p ⬍ 0.01) compared with BSA placebo, as was observed for the C3 convertase product, C3c (Fig 3). This indicates that APT070 is having an effect on C4 cleavage, in addition to an effect on C3 convertase. MAC deposition also was found to be significantly reduced ( p ⬍ 0.01) compared with the BSA-treated placebo and to a greater extent than C3c or C4. This indicates that under these conditions APT070 is inhibiting C5 convertase to a greater extent than C3 convertase. APT070 is known to be capable of inhibiting both C3 and C5 convertases. As expected, immunoreactive NF levels also were significantly greater in the APT070 incubated preparation compared with the BSA-treated placebo control ( p ⬍ 0.01), indicating a level of neuronal protection with APT070. pSC death was also significantly reduced in the APT070-treated tissue (5% of NMJs with one or more EthD-1–positive nuclei; n ⫽ 1,096) compared with the BSA placebo control sample (45% of NMJs with one or more EthD-1–positive nuclei; n ⫽ 1,237; 2 test, p ⬍ 0.01). To ensure that APT070 was also protective when the source of anti-GQ1b antibody was human MFS serum, we conducted a further study by preincubation of hemidiaphragm tissue (in the presence of APT070) with MFS serum in vitro, then NHS as described earlier. In this study, APT070 was protective, as expected (Fig 4), comprising a reduction in C3c deposits, complete inhibition of MAC formation, and preservation of NF levels at control values. Protective Effects of APT070 In Vivo after Passive Immunization To demonstrate that any neuroprotective effects observed in vitro could also occur in vivo, we performed passive immunization studies (Fig 5). Mice were injected IP with CGM3, followed by either APT070 (580g IV, n ⫽ 5) or BSA (580g IV, n ⫽ 4), before NHS administration IP as a source of complement. As- Fig 3. Mouse hemidiaphragm in vitro preparations incubated with APT070 or bovine serum albumin (BSA) at 48g/ml. C4 deposits are reduced at the neuromuscular junction (NMJ) in APT070-treated tissue compared with BSA (p ⬍ 0.01), indicating that APT070 retains the similar characteristics as CR1. Membrane attack complex (MAC) deposition is greatly reduced in APT070treated tissue compared with BSA placebo control (p ⬍ 0.01). The neurofilament (NF) signal was preserved in APT070-incubated tissue compared with BSA tissue (p ⬍ 0.01), in which the signal over the NMJ was abolished. nAChR ⫽ nicotinic acetylcholine receptor. Mann–Whitney U test: asterisk indicates significantly different from BSA. Fig 4. Mouse hemidiaphragm in vitro preparations were incubated with an anti-GQ1b antibody positive human Miller–Fisher syndrome (MFS) serum plus normal human serum (NHS) in the presence of APT070 or bovine serum albumin (BSA) as placebo. Complement C3c and membrane attack complex (MAC) deposits were significantly reduced over the neuromuscular junctions (NMJs) in preparations treated with APT070 compared with BSA. The neurofilament (NF) signal was similarly preserved in the presence of APT070. nAChR ⫽ nicotinic acetylcholine receptor. Mann–Whitney U test: asterisk indicates significantly different from control, and number sign indicates significantly different from BSA. Halstead et al: Complement Inhibition in MFS 207 Fig 5. Analysis of diaphragm tissue from mice passively immunized with CGM3 and normal human serum (NHS), treated with a single intravenous injection of APT070 or bovine serum albumin (BSA; 580g, after CGM3 and before NHS administration). C3c and membrane attack complex (MAC) are reduced at the neuromuscular junctions (NMJs) of mice treated with APT070 compared with BSA as placebo (p ⬍ 0.05). Neurofilament (NF) signal over the NMJ was significantly greater in APT070-treated tissue compared with BSA (p ⬍ 0.05). nAChR ⫽ nicotinic acetylcholine receptor. Asterisk indicates significantly different from BSA. suming a completely even volume of distribution (20ml) of APT070 throughout the mouse (20gm) with no metabolism or excretion, this equates to an in vitro dose of 29g/ml. C3c deposits at NMJs were abundant in both samples, although significantly lower in APT070-injected mice ( p ⬍ 0.01). MAC deposits were massively reduced in APT070-treated mice compared with BSA placebo control animals ( p ⬍ 0.01). By corollary, the NF signal was also significantly greater in APT070-treated animals compared with BSA-treated animals ( p ⬍ 0.01), indicating a neuroprotective effect. pSC death was significantly reduced in the APT070treated tissue (4% of NMJs with one or more EthD1–positive nuclei; n ⫽ 1,104) compared with BSA placebo control samples (36% of NMJs with one or more EthD-1–positive nuclei; n ⫽ 819; 2 test, p ⬍ 0.01), thus demonstrating that APT070 also has a glial protective effect in vivo. The above experiments demonstrated a prophylactic protective effect of APT070 in vivo. To demonstrate that APT070 could be an immediately effective therapy from the point of administration onward, we conducted a further experiment in which APT070 was coadministered with NHS. This was compared with an APT070 preadministration period of 1 hour and with BSA as a placebo control. Immunohistological parameters are shown in Figure 6. Under both APT070 preadministration and coadministration conditions, C3c and MAC levels were significantly reduced compared with placebo, with APT070 preadministration 208 Annals of Neurology Vol 58 No 2 August 2005 being more effective in attenuating MAC deposition than APT070 coadministration. Similarly, NF levels were significantly protected by both preadministration and coadministration of APT070, but more so in the former condition. Discussion We have demonstrated previously that functional failure and pathological injury to terminal motor axons and pSCs is entirely dependent on the deposition of MAC, directed to the NMJ by anti-GQ1b antibody binding to nerve terminal gangliosides.11 Here, we have used APT070 to inhibit the progression of the complement cascade and preserved NMJ components. APT070 combines the three N-terminal short consensus repeats of human CR1 with the membranetargeting peptide derivative “2-addressin.”22 CR1 has both classical and alternative C3 and C5 convertase decay accelerating activity and is also a cofactor for Factor I cleavage of C4b and C3b. The N-terminal region of CR1 shares these activities, although not in the same ratio as found in intact CR1.23 Initial studies were undertaken to assess the effectiveness of APT070 in vitro, and they demonstrated a dose-dependent effect on complement inhibition. The C5 convertase regulatory activity of APT070 was more striking at the concentration range investigated than C3 convertase or C4 cleavage activity as manifested by the dramatic reduction in MAC deposits relative to C3c/C4 deposits. In these in vitro studies, administra- Fig 6. Analysis of diaphragm tissue from mice passively immunized with CGM3 (1.5mg total dose for 16 hours) followed by a single intravenous bolus of APT070 (580g total dose) either 1 hour before or concomitantly with NHS (0.5ml intraperitoneally). Complement C3c and membrane attack complex (MAC) were significantly reduced when APT070 was either preadministered or coadministered, but they were significantly more so in the former. Similarly, neurofilament (NF) immunoreactivity was significantly better preserved with preadministered APT070 compared with coadministered APT070, with both conditions being significantly superior to placebo. nAChR ⫽ nicotinic acetylcholine receptor. Mann–Whitney U test: asterisk indicates significantly different from BSA, and number sign indicates significantly different from APT070 preadministration. tion of APT070 at 48g/ml protected both axons and pSCs at the NMJ, and a similar dose was used in subsequent in vivo studies. A passive immunization model was used to investigate APT070 in vivo. In this acute model, complementmediated damage occurs rapidly after NHS administration (ie, within 1 hour). In this model, although small amounts of mouse complement products, including MAC, may be deposited at nerve terminals, the majority of MAC has its origin from the heterologous NHS that is required for disease induction in the mouse. Thus, a particular value of this model is that APT070 effects are being largely assessed on human complement activation and are therefore of greater relevance to its possible therapeutic effects in human disease. In initial in vivo studies, APT070 was given prophylactically, 2 hours before dosing with NHS, to ensure a wide distribution throughout the mouse and to create optimal conditions for a maximal protective effect. Subsequently, APT070 was given at the same time as NHS as the source of complement to create a more realistic clinical situation. In these passively immunized mice, the effect of APT070 was that MAC deposition at nerve terminals was greatly reduced, both when preadministered and coadministered with NHS, and both neuronal and pSC integrity was significantly preserved. Importantly, no treatment could be expected to acutely reverse the severe membrane injury mediated by MAC pores, but that would ideally be able to halt the progression of any ongoing MAC pore formation, thereby allowing initiation of intrinsic repair mechanisms. Clinical relevance was also assessed by using MFS serum as a source of complement-fixing anti-GQ1b antibody, and APT070 was found to be similarly protective. Comparisons with other available treatments, including IV immunoglobulin,9,24,25 were not assessed in this study but would form important considerations when designing clinical trials. APT070 contains a membrane-localizing motif, which is believed to target the complement regulatory elements of the compound to cell membranes, especially those containing acidic phospholipids.22 Previous studies on APT070 have shown that it can be incorporated into renal membranes.17 In this study, we have not directly demonstrated APT070 incorporation into mouse motor nerve terminal membranes. However, by inference from the therapeutic benefit, it is likely that APT070 has been incorporated at this site. APT070 may also have some complement regulatory activity in a soluble phase that would be occurring in the extracellular fluid environment surrounding the motor nerve terminal. Unlike molecules such as cobra venom factor, APT070 does not deplete the serum or tissues of complement components, but rather simply inhibits the progression of the activation pathway, and therefore is capable of acting locally at sites of complement activation.16 Regardless of the precise location of complement inhibition, APT070 clearly has a dramatically beneficial effect in this model system. Current treatment of GBS uses the multimechanistic approaches of plasma exchange and IV immunoglobulin therapy.26 By contrast, these data demonstrated a clearly protective role through complement inhibition for APT070 at the NMJ that may augment other approaches such as inhibition of antibody binding or se- Halstead et al: Complement Inhibition in MFS 209 lective plasma exchange.19 APT070 may be a valuable drug for the treatment of GBS, as well as other acute and chronic complement-mediated neuropathies and nerve terminal disorders in which active complement deposition is a pathophysiological feature. Disclosure Richard A. Smith has a commercial interest in Inflazyme Pharmaceuticals Ltd that freely provided the APT070 to the Willison Laboratory where these experiments were conducted. This work was supported by grants from the Wellcome Trust (060349, S.K.H., P.D.H., E.R.W., H.J.W.) and Guillain–Barré Syndrome Support Group UK (J.A.G., H.J.W.). References 1. Griffin JW, Li CY, Macko C, et al. Early nodal changes in the acute motor axonal neuropathy pattern of the Guillain-Barre syndrome. J Neurocytol 1996;25:33–51. 2. Hafer-Macko C, Hsieh ST, Li CY, et al. Acute motor axonal neuropathy: an antibody-mediated attack on axolemma. Ann Neurol 1996;40:635– 644. 3. Hafer-Macko CE, Sheikh KA, Li CY, et al. Immune attack on the Schwann cell surface in acute inflammatory demyelinating polyneuropathy. Ann Neurol 1996;39:625– 635. 4. Hartung HP, Schwenke C, Bitter-Suermann D, et al. GuillainBarre syndrome: activated complement components C3a and C5a in CSF. Neurology 1987;37:1006 –1009. 5. Koguchi Y, Yamada T, Kuwabara S, et al. Increased CSF C4d in demyelinating neuropathy indicates the radicular involvement. Acta Neurol Scand 1995;91:58 – 61. 6. Koski CL, Sanders ME, Swoveland PT, et al. Activation of terminal components of complement in patients with GuillainBarre syndrome and other demyelinating neuropathies. J Clin Invest 1987;80:1492–1497. 7. Sanders ME, Koski CL, Robbins D, et al. Activated terminal complement in cerebrospinal fluid in Guillain-Barre syndrome and multiple sclerosis. J Immunol 1986;136:4456 – 4459. 8. Wanschitz J, Maier H, Lassmann H, et al. Distinct time pattern of complement activation and cytotoxic T cell response in Guillain-Barre syndrome. Brain 2003;126:2034 –2042. 9. Jacobs BC, O’Hanlon GM, Bullens RW, et al. Immunoglobulins inhibit pathophysiological effects of anti-GQ1b-positive sera at motor nerve terminals through inhibition of antibody binding. Brain 2003;126:2220 –2234. 10. Plomp JJ, Molenaar PC, O’Hanlon GM, et al. Miller Fisher anti-GQ1b antibodies: alpha-latrotoxin-like effects on motor end plates. Ann Neurol 1999;45:189 –199. 11. Halstead SK, O’Hanlon GM, Humphreys PD, et al. Antidisialoside antibodies kill perisynaptic Schwann cells and damage motor nerve terminals via membrane attack complex in a murine model of neuropathy. Brain 2004;127:2109 –2123. 210 Annals of Neurology Vol 58 No 2 August 2005 12. Turnberg D, Botto M. The regulation of the complement system: insights from genetically-engineered mice. Mol Immunol 2003;40:145–153. 13. Feasby TE, Gilbert JJ, Hahn AF, et al. Complement depletion suppresses Lewis rat experimental allergic neuritis. Brain Res 1987;419:97–103. 14. Vriesendorp FJ, Flynn RE, Pappolla MA, et al. Soluble complement receptor 1 (sCR1) is not as effective as cobra venom factor in the treatment of experimental allergic neuritis. Int J Neurosci 1997;92:287–298. 15. Smith GP, Smith RA. Membrane-targeted complement inhibitors. Mol Immunol 2001;38:249 –255. 16. Linton SM, Williams AS, Dodd I, et al. Therapeutic efficacy of a novel membrane-targeted complement regulator in antigeninduced arthritis in the rat. Arthritis Rheum 2000;43: 2590 –2597. 17. Pratt JR, Jones ME, Dong J, et al. Nontransgenic hyperexpression of a complement regulator in donor kidney modulates transplant ischemia/reperfusion damage, acute rejection, and chronic nephropathy. Am J Pathol 2003;163:1457–1465. 18. Goodyear CS, O’Hanlon GM, Plomp JJ, et al. Monoclonal antibodies raised against Guillain-Barre syndrome-associated Campylobacter jejuni lipopolysaccharides react with neuronal gangliosides and paralyze muscle-nerve preparations. J Clin Invest 1999;104:697–708. 19. Willison HJ, Townson K, Veitch J, et al. Synthetic disialylgalactose immunoadsorbents deplete anti-GQ1b antibodies from autoimmune neuropathy sera. Brain 2004;127:680 – 691. 20. Bullens RW, O’Hanlon GM, Wagner E, et al. Complex gangliosides at the neuromuscular junction are membrane receptors for autoantibodies and botulinum neurotoxin but redundant for normal synaptic function. J Neurosci 2002;22:6876 – 6884. 21. O’Hanlon GM, Humphreys PD, Goldman RS, et al. Calpain inhibitors protect against axonal degeneration in a model of anti-ganglioside antibody-mediated motor nerve terminal injury. Brain 2003;126:2497–2509. 22. Smith RA. Targeting anticomplement agents. Biochem Soc Trans 2002;30:1037–1041. 23. Mossakowska D, Dodd I, Pindar W, et al. Structure-activity relationships within the N-terminal short consensus repeats (SCR) of human CR1 (C3b/C4b receptor, CD35): SCR 3 plays a critical role in inhibition of the classical and alternative pathways of complement activation. Eur J Immunol 1999;29: 1955–1965. 24. Buchwald B, Ahangari R, Weishaupt A, et al. Intravenous immunoglobulins neutralize blocking antibodies in Guillain-Barre syndrome. Ann Neurol 2002;51:673– 680. 25. Zhang G, Lopez PH, Li CY, et al. Anti-ganglioside antibodymediated neuronal cytotoxicity and its protection by intravenous immunoglobulin: implications for immune neuropathies. Brain 2004;127:1085–1100. 26. Czaplinski A, Steck AJ. Immune mediated neuropathies— an update on therapeutic strategies. J Neurol 2004;251: 127–137.