Myotoxic Effects of the Skeletal Muscle-Specific Immunotoxin, Ricin-mAb35, on Orbicularis Oculi Muscle After Eyelid Injections in Rabbits

Transcription

1 Ophthalmic Plastic and Reconstructive Surgery Vol. 20, No. 4, pp The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc. Myotoxic Effects of the Skeletal Muscle-Specific Immunotoxin, Ricin-mAb35, on Orbicularis Oculi Muscle After Eyelid Injections in Rabbits Andrew R. Harrison, M.D.*, Stephanie Skladzien, B.A.*, Stephen P. Christiansen, M.D.*, and Linda K. McLoon, Ph.D.* *Departments of Ophthalmology, Otolaryngology, Pediatrics, and Neuroscience, University of Minnesota, Minneapolis, Minnesota, U.S.A. Purpose: The authors recently demonstrated that a single injection of the immunotoxin ricin-mab35 has potent and long-lasting myotoxic effects in extraocular muscles. The myotoxicity of injected ricin-mab35 was tested in the eyelids of rabbits to determine its potential for use in the treatment of benign essential blepharospasm and other dystonias. Methods: The immunotoxin ricin-mab35 was injected in one eyelid of adult rabbits. After 1 week, 1 month, or 6 months, the rabbits were euthanized, and the eyelids were prepared for histologic examination of inflammatory cell infiltrate with immunohistochemical localization of cd11b and myosin heavy chain isoform expression. Muscle loss was quantified by analysis of muscle fiber cross-sectional area and total myofiber number. Results: Within the first week after a single injection of ricin-mab35, some edema developed, which resolved by the second week. Otherwise, the eyelids were normal in appearance. A short-lived inflammatory response was seen at 1 week, but this resolved 1 month after treatment. One week after injection, there was a significant decrease in the total number of orbicularis oculi myofibers in the ricin-mab35 treated eyelids. This myofiber loss remained significant 1 month later and was maintained 6 months after the initial injection. Conclusions: Direct injection of the immunotoxin ricin-mab35 resulted in significant, acute muscle loss in the orbicularis oculi of rabbits that was maintained for up to 6 months. Physiologic studies are needed to demonstrate concomitant loss of muscle strength, but these results suggest that ricin-mab35 injection holds promise as a muscle-weakening agent in the eyelid. Facial botulinum toxin injections have become the mainstay treatment for patients with muscle spasm diseases that involve the eyes and face, including benign Accepted February 25, Supported by the Benign Essential Blepharospasm Research Foundation, the Minnesota Lions and Lionesses, and an unrestricted grant to the Department of Ophthalmology from Research to Prevent Blindness, Inc., New York, New York, U.S.A. Address correspondence and reprint requests to Dr. Linda McLoon, Departments of Ophthalmology and Neuroscience, University of Minnesota, Room 374, Lions Research Building, th Street SE, Minneapolis, MN 55455, U.S.A. mcloo001@tc.umn.edu DOI: /01.IOP D essential blepharospasm and hemifacial spasm. 1,2 The overwhelming acceptance by patients and their physicians attests to the therapeutic effects of these treatments and their safety and ease of administration. Although effective, one of the concerns with the use of botulinum toxin injections is that injections often become less effective over time, and patients require decreasing time intervals between subsequent injections. 3,4 In some patients, antibodies to botulinum toxin develop, and this correlates with decreasing effectiveness of the treatments. 5 There also is a group of patients who are unresponsive to botulinum toxin

2 IMMUNOTOXIN TREATMENT OF RABBIT EYELIDS 313 Recent studies in extraocular muscle suggest that ricin-mab35, an immunotoxin targeted to skeletal muscle, might have a longer treatment effect than botulinum toxin. 7,8 The ricin-mab35 molecule is composed of the toxin ricin, which has been conjugated to an antibody to the nicotinic acetylcholine receptor, which targets the immunotoxin specifically to adult skeletal muscle fibers. 9 The mechanism of action for muscle weakening caused by the ricin-mab35 is direct myotoxicity to muscle fibers. A single injection of ricin-mab35 in rabbit extraocular muscles results in a significant longterm muscle fiber loss 8 and a significant reduction in muscle force compared with the contralateral control muscles for up to 24 weeks. 10 Thus, injection of this immunotoxin could add a potential therapeutic modality for treatment of patients with muscle spasm diseases of the eyelids and face, especially for those in whom a longer-term treatment effect is desired or for those who are or have become nonresponders to botulinum toxin treatment. The purpose of this study was to examine the histologic effects of ricin-mab35 injection in the eyelids of rabbits to determine the potential usefulness of this agent for the treatment of muscle spasm diseases of the face and other dystonias. METHODS This study involved the injection of ricin-mab35 in the upper eyelids of adult New Zealand White rabbits obtained from Bakkon Rabbitry (Viroqua, WI, U.S.A.) and housed with Research Animal Resources at the University of Minnesota. This study was approved by the Institutional Animal Care Committee at the University of Minnesota and adhered to the Guidelines of the National Institutes of Health for Use of Animals in Research. Rabbits were anesthetized with a 1:1 volume of ketamine (10 mg/kg) and xylazine (2 mg/kg). After a suitable level of anesthesia was achieved, proparacaine HCl was placed in the conjunctival cul-de-sac to reduce blinking. Ricin mab35, 9 in a volume of 1 ml, was injected from the medial to lateral canthus of the upper eyelids with a 30-gauge needle. Based on our previous studies, ricinmab35 was injected in the upper eyelids of 4 rabbits at a dose of 1/5 and 8 rabbits at 1/10 maximum tolerated dose for mice. 7 An identical procedure was performed on the contralateral eyelids, substituting an equal volume of normal saline for the immunotoxin. The treatment and control eyelids were randomized. The rabbits were monitored daily for eyelid changes and any systemic reactions to the immunotoxin. One week, 1 month, and 6 months after a single injection of the ricin-mab35, the treated rabbits were euthanized with an overdose of barbiturate anesthesia. The eyelids were removed, embedded in tragacanth gum, and frozen in methylbutane that was chilled to a slurry on liquid nitrogen. The eyelids were sectioned at 12 m and stored at 80 C. The sections were processed for the immunohistochemical localization of cd11b-positive cellular infiltrate, specifically neutrophils, monocytes, and macrophages. The tissue sections were fixed for 10 minutes in 10% formalin and quenched in hydrogen peroxide to remove endogenous peroxidase. After a phosphate-buffered saline rinse, the sections were blocked with normal horse serum and incubated with an antibody to cd11b at a dilution of 1:10 (Serotec, Raleigh, NC, U.S.A.). The tissue was incubated with the peroxidase Vectastain Elite ABC kit (Vector Labs, Burlingame, CA, U.S.A.) that was visualized by incubation with diaminobenzidine and heavy metals. The sections were immunostained for fast, developmental, and neonatal myosin heavy chain isoforms. Unfixed sections were blocked in normal horse serum and incubated with an antibody to pan-fast myosin heavy chain (NovoCastra, Burlingame, CA, U.S.A.; 1:40), neonatal (NovoCastra, 1:20), or developmental myosin (NovoCastra, 1:20) and then processed as for the cd11b antibody. The eyelid sections were analyzed with the use of Bioquant Nova Prime software (Bioquant, Nashville, TN, U.S.A.). CD11b-positive cells were counted in the preseptal region of each eyelid cross section, and at least 4 fields (0.51 mm 2 ) were measured in this region in each cross section. A standard area measurement for cd11b-positive cells was determined morphometrically by calculation of the average area of 20 to 40 cd11b-positive cells. The computer used this standard area measurement for a cd11bpositive cell to determine the total number of cd11b-positive cells for each field measured. Wherever the total area of label was greater than the area measurement for a single cell, the computer divided this by the standard cd11bpositive cell area measurement. This number was the standardized count per field. The validity of this technique was manually confirmed by examination of representative sections. Three to 4 eyelid cross sections were counted in each of the 3 regions of the treated eyelids (12 sections per eyelid), and at least 4 eyelids were examined for each injection parameter. The sections at each postinjection interval were analyzed to determine total muscle cross-sectional areas, individual myofiber cross-sectional areas, and total myofiber number in the palpebral portion of the eyelids, using the Bioquant Nova Prime and Topographer software (Bioquant, Nashville, TN U.S.A.). We also assessed the integrity of the skin, vasculature, and peripheral nerves within the injected tissue. Nearby structures in the orbit were examined for histologic changes, to

3 314 A. R. HARRISON ET AL. ensure that there was no spread of the injected drug. All data were analyzed statistically for significance by either Student t tests or analysis of variance and were performed with the use of Prism and Statmate software (Graphpad, San Diego, CA, U.S.A.). Statistical significance was defined as p RESULTS In the first week after a single eyelid injection of the ricin-mab35, the eyelids were edematous, but no hair loss or skin lesions developed. The edema resolved by the second week, and the eyelids were normal in appearance during the remainder of the study. There were no apparent systemic reactions to the toxin. One week after immunotoxin injection, there was a significant decrease of 68% (p ) and 71% (p ), respectively, in the total number of orbicularis oculi myofibers at the 1/10 and 1/5 maximum tolerated dose for eyelids injected with ricinmab35 compared with the saline-injected controls (Fig. 1). There was no significant difference between the overall muscle toxicity when the two doses were compared (p ) (Fig. 1). However, there was a more even distribution of muscle loss across the treated eyelid at the higher dose of ricin-mab35 when sections were examined from the medial, central, and lateral portions of the treated eyelids from the medial to lateral canthal dimension (Fig. 2). This suggests that it might be possible to increase myotoxicity in the eyelid with the lower dose if access to the muscle fibers was improved. One month after a single injection of ricinmab35, muscle loss was still significant compared with the saline-injected control muscles, with 58% loss of muscle fibers compared with the normal eyelid control (Fig. 3). There was no statistical difference between the total muscle FIG. 2. Quantification of muscle loss from sections taken close to the medial canthus, from the center, and from close to the lateral canthus in eyelids injected with 1/10 maximum tolerated dose ricin-mab35. loss at 1 week after treatment compared with that seen at 1 month (p 0.09). At 1 month, the myofibers were small in cross-sectional area, and expression of developmental myosin was upregulated in these muscles compared with normal controls (Fig. 4). No changes were seen in the skin or other tissues within the eyelid sections at either of the time intervals examined. Six months after a single injection of ricin-mab35 in the eyelid, muscle loss was still significant and similar to that seen at 1 month when compared with the contralateral control (Fig. 3). The muscles were examined at both 1 week and 1 month for evidence of inflammatory cell infiltrate as FIG. 1. Quantification of muscle loss of eyelids treated with ricin-mab35 at 1/10 and 1/5 maximum tolerated dose as determined in mice. *Significant difference from control. FIG. 3. Quantification of muscle loss at 1 week and 1 month after single injection of 1/10 maximum tolerated dose ricinmab35 compared with normal saline injected control eyelids. *Significant difference from control.

4 IMMUNOTOXIN TREATMENT OF RABBIT EYELIDS 315 visualized by positive staining for cd11b (Fig. 5). There was a significant increase in inflammatory cell infiltrate at 1 week after a single eyelid injection of ricin-mab35 (p ). By 1 month, this was reduced to a level that was only slightly elevated from that seen in normal eyelids. DISCUSSION Ricin-mAb35, when injected directly in the eyelids of adult rabbits, resulted in significant acute muscle loss that was maintained for 6 months after a single injection. At 1 month, there was slight evidence for some regeneration. This is confirmed by myofibers with very small cross-sectional areas and by expression of developmental myosin heavy chain isoform in the treated muscle. 11 Despite some evidence of ongoing regeneration, muscle loss was still significant at 6 months after a single injection of the ricin-mab35. The lack of significant regeneration at 1 and 6 months is very promising and indicates that ricin-mab35 may have a much longer therapeutic effect than botulinum toxin A for the treatment of muscle spasm diseases in the face. However, our experience with ricin-mab35 in extraocular muscle suggests that orbicularis oculi muscle will slowly regenerate with time after treatment. 8 Other agents, including the local anesthetic bupivacaine 12,13 or injection with dihematoporphyrin ether and laser photochemomyectomy, 14 can significantly injure muscle acutely, but the muscle regener- FIG. 4. Photomicrograph of preseptal region of ricin-mab35 treated eyelid 1 month after single injection. A. Myofibers are extremely small in cross-sectional area. B. Small groups of myofibers in treated eyelids are positive for developmental myosin heavy chain isoform. C. Saline-injected control eyelids immunostained for fastmyosin. Bar is 100 m. FIG. 5. Quantification of inflammatory cell infiltrate as visualized by antibody to cd11b, which is positive for macrophages, neutrophils, and lymphocytes. There is a significant elevation of inflammatory cell infiltrate 1 week after single injection of ricin mab35, which is largely back to control levels by 1 month. *Significantly different from control. **Control levels of cd11bpositive infiltrate. These average between 1 and 3 cd11b-positive cells in each eyelid cross section.

5 316 A. R. HARRISON ET AL. ates rapidly. In contrast, both orbicularis myectomy surgery 15 and doxorubicin injection in the eyelid 16 result in permanent muscle loss, but both are associated with eyelid skin injury that many patients find unsatisfactory. Botulinum toxin A is a well-established and wellaccepted treatment for various focal dystonias, including blepharospasm. 1,2 The action of botulinum toxin A is to paralyze the muscle by preventing release of acetylcholine in the neuromuscular junctions. This treatment does not cause permanent muscle weakness, and muscle spasms return in several months. 6 There is also a group of patients, estimated to be between 10% and 25%, who do not respond to botulinum toxin treatment. 6,17 20 Our studies in extraocular muscles demonstrated that a single injection of ricin-mab35 resulted in both significant long-term muscle loss 8 and muscle weakening. 10 Although further studies are required to determine the long-term efficacy of ricin-mab35 in weakening the orbicularis oculi muscle, it holds promise as a muscleweakening agent that might be longer lasting than botulinum toxin treatments. The differential muscle loss across the eyelid from the medial to lateral canthus at the lower dose of the ricinmab35 suggests that it may be possible to increase the myotoxicity of a single injection of the ricin-mab35 by increasing the infiltration of the drug within the eyelid with the use of hyaluronidase. 21 This might increase the effectiveness of its myotoxicity because the ricin-mab35 is targeted directly to myofibers by the antibody to the nicotinic acetylcholine receptor. It should be pointed out that although neuromuscular junctions in the orbicularis oculi of rabbits are concentrated in the medial and lateral canthal regions, they are also diffusely located across the entire medial to lateral extent of the muscle. 22 This is particularly true in the pretarsal portion of this muscle. Thus, access to the muscle fibers is critical for its effectiveness. Drugs such as collagenase injected together with the ricin-mab35 should increase access to the myofibers in the eyelid, which are surrounded by dense connective tissue. Ricin-mAb35 appears to be as selectively myotoxic in the eyelid as it is in the orbit. Light microscopic examination demonstrated that the nerves, skin, and capillaries within the ricin-mab35 treated tissue were normal. This suggests that neurotoxicity will not play a significant role in the pharmacologic effects of the immunotoxin. Muscle-specific binding of the immunotoxin should also limit its spread outside of the direct injection site, making it unlikely that orbital or ocular cytotoxicity will be a significant concern. In summary, a single injection of ricin-mab35 in the eyelid results in significant muscle loss in the treated lids and a limited inflammatory reaction. We are currently in the process of examining its ability to weaken the orbicularis muscle and its long-term effectiveness, both of which are critical to its potential use as a treatment for blepharospasm and other dystonias. REFERENCES 1. Scott AB, Kennedy RA, Stubbs HA. Botulinum toxin injection as a treatment for blepharospasm. Arch Ophthalmol 1985;103: Harrison AR. Chemodenervation for facial dystonias and wrinkles. Curr Opin Ophthalmol 2003;14: Patrinely JR, Whiting AS, Anderson RL. Local side effects of botulinum toxin injections. Adv Neurol 1988;49: Perman KI, Baylis HI, Rosenbaum AL, Kirschen DG. The use of botulinum toxin in the medical management of benign essential blepharospasm. Ophthalmology 1986;3: Siatkowski RM, Tyutyunikov A, Biglan AW, et al. Serum antibody production to botulinum A toxin. Ophthalmology 1993;100: Brin MF, Fahn S, Moskowitz C, et al. Localized injections of botulinum toxin for the treatment of focal dystonia and hemifacial spasm. Mov Disord 1987;2: Christiansen S, Sandnas A, Prill R, et al. Acute effects of the skeletal muscle-specific immunotoxin, ricin-mab35, on extraocular muscles of rabbits. Invest Ophthalmol Vis Sci 2000;41: Christiansen S, Peterson D, To T, et al. Long-term effects of the skeletal muscle-specific immunotoxin, ricin-mab35, on extraocular muscles of rabbits: potential treatment for strabismus. Invest Ophthalmol Vis Sci 2002;43: Hott JS, Dalakas MC, Sung C, et al. Skeletal muscle-specific immunotoxin for the treatment of focal muscle spasm. Neurology 1998;50: Christiansen SP, Becker BA, Iaizzo PA, McLoon LK. Extraocular muscle force generation after ricin-mab35 injection: implications for strabismus treatment. J AAPOS 2003;7: Fitzsimmons RB, Hoh JF. Embryonic and foetal myosins in human skeletal muscle: the presence of foetal myosins in Duchenne muscular dystrophy and infantile spinal muscular dystrophy. J Neurol Sci 1981;52: McLoon LK, Wirtschafter JD. Regional differences in the subacute response of rabbit orbicularis oculi to bupivacaine induced myotoxicity as quantified with an N-CAM immunohistochemical marker. Invest Ophthalmol Vis Sci 1993;34: McLoon LK, Nguyen LT, Wirtschafter JD. Time course of the regenerative response in bupivacaine injured orbicularis oculi muscle. Cell Tissue Res 1998;294: Wirtschafter JD, Slovut DP, Stordal L, et al. Severe but temporary injury to rabbit orbicularis oculi muscle using hematoporphyrin derivative and laser photochemomyectomy. Mov Disord 1992;7: Gillum WN, Anderson RL. Blepharospasm surgery: an anatomical approach. Arch Ophthalmol 1981;99: McLoon LK, Wirtschafter J, Cameron JD. Muscle loss from doxorubicin injections into the eyelids of a patient with blepharospasm. Am J Ophthalmol 1993;116: Elston JS. Long-term results of treatment of idiopathic blepharospasm with botulinum toxin injections. Br J Ophthalmol 1987;71: Grandas F, Elston J, Quinn N, Marsden CD. Blepharospasm: a review of 264 patients. J Neurol Neurosurg Psychiatry 1988;51: Jankovic J, Schwartz KS. Clinical correlates of response to botulinum toxin injections. Arch Neurol 1991;48: Jankovic J, Schwartz KS. Response and immunoresistance to botulinum toxin injections. Neurology 1995;45: McLoon LK, Wirtschafter JD. Doxorubicin chemomyectomy in orbicularis oculi: increasing drug infiltration at the injection site. Curr Eye Res 1996;15: Lander T, McLoon LK, Wirtschafter JD. The orbicularis oculi muscle fibers are relatively short and heterogeneous in length. Invest Ophthalmol Vis Sci 1996;37: