European Journal of Pain (I 998) 2: 143-l 51

Immunoisolating encapsulation of intrathecally implanted bovine chromaffin cells prolongs their survival and produces anti-allodynic effect in spinally injured rats

Wei Yua, Jing-Xia Haoa, Xiao-Jun Xua, Joel Saydoffb, Sandy Shermanb, Anders Eriksson”, Anders Haegerstrand” and Zsuzsanna Wiesenfeld-Hallin”

“Karolinska Institute, Department of Medical Laboratory Sciences and Technology, Division of Clinical Neurophysiology, Huddinge University Hospital, Huddinge, Sweden; bCytoTherapeutics Inc., Providence, Rhode Island, USA; and “Astra Pain Control AB, Huddinge, Sweden

We have previously reported that intrathecal (i.t.) implantation of bovine chromaffin cells has an anti-allodynic effect in a rat model of mechanical and cold allodynia-like neuropathic pain after spinal cord injury. The technique of encapsulation of the cells by a semipermeable membrane has been developed recently. The present study was undertaken to investigate the effects of encapsulated bovine chromaffin cells on the allodynia-like pain in the same model. Capsules with bovine chromaffin cells or control capsules were implanted in the spinal subarachnoidal space in rats. Their response in behavioural tests were recorded for 2 months. At termination, the capsules were explanted and examined morphologically with tyrosine hydroxylase immunohistochemistry. The mechanical allodynia was totally abolished from week 2 after implantation of the cells and throughout the 8-week test period. The abnormal cold response was also attenuated in about half of the animals. The threshold to acute nociceptive stimulation was not affected. Eight weeks after implantation, 60-80% of the encapsulated chromaffin cells were still tyrosine hydroxylase positive. No effects were observed with control capsules. The results indicate that spinal implantation of encapsulated xenogeneic chromaffin cells may be useful in treating some refractory painful states associated with spinal cord injury. Immunoisolation of chromaffin cells by a semipermeable membrane may inhibit immunorejection, prolong the survival of the cells and enhance their anti-allodynic effect.

INTRODUCTION

Transplantation of allografted adrenal medullary tissue or xenogeneic bovine chromaffin cells into the spinal subarachnoidal space produces anti- nociception and reduces abnormal pain-related behaviours in some animal models (Sagen et al.,

Paper received 20 November 1997 and accepted in revised form 10 March 1998. Correspondence to: Prof. Zsuzsanna Wiesenfeld-Hallin, Karolinska Institute, Department of Medical Laboratory Sciences and Technology, Division of Clinical Neurophysiology, Huddinge University Hospital, Huddinge, Sweden.

1986a,b, 1990; Hama & Sagen, 1993; Wang & Sagen, 1995; Siegan & Sagen, 1996). In a model of chronic central pain after ischaemic spinal cord injury, rats exhibit pain-like response to innocuous mechanical and cold stimuli applied to skin areas rostra1 to the dermatomes of the injured spinal segments, which mimics some char- acteristics of central pain in patients with spinal cord injury (Xu et al., 1992). In a recent study with this model, intrathecal (i.t.)-free bovine chromaffin cells caused alleviation of allodynia (Yu et al., 1998). However, very few chromaffin cells survived 5 weeks after implantation, and the analgesic effect declined thereafter.

1090-3801/98/020143+09 $12.00/O 0 1998 European Federation of Chapters of the International Association for the Study of Pain

144 w. YU ET AL.

Immunological isolation of the cells by a semi- permeable membrane can block the passage of large immunoglobulins, complement and host white cells, and thereby prevent graft rejection (Emerich et al., 1997). Initial studies in normal rats demonstrated that encapsulated bovine chro- maffin cells reduced nociception when stimulated with nicotine, and after 3 months the retrieved capsules could still release met-enkephalin and catecholamines (Sagen et al., 1993). Recent stud- ies in sheep have verified the safety and viability for implantation (Joseph, 1994; Kaplan et al., 1996). Initial clinical studies in patients with cancer pain also confirmed patient safety and gave encouraging results (Buchser et al., 1996). Therefore, it is of interest to test the effect of i.t.- implanted encapsulated bovine chromaffin cells on the chronic allodynia-like pain and compare it to the effects of i.t.-free cells. The viability of the chromaffin cells was also investigated with immunohistochemical staining.

MATERIALS AND METHODS

Female Sprague-Dawley rats (B&K Universal, Stockholm, Sweden) weighing 200-250g at the start of the experiments were used. All ex- perimental protocols were approved by the local research ethics committee.

Photochemically-induced spinal cord injury

The details of the photochemical spinal cord lesion have been described previously (Xu et al., 1992). Briefly, rats were anesthetized with chloral hydrate (Sigma, USA, 300mg/kg i.p.) and ver- tebrae T12-Ll were exposed after a midline in- cision of the skin on the back. The animals were positioned beneath a tuneable argon ion laser (Innova, Model 70, Coherent Laser Products, California, USA) and irradiated with a knife edge beam, which covered the middle of the T13 vertebra, corresponding to spinal cord segment L4, with an average power of 0.17 W for 10 min. Immediately before the irradiation, erythrosin B (Red No. 3, Aldrich-Chemie, Steinhem, Ger- many, 32.5 mg/kg dissolved in 0.9% saline) was injected i.v. and the dose was repeated after 5 min.

After spinal irradiation, the wound was closed and the animals were kept warm for 2 h. Their bladders were emptied manually two to three times a day for l-2 weeks. Previous studies have shown that sham-operated rats injected with sa- line instead of Erythrosin B and then irradiated expressed no allodynia (Xu et al., 1992). There- fore, no sham-operated animals were included in the study.

Behavioural pharmacology

Response thresholds were recorded before the implantation of the encapsulated cells. After the implantation, the thresholds were recorded one to two times a week for 8 weeks. The behavioural tests used were the following:

Von Frey hair tests

A set of calibrated von Frey hairs (Stoelting, Illinois, USA) was used to test the vocalization threshold in response to graded mechanical pres- sure applied to the skin. The rats were gently restrained in a standing position and the von Frey hair was pressed onto the skin until the filament became bent. The stimulation was ap- plied 5-10 times at each intensity, beginning with the lowest, at about once every 2-3 s. The pres- sure which induced consistent vocalization (>75% response rate) was considered as pain threshold.

Cold spray test

Ethyl chloride (Medikema AB, Perstorp, Sweden) was sprayed on shaved skin areas, and the re- sponse of the animal was recorded. The response of the rat was graded with a score of 0 =no response; 1 = localized response (transient skin twitch); 2 = transient vocalization; and 3 = sus- tained vocalization. The spray produced intense cold, but not painful, sensation when applied to the experimenter’s forearm, and skin temperature was reduced to about 0°C immediately after application, with rapid recovery.

Tail flick test

Nociceptive threshold was monitored with the tail flick test. The light beam from a projector

European Journal of Pain (19981, 2

EFFECTS OF CHROMAFFIN CELLS ON CENTRAL PAIN 145

3’ 12 IO.5 mm

2mm

FIG. 1. Schematic illustration of the implanted device. 1, semipermeable membrane; 2, acrylic glue; 3, alginate with cells; 4, holder.

bulb was focused on the tail l-2 cm from the tip, and the latency of the tail flick was measured. The intensity of the thermal stimulus was adjusted so that the baseline latency was 3-4 s and a 10-s cut-off was employed to avoid tissue damage.

Encapsulation and implantation of cells

Bovine chromaffin cells were isolated and en- capsulated by CytoTherapeutics Inc., as in earlier studies in sheep (Joseph et al., 1994; Kaplan et al., 1996), with modification for rats. The capsules of semipermeable membrane (o.d. 665 pm, i.d. 545 pm, length 15 mm, blockade of molecular weight 99 kD) contained 4-5 ~1 bovine chro- maffin cells (50 000-60 000 cells/$) in 1.5% so- dium alginate (Ultrapure grade, Protan, Norway) and was sealed by acrylic glue. One end of the capsule was attached to a tiny plastic holder (Fig. 1). The cells were estimated to have >90% viability as assessed by trypan blue, which stained the non-viable cells with broken cytoplasmic membrane (Singh et al., 1985). The devices were shipped in CHO-SFM + medium (Gribco, USA) at 37°C and cultured for 1 week before im- plantation. A total of 10 animals with robust chronic allodynia were selected from a large pool of rats irradiated 3-5 months prior to the ex- periment. Five animals received capsules with cells, and the other five received capsules with only alginate as controls. Under chloral hydrate anesthesia, the dura and subarachnoidal mem- brane were lifted up and incised for 2-3 mm after laminectomy of vertebrae Ll. The capsule was then inserted under the membrane and pushed forward to L3-L4 level of the spinal cord. Care was taken to make it slip smoothly on the dorsal

surface without meeting any resistance. The wound was closed in layers and the rats were allowed to recover for 1 week before testing.

Histology

At the termination of the experiment, the capsules were explanted for histological examination. Fol- lowing fixation in 4% paraformaldehyde for 48 h and paraffin embedding, the retrieved capsules were cut longitudinally in 5-pm sections on a rotary microtome. Immunohistochemical stain- ing by monoclonal antibodies to tyrosine hydroxylase (TH, 1:6000, Incstar USA) and the avidin-biotin peroxidase reaction (Vector Labs, California, USA) with eosin (Polysciences, USA) counterstaining was used to reveal cell viability and morphology.

Statistical analysis

The responses to von Frey hairs were expressed as medians & median-derived absolute value. Data of cold spray test were expressed by fre- quency distribution. The tail flick latency data was expressed as meanf SEM. Whenever ap- propriate, the Kruskal-Wallis ANOVA was used, followed by the Wilcoxon signed-ranks test.

RESULTS

One to three months after the ischemic injury, about 50% of rats showed mechanical and cold allodynia-like behaviours, similar to earlier ob- servations (Xu et al., 1992). The vocalization threshold to pressure exerted by von Frey hairs in normal or non-allodynic rats was 73-95 g. In contrast, there was a marked decrease in vocalization threshold to mechanical pressure (median 0.9 g, range 0.6-4.1 g) in allodynic rats. Other behaviours, such as skin twitch, jumping and escaping were also evoked by low intensity mechanical stimuli. In response to the cold spray, normal or non-allodynic rats exhibited slight skin twitch (score 1). In contrast, the allodynic rats exhibited transient (score 2) or sustained (score 3) vocalization, licking of the stimulated area

European Journal of Pain (1998), 2

146

100

30 9 % $ 10 3 j

6 3

E=

? 1

0.3 I I I I I I I I I 0 1 2 3 4 5 6 7 8

Time (weeks)

FIG. 2. The effect of encapsulated chromaffin cells (U) and control capsules (0) on the vocalization threshold to mechanical stimuli in allodynic rats. Kruskal-Wallis ANOVA revealed that encapsulated bovine chromaffin cells, but not controls, significantly raised the vocalization threshold (pcO.05). Individual analyses were performed with the Wilcoxon sign-ranks test, *= ~0.05, compared to pre-implantation values. Note that the y-axis is logarithmic.

and escaping. The allodynic dermatomes at the mid or lower back and abdomen corresponded to the spinal segments just rostra1 to the lesioned lumbar cord. The allodynic behaviours occurred only when the rats were tested and the animals otherwise exhibited no signs of discomfort.

The chronic allodynia-like behaviour was usually stable for months without signs of re- mission. The control capsules did not influence the allodynic response to mechanical (Fig. 2) or cold stimuli [Fig. 3(a)]. In contrast, capsules with chromaffin cells totally reversed the mechanical allodynia-like behaviour in four of five rats (Fig. 2). The onset of the reversal of mechanical al- lodynia to normal response was around week 2 after implantation, and lasted until the end of the experiment (week 8). The abnormal response to cold was also attenuated [Fig. 3(b)]. Also around week 2, one of the rats recovered from cold allodynia; and during weeks 5 and 6, three out of the five rats were relieved from the cold allodynia. The rats showed no changes in tail flick latency (Fig. 4).

w. YU ET AL.

(a)

0 1 2 3 4 5 6 7 8 Time (weeks)

6)

0 1 2 3 4 5 6 7 8 Time (weeks)

FIG. 3. Distribution of percent of animals with abnormal (hatched columns, cold score=21 and normal cold response (empty columns, cold score= I) over the test period. The allodynic rats implanted with control capsules (a) remained hypersensitive, whereas 40-60% of the animals implanted with encapsulated bovine chromaffin cells were relieved from cold allodynia during weeks 4-7 (b).

At explantation, the capsules were usually loc- ated at the original implantation site, and no damage to the capsules or the spinal cord was observed. Eosin-stained cells were gathered in clusters and spread out in the whole matrix. The majority of the cells had well preserved viable cellular morphology of round to cubic forms similar to in vitro chromaffin cells. About 60-80%

European Journal of Pain 11998), 2

EFFECTS OF CHROMAFFIN CELLS ON CENTRAL PAIN

I I I I I I I I I 0 1 2 3 4 5 6 I 8

Time (weeks)

FIG. 4. The latency of the tail flick test in rats with active (0) and control capsules (0). Neither group of animals showed significant changes in tail flick latency throughout the test period. Kruskal-Wallis ANOVA, p>o.o5.

of these cells were tyrosine hydroxylase im- munopositive (Fig. 5), indicating their functional secretory capacity.

DISCUSSION

Encapsulated bovine chromaffin cells survived and alleviated chronic allodynia-like symptoms in spinally injured rats for the entire duration of the experiment. The results are in agreement with an earlier study with free cells (Yu et al., 1998). However, few cells survived after 5 weeks of implantation (Yu et al., 1998). In contrast, in the present study, the majority of the cells in the capsules were still viable and 60-80% had TH positive immunoreactivity, indicating functional capacity. This could explain why the analgesic effect of the free cells was attenuated after 4 weeks, whereas the encapsulated cells maintained their efficacy after 2 months. Thus, in comparison to the free cells, the encapsulated chromaffin cells survived longer in the subarachnoidal space and thereby reversed allodynia for a longer period.

Although the central nervous system has been considered immunologically privileged (Mason et al., 1986; Widner & Brundin, 1988), a number

FIG. 5. Photomicrographs showing bovine chromaffin cells 8 weeks after implantation. (A) Eosin-stained clustered chromaffin cells within the semipermeable membranes (arrowheads), more than 60% are dark TH immunoreactive (arrows), scale bar=80pm. (B) Heavy dark granules of TH immunoreactivity are present in the cytoplasm (arrows). The hallow areas surrounding the cell clusters resulted from the shrinkage of the matrix (asterisks) after dehydration. Scale bar=8 burn.

of studies have indicated that immuno- suppression by cyclosporin, for example, is neces- sary for longer survival of chromaffin cells (Ortega et al., 1992). Since the semipermeable membrane selectively blocks the passage of large proteins such as immunoglobulins (MW= 150 OOOD), complement factors (MW = 220 OOOD) and host white blood cells, but lets through freely nutrients and analgesic substances of small molecular weight, such as nor- epinephrine (MW = 180D) and met-enkephalin (MW =573D), it provides the necessary nu- trients, but does not influence the release of functional substances. The membrane itself seems to have no toxic effect in either animals or humans (Buchser et al., 1996; Joseph et al., 1994; Kaplan et al., 1996).

European Journal of Pain (19981, 2

148 w. YU ETAL.

In agreement with earlier data (Yu et al., 1998), Biomed II Programme of the European Com- the onset of the analgesic effect was about 2 mission (project BMH4-CT950 172) and the weeks after implantation. Thus, the cells may Bank of Sweden Tercentenary Foundation. require time to recover from the handling, to adapt to the new environment or to build up a steady level of substances. Furthermore, no REFERENCES

analgesic tolerance was observed in the current study, confirming earlier results (Yu et al., 1998). Previous studies demonstrated that the analgesia was primarily mediated by opioid and adrenergic receptors (Sagen et al., 1986a, 1993; Yu et al., 1998). Other substances released from the chro- maffin cells, such as galanin, NPY and neuro- trophic factors may also exert sensory effects (Unsicker, 1993). The synergy of several sub- stances released at fairly low levels may underly the mechanism of the analgesia without the de- velopment of tolerance.

As observed earlier with free chromaffin cells, the encansulated chromaffin cells had no effect on the tail flick latency. This may result from an insufficient concentration of analgesic substances on sacral spinal cord, which mediates the tail flick response. Previous work showed that stimu- lation of nicotinic receptors was needed to pro- duce acute antinociception (Sagen et al., 1986a, 1993), whereas no nicotine was used in the current experiment, and therefore the general i.t. con- centration might be low. Furthermore, free cells tended to spread along the spinal cord from the injection site, while the capsules were located on the middle of the lumbar cord. The stable location of the capsules, however, provides a highly-loc- alized analgesic effect.

In conclusion, application of chromaffin cells encapsulated in a semipermeable membrane is preferable to free cells due to reduction of im- munorejection by the host tissue. This may pro- vide a new therapeutic approach for the long- term relief of chronic neuropathic pain without the need for immunosuppression.

ACKNOWLEDGEMENT

We thank MS G. Hovanesian for the help in histological work. This work is supported by Astra Pain Control AB, the Swedish Medical Research Council (Project no. 07913, 11038) the

Buchser E, Goddard M, Heyd B, Joseph JM, Favre J, Tribolet Nd, Lysaght M, Aebischer P. Immunoisolated xenogeneic chromaffin cell therapy for chronic pain. An- esthesiology 1996; 85: 1005-1012.

Emerich DF, Lindner ML, Saydoff JA, Gentile FG. Treat- ment of central nervous system diseases with polymer- encapsulated xenogeneic cells. In: Freeman TB, Winder H, editors. Fetal Transplantation in Neurological Disorders. Humana Press, 1997: in press.

Hama AT, Sagen J. Reduced pain-related behavior by ad- renal medullary transplants in rats with experimental painful peripheral neuropathy. Pain 1993; 52: 223-23 1.

Joseph J, Goddard M, Mills J, Padrun V, Zurn A, Zeilinski B, Favre J, Gardaz J, Mosimann F, Sagen J, Christenson L, Aebischer P. Transplantation of encapsulated bovine chromaffin cells in the sheep subarachnoid space: a pre- clinical study for the treatment of cancer pain. Cell Tramp1 1994; 3: 355-364.

Kaplan FA, Krueger PM, Goddard MB. Peripheral xeno- geneic immunological response to encapsulated bovine adrenal chromaffin cells implanted within the sheep lumbar intrathecal space. Transplantation 1996; 61: 121% 1221.

Mason D, Charlton H, Jones A, Lavy C, Puklavec M, Simmonds S. The fate of allogeneic and xenogeneic neur- onal tissue transplanted into the third ventricle of rodents. Neuroscience 1986; 19: 685694.

Ortega JD, Sagen J, Pappas GD. Short-term immuno- suppression enhances long-term survival of bovine chro- maflin cell xenografts in rat CNS. Cell Transpl 1992; 1: 3341.

Sagen J, Pappas GD, Perlow MJ. Adrenal medullary tissue transplants in the rat spinal cord reduce pain sensitivity. Brain Res 1986a; 384: 189-194.

Sagen J, Pappas GD, Pollard HB. Analgesia induced by isolated bovine chromaffin cells implanted in rat spinal cord. Proc Nat1 Acad Sci 1986b; 83: 7522-7526.

Sagen J, Wang H, Pappas GD. Adrenal medullary implants in the rat spinal cord reduce nociception in a chronic pain model. Pain 1990; 42: 69-79.

Sagen J, Wang H, Tresco PA, Aebischer P. Transplants of immunologically isolated xenogeneic chromaffin cells provide a long-term source of pain-reducing neuroactive substances. J Neurosci 1993; 13: 2415-2423.

Siegan JB, Sagen J. Adrenal medullary transplants attenuate acute and tonic nociceptive phases of the formalin re- sponse in rats. IASP Abstr. 8th World congress on pain, 1996: 223.

Singh G, Bohnke M, von DD, Draeger J, Lindstrom RL, Doughman DJ. Vital staining of cornea1 endothelium. Cornea 1985; 4: 80-91.

Unsicker K. The trophic cocktail made by adrenal chro- maffin cells. Exp Neural 1993; 123: 167-173.

Wang H, Sagen J. Attenuation of pain-related hyper- ventilation in adjuvant arthritic rats with adrenal me- dullary transplants in the spinal subarachnoid space. Pain 1995; 63: 313-320.

European Journal of Pain ( 1998),2

EFFECTS OF CHROMAFFIN CELLS ON CENTRAL PAIN 149

Widner H, Brundin P. Immunological aspects of grafting in the mammalian central nervous system. A review and speculative synthesis. Brain Res Rev 1988113: 218-324.

Xu X-J, Hao J-X, Aldskogius H, Sieger A, Wiesenfeld- Hallin Z. Chronic pain-related syndrome in rats after ischemic spinal cord lesion: a possible animal model for pain in patients with spinal cord injury. Pain 1992; 48: 279-290.

Yu W, Hao J-X, Xu X-J, Saydoff J, Haegerstrand A, Wiesenfeld-Hallin Z, Hiikfelt T. Long term alleviation of allodynia-like behaviors by intrathecal implantation of 1 bovine chromaffin cells in rats with spinal cord injury. Pain 1998; 74: 115-122.

Commentary

The above paper by Yu et al. is an important contribution to the evolving story of how adrenal chromaffin cell transplants can be used to deliver neuro-active substances for the treatment of chronic pain. The results are consistent with previous reports describing the effects of adrenal transplants in reducing pain sensitivity (Czech & Sagen, 1995; Sagen, 1996), alleviating pain behaviours in experimental models of chronic pain (Czech & Sagen, 1995; Sagen, 1996; Yu et al., 1998), and with clinical studies in pain management (Pappas et al., 1997; Burgess, 1996; Sagen, 1996). Although the antinociceptive value of chromaffin cell transplants has been well docu- mented following peripheral nerve or tissue dam- age, the studies by Yu et al. (1998 and present study) demonstrate significant effects in a model of pain following injury to the central nervous system, i.e. central pain and, specifically, spinal cord injury (SCI) pain.

The results of the present study, using the technique of immuno-isolation, should be viewed as an important step towards the development of an effective treatment strategy for SC1 pain. This technique has been shown clinically to be efficacious in the management of cancer pain (Burgess, 1996). The importance of developing an intervention for SC1 pain is evident when one examines the clinical characteristics and pre- valence of this condition (Yezierski, 1996). Pain associated with spinal injury affects nearly 70% of the SC1 population, rarely diminishes with time, in most cases is not effectively treated with pharmacotherapy, and often interferes with efforts to re-establish an acceptable quality of

life. While the successful use of adrenal chro- maffin cells in the ischemic model of SC1 pain is significant, of equal importance is the fact that the results are consistent with those obtained in other experimental models (Czech & Sagen, 1995; Sagen, 1996), including a recently described excitotoxic model of SC1 (Brewer & Yezierski, 1998). This fact underscores the widespread use- fulness of adrenal chromaffin cells which can be attributed to their phenotypic and neurochemical plasticity in response to environmental factors, and their production of a host of potentially therapeutic substances (Unsicker, 1993; Sagen, 1996). These cells constitutively release ca- techolamines, indolamines, opioid peptides and a virtual ‘cocktail’ of neurotrophic factors. Al- though the use of adrenal chromaffin cells as ‘cellular minipumps’ ‘provides’ a promising new approach for the treatment of chronic pain, the possibility also exists that there is a substantially expanded list of neuro-active substances that may play an important role in pain management.

In spite of the well-documented anti-nociceptive properties of adrenal chromaffin cells, the specific substances responsible for the analgesic efficacy of these cells is not known. The effectiveness of adrenal transplants across different experimental models and injury-induced behavioural changes, however, is a strong endorsement for the use of chromaffin cells in patients with severe and/or debilitating pain, including SC1 pain. Previous studies have shown the antinociceptive effects of chromaffin cells are partially blocked by both the a-adrenergic antagonist, phentolamine, and the opioid antagonist, naloxone (Sagen et al., 1993; Yu et al., 1998). These results demonstrate that catecholamines and opioids released from chro- maffin cells play an important role in the reversal of pain behaviours, but are not completely re- sponsible for the effects. Since central pain, in- cluding SC1 pain, is known to be refractory to opioids, other substances may prove to be more important in the alleviation of central pain. For example, the various trophic factors including FGF-2, CNTF, TGF-P, interleukins, neuro- trophins and neuropeptides secreted by these cells may influence neuronal recovery from peripheral and/or central nervous system (CNS) injury, and thus impact on the onset of injury-induced pain.

European Journal of Pain (1998), 2

150 w. YU ET AL.

Another potential action of adrenal chromaffin cells involves effects on the N-methyl-D-as- partate-nitric oxide cascade (Czech & Sagen, 1995), and effects on second messengers, e.g. cyclic guanosine 3’ S-monophosphate (Siegan et al., 1996).

Recently, it was proposed that a useful strategy for the treatment of central pain may be opposing glutamate and potentiating GABAergic neuro- transmission (Canavero & Bonicalzi, 1998). While there is evidence to support this claim, one must not discount reports related to the wide range of drugs used in the management of central pain, and especially that following SC1 (Rag- narsson, 1997). For example, anticonvulsive agents, psychotropic medications with anti- depressant effects, and tricyclic antidepressants have all been used to provide transient relief of SC1 pain. The results associated with the success of adrenal chromaffin cells, however, offers an exciting new approach for the treatment of SC1 pain. For this reason, it is likely that future therapies will involve a combination of strategies targeting specific transmitter and/or second mes- senger systems, as well as injury cascades known to have a negative influence on cell survival.

In the present study, the method of immuno- isolation was used to provide a delivery system for bio-active substances secreted by adrenal chromaffin cells. This method was developed to enhance the delivery of chromaffin cell products to the CNS, and to prolong the survivability of these cells (Burgess, 1996). With this method, cells are encapsulated within a selectively permeable membrane which provides protection from circu- lating lymphocytes and antibodies while allowing the passage of nutrients and secreted factors into the CNS (Sagen, 1996). Encapsulated cells also provide a degree of control over the trans- plantation process by allowing localization of the device to a targeted region, e.g. spinal cord, and retrievability of capsules in the event of complications. Based on the results of the present study, it is suggested that an approach similar to that used in the treatment of cancer pain (Burgess, 1996) could prove efficacious in the treatment of various pain syndromes associated with SCI. While cellular implantation offers a novel thera- peutic strategy, there are technical and safety

aspects of this approach that should be con- sidered. Dose management, toxic effects of re- leased substances and the potential biohazard of using xenografts should not be overlooked.

In conclusion, the challenge confronting health-care professionals to manage chronic pain necessitates not only further research into the pathophysiology and neurochemical changes as- sociated with these conditions, but also the de- velopment of novel treatments. The use of cellular transplants, of primary cells or cell lines (Eaton et al., 1997) and technology to ensure that these ‘cellular minipumps’ remain viable offers the kind of innovative approach that could significantly influence the future design of more effective in- terventions for previously intractable chronic pain syndromes. Continued research evaluating the efficacy of primary cells and genetically en- gineered cell lines capable of delivering peptides, mono-amines, amino acids, neurotrophins and/ or growth factors should provide further insight into the potential therapeutic value of this treat- ment strategy.

R. P. YEZIERSKI University of Miami School

of Medicine, USA

REFERENCES

Brewer KL, Yezierski RP. Effects of adrenal medullary transplants on pain-related behaviors following ex- citotoxic spinal cord injury. Brain Res 1998 (in press).

Burgess FW. Cell therapy for cancer pain management. Bull Am Pain Sot 1996; 6: 4-7.

Canavero S, Bonicalzi V. The neurochemistry of central pain: evidence from clinical studies, hypothesis and thera- peutic implications. Pain 1998; 74: 109-l 14.

Czech KA, Sagen J. Update on cellular transplantation into the CNS as a novel therapy for chronic pain. Prog Neurobiol 1995; 46: 507-529.

Eaton MJ, Santiago DJ, Dancausse HA, Whittemore SR. Lumbar transplants of immortalized serotonergic neurons alleviate chronic pain. Pain 1997; 72: 59-69.

Pappas GD, Lazorthes Y, Bes JC, Tafani M, Winnie AP. Relief of intractable cancer pain by human chromaffin cell transplants: experience at two medical centers. Neurolog Res 1997; 19: 71-77.

Ragnarsson KT. Management of pain in persons with spinal cord injury. J Spinal Cord Med 1997; 20: 186-199.

Sagen J, Wang H, Tresco PA, Aebischer P. Transplants of immunologically isolated xenogenic chromaffin cells provide long-term source of pain-reducing neuroactive substances. J Neuroscience 1993; 13: 2415-2423.

Sagen J. Chromaffin cell transplants in the CNS: basic and clinical update. In: Lanza RP, Chick WL, editors.

European Journal of Pain (1998). 2

EFFECTS OF CHROMAFFIN CELLS ON CENTRAL PAIN 151

Yearbook of Cell and Tissue Transplantation. Netherlands: Yezierski RP. Pain following spinal cord injury: the clinical Kluwer Academic, 1996: 71-89. problem and experimental studies. Pain 1996; 68: 185-194.

Siegan JB, Hama AT, Sagen J. Alterations in rat spinal cord Yu W, Hao X-J, Xu X-J, Saydoff J, Haegerstrand A, cGMP by peripheral nerve injury and adrenal medullary Weisenfeld-Hallin Z. Long-term alleviation of allodynia- transplantation. Neuroscience Lett 1996; 215: 49-52. like behaviors by intrathecal implantation of bovine chro-

Unsicker K. The trophic cocktail made by adrenal chro- maffin cells in rats with spinal cord injury. Pain 1998; 74: maffin cells. Exp Neurology 1993; 123: 167-173. 115-122.

European Journal of Pain (1998), 2