Extracorporeal Whole Blood Immunoadsorption of Autoimmune Myasthenia Gravis by Cellulose Tryptophan Adsorbent

Abstract

Whole blood immunoadsorption (WBIA) system, using an adsorbent to remove pathogenic antibodies of myasthenia gravis (MG), was studied. Cellulose-tryptophan adsorbent was synthesized and its adsorption capacity of binding with acetylcholine receptor in the plasma of MG patient was evaluated. Experimental autoimmune myasthenia gravis (EAMG) rabbits were induced by Ta183–200 peptide. The rabbits underwent extracorporeal whole blood adsorption for 2 h. Results showed no significant damages on blood cells and no changes in the concentration of electrolytes. Total protein decreased by 12.0% (P < 0.05), and globulin protein decreased 23.9 ± 5.6% (P < 0.05). The mean overall removal of antibodies against Ta183–200 was 41.12%. The percentage of decrement of compound muscle action potential in 3, 5, 10 Hz of EAMG rabbits all dropped down after the treatment. In conclusion, the adsorbent is biocompatible, was safe for whole blood immunoadsorption, and can remove antibodies in an MG patient effectively. Whole blood immunoadsorption improved clinical manifestation and neuromuscular function of the EAMG rabbits.

• Experimental autoimmune myasthenia gravis
• Whole blood immunoadsorption
• Biocompatibility
• Antibody
• Neuromuscular transmission function

Introduction

Myasthenia gravis (MG) is an autoimmune disorder characterized by a disturbance in neuromuscular transmission that results in muscle weakness. The damage is attributed to injury of acetylcholine receptors by antibodies (AchR-ab), which can cause the loss of receptors, or directly inhibit their function (Lindstrom, [2000]). Experimental autoimmune myasthenia gravis (EAMG) is a perfect animal prototype of myasthenia gravis (Premkumar et al., [2000]). It can be induced by purified AchR, and also can be made by the myasthenogenic peptides on the a-subunit of AchR. a183–200 segment is one of the immunogens for inducing the animal to produce blocking antibodies of AchR (Masaharu et al., [1988]).

Oral anticholinesterase can control the symptoms of MG, but high concentration of these drugs can precipitate a cholinergic crisis. Thymectomy is indicated for thymoma, but does not usually improve myasthenia. Immunosuppressive therapy is effective in most of the patients, but some important drawbacks, such as it inhibits the whole immune system and severe complications resulting from the long-term therapy, also confined its use in some MG patients (Angela et al., [2001]). Plasmapheresis has been used since 1975 in the treatment of myasthenia gravis and is a method of temporarily improving myasthenic crisis. Hemoperfusion using an immunoadsorbent was developed and can effectively remove pathogenic components in MG.

The literature reported that the AchR-ab could be removed by polyvinyl alchol linked with tryptophan in plasma, which was used to treat 20 MG patients by Shibuya in 1992 (Shibuya et al., [1994]) and 16 MG patients by Grob in 1995 (Grob et al., [1995]). We have studied the adsorption capacities of various ligands for the removal of pathogenic antibodies in MG in vitro and found that immunoadsorbent prepared by coupling l-tryptophan to epichlorohydrin-activated cellulose bead showed the best result (Yan et al., [2002]).

This article will develop a whole blood immunoadsorptior (WBIA) system as a treatment modality for MG and study the therapeutic efficacy and safety of the WBIA system with cellulose-tryptophan adsorbent on an animal model.

Materials and Methods

Peptide Synthesis

The residual sequence of acetylcholine receptor of Torpedo Californica, Ta183–200, was synthesized by solid-phase procedure and purified by reverse-phase HPLC. In order to promote peptide solubility, an additional sequence of Lys-Lys at the N-terminus and Lys-Lys-Gly at the C-terminus (Masaharu et al., [1988]) was introduced (Fig. 1).

Figure 1. Amino acid sequence of synthetic peptides.

Induction of Experimental Autoimmune Myasthenia Gravis Animal Models

Female rabbits weighing approximately 2 kg were injected intradermally at multiple sites with 500 µg of Ta183–200, which was emulsified with an equal volume of Freund's complete adjuvant. A booster injection of 500ug Ta183–200 with Freund's incomplete adjuvant was administered after 4 weeks.

Clinical Assessment of MG in Animal Model

The immunized rabbits were examined daily for clinical signs of EAMG. On the day before and the 5th day after the perfusion, the disease was graded (Karussis et al., [1994]) as: mild (+) (weight loss and ear weakness), moderate (++) (generalized weakness and difficulty in walking), and severe (+++) (flaccid quadriplegia, severe weight loss and respiratory insufficiency).

Assay of Serum Antipeptide Antibodies

The antibody against Ta183–200 was determined by indirect ELISA. Briefly, 96 well-flat-bottomed plates were coated with AchR at 10 ug/mL in PBS overnight at 4°C. The plates were then washed with PBS and incubated with serum samples overnight at 4°C. Then the plates were incubated with horseradish peroxidase-labeled goat Ab against rabbit immunoglobulins at 37°C for 2 h. After washing with PBS, p-nitrophenyl phosphate was added, and the optical density (OD) at 492 nm was recorded as the titres of antibody.

Assessment of Electrophysiological Function

Repetitive nerve stimulation was performed under sodium pentobarbital anesthenia using a conventional clinical electromyography apparatus (Counterpoint MK2, DANTEC). The compound muscle action potential (CMAP) of deep peroneal nerve was recorded in response to supramaximal stimuli at 3, 5, 10 Hz. The decrement was evaluated by comparing the amplitude of the fifth to first recorded muscle action potential.

Measurement of Blood Components

Blood cell counts, concentration of proteins and electrolytes in plasma were all autoanalyzed by conventional methods (Automatic Analyzer, 7170A Hitachi and Blood Cell Analyzer, XE-2100 SYSMEX).

Preparation of Adsorbents

Cellulose beads as the matrice were synthesized according to the methods reported (Kong et al., [1998]). The beads were activated by epichlorohydrin, and tryptophan was coupled on the beads as ligand.

Determination of the Adsorption Capacity for Achr-ab

One milliliter cellulose-tryptophan beads was incubated with 3.0 mL AchR positive serum and stirred for 3 h at 37°C. AchR-ab titres were determined by ELISA. The adsorption capacity was calculated as the percentage decrease of the plasma OD at 492 nm.

Whole Blood Extracorporeal Immunoadsorption

The WBIA circuit is shown in Fig. 2. The column was packed with 3 g cellulose-tryptophan adsorbents in the bead type,which was retained by a 40 mesh filter fitted at the bottom of the column and equilibrated with sterile heparinized saline (3 µL/mL) before use. Blood was pumped from the carotid arterial catheter through an adsorbent column at a flow rate of 10–15 mL/min and then returned back via the cannula in the jugular vein. The rabbits were fully heparinzed by injection of heparin and maintained in an anticoagulated state by discrete addition as needed according to clotting times determined during the course. The procedure was carried out for 2 h. The extracorporeal blood was maintained in a water bath at a constant temperature of 39.4°C.

Figure 2. Schematic representation of WBIA.

Experimental Setup

Hemoperfusion was performed with 8 healthy rabbits and 8 EAMG rabbits. Blood samples from the healthy rabbits were withdrawn before and after the procedure for test of elements and soluble components in plasma.

Blood was withdrawn form EAMG rabbits at intervals from the beginning to the end of the course for assessment of the titre of anti-peptide antibody in serum. Just before and on the 5th day after the procedure, the disease rabbits were graded and electrophysiological functions were assayed. Student's t test was used for statistical analysis.

Results

Biocompatibility of the Adsorbent

Changes in blood composition from a 2 h course are shown in Table 1. There was no significant change in blood cell counts, haematocrits, haemoglobin, and plasma electrolytes (p > 0.05). Although there was no significant decrease in albumin, the immunoadsorption did result in a significant decrease in total protein concentration. The mean decrease of total protein was approximately 12%, mainly of IgG removal.

Table 1. Changes in blood components after 2 h WBIA

	Before WBIA	After WBIA	P	Reduction in %
WBC (×10^9/L)	7.86±2.99	7.65±2.73	0.535	0.9±0.7
RBC (×10^12/L)	3.89±0.88	3.80±0.85	0.177	2.1±4.4
Hgb (g/L)	79.50±12.49	79.00±13.03	8.844	0.5±8.7
Hct (%)	26.89±4.53	26.80±4.20	0.812	0.1±3.8
Plt (×10^9/L)	332.1±197.2	320.6±193.3	0.274	4.1±9.6
TP (g/L)	62.38±6.63	54.75±4.98	0.000	12.0±4.3
Glo (g/L)	28.63±4.21	21.63±2.20	0.000	23.9±5.6
Alb (g/L)	33.75±3.45	32.63±3.25	0.259	3.1±7.1
K^+ (mmol/L)	3.50±0.62	3.59±0.64	0.133	−2.5±4.0
Na^+ (mmol/L)	139.5±6.8	137.6±6.6	0.054	1.3±1.6
Cl^+ (mmol/L)	103.5±6.0	102.1±7.5	0.092	1.4±1.8

Data are the means ± SD for eight rabbits.

Observation of Antibody Adsorbed

The adsorption capacity of cellulose-tryptophan resin was evaluated in vitro with serum of MG patient. After 3 h perfusion, 1 mL resin could remove 44.79 ± 1.49% (n:3) AchR-ab in 3 mL MG patient serum. Figure 3 shows the performance of the resin as a function of time during the course of WBIA on EAMG rabbits. Optical density of antibody at the beginning of WBIA was regarded as 100%; each data point represents the mean titres as percentage of antipeptide antibody in plasma of 8 EAMG rabbits.

Figure 3. Kinetic observations of antibody removal.

Figure 3 showed that the majority of the antibody was removed during the first one hour, that was 41.12 ± 3.14%. During the second hour, the antibody titre in the plasma reduced slowly. Immunoadsorption reached the maximum from 90 to 120 min. The overall removal of antibodies within 2 h was 48.54 ± 2.47%.

Therapeutic Effect on EAMG by WBIA Assessment of Clinical Manifestation

All 8 EAMG rabbits after therapy had significant improvement in strength, starting from 24–48 h WBIA treatment and continued to improve after the perfusion. At the 5th day after WBIA, the rabbits underwent the 4 levels (Table 2) graded of clinical manifestation, and were compared with that on the day before WBIA.

Table 2. Grading of clinical manifestation pre and 5th day after WBIA

No. of rabbits	Pre WBIA	Fifth day after WBIA
1	+	0 – +
2	+	0
3	+	0
4	+	0
5	+	0
6	+	0
7	++	+
8	+	0

0 stand for the rabbits recovered normal after therapy.

Assement of Neuromuscular Transmission Function

On the 5th day after WBIA, stimulation of the deep peroneal nerve showed that the mean decrement of potentials evoked from the anterior tibial muscle at three stimulation frequencies decreased. At 3 Hz, the potential decreased from 21.8 to 16.0%, at 5 Hz, it decreased from 23.8 to 16.88%, and at 10 Hz, it decreased from 25.7 to 17.5%.

Discussion

Therapeutic plasmapheresis is a recognized and recommended treatment for patients with MG. Extracorporeal immunoadsorption from plasma constitutes a more refined version of conventional plasmapheresis, in that the removal of the putative pathogenic plasma constituents can be achieved more selectively without any need for replacement fluids derived from human plasma. Our experimental results show that WBIA is a reliable and cost-effective treatment for MG.

The adsorbent used has good biocompatibility properties. In our trials, the effect on the blood cells was satisfactory. WBC reduced 0.9 ± 0.7%, RBC reduced 2.1 ± 0.4%, and Plt depletion was 4.1 ± 9.6%. The soluble blood components, K⁺, Na⁺, Cl⁻, were unaffected by the WBIA system. Likewise, albumin concentration remained constant. The total protein decreased mainly due to lost of globulin protein (23.9 ± 5.6%). The reduction of globulin protein indicates that tryptophan as a ligand is not of high selectivity in binding with AchR antibody, because it can also bind with other immunoglobulins through the hydrophobic binding force (Yoshida et al., [2000]). Although the globulin protein decreased significantly (p < 0.001), there was no need for substitute fluids.

After 2 h WBIA, the removal of Ab was effective and the neurotransmission function of EAMG rabbits improved. The decrement of CMAP reduced and the pronounced rapid recovery from MG weakness was noted in almost all the EAMG rabbits. Only one rabbit improved slightly (No. 1).

In conclusion, our system could efficiently remove the AchR-ab and so the manifestation of EAMG rabbits improved quickly after the WBIA therapy, which indicates the potency of future application.

Acknowledgments

The support of National Key Project of Fundamental Research and Advances (No. G1999064707), grants from the Health Ministry Scientific Foundation of PR China and grants for biomaterials project of Tianjin-Nankai University co-construction are acknowledged.