The most commonly used immunogen to induce experimental autoimmune encephalomyelitis is MOG35-55, a 21-residue peptide derived from myelin oligodendrocyte glycoprotein (MOG). In most studies, mice exhibit a chronic disease; however, in some studies mice show a transient disease. One variable that is not often controlled for is the peptide fraction of the purified MOG material, which can vary from less than 50% to over 90%, with the remainder of mass primarily comprised of the counter ion used for peptide purification. We compared the development of clinical signs in female C57Bl6 mice immunized with two commercially available MOG35-55 peptides of similar purity but different peptide fraction (MOG-A being 45%; MOG-B being 72%). A single immunization with MOG-A induced a chronic disease course with some recovery at later stages, whereas immunization with MOG-B induced a similar course of disease but with significantly lower average clinical scores despite a higher peptide content. The addition of a booster immunization significantly increased clinical severity with both preparations, and significantly reduced the average day of onset using MOG-A. To determine if the counter ion could influence disease, we compared MOG-B-containing trifluoroacetate with MOG-B-containing acetate. Although disease incidence and severity were similar, the average day of disease onset occurred approximately 5 days earlier with the use of MOG-B-containing trifluoroacetate. These results demonstrate that differences in peptide fraction influence the course of encephalomyelitis disease, which may be due in part to the levels of counter ions present in the purified material. These findings underscore the fact that a knowledge of peptide fraction is as critical as knowledge of peptide purity when using peptides from different sources.
The most commonly used immunogen to induce experimental autoimmune encephalomyelitis (EAE) is MOG35-55, a 21-residue peptide derived from myelin oligodendrocyte glycoprotein. Subtle variations in MOG immunization protocols can influence the type of disease, the average day of onset, disease severity, and overall incidence. We hypothesized that variance in peptide fraction and counter ion may contribute to those differences. Surprisingly, immunization with a MOG preparation of high (72%) peptide fraction resulted in lower average EAE clinical scores and later disease onset than immunization with a preparation having only 45% peptide fraction. Replacement of trifluoroacetate (TFA), the counter ion used for MOG peptide purification, with acetate delayed disease onset. These results show that knowledge of peptide fraction and the counter ion present are as critical as peptide purity, particularly when using peptides from different sources. The image shows TFA molecules (not to scale) hypothetically binding to the positively charged residues of MOG.
Experimental autoimmune encephalomyelitis (EAE) is the paramount model of multiple sclerosis, in which immunization with brain specific proteins or peptides are used to induce an autoimmune response. A commonly used immunogen is MOG35-55, a 21-residue peptide spanning amino acids 35–55 of the myelin oligodendrocyte glycoprotein (MOG) in C57BL/6 mice (Stefferl et al. 2000). In most studies, mice develop a chronic disease, begin to exhibit clinical signs about 10 days after immunization, reach maximum severity after about 3 weeks, and maintain maximal scores or show slight improvement over the following weeks (Malipiero et al. 1997; Suen et al. 1997; Heneka et al. 2001; Feinstein et al. 2002; Polak et al. 2005, 2012a,b; Sharp et al. 2008; Kuwabara et al. 2009; Simonini et al. 2010; Kim et al. 2011; Recks et al. 2011; Guo et al. 2012; Ji et al. 2012; Ruffini et al. 2012). Occasionally, the course of disease was chronic but included relapses (Johns et al. 1995; Berard et al. 2010); and in some studies, mice developed a moderate to severe disease by 3 weeks, after which scores improved and mice showed gradual recovery (Ren et al. 2011; Zaheer et al. 2012). Although numerous factors can influence the course of EAE, only few studies have carefully examined possible causes for variation (Berard et al. 2010).
Two parameters that have not been well characterized are the peptide fraction of the MOG preparation used, and the nature and amount of the counter ion used during the final stages of peptide purification. In contrast to peptide purity which typically is 95% or greater, the peptide fraction (the percentage of protein present in the final lyophilized mass) can range from less than 50% to over 90%. The remaining material is comprised of primarily the counter ion used for purification, which for small peptides is usually trifluoroacetic acid (TFA), as well as some water and trace amounts of other salts. During the course of comparing different commercial sources of MOG35-55 peptide, we became aware that the peptide fraction differed substantially between preparations. As immunization with different doses of MOG can influence EAE disease (Sestero et al. 2012; Naves et al. 2013), and as TFA can influence cell growth and survival (Ma et al. 1990; Cornish et al. 1999), T-cell responses (Trudell et al. 1991; Furst et al. 1997; You et al. 2006, 2010), and the activity of certain receptors (Tipps et al. 2012) and channels (Han et al. 2001), results obtained with different MOG preparations may not be directly comparable. In this study, we report that immunization with a MOG35-55 preparation of low peptide fraction led to significantly higher clinical scores than immunization with one of higher peptide fraction. We also show that TFA, present at higher levels in the lower peptide fraction preparation, influences EAE disease.
Material and methods
Two sources of MOG35-55 peptide were used in this study. MOG-A was from AnaSpec (AnaSpec Inc., Eurogentec, Fremont, CA, USA) and MOG-B from CPC (CPC Scientific Inc., Sunnyvale, CA, USA). Both peptides have the sequence MEVGWYRSPFSRVVHLYRNGK. The MOG-A preparation was > 95% full-length MOG35-55 as determined by reverse phase HPLC, and had a net protein content of 45.4%. The remaining 54.6% consist primarily of the counter ion TFA, some water, and trace amounts of other salts. MOG-B was > 95% pure by reverse phase HPLC, and had a net protein content of 72% (studies 1 and 2) or 80% (study 3). Again, the remaining material consists primarily of TFA, water, and trace amounts of salts. An aliquot of the MOG-B peptide was also used (a gift of CPC Scientific), in which the TFA was exchanged for acetate, resulting in MOG-B-acetate which was 95.1% pure peptide by HPLC, with a protein content of 88%.
Female C57BL/6 mice aged 56 ± 3 days were purchased from Charles River Breeding (Cambridge, MA, USA). Mice were housed five per cage, and kept in a controlled 12:12 h light/dark environment, and provided food ad libitum. All procedures were approved by local IACUC.
Induction of EAE
For study 1, a 3 mg/mL solution of MOG-A or MOG-B was prepared by dissolving 3 mg of the peptide preparation in 1 mL of saline. A 100-μL aliquot was emulsified with 100 μL of complete Freund's adjuvant (CFA) containing 500 μg of Mycobacterium tuberculosis (Difco, Detroit, MI, USA). After emulsification, a 200-μL aliquot was injected subcutaneously into a single area over the hind limb. Immediately after injection, the animals received an intraperitoneal (i.p.) injection of pertussis toxin (PT, 200 ng in 200 μL phosphate-buffered saline). Two days later, the mice received a second PT injection. In study 2, some mice received a booster immunization 7 days later that was identical to the first immunization. In study 3, mice were immunized as above using MOG-B containing TFA or with MOG-B containing acetate, and a booster immunization was given. In this study, the amount of material injected was normalized based on the peptide fraction, such that each mice was immunized with 60 μL of a solution containing 4 mg/mL MOG35-55 peptide, for a total of 240 μg of peptide. In this study, some mice were immunized with MOG-B in which additional TFA was added (MOG-B xTFA) to the final emulsion to bring the final concentration to 2.75 mg/mL, and so the mice received 240 μg peptide and 165 μg TFA. Clinical signs were scored on a five-point scale: grade 0, no clinical signs; (i) limp tail; (ii) impaired righting; (iii) paresis of one hind limb; (iv) paresis of two hind limbs; (v) death. When a mouse died, it was assigned a score of 5 which was maintained for the remainder of the study for calculation of average daily disease score and average of the highest scores exhibited.
Comparison of the development of clinical signs was done by two-way repeated measures anova. Pair-wise comparisons (average day of onset, average maximum clinical score) were done by non-parametric t-test. Significance was taken at p < 0.05.
We first compared the development of clinical scores in mice immunized a single time between two different commercially bought MOG peptide preparations (Table 1, Study 1). In this study, the amount of material injected was based upon the total mass of the MOG preparation and did not take into account differences in peptide fraction. As such, 300 μg of MOG-A contained 135 μg of MOG peptide; whereas 300 μg of MOG-B contained 216 μg of peptide. The average day of disease onset was slightly earlier in MOG-A immunized mice (10.5 ± 0.7 days) versus MOG-B immunized mice (12.2 ± 0.5 days); however, this difference did not reach statistical significance (p = 0.07). The overall disease incidence was similar in the two groups, reaching 100% (16/16) by day 12 in the MOG-A group, and 86.7% (13/15) at day 16 in the MOG-B group (Fig. 1a). However, maximum clinical scores and subsequent recovery differed (Fig. 1b). In the MOG-A group, the maximum average daily clinical score (MACS) was 2.8 ± 0.2 on day 17, after which eight mice showed some recovery. In contrast, in the MOG-B group, the MACS reached only 1.9 ± 0.3 on day 16. After that, six mice showed signs of recovery of which four reached scores of 0. At the end of the study, the average clinical score was 2.1 ± 0.2 in the MOG-A group, and the incidence remained at 100% (16/16). In the MOG-B group, at the end of the study the average clinical score decreased to 1.1 ± 0.3 and incidence to 60% (9/15). There was a statistically significant difference in the evolution of clinical scores in the two groups (F[12,1] = 2.78, p = 0.0012). In addition, the average of the maximum clinical scores (AMCS) attained by mice in the MOG-A group was significantly greater than that those in the MOG-B group (3.1 ± 0.2 vs. 2.0 ± 0.3, p < 0.005). These results show that mice receiving a single immunization with the MOG-A preparation of lower peptide content (135 μg vs. 216 μg in the MOG-B preparation), showed a trend toward an earlier day of onset, had significantly higher average clinical scores, and showed reduced magnitude of recovery.
Table 1. Summary of disease parameters
Day of onset
Score at end
Inc at end
Inc Max, maximum disease incidence during study; MACS, maximum average clinical score, the highest average daily score for all mice in the group; AMCS, average maximum clinical score, the average of the highest scores for all mice in group across entire study.
p = 0.0012 versus Study 1, MOG-A (time × group); F[12,1] = 2.78
p < 0.0001 versus Study 1, MOG-A (time × group); F[12,1] = 3.72
p = 0.0014 versus Study 1, MOG-B (time × group); F[12,1] = 2.14
p < 0.0001 versus Study 2, MOG-A (time × group); F[13,1] = 5.02
We tested if differences in disease course owing to different MOG preparations were maintained if mice were given a booster immunization (Table 1, Study 2). Immunizations were carried out as above, with a second booster immunization given 7 days after the first immunization. Although the average day of disease onset was similar to that observed in the single-immunized groups; in the booster immunized mice the average day of onset was significantly lower in the MOG-A group (10.1 ± 0.6 day) versus the MOG-B group (13.4 ± 0.2 days for MOG-B). Disease incidence was also similar to the single-immunization groups, reaching 100% (15/15) by day 12 in the MOG-A group, and 82.4% (14/17) at day 21 in the MOG-B group (Fig. 2a). In the MOG-A group, the MACS was 3.9 ± 0.1 on day 21, after which two mice showed slight recovery; and two mice died. At the end of the study, the average clinical score was 3.7 ± 0.2 and incidence 100% (13/13). The booster immunization significantly influenced the development of clinical scores using MOG-A (F[12,1] = 3.72, p < 0.0001 vs. MOG-A no booster). In the MOG-B group, the additional booster also significantly increased the MACS which was 2.6 ± 0.4 on day 21 (F[12,1] = 2.14, p = 0.0014 vs. MOG-B no booster). After that, five mice showed slight (one point) recovery, and three mice worsened. At the end of the study, the average clinical score decreased slightly to 2.3 ± 0.4 and incidence remained at 82.4% (14/17). There was a statistically significant difference in the evolution of clinical scores in the MOG-A booster versus MOG-B booster groups (F[13,1] = 5.03, p < 0.0001) (Fig. 2b). In addition, booster immunization increased the maximum scores reached by all mice versus singly immunized mice (Table 1, AMCS scores); and the AMCS of mice in the MOG-A group with booster (4.1 ± 0.1) was significantly greater than those in the MOG-B group with booster (2.8 ± 0.4, p < 0.005). These results show that booster immunizations significantly increased disease scores, and that booster immunization with MOG-A, containing the lower peptide fraction, accelerated disease onset.
The addition mass present in MOG preparations comprised primarily the counter ion used for the final step of purification, in this case TFA. Thus, 300 μg of MOG-A material contained approximately 55% TFA (roughly 165 μg), whereas 300 μg of MOG-B preparation contained approximately 28% TFA (roughly 84 μg). TFA has been reported to covalently modify proteins, which could alter their immunogenicity. However, HPLC/MS analysis of the MOG-A and MOG-B preparations revealed a single major peak of molecular weight 2851.4 daltons, equivalent to the theoretical weight of the MOG35-55 peptide (data not shown). TFA has been reported to exert inflammatory responses, suggesting that the higher TFA in MOG-A may have influenced the development or severity of EAE. To test this, we compared MOG-B containing approximately 20% TFA with MOG-B containing approximately 12% acetate. We immunized mice with these two preparations, taking into account the difference in peptide fraction so that the mice received equivalent amounts (240 μg) of the peptide (Fig.3). The average day of disease onset was significantly earlier in the TFA versus the acetate group (11.6 ± 0.8 vs. 16.4 ± 1.7; p < 0.05). However, the subsequent disease course was similar in the two groups. The incidence of disease reached 78% (7/9) in the TFA group at day 16 and on day 18 in the acetate group, although it increased to 100% (9/9) in the latter group at day 28. The average maximum clinical score was 2.2 ± 0.6 at day 16 for the TFA group, after which three showed slight recovery and the clinical score decreased to 1.9 ± 0.8 and the end of the study (day 35). In the acetate group, the maximum clinical score was similar (2.0 ± 0.5 at day 18; which transiently increased at day 28 for 1 day to 2.1 ± 0.3). After that two mice showed some recovery and the final average clinical score was 1.8 ± 0.8 on day 35 and final incidence 78%. The average of the maximum clinical scores was similar in the two groups. These results suggest that TFA influences an early stage of EAE leading to more rapid disease onset.
To test if TFA could directly influence EAE disease, a group of mice were immunized with 60 μL of a 4 mg/mL solution of MOG-B to which an additional 1.75 mg/mL of TFA was added, bringing the final TFA concentration to 2.75 mg/mL. These mice therefore received 240 μg of MOG peptide and a similar amount of TFA (165 μg) as the mice immunized with MOG-A in study 1. However, the additional TFA did not significantly affect disease incidence or severity (Fig. 4), or the average day of onset or maximal clinical scores (data not shown).
The earliest EAE studies observed that a single injection of a monoclonal antibody to MOG accelerated a chronic relapsing form in SJL mice that had been induced with the myelin basic protein (Schluesener et al. 1987). After successful purification of human MOG, it was quickly shown to induce a chronic relapsing remitting disease in Lewis rats (Abo et al. 1993; Johns et al. 1995). Encephalitogenic epitopes of MOG were then characterized which could induce by direct immunization a relapsing remitting disease in Biozzi antibody high and SJL mice (Amor et al. 1994). Later the MOG35-55 peptide was shown to induce a highly reproducible chronic disease in C57BL/6 and C3H.SW mice (Mendel et al. 1995).
Since that time the MOG35-55 peptide has been used extensively. The protocol used in this study, immunization of C57BL/6 mice with MOG35-55 peptide in CFA, followed by two injections of PT, is now the most prevalent (Ransohoff 2012). In the majority of studies, the disease has a monophasic chronic course; however, in some reports the disease shows a transient course or a chronic relapsing remitting course. The basis for these differences is not well understood; however, in one recent study the effects of peptide and M. tuberculosis dose on disease were examined (Berard et al. 2010). The authors reported that immunization with 300 μg MOG35-55 in CFA containing 400 μg M. tuberculosis yielded a chronic course, whereas immunization with 50 μg MOG35-55 in CFA containing only 50–100 μg M. tuberculosis led to a relapsing remitting course. The authors found that chronic disease is associated with a dominance of CD8+ T cells and more efficient activation of naïve myelin-reactive T cells (Berard et al. 2010). Relapsing EAE has also been associated with increased MOG antibody response (Zhang et al. 2004).
In this study, we found that after a single immunization using MOG-A, containing a lower amount of peptide, mice exhibited a chronic disease with moderate severity, and with little recovery. In contrast, after a single immunization with MOG-B which contained about 60% more peptide (216 vs. 135 μg), the mice showed less severe clinical signs, and by the end of the study four of the 15 mice recovered completely, reducing the incidence to 60%. In contrast, when mice were given a booster immunization, the disease course for both MOG preparations was again chronic but with little recovery; and the clinical scores were significantly higher. However, the booster immunization with MOG-A resulted in a more rapid appearance of symptoms and more severe disease, consistent with our previous experience using this peptide preparation. The current studies were carried out to 30 days; whether differences in disease severity or incidence between the two different MOG preparations are maintained for longer times, or converge, is therefore not known.
The increased disease severity following immunization with MOG-A could be because of several factors. The dependency of EAE disease on the amount of MOG peptide used for immunization has not been well characterized. However, a recent study (Sestero et al. 2012) showed that in wild-type male C57Bl6 mice, immunization using 450 μg of MOG35-55 resulted in a less severe disease than using 150 μg of peptide. This was hypothesized to involve increased T-cell anergy, the induction of immune non-responsiveness or activation of induced cell death (AICD) following exposure to high doses of antigen. In contrast, in mouse lines where the threshold for AICD was modified by knockout of casein kinase 1 (Sestero et al. 2012) or of the IFNα-R (Naves et al. 2013), disease severity increased with immunization dose. Consistent with a role for AICD in disease progression, exposure of myelin basic protein-reactive TCR transgenic T cells to a superagonist peptide led to significant AICD mediated through Fas signaling (Ryan et al. 2005); and spontaneous remissions in proteolipiod protein peptide induced EAE involved Fas mediated AICD (Suvannavejh et al. 2000). It is therefore possible that lower disease severity following immunization with MOG preparations of higher peptide fraction is due in part to induction of Tcell AICD, which itself can differ between transgenic mouse lines.
In this study, we focused attention on differences in the other material present in the final purified preparations, which in this case is primarily TFA, the counter ion used for purification, in chromatography to purify the peptide. In our studies, immunization with 300 μg MOG-A therefore contained approximately 165 μg TFA; whereas 300 μg of MOG-B contained approximately half, only 84 μg TFA. TFA is known to have significant effects on cells and tissues, both in vitro and in vivo. TFA inhibits cell proliferation at concentrations as low as 10 nM (Cornish et al. 1999); and at higher doses (0.5–7.0 mM) stimulates growth of glioma cells (Ma et al. 1990). TFA has recently been shown to act as an allosteric modulator of glycine receptors, increasing receptor activity at lower glycine concentrations (Tipps et al. 2012) and previously was shown to activate ATP sensitive potassium channels (Han et al. 2001). In vivo, TFA can trifluoroacetylate amino groups in proteins and phospholipids (Satoh et al. 1985), which can elicit antibody responses (Trudell et al. 1991; Furst et al. 1997), increase pro-inflammatory cytokine production (You et al. 2006), and induce T-cell responses (You et al. 2010). These observations raised the possibility that the higher TFA in MOG-A was contributing to increased disease severity. We found that reducing the TFA content by exchanging it for acetate caused a delay in the average day of onset (Fig. 3b). This is consistent with our findings that the average day of onset for MOG-B was delayed compared to MOG-A which contained less TFA. Although we did not observe increased disease severity, this may be due in part since the MOG-A contained close to 55% TFA, while MOG-B in study 3 contained closer to 20%. In contrast, immunization with MOG-B in which additional TFA was directly added to the emulsion (Fig. 4) did not alter disease progression. It remains possible that other material present at different levels in the MOG preparations contribute to the difference in EAE, however, since these are present in trace quantities, this seems unlikely.
In summary, our findings demonstrate that immunization with two distinct preparations of MOG35-55, having similar degrees of purity but different peptide fractions, leads to distinct patterns of clinical disease. A knowledge of final peptide content and counter ions present is therefore critical when comparing data obtained with peptides from different sources as well as from different peptide lots.
Acknowledgments and conflict of interest disclosure
This work was supported in part by a grant from the National Multiple Sclerosis society (DLF), from the VA Merit Review (DLF) and by a Research Career Scientist award (DLF). We wish to thank Dr Susan McGuire for fruitful conversations.
All experiments were conducted in compliance with the ARRIVE guidelines. The authors have no conflict of interest to declare.