autoantigens with 4-hydroxy-2-nonenal (HNE) provide ideal substrates for detection of anti-HNE antibodies and peptide antioxidants

Journal of Immunological Methods 303 (2005) 66 – 75 www.elsevier.com/locate/jim

Research paper

In vitro modification of solid phase multiple antigenic peptides/ autoantigens with 4-hydroxy-2-nonenal (HNE) provide ideal substrates for detection of anti-HNE antibodies and peptide antioxidants Biji T. Kurien a,*, R. Hal Scofield a,b,c b

a Arthritis and Immunology Program, 825 NE 13th Street, Oklahoma City, OK 73104, USA Department of Medicine, University of Oklahoma Health Sciences Center, Oklahoma City, OK 73104, USA c Department of Veterans Affairs Medical Center, Oklahoma City, OK 73104, USA

Received 25 March 2005; received in revised form 23 May 2005; accepted 25 May 2005 Available online 12 July 2005

Abstract The role of free radicals in protein modification and the importance of anti-4-hydroxy-2-nonenal (HNE) antibodies as marker of HNE-mediated cell toxicity has been well documented. Proteins modified by HNE in vitro, prior to immobilization on ELISA plates, have served as substrates for assaying these antibodies. We found preferential binding of HNE-modified versus unmodified proteins to ELISA plates and this prompted us to seek a more reliable assay. We report a method to HNE-modify any cysteine/ histidine/lysine-containing protein or multiple antigenic peptide (MAP) following their immobilization on an ELISA plate. To a set of wells, HNE (200 AM) dissolved in PBS is added and incubated for 4 h, followed by regular ELISA. Since HNE was supplied dissolved in ethanol, PBS with appropriate amount of ethanol added was used as control. For inhibition experiments, HNE is incubated with or without inhibitors and then added to the wells. The commercial anti-HNE serum bound only to HNE-modified antigens. Sera from rabbits and mice immunized with HNE-modified 60 kDa Ro autoantigen preferentially bound the modified antigens. Modification of solid phase antigens in this manner makes assaying anti-HNE antibodies unambiguous. Lengthy dialysis procedures or the use of spin columns that lead to antigen loss becomes unnecessary for the separation of free HNE. We were able to HNE-modify various antigens (BSA, the autoantigens Ro, La and Sm/nRNP, 60 kDa Ro and Sm MAPs) using this procedure. Using MAPs, we confirmed the importance of histidine, lysine and cysteine residues in HNE modification. In addition, this method allowed identification of inhibitors of HNE-modification. We obtained 61%, 70% and 74% inhibition of HNEmodification of solid phase Ro MAP 166 substrate using BSA, Ro MAP 482 and Ro MAP 166, respectively. Glycyl-proline dipeptide and a MAP from the Sm autoantigen (PPPGMRPP) showed 0% inhibition of HNE-modification. D 2005 Elsevier B.V. All rights reserved. Keywords: Free radicals; Lupus; Autoimmunity; Epitopes; Autoantibodies; Antigens; Hydroxynonenal

* Corresponding author. Tel.: +1 405 271 7394; fax: +1 405 271 4110. E-mail address: [email protected] (B.T. Kurien). 0022-1759/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jim.2005.05.012

B.T. Kurien, R. Hal Scofield / Journal of Immunological Methods 303 (2005) 66–75

1. Introduction In 1954 Gerschman et al. (Finkel and Holbrook, 2000) proposed that hyperoxic injury results in the formation of reactive oxygen species (ROS). Since then our knowledge about the physiology of oxygen and its reduced metabolites (ROS) has grown phenomenally. A variety of intracellular processes leads to the production of ROS like superoxide anions, hydrogen peroxide and hydroxyl radicals. Under normal circumstances, cellular defense systems comprising antioxidant enzymes (superoxide dismutases, catalase, glutathione peroxidase) and antioxidants (reduced glutathione, ascorbic acid, vitamin E) are able to overwhelm the ROS flux. However, when the proxidant/antioxidant balance is compromised under oxidative stress, the balance tilts in favor of the ROS, leading to oxidative damage. ROS chemically modifies a variety of bio-molecules such as polyunsaturated fatty acids that form part of membrane phospholipids, leading to the formation of hydroperoxides. Also, bioactive lipid hydroperoxides are formed via the lipoxygenase enzyme systems. The hydroperoxides formed from the combined chemical and enzymatic processes form the substrates for the Hock cleavage that generates an array of a, h acyl aldehydes. Of these, the most studied one is the cytotoxic aldehyde, 4-hydroxy-2-nonenal (HNE). Thus, lipid peroxidation mediated by free radicals is always associated with the formation of reactive aldehydes (Van Kuijk et al., 1990; Esterbauer et al., 1991; Finkel and Holbrook, 2000; Bennaars-Eiden et al., 2002; Thannickal, 2003; Siems and Grune, 2003; Kurien and Scofield, 2003). HNE, a comparatively stable, diffusible and longlived molecule, is considered to be cytotoxic, mutagenic and genotoxic on account of its ability to covalently modify cellular target proteins/peptides, phospholipids and nucleic acids, thereby bringing about conformational change and altered biological function. The three nucleophilic amino acids, cysteine, histidine and lysine, have been shown to be the target of modification by HNE (Esterbauer et al., 1991). HNE-modified proteins have been shown to be present in the sera of children with systemic lupus

67

erythematosus (SLE), an autoimmune disorder (Grune et al., 1997). We, and others, have shown increased oxidative damage in SLE (Grune et al., 1997; Jiang and Chen, 1992; Kurien and Scofield, 2003). We have also shown that immunization of rabbits with an autoantigen modified in vitro with HNE, results in rapid loss of tolerance to the immunogen compared to immunization with the unmodified autoantigen (Scofield et al., 2005). The cellular targets for HNE modification have been the focus of intense investigation thus far. Evidence has accumulated for the presence of antibodies against HNEmodified proteins (Zheng et al., 1992; Maggi et al., 1995; Bergmark et al., 1995; Holvoet et al., 1995; Rhodes et al., 1995; Itabe et al., 1998; Chang et al., 1999; Allison and Fearon, 2000; Mottaran et al., 2002; Willis et al., 2002; Scofield et al., 2005). Therefore it is vital to have a reliable in vitro assay to detect antibodies against HNE-modified proteins. Proteins modified in vitro by HNE (prior to immobilization on ELISA plates), in the presence (Scofield et al., 2005) or absence (Bennaars-Eiden et al., 2002) of sodium cyanoborohydride, have been used to coat ELISA plates for detection of anti-HNE autoantibodies. However, we were faced with the question of differential immobilization (on ELISA plates) of HNE-modified proteins compared with unmodified proteins (since binding to plastic is charge/conformation-dependent). With the modification of histidines and lysines by HNE, the proteins tend to lose positive charges due to these amino acids, and thereby become more negatively charged. This enables the modified proteins to bind to ELISA plates better than the unmodified proteins, thereby skewing the results obtained. In addition, the conformation of the protein also changes following modification, which may also affect the binding of the protein to the solid surface. By modifying a solid phase protein, as reported here, removes the ambiguity associated with the binding of proteins that have been modified by HNE prior to immobilization. This modification procedure just requires an additional step in the normal ELISA format and takes away the additional step of dialyzing the unbound HNE away or the use of spin column to separate the free HNE We were also able to confirm that indeed histidines, cysteines or lysines

68

B.T. Kurien, R. Hal Scofield / Journal of Immunological Methods 303 (2005) 66–75

are essential for modification by HNE using multiple antigenic peptides (MAPs) that did or did not have these residues. In addition, we were able to identify soluble phase peptides/antigens that inhibited HNE from modifying solid phase antigens (antigens immobilized on ELISA plates).

2.5. Preparation of 60 kDa Ro antigen for immunization

2. Materials and methods

Purified 60 kDa Ro was modified by the addition of 10 mM 4-hydroxy-2-nonenal at room temperature for 24 h in the presence of 10 mM sodium cyanoborohydride (NaCNBH3) and then dialyzed against 0.1 N NaCl using a 10 kDa molecular weight cut-off membrane (Scofield et al., 2005).

2.1. Materials

2.6. Rabbit immunization

4-hydroxy-2-nonenal was from Cayman Scientific (Ann Arbor, MI); alkaline phosphatase conjugated anti-rabbit and anti-mouse IgG were from Jackson Laboratories, Bar Harbor, ME. Serocluster bUQ vinyl ELISA plates were from Costar, Cambridge, MA. Glycyl-proline, bovine serum albumin and phosphate-buffered saline tablets were from Sigma Chemical Co, MO. All other chemicals were reagent grade.

Rabbit immunization with HNE-modified and unmodified 60 kDa Ro was carried out as reported earlier (Scofield et al., 2005).

2.2. 60 kDa Ro and Sm-RNP autoantigens Purified bovine 60 kDa Ro (Scofield et al., 1999, 2004) and Sm-RNP antigens were a generous gift from Immunovision (Springdale, AK).

2.7. Mice immunization Five groups of BALB/c mice with 10 mice in each group were used. Group I was immunized with purified 60 kDa Ro treated with 1 mM NaCNBH3 alone. Groups II to IV were immunized with 60 kDa Ro modified with 0.4 mM, 2 mM and 10 mM HNE, respectively, in the presence of 1 mM NaCNBH3. Group V mice received only saline. The anti-sera from these groups of mice were used to determine binding to Ro and multiple antigenic peptides (MAPs) from 60 kDa Ro.

2.3. Multiple antigenic peptides (MAPs) MAPs were constructed from the sequence of 60 kDa Ro (Deutscher et al., 1988; Ben-Chetrit et al., 1989; Scofield et al., 2004) as mentioned (Scofield et al., 2005). MAPs contain multiple copies of the same peptide sequence (eight copies) coupled covalently to a hepta lysine backbone (Kurien et al., 1998). Ro MAP 166–180 (LAVTKYKQRNGWSHK), Ro MAP 310–323 (NEKLLKKARIHPFH), Ro MAP 482–495 (ALREYRKKMDIPAK) and an unrelated MAP from Sm autoantigen with the sequence PPPGMRPP (James et al., 1995) were used for modification with HNE in this study. 2.4. La autoantigen Recombinant La protein was expressed and purified as described earlier (Bachmann et al., 1997).

2.8. In vitro modification of solid phase proteins by HNE Purified 60 kDa Ro/La/Sm-RNP/BSA antigens and 60 kDa Ro/Sm MAPs were coated at room temperature for 2 h on ELISA plates in carbonate coating buffer. The plates were washed twice with phosphate buffered saline (150 Al each time), pH 7.4. The plates were tapped dry and 50 Al of HNE (200 AM) in PBS was added per well. Since HNE was supplied dissolved in ethanol, PBS with appropriate amount of ethanol added was used as the control. This PBS solution (50 Al/well) was added for the protein fraction that served as unmodified controls. The plates were incubated at room temperature in a humid chamber for 4 h. The plates were washed with PBS containing 0.05% Tween (PBST). The plates were blocked for 2 h at room temperature with 3% milk/PBS and incubated with sample overnight at 4 8C. The plates

B.T. Kurien, R. Hal Scofield / Journal of Immunological Methods 303 (2005) 66–75

were then washed with PBST and incubated with appropriate secondary antibody for 2 h at room temperature. Following washing, color was developed with p-nitrophenol phosphate substrate. For kinetic studies, Ro/La/Ro MAP coated plates were treated with different concentrations of HNE for varying periods of time at room temperature or at 4 8C. For inhibition experiments HNE was incubated for 75 min at room temperature with or without glycyl-proline, BSA, Ro MAPs 166, 310 or 482 and a PPPGMRPP sequence containing MAP from the Sm autoantigen.

3. Results Our laboratory has been using 60 kDa Ro modified in vitro with HNE (modified in bulk and stored at 80 8C in aliquots) to coat ELISA plates for the detection of anti-HNE antibodies. Since there was the possibility of differential binding of HNE-modified protein to these plates compared to the unmodified protein, we first tested the binding of a commercial anti-HNE antibody to these two sets of

69

proteins (HNE-modified proteins immobilized on ELISA plates after HNE-modification and proteins modified with HNE following immobilization). HNE-modified 60 kDa Ro (modified in vitro with 200 AM HNE in solution; Scofield et al., 2005) was immobilized on an ELISA plate. Unmodified 60 kDa Ro was also immobilized and one half of the unmodified Ro antigen was modified with 200 AM HNE (for 1 or 2 h). The other half was not modified with HNE and was used as the control. There was no significant difference in anti-HNE binding between the 60 kDa Ro modified with HNE for 1 or 2 h. We found that there was a 6-fold increase in binding by a commercial anti-HNE antibody to the HNE-modified 60 kDa Ro (modified prior to immobilization on the plate) compared to the unmodified antigen that was immobilized first and then modified with HNE (Fig. 1). Anti-Ro 60 containing human serum bound to unmodified 60 kDa Ro with an OD of 3.6 F 0.8 while a normal serum did not bind at all. The increase in binding seen in Fig. 1 is entirely due to anti-HNE binding to the HNE on the modified protein. AntiHNE bound much faster to the antigen that had been modified in solution compared to the antigen that was

2

Non-immune control 1.6

Anti-HNE

1.2

OD 0.8 0.4 0 Ro, HNE-modified and bound to plate

Ro bound to plate and HNE-modified

Unmodified Ro bound to plate

60 kDa Ro Fig. 1. The difference in binding of a commercial anti-HNE antibody to HNE-modified 60 kDa Ro (modified prior to immobilization) immobilized on an ELISA plate compared to binding to 60 kDa Ro modified with HNE for 1 h following immobilization on the plate. 60 kDa Ro was modified with 200 AM HNE for 24 h at room temperature in the presence of 10 mM sodium cyanoborohydride and the unbound HNE was removed by dialysis. This was coated on the plate at 5 Ag/ml. Unmodified 60 kDa Ro was also coated at 5 Ag/ml and modified with HNE for 1 h (post-immobilization). The HNE-modification was carried out as mentioned in Materials and methods except for the fact that the incubation with HNE was carried out for 1 h/2 h instead of 4 h. 60 kDa Ro that was not modified with HNE post-immobilization was used as the control. These antigens were tested for their ability to bind a commercial anti-HNE antiserum. Anti-Ro 60 containing human serum bound to unmodified 60 kDa Ro with an OD of 3.6 F 0.8 while a normal serum did not bind at all. The values are means F S.D. for 6 determinations.

70

B.T. Kurien, R. Hal Scofield / Journal of Immunological Methods 303 (2005) 66–75

the best based on the binding by anti-HNE. Incubation with increasing amounts of HNE gave a progressively lower response (Fig. 2A). The anti-HNE did not bind to the unmodified 60 kDa Ro protein. Next we determined the optimal time required for obtaining the best HNE-modification of an antigen. 60 kDa Ro incubated with HNE for 4 h yielded the highest binding by the anti-HNE compared to that obtained with the protein incubated for 10 min and 1, 2, 8, 16 or 24 h with HNE (data not shown). Since modification with 200 and 400 AM HNE gave almost similar binding (Fig. 2A), we investigated this further using other HNE-antibodies and an anti-Ro containing human serum. Fig. 2C and D show the binding of rabbit and human sera to unmodified and HNE-modified Ro (Fig. 2C—200 AM HNE; Fig. 2D—400 AM

modified post-immobilization even when the antigen was modified for 4 h as seen in Fig. 2. This indirectly showed that HNE-modified 60 kDa Ro bound better to the ELISA plate compared to the unmodified protein. Therefore, we worked on perfecting this method of modifying the antigen with HNE following immobilization to detect anti-HNE antibodies by ELISA unambiguously. Consequently we observed that it was feasible to modify the antigen following its immobilization on the plate. First we determined the optimal HNE required for modifying an antigen. Ro autoantigen was first immobilized and incubated with different amounts of HNE. The amount of HNE covalently bound to the protein was determined using a commercial anti-HNE serum. It was found that modification with 200 AM HNE was

A

B 2

OD

1 - 100 µM HNE 2 - 200 µM HNE

3 - 500 µM HNE

1.2

4 - 1 mM HNE

1.6

5 - 5 mM HNE

1.2

6 - 10 mM HNE

0.8

*

Non-immune Anti-Ro 60 Anti-HNE Ro 60 Anti-HNE

0.8 0.4 0.4 0 Ro

1

2

3

4

5

6

CC

0

Ro modified with HNE

C

D 3

Non-immune Anti-Ro Anti-HNE Ro

2

Anti-HNE Human anti-Ro Human normal

La

HNE modified La

Non-immune Anti-Ro Anti-HNE

Human anti-Ro Human normal

Anti-HNE Ro

OD 1

0 Ro

HNE (200 µM) modified Ro

Ro

HNE (400 µM) modified Ro

Fig. 2. Binding of anti-Ro, anti-HNE-Ro and anti-HNE antibodies to unmodified Ro/La and HNE modified Ro/La. The antigens were modified with HNE (post-immobilization) as mentioned in Materials and methods and tested for their ability to bind anti-Ro, anti-HNE-Ro and anti-HNE antibodies. (A) Anti-HNE binding to unmodified Ro and Ro modified with different amount of HNE. (B) Anti-Ro, anti-HNE-Ro, anti-HNE antibodies and preimmune serum binding to unmodified and HNE-modified La. (C and D) Anti-Ro, anti-HNE-Ro, anti-HNE antibodies and preimmune serum binding to unmodified Ro and Ro modified with 200 (C) and 400 AM (D) of HNE, respectively. Values are means F S.D. for four determinations for panel B. Values are means for two determinations for panels A, C and D. CC refers to conjugate control. *t = 9.49, p b 0.001.

B.T. Kurien, R. Hal Scofield / Journal of Immunological Methods 303 (2005) 66–75

of the commercial anti-HNE serum as well as the sera of rabbits immunized with unmodified Ro or HNEmodified Ro is shown in Fig. 3. It can be seen that the commercial anti-HNE binds only to the HNE-modified peptide just like earlier results (Fig. 3B, C and D). It is very interesting to see that anti-HNE antibody does not bind to HNE-modified PPPGMRPP MAP from Sm (Fig. 3A). The increased binding by antiHNE Ro or anti-Ro sera to the modified PPPGMRPP in not significant compared to the binding by the preimmune sera (Fig. 3A). Anti-HNE Ro bound significantly higher to the HNE-modified Ro MAP 310 compared to the unmodified Ro MAP 310 (t = 2.97, p = 0.025). Anti-HNE bound twice as much to the HNE-modified Ro MAP 310 compared to the unmodified MAP (Fig. 3B). Similarly anti-HNE Ro sera bound significantly to Ro MAPs 166 and 482 compared to the unmodified MAPs (t = 8.62, p = b 0.001 and t = 3.44, p = 0.014, respectively) (Fig. 3C and D).

HNE). Ro modified by HNE at 200 AM was found to be bound better by anti-HNE sera (both commercial anti-HNE as well as anti-HNE Ro sera), while rabbit anti-Ro and human anti-Ro sera bound less to HNEmodified 60 kDa Ro compared to unmodified 60 kDa Ro. Similar results were obtained with the La autoantigen also (Fig. 2B). Anti-HNE Ro bound significantly higher to unmodified La and HNE-modified La compared to anti-Ro rabbit serum. Importantly, commercial anti-HNE serum bound highly significantly to HNE-modified La compared to unmodified La (t = 9.49, p b 0.001). Next we used MAPs with or without histidine, lysine or cysteine residues to check their susceptibility to modification by HNE, following their immobilization on the plate. Three MAPs constructed from the 60 kDa Ro sequence containing multiple histidines/ lysines and one MAP from Sm autoantigen that lacked these three residues were used. The binding

A

OD

71

B 1.2

Non-immune Anti-Ro 60

0.8

Anti-HNE Ro 60 Anti-HNE

Non-immune Anti-Ro 60 Anti-HNE Ro 60 Anti-HNE

*

0.4 0 0

Sm-MAPPPPGMRPP

HNE-modified SmMAP-PPPGMRPP

Ro-MAP-310

C

D

**

1.2

OD

HNE-modified Ro-MAP-310

*** #

0.8

##

0.4 0 Ro-MAP-166

HNE-modified Ro-MAP-166

Ro-MAP-482

HNE-modified Ro-MAP-482

Fig. 3. Binding of preimmune serum, anti-Ro, anti-HNE-Ro and commercial anti-HNE antibodies to unmodified MAPs and HNE modified Ro MAPs and a MAP from Sm autoantigen. The antigens were modified with HNE (post-immobilization) as mentioned in Materials and methods and tested for their ability to bind anti-Ro, anti-HNE-Ro and anti-HNE antibodies. (A) The binding of all four sera to modified and unmodified Sm MAP PPPGMRPP. (B) The binding of all four sera to modified and unmodified Ro MAP 310. (C) The binding of all four sera to modified and unmodified Ro MAP 166. (D) The binding of all four sera to modified and unmodified Ro MAP 482. Values are means F S.D. for four determinations. *t = 2.97, p = 0.025; **t = 8.62, p b 0.001; ***t = 16.4, p b 0.001. #t = 16.4, p b 0.001; # #t = 3.9, p = 0.03.

72

B.T. Kurien, R. Hal Scofield / Journal of Immunological Methods 303 (2005) 66–75

1- α-HNE binding unmodified Ro MAP166 2- Control sera binding unmodified Ro MAP166

3

3- α-HNE binding HNE-modified Ro MAP166 4- Control sera binding HNE-modified Ro MAP166 2

OD

**

1

*

0 1

2

3

4

Uninhibited

1

2

3

4

Ro MAP166inhibited

1

2

3

4

Ro MAP 482inhibited

1

2

3

4

PPPGMRPP MAP-inhibited

Ro MAP 166 Fig. 4. Inhibition of binding of HNE to Ro MAP 166 used as the solid phase antigen. Ro MAP 166 was immobilized on ELISA plates and tested for modification by HNE in the presence or absence of Ro MAP 166, Ro MAP 482, PPPGMRPP, glycyl-proline and BSA. HNE was incubated with or without each these inhibitors at room temperature for one hour and then added to Ro MAP 166 immobilized wells of ELISA plates. The degree of modification was checked with anti-HNE sera. Figure shows the binding of anti-HNE and preimmune sera to unmodified and HNEmodified Ro MAP 166 in the presence or absence of the various inhibitors used. Values are means F S.D. for four determinations. *t = 20, p b 0.001; **t = 16.8, p b 0.001) For BSA (data not shown), t = 8.45, p b 0.001.

In addition, commercial anti-HNE bound highly significantly to the HNE-modified Ro MAPs 166 and 482 compared to the unmodified MAPs (t = 16.4, p b 0.001 and t = 3.9, p = 0.03, respectively) (Fig. 3C and D). The mice immunized with different levels of HNEmodified Ro also behaved almost similarly. The saline immunized group did not bind to either modified or unmodified antigens, while the mice sera from the groups immunized with HNE-modified Ro and the commercial rabbit anti-HNE serum bound to the HNE-modified MAPs and not to the PPPGMRPP MAP (data not shown). We then investigated the ability of Ro MAP 166, Ro MAP 480, Sm MAP PPPGMRPP, glycyl-proline and BSA to act as inhibitors of HNE-modification of Ro MAP 166 immobilized on ELISA plates. We found that PPPGMRPP and glycyl-proline did not inhibit HNE modification of the substrate (0% inhibition in both cases). However, there was a 61%, 70% and 74% inhibition of HNE modification of Ro MAP 166 by BSA (data not shown), Ro MAP 482 and Ro MAP 166, respectively (Fig. 4). The inhibition of

binding of anti-HNE to HNE-modified Ro MAP 166 immobilized on the plate was highly significant by BSA (t = 8.45, p b 0.001), Ro MAP 482 (t = 16.8, p b 0.001) and Ro MAP 166 (t = 20, p b 0.001) compared to the binding of anti-HNE to Ro MAP 166 in the absence of inhibitors.

4. Discussion The question of preferential binding of HNE-modified proteins (on account of the modification of lysine/ histidine) compared to unmodified proteins to ELISA plates prompted us to investigate the feasibility of modifying proteins immobilized to a solid surface. In this manner, only that portion of the molecule not in contact with the plate is liable to become modified, and thereby allows a more reliable assay to detect antibodies to HNE. Previous studies have demonstrated that the addition of HNE (in the absence of sodium cyanoborohydride) to a protein in solution is a biphasic process, with the initial process of adding a single molecule of

B.T. Kurien, R. Hal Scofield / Journal of Immunological Methods 303 (2005) 66–75

HNE happening in less than a minute. This was followed by the slower addition of a second and third molecule within 20 to 120 min and the addition of a total of up to 6 molecules of HNE in 16 h (Bennaars-Eiden et al., 2002). We have a different scenario in our hands, since this procedure involves modification of the protein post-immobilization on a solid surface. In our assay, the best HNE modification was obtained in 4 h. The peptide was incubated with a 320-fold molar HNE excess while the protein was incubated with a 1175-fold excess (200 AM HNE). When a 600-fold (100 AM HNE) molar excess of HNE was used for the protein, the binding of antiHNE sera was less than half that was obtained with 1175 fold excess of HNE (Fig. 2A). Any further boost to binding by anti-HNE was not obtained by incubating the solid phase protein with 200 AM HNE for 8, 16 or 24 h at room temperature. HNE used above the 200 AM level proved counterproductive We also found that modifying the protein at room temperature was more efficient than modifying at 4 8C. In our earlier study (Scofield et al., 2005) where we had used previously modified proteins as the solid phase antigen, we had used 10 mM HNE to modify the antigen. The modification of proteins by HNE could alter the protein conformation as well as its net charge. This effect could be compounded, especially during modification in solution phase, on account of the fact that HNE could act as a heterobifunctional cross-linking agent. There are two very reactive electrophilic sites, the alkene bond and the aldehyde group in HNE and related compounds with longer alkyl chains. The alkene bond reacts through Michael-type addition with the three nucleophilic amino acids cysteine, histidine and lysine. This reaction results in a product that is generally thought to be a cyclic hemiacetal, which equilibrates with an open-chain form of the protein adduct. The free aldehyde in the open-chain form of the alkenal can react with a second lysine, histidine or cysteine to act as heterobifunctional cross-linking reagents (Esterbauer et al., 1991). Rabbit anti-HNE Ro antibodies bind excellently ˆ HNE-modified 60 kDa Ro or its MAPs as well toA as the unmodified antigen compared to anti-60 kDa Ro rabbit antibodies. Our earlier studies have shown that immunization with HNE-modified 60 kDa Ro increases antigenicity and facilitates epitope spreading. A similar result was seen for anti-HNE Ro

73

binding to unmodified La (Scofield et al., 2005). We used 60 kDa Ro that was already HNE-modified (modified using 10 mM HNE in this case) to coat ELISA plates and assay anti-HNE antibodies in that study (Scofield et al., 2005). We observe almost the same kind of differential reactivity by anti-HNE Ro sera towards unmodified 60 kDa Ro and La, as well as HNE-modified Ro using this method. Binding of human anti-Ro sera to the HNE-modified Ro was significantly different compared to binding to the unmodified Ro in ELISA. This is probably because the binding of anti-Ro sera is largely conformational. These sera do not recognize 60 kDa Ro on a HeLa cell lysate immunoblot, owing to the denaturation of the protein on SDS-PAGE. It is therefore possible that the conformation of 60 kDa Ro is altered by HNE modification, thus accounting for lower binding of human anti-Ro antibodies to HNE-modified Ro. Fig. 1 compares the binding of 60 kDa Ro immobilized on ELISA plate following HNE-modification in solution with 60 kDa Ro that has been HNEmodified for 1 h following immobilization. AntiHNE interaction with the former is very fast compared to the latter on account of the fact that (a) there is probably more HNE on the protein (this protein was modified for 24 h at room temperature with 0.2 mM HNE in the presence of sodium cyanoborohydride) compared to the 60 kDa Ro protein that has been HNE-modified for 1 h post-absorption and (b) more antigen is immobilized on the plate on account of charge changes arising as a consequence of modification. Consequently the time taken for the color to develop is much higher (couple of hours more at least) for antigens modified post-immobilization (Figs. 2–4) compared to antigens modified in solution and immobilized on plate (Fig. 1). This is the reason for the low OD values obtained for the 60 kDa Ro that was modified with HNE for 1 h as seen in Fig. 1. The fact that the commercial anti-HNE sera do not bind the PPPGMRPP peptide reflects the fact that there are no histidines, cysteines or lysines in its sequence. This MAP is thus not susceptible to modification by HNE. Also, it is interesting to find that it does not bind to the seven lysines that form the core of the MAP, anchoring eight sequences of the peptide (Kurien et al., 1998). Thus the backbone portion of MAPs does not interfere with this assay.

74

B.T. Kurien, R. Hal Scofield / Journal of Immunological Methods 303 (2005) 66–75

It was interesting to find that PPPGMRPP MAP and glycyl-proline, both of which lack the required amino acids for HNE-modification also fail to act as an HNE-sink. In other words, they do not prevent HNE from modifying the substrate. On the other hand there was a significant inhibition of HNE binding to the substrate by the lysine and histidine containing Ro MAP 166 and 482 in addition to BSA. Thus, this procedure to modify antigens following immobilization on solid surfaces will be very useful to identify antibodies against the modifying agent as well as identifying inhibitors of modification. This will help enable the identification of appropriate antioxidants for therapeutic purposes. This procedure is also simple and fast in that it does not require timeconsuming dialysis to remove unbound HNE compared to the modification of antigens in the soluble phase.

Acknowledgements Supported by NIH grant ARO1844 to RHS and Oklahoma Center for the Advancement of Science and Technology to RHS and BTK. We thank Dr. Kenneth Hensley for modifying 60 kDa Ro in solution for the assays used in Fig. 1.

References Allison, M.E., Fearon, D.T., 2000. Enhanced immunogenicity of aldehyde-bearing antigens: a possible link between innate and adaptive immunity. Eur. J. Immunol. 30, 2881. Bachmann, M., Gro¨lz, D., Bartsch, H., Klein, R.R., Tro¨ster, H., 1997. Analysis of expression of an alternative La (SS-B) cDNA and localization of the encoded N- and C-terminal peptides. Biochim. Biophys. Acta 1356, 53. Ben-Chetrit, E., Gandy, B.J., Tan, E.M., Sullivan, K.F., 1989. Isolation and characterization of a cDNA clone encoding the 60 kDa component of the human SS-A/Ro ribonucleoprotein autoantigen. J. Clin. Invest. 83, 1284. Bennaars-Eiden, A., Higgins, L., Hertzel, A.V., Kapphahn, R.J., Ferrington, D.A., Bernlohr, D.A., 2002. Covalent modification of epithelial fatty acid-binding protein by 4-hydroxynonenal in vitro and in vivo. Evidence for a role in antioxidant biology. J. Biol. Chem. 277, 50693. Bergmark, C., Wu, R., de Faire, U., Lefvert, A.K., Swedenborg, J., 1995. Patients with early-onset peripheral vascular disease have increased levels of autoantibodies against oxidized LDL. Arterioscler. Thromb. Vasc. Biol. 15, 441.

Chang, M.K., Bergmark, C., Laurila, A., Horkko, S., Han, K.H., Friedman, P., Dennis, E.A., Witztum, J.L., 1999. Monoclonal antibodies against oxidized low-density lipoprotein bind to apoptotic cells and inhibit their phagocytosis by elicited macrophages: evidence that oxidation-specific epitopes mediate macrophagerecognition. Proc. Natl. Acad. Sci. U. S. A. 96, 6353. Deutscher, S.L., Harley, J.B., Keene, J.D., 1988. Molecular analysis of the 60 kDa human Ro ribonucleoprotein. Proc. Natl. Acad. Sci. U. S. A. 85, 9479. Esterbauer, H., Schaur, R.J., Zollner, H., 1991. Chemistry and biochemistry of 4-hydroxynonenal, malonaldehyde and related aldehydes. Free Radic. Biol. Med. 11, 81 (Review). Finkel, T., Holbrook, N.J., 2000. Oxidants, oxidative stress and the biology of ageing. Nature 408, 239. Grune, T., Michel, P., Sitte, N., Eggert, W., Albrecht-Nebe, H., Esterbauer, H., Siems, W.G., 1997. Increased levels of 4-hydroxynonenal modified proteins in plasma of children with autoimmune diseases. Free Radic. Biol. Med. 23, 357. Holvoet, P., Perez, G., Zhao, Z., Brouwers, E., Bernar, H., Collen, D., 1995. Malondialdehyde-modified low density lipoproteins in patients with atherosclerotic disease. J. Clin. Invest. 95, 2611. Itabe, H., Jimi, S., Kamimura, S., Suzuki, K., Uesugi, N., Imanaka, T., Shijo, H., Takano, T., 1998. Appearance of cross linked proteins in human atheroma and rat pre-fibrotic liver detected by a new monoclonal antibody. Biochim. Biophys. Acta 1406, 28. James, J.A., Gross, T., Scofield, R.H., Harley, J.B., 1995. Immunoglobulin epitope spreading and autoimmune disease after peptide immunization: Sm B/BV-derived PPPGMRPP and PPPGIRGP induce spliceosome autoimmunity. J. Exp. Med. 181, 453. Jiang, X., Chen, F., 1992. The effect of lipid peroxides and superoxide dismutase on systemic lupus erythematosus: a preliminary study. Clin. Immunol. Immunopathol. 63, 39. Kurien, B.T., Scofield, R.H., 2003. Free radical mediated peroxidative damage in systemic lupus erythematosus. Life Sci. 73, 1655. Kurien, B.T., Jackson, K., Scofield, R.H., 1998. Immunoblotting of multiple antigenic peptides. Electrophoresis 19, 1659. Maggi, E., Chiesa, R., Melisano, G., Astore, G., Grossi, A., Finardi, G., Bellomo, G., 1995. LDL oxidation in patients with atherosclerosis: a study of in vitro and In vivo oxidation markers. Arterioscler. Thromb. Vasc. Biol. 14, 1892. Mottaran, E., Stewart, S.F., Rolla, R., Vay, D., Cipriani, V., Moretti, M., Vidali, M., Sartori, M., Rigamonti, C., Day, C.P., Albano, E., 2002. Lipid peroxidation contributes to immune reactions associated with alcoholic liver disease. Free Radic. Biol. Med. 32, 38. Rhodes, J., Chen, H., Hall, S.R., Beesley, J.E., Jenkins, D.C., Collins, P., Zheng, B., 1995. Therapeutic potentiation of the immune system by costimulatory Schiff-base-forming drugs. Nature 377, 71. Scofield, R.H., Kurien, B.T., Zhang, F., Mehta, P., Kaufman, K., Gross, T., Bachmann, M., Gordon, T., Harley, J.B., 1999. Protein–protein interaction of the Ro–ribonucleoprotein particle using multiple antigenic peptides. Mol. Immunol. 36, 1093. Scofield, R.H., Kurien, B.T., Gross, J., Reed, N.B., Taylor, A.K., Dominguez, N., Mehta, P., Bachmann, M., 2004. Interaction of

B.T. Kurien, R. Hal Scofield / Journal of Immunological Methods 303 (2005) 66–75 calcium and Ro60: increase of antigenicity. Mol. Immunol. 41, 809. Scofield, R.H., Kurien, B.T., Ganick, S., McClain, M.T., Pye, Q., James, J.A., Schneider, R.I., Broyles, R.H., Bachmann, M., Hensley, K., 2005. Modification of upus-associated 60 kDa Ro protein with the lipid oxidation product 4-hydroxy-2-nonenal increases antigenicity and facilitates epitope spreading. Free Radic. Biol. Med. 38, 719. Siems, W., Grune, T., 2003. Intracellular metabolism of 4-hydroxynonenal. Mol. Aspects Med. 24, 167 (Review). Thannickal, V.J., 2003. The paradox of reactive oxygen species: injury, signaling, or both? Am. J. Physiol., Lung Cell. Mol. Physiol. 284, L24.

75

Van Kuijk, F.J., Holte, L.L., Dratz, E.A., 1990. 4-Hydroxyhexenal: a lipid peroxidation product derived from oxidized docosahexaenoic acid. Biochim. Biophys. Acta 1043, 116. Willis, M.S., Klassen, L.W., Tuma, D.J., Sorrell, M.F., Thiele, G.M., 2002. Adduction of soluble proteins with malondialdehyde– acetaldehyde (MAA) induces antibody production and enhances T-cell proliferation. Alcohol., Clin. Exp. Res. 26, 94. Zheng, B., Brett, S.J., Tite, J.P., Lifely, M.R., Brodie, T.A., Rhodes, J., 1992. Galactose oxidation in the design of immunogenic vaccines. Science 256, 1560.