Human CRP was purified from discarded body fluids employing PCh-affinity chromatography, exactly as described previously (46).

Limulus hemolymph was obtained from Marine Biological Laboratory. Li-CRP was purified from the hemolymph as described previously, with some modifications (33). Hemolymph was first centrifuged at 15,000 g for 15 min to collect the clear blue plasma. The plasma was passed through a Sepharose 4B (MilliporeSigma, 4B200) column in 10 mM Tris-HCl, pH 7.2, containing 150 mM NaCl (TBS) and 2 mM CaCl₂ to remove carbohydrate-binding proteins. Next, polyethylene glycol-8000 was added to the plasma at a final concentration of 3% and incubated at 4°C for 16 h with shaking to remove hemocyanin. The plasma was then centrifuged at 30,000 g for 30 min. The blue pellet was discarded, and the transparent plasma was recovered and used to purify Li-CRP by employing two different affinity chromatography methods, as follows.

Method 1: Li-CRP was isolated from plasma by employing Ca²⁺-dependent PCh-affinity chromatography as described previously for purification of human CRP (46). Briefly, plasma was passed through a PCh-Sepharose column in a Ca²⁺-containing buffer. After washing the column, Li-CRP was eluted by using EDTA. The EDTA eluate was concentrated and Li-CRP was further purified by using gel filtration chromatography on a Superose 12 column in TBS containing 5 mM EDTA. Fractions containing Li-CRP were pooled and dialyzed against TBS containing 2 mM CaCl₂. This preparation of pure Li-CRP which involved purification by PCh-affinity chromatography was named Li-CRP-I.

Method 2: Li-CRP was isolated from plasma by employing Ca²⁺-dependent PEt-affinity chromatography as described previously for purification of recombinant human CRP mutants (46). Briefly, plasma was passed through a PEt-Sepharose column in a Ca²⁺-containing buffer. After washing the column, Li-CRP was eluted by EDTA. The EDTA eluate was concentrated and Li-CRP was further purified as described above in method 1. This preparation of pure Li-CRP which involved purification by PEt-affinity chromatography was named Li-CRP-II.

The purity of Li-CRP preparations was determined by using denaturing SDS-PAGE under reducing conditions in a 4-20% gradient gel and by N-terminal sequencing. The N-terminal amino acid sequencing of Li-CRP was performed at the Molecular Structure Facility, University of California, Davis. The concentrations of Li-CRP-I and Li-CRP-II were determined by using the extinction coefficient of 15.49 at A₂₈₀ (36).

The molecular weight of native Li-CRP was determined by employing gel filtration chromatography on a calibrated Superose 12 column (10/300 GL, GE healthcare), as described previously for human CRP (46). The gel filtration column was equilibrated with TBS containing 5 mM EDTA. Li-CRP was injected into the column and eluted with TBS containing 5 mM EDTA at a flow rate of 0.3 ml/min. Fractions (60 fractions, 250 μl each) were collected and absorbance at 280 nm measured to locate the elution volume of Li-CRP. The molecular weight of the subunits of Li-CRP was determined by employing denaturing SDS-PAGE under reducing conditions in a 4-20% gradient gel. The composition of Li-CRP was further evaluated by employing anion exchange chromatography on a MonoQ column (5/50 GL, GE healthcare), as described previously for human CRP (46). The fractions collected after the chromatography were subjected to denaturing SDS-PAGE under reducing conditions in a 4-20% gradient gel.

Deglycosylation of Li-CRP was performed using the Glycoprofile IV Chemical Deglycosylation kit (MilliporeSigma, PP0510) according to manufacturer’s instructions. Briefly, 150 µl of chilled trifluoromethanesulfonic acid (MilliporeSigma, 34781-7) was added to 1 mg of cold lyophilized Li-CRP and incubated on ice for 25 min with occasional shaking. Bromophenol blue (0.2%; 4 µl) was added to the reaction followed by dropwise addition of 60% pyridine solution (MilliporeSigma, P5496) in a methanol-dry ice bath until the color of the reaction changed from red to light purple or blue. Samples were then dialyzed against TBS overnight. The deglycosylation of Li-CRP was verified and the molecular weight of the deglycosylated subunits determined by employing denaturing SDS-PAGE under reducing conditions in a 4-20% gradient gel.

Li-CRP-I was purified as described above. Rabbit polyclonal antibodies against purified Li-CRP-I were generated commercially by Thermo Fisher Scientific. The antiserum obtained from the company was used as such after titrating to determine the dilution to be used in the assays.

The immunological cross-reactivity between Li-CRP-I and Li-CRP-II was determined by employing anti-Li-CRP-I antibodies and purified Li-CRP-I and Li-CRP-II, as follows. Microtiter wells were coated with increasing amounts of Li-CRP-I or Li-CRP-II and incubated at 37°C for 2 h. The unreacted sites in the wells were blocked with TBS containing 0.5% gelatin for 45 min at room temperature. Next, anti-Li-CRP-I antibodies diluted in TBS was added to the wells and incubated at 37°C for 1 h. HRP-conjugated donkey anti-rabbit IgG, diluted in TBS, was used as the secondary antibody. Color was developed using ABTS as the substrate and the OD was read at 405 nm in a microtiter plate reader.

Binding activity of Li-CRP for PCh was measured by using pneumococcal C-polysaccharide (PnC; purchased from Statens Serum Institut) as the PCh-containing ligand, exactly as described previously (9).

Lyophilized Aβ peptide 1-42 was purchased from Bachem (H-1368) and reconstituted according to manufacturer’s instructions. Reconstituted Aβ peptide solution should contain Aβ monomers; however, oligomers of Aβ monomers are also present in the reconstituted Aβ peptide solution (47). Aβ monomers were prepared from Aβ peptides according to a published method (48) and exactly as described previously (21). In brief, lyophilized Aβ peptides were dissolved in hexafluoroisopropanol (MilliporeSigma) and incubated at 37°C for 2 h for monomerization of the oligomers present in the Aβ peptide solution. After removing hexafluoroisopropanol by evaporation overnight, the vials containing the film of Aβ monomers were stored at -20°C. Aβ fibrils were prepared from Aβ peptides according to a published method (48) and exactly as described previously (21). In brief, when needed, ice-cold TBS was added to a vial containing the film of Aβ monomers to obtain a 200 μg/ml solution of Aβ monomers. To prepare Aβ fibrils, the Aβ monomers were vortexed and transferred to a 96-well microtiter plate (Immunochemistry Technologies, Costar 266). The plate was incubated at 37°C for 3 h with shaking at 300 rpm. The resulting Aβ fibrils were stored at -20°C until needed.

Binding activity of Li-CRP for immobilized Aβ was evaluated as described earlier (21). Briefly, lyophilized Aβ peptides were reconstituted in TBS. Microtiter wells were coated with 10 μg/ml of Aβ at 4°C overnight. The unreacted sites in the wells were blocked with TBS containing 0.5% gelatin. Li-CRP, diluted in TBS containing 0.1% gelatin, 0.02% Tween-20 and 2 mM CaCl₂ (TBS-Ca), was added in duplicate wells and incubated at 37°C overnight. After washing the wells, rabbit polyclonal anti-Li-CRP-I antibody was used to detect bound Li-CRP. HRP-conjugated donkey anti-rabbit IgG (GE Healthcare) was used as the secondary antibody. Color was developed, and the OD was read at 405 nm.

Binding activity of human SAP (Calbiochem, 565190) for immobilized Aβ was evaluated as follows. Microtiter wells were coated with 10 µg/ml of Aβ peptide, monomer and fibrils diluted in TBS. The unreacted sites in the wells were blocked with TBS containing 0.5% gelatin for 45 min at room temperature. SAP, diluted in TBS-Ca was then added in duplicate wells and incubated for 2 h at 37°C. Bound SAP was detected by using rabbit anti-human SAP (Calbiochem, 565191) by adding the antibody (5 µg/ml) to the wells and incubating at 37°C for 1 h. HRP- conjugated donkey anti-rabbit antibody (Southern Biotech, 6441-05) was used as the secondary antibody. Color was developed, and the OD was read at 405 nm.

The fibrillation of Aβ monomers was monitored by employing ThT assay according to a published method (49, 50) and exactly as described previously (21). In brief, the effects of Li-CRP on Aβ fibrillation were determined by adding CRP to Aβ at time zero, that is, CRP and Aβ were mixed together before the beginning of the fibrillation reaction. The fibrillation reaction mix was prepared with and without CRP. In CRP-containing mixture, the final concentrations of CRP were 1, 10 and 100 μg/ml. After vortexing, 240 μl of each mixture was transferred in triplicate wells in a 96-well microtiter plate (Immunochemistry Technologies, Costar 266). Fluorescence was measured by using the Synergy H1 microplate reader (BioTek) with excitation at 440 nm and emission at 480 nm. After the first measurement at 5 min, the plate was incubated at 37°C for 3 h with shaking at 300 rpm; fluorescence was measured every 15 min for 3 h.

Gel filtration and SDS-PAGE were performed to determine the molecular weight of native Li-CRP, the number of subunits in each molecule and the molecular weight of the subunits. As shown in Figure 1A , the elution volumes from the calibrated gel filtration column for both Li-CRP-I purified by PCh-affinity chromatography and Li-CRP-II purified by PEt-affinity chromatography were 10 ml. The molecular weights of both Li-CRP-I and Li-CRP-II were calculated to be ~300 kDa. SDS-PAGE ( Figure 1B ) revealed that Li-CRP-I was composed of two types of subunits of molecular weight 27.3 ± 1.7 kDa (band A) and 25.8 ± 2.7 kDa (band B). The band A was a doublet of two bands. Li-CRP-II was also composed of two types of subunits of molecular weight 29.6 ± 2.6 kDa (band C) and 24.7 ± 2.5 kDa (band D), although a third subunit of molecular weight 28.2 ± 2.5 kDa was also seen as a faint band in between bands C and D. In another preparation of Li-CRP-II, there was an additional faint band of molecular weight 54.8 ± 0.01 (band E, Figure 1D ). Thus, both Li-CRP-I and Li-CRP-II were composed of twelve subunits, assuming an average molecular weight of 25 kDa for each subunit.

We checked the purity of Li-CRP-I eluted by using either EDTA (lane 2) or PCh (lane 3) from the PCh-affinity column ( Figure 1C ). We also compared the purity of Li-CRP-I purified by either affinity chromatography (lanes 2-3) only or by affinity chromatography followed by gel filtration (lanes 4-5). As shown, there was no difference in the purity of Li-CRP-I eluted by either EDTA or PCh from the affinity column. Similarly, there was no difference in the purity of Li-CRP-I purified by just affinity chromatography or by affinity chromatography followed by gel filtration. These results indicated that the affinity-purified Li-CRP-I was pure; there was no need for further purification.

The purity of Li-CRP-II eluted by using either EDTA (lanes 2-4) or PEt (lanes 5-7) from the PEt-affinity column was also investigated ( Figure 1D ). As shown, when Li-CRP-II was eluted from the affinity column by using PEt (lane 5), the top band C (as shown in Figure 1B ) was absent; instead, there was a doublet of the middle band. We also compared the purity of Li-CRP-II purified by either affinity chromatography (lanes 2 and 5) alone or by affinity chromatography followed by ion-exchange chromatography (lanes 3 and 6) or by affinity chromatography followed by gel filtration (lanes 4 and 7). There was no difference in the purity of Li-CRP-II purified by just affinity chromatography or by affinity chromatography followed by either ion-exchange chromatography or gel filtration. These results indicated that the affinity-purified Li-CRP-II was also pure; there was no need for further purification.

To remove the contaminant protein (band E) present in Li-CRP-II, each fraction collected after ion-exchange chromatography was subjected to SDS-PAGE individually ( Figure 2A ), instead of pooling the fractions (lane 2, Figure 1D ). As shown in Figure 2A , the intensity of band E was directly proportional to the intensity of band C and that the compositions of Li-CRP-II present in each fraction were different from each other. Some fractions had more of subunit C and some had more of subunit D. That was not the case with Li-CRP-I ( Figure 2B ); when individual fractions of Li-CRP-I after ion-exchange chromatography were subjected to SDS-PAGE, the composition of Li-CRP-I in each fraction was the same.

Deglycosylation of Li-CRP followed by SDS-PAGE were performed to determine whether each type of subunit present in Li-CRP-I and Li-CRP-II was glycosylated. As shown in Figure 2C , deglycosylation of Li-CRP-I resulted in a single band of molecular weight 23.2 kDa (lanes 1 and 2). Similarly, deglycosylation of Li-CRP-II also resulted in a single band of molecular weight 23.2 kDa (lanes 4 and 5). These data suggest that the difference in the electrophoretic mobility between subunits A and B and between subunits C and D was due to the difference in the extent of glycosylation of each type of subunit. Bands A and B were the products of a single gene and bands C and D were also the products of a single gene.

Native PAGE was also performed to check the purity of Li-CRP preparations. As shown in Figure 2D , there was only one band for Li-CRP-I, further indicating that it was a pure preparation of Li-CRP-I. In contrast, two bands were observed in the native PAGE of Li-CRP-II. This was not expected. Likely, the top band seen in Li-CRP-II lane contained ligand-bound Li-CRP-II and the ligand could be the protein seen as band E in SDS-PAGE gels.

The N-terminal sequences of the first ten amino acids of Li-CRP-I and Li-CRP-II were LEEGEITSKV and LEEGEITSKI, respectively. The amino acid sequences of Li-CRP-I and Li-CRP-II were identical for the first nine residues. Sequencing of Li-CRP-II, however, revealed an additional protein with the sequence MLTTKVRFFH of the first 10 residues, probably reflecting the band E in Li-CRP-II.

The preparations of Li-CRP-I and Li-CRP-II purified by affinity chromatography and gel filtration as shown in Figure 1B were used in the subsequent experiments involving Li-CRP.

Immunological cross-reactivity assays for Li-CRP-I and Li-CRP-II showed that the antibodies to Li-CRP-I reacted with both Li-CRP-I and Li-CRP-II with almost equal affinity ( Figure 3 ). These data suggested that anti-Li-CRP-I antibodies could be used to detect both Li-CRP-I and Li-CRP-II. These results were not surprising considering the similarity in the amino acid sequences of Li-CRP-I and Li-CRP-II.

We next evaluated whether Li-CRP-II which was purified by PEt-affinity chromatography bound to PCh also and, if so, then whether the PCh-binding ability of Li-CRP-II was different from that of Li-CRP-I and human CRP. As shown in Figure 4 , both Li-CRP-I and Li-CRP-II and human CRP bound to PCh in a CRP concentration-dependent manner. However, the binding of Li-CRP-I and Li-CRP-II to PCh was not comparable to each other and was different from human CRP. For equivalent binding (OD at 405 nm equal to 1) of Li-CRP-I, Li-CRP-II and human CRP to PCh, the required concentrations of Li-CRP-I, Li-CRP-II and human CRP were 1.2 μg/ml, 12 μg/ml and 0.009 μg/ml, respectively. Thus, for equivalent binding of Li-CRP-II and Li-CRP-I to PCh, 10-times more of Li-CRP-II was required compared to Li-CRP-I, indicating that the PCh-binding avidity of Li-CRP-II was 90% less than that of Li-CRP-I. For equivalent binding of Li-CRP-II and human CRP to PCh, ~1300-times more of Li-CRP-II was required compared to human CRP, indicating that the PCh-binding avidity of Li-CRP-II was ~99% less than that of human CRP. Similarly, for equivalent binding of Li-CRP-I and human CRP to PCh, ~130-times more of Li-CRP-I was required compared to human CRP, indicating that the PCh-binding avidity of Li-CRP-I was also ~99% less than that of human CRP.

Employing solid-phase Aβ-binding assays, the binding of Li-CRP to immobilized Aβ, in the presence and absence of Ca²⁺, was determined. As shown in Figure 5 , Li-CRP-I bound to immobilized Aβ peptides, monomers and fibrils, at physiological pH, in a CRP concentration-dependent manner. The binding of Li-CRP-I to Aβ did not require Ca²⁺ since the binding also occurred in the absence of Ca²⁺. Similarly, Li-CRP-II also bound to Aβ peptides, monomers and fibrils, at physiological pH, in a CRP concentration-dependent manner. However, in contrast to Li-CRP-I, the binding of Li-CRP-II to Aβ was Ca²⁺-dependent since the binding of Li-CRP-II to Aβ was either absent or drastically reduced in the absence of Ca²⁺. These data suggest that Aβ is a ligand of both native Li-CRP-I and native Li-CRP-II although the undefined Aβ-binding site may be located on different regions in Li-CRP-I and Li-CRP-II.

Next, to compare Li-CRP with human SAP, the binding of human SAP to immobilized Aβ, in the presence and absence of Ca²⁺, was investigated ( Figure 6 ). As shown, human SAP bound to immobilized Aβ peptides, monomers and fibrils, at physiological pH, in the presence of Ca²⁺ and in an SAP concentration-dependent manner. However, like Li-CRP-II, human SAP did not bind to Aβ in the absence of Ca²⁺, suggesting that Li-CRP-II and human SAP recognize Aβ in similar environment.

Next, we investigated whether Li-CRP capable of binding to immobilized Aβ can also bind to Aβ when both Li-CRP and Aβ are in the fluid phase and inhibit the formation of Aβ fibrils. Aβ monomers were employed in the fibrillation assays ( Figure 7 ). In the absence of Li-CRP, the fibrillation of Aβ began within 5 min and continued until ~2 h when further fibrillation stopped (black curves in both panels). Both Li-CRP-I and Li-CRP-II inhibited the fibrillation of Aβ in a CRP concentration-dependent manner. In case of Li-CRP-I, there was no statistically significant difference in fibrillation with or without 1 μg/ml of Li-CRP-I and there was no statistically significant difference in fibrillation with 1 μg/ml and 10 μg/ml of Li-CRP-I. There were statistically significant differences in the fibrillation when Li-CRP-I was used at 10 μg/ml and 100 μg/ml. In case of Li-CRP-II, the inhibition of fibrillation was more efficient; even 1 μg/ml of Li-CRP-II was able to significantly inhibit the fibrillation of Aβ and the inhibition of fibrillation was significantly more at higher concentrations of Li-CRP-II. Thus, both Li-CRP-I and Li-CRP-II bound to Aβ in the fluid phase, inhibited the rate of fibrillation of Aβ over time, and also inhibited the total amount of the fibrils formed by the end of the fibrillation reaction.