25-Hydroxyl-cholesterol Binds and Enhance Anti-viral Effects of Zebrafish Monomeric C-reactive Proteins

C-reactive proteins (CRP) are among the faster acute-phase inflammation-responses coded by one gene in humans (hcrp) and seven genes (crp1-7) in zebrafish (Danio rerio). In this study, preferential 25-hydroxycholesterol (25HOCh) binding to zebrafish CRP1-7 compared to other lipids were predicted by in silico docking and confirmed by solid-phase binding-assays. In addition, 25HOCh enhanced methyl-betacyclodextrin-sensitive (Cholesterol-dependent) CRP1-7 anti-viral effects in a fine-tunned isoform-dependent manner. In silico and structural studies suggested that the crosstalk between the anti-viral enhancements of both 25HOCh and CRP1-7 were dependent on protein monomers rather than oligomers but differred among isoforms. The presence of oxidized cholesterols in human atherosclerotic plaques amplifies the importance that similar interactions may have for vascular and/or neurodegenerative diseases during viral infections. In this context, the zebrafish model offers a genetic tool to further investigate how the expression and functions of different CRP isoforms and/or transcript variants may be controlled.

10 233 resulted selected for the automatical modelling corresponded to zebrafish CRP5 ± Ca++ (4PBP.pdb and 234 4PBO.pdb) [6] (RCSB data bank at http://www.rcsb.org/pdb/home/home.do). To predict their docking ΔG energies to CRP1-7, phosphocholine head (PC), other phospholipid 239 heads [43][44][45] and cholesterol (Ch) [16] were selected because of their hCRP ligand properties. Interestingly, 240 the results predicted the lowest ΔG (prediction of stronger binding) for Ch (ΔG ranges from -7.5 to -9 241 Kcal/mol) compared to phospholipid-heads (ΔG ranges from -4 to -5. 5 Kcal/mol). The addition of a glycerol 242 molecule to those phospholipid-heads did not changed their predicted ΔG (not shown). Results also predicted 243 Ch docking energies more Ca ++ -independent than most other lipid-heads ( Figure 1A), and alternative docking 244 locations for Ch and other lipid-heads (not shown).  [46], ii) Ch is a key molecule involved in coronary 253 diseases and iii) Ch-related physiological compounds are highly diverse, an screening for other physiological 254 Ch-related compounds was performed before studying any possible anti-viral effects.

256
When 23 Ch-related physiological compounds were explored for docking to CRP1-7, the stronger 257 predictions (ΔG ranges between -7.5 to -9.3 Kcal/mol) were obtained for most hydroxy derivatives for most 12 284 the lower 25HOCh concentrations assayed (<10 µM) ( Figure 3B). On the other hand, although CRP7 285 showed slightly higher bindings at >100 µM 25HOCh, similar values were obtained for all ssCRP1-7 at those 286 higher concentrations. Binding of ssCRP1-7 to solid-phase 25HOCh showed relatively lower values than to 287 rCRPs, most probably due to the lower CRP concentrations in the ssCRP1-7 (compare ordinate values of 288 Figure 3A and B).
289 Mapping of CRP5 binding and docking to 25HOCh

290
To further clarify the 25HOCh binding to CRP1-7, we mapped such interaction. Because m-hCRP rather 291 than p-hCRP is the conformation that preferentially binds Ch [16,17,48], some non-conformational motifs may 292 conserve Ch-/ 25HOCh-binding activity. Therefore, we selected a pepscan to explore for possible non-293 conformational interactions with 25HOCh by solid-phase binding assays and by docking predictions.

294
For the peptide binding assays, each of the synthetically biotinylated 15-mer peptides derived from the 295 CRP5 amino acid sequence were incubated with 25HOCh-coated solid-phases. Results showed maximal binding 331 Insect-made rCRPs suggested different oligomerization states 332 E.coli-made zebrafish c-reactive protein CRP5 isoform (rCRP5) crystalized as trimers (t-CRP5) as 333 shown by X-rays [6]. However, it is not yet known whether or not trimers are the physiological form or that 14 335 Our first attempts to characterize CRP1-7 isoforms included expression in E.coli. However, 336 numerous experiments met with irreproducibility, expression failure, high CRP denaturation or low 337 yields, despite reduction of autoinduction or temperature, and/or re-cloning of the best producing 338 clones (not shown). Most probably some of those results could be explained by the toxicity of the 339 rCRPs to E.coli.
341 Results showed that while insect-made rCRP2 / rCRP5 / rCRP7 could be expressed and purified by 342 non-denaturing affinity chromatography, all attempts to purify rCRP1 were unsuccessful. Western blot 343 analysis using anti-polyH antibodies indicated that although small amounts of rCRP1 were present, they were 344 not retained by the affinity columns, most probably due to polyH tail inaccessibility (not shown), perhaps 345 becuase a different conformation of rCRP1 compared to the other rCRPs.   Figure 6B, right). The slightly 360 different positions of the monomeric forms could be most probably due to differences in their glycosylation, 361 although recently different post-transcriptional deimidation has been also described in cod CRPs [53].

362
The most probable explanation for all the above commented data suggests that while insect-made 363 rCRP2/rCRP5 may exist as an equilibrium among trimers, dimers and monomers, rCRP7 has a stronger 364 tendency to form monomers. 369 Similar CRP2-7 levels were present in ssCRP2-7 as shown using actin as an internal marker ( Figure 6C, up). In 370 these experiments, it was not possible to detect the presence of any CRP1 band, most probably because its lower 371 concentration, since previous results demonstrated its presence by dot-blot when using concentrated ssCRP1 [7].
372 Most probably, all ssCRP1-7 were secreted from EPC transfected cells mainly as monomers.

374
To obtain more data on the possible structures of CRP1-7, their amino acid sequences were modelled 375 using the SWISS-MODEL web program. Automatic modelling showed that only CRP2 / CRP5 rendered 376 trimers, while the rest of the CRP1-7 modelled as monomers only (Table 2). CRP2 / CRP5 have differences 377 in most of the modelling parameters, specially in their torsion-angle potentials, compared to other CRP1-7 378 ( Table 2). Because the existence of ~ 70 EST from zebrafish in the UniGene Bank classified as CRP5 379 transcript variants [6] offered another opportunity to test the reliability of the trimer/monomer predictions 380 mentioned above, we explored also these data. The corresponding results predicted that 97.

388
The PAGE/Western data and the in silico predictions, together with the results of 25HOCh binding 389 ( Figure 3A) and enhancement of anti-SVCV effects ( Figure 5A) by ssCRP1-7, may implicate more the m-390 CRP1-7 rather than t-CRP1-7 in those functions. However, CRP1-7 may also physiologically exist as an 391 equilibrium of trimers, dimers and monomers, as shown in the cases of CRP2 /CRP5 and to lower extent in 392 CRP7. On the other hand, because m-hCRP can also be produced in vitro, for instance by treatments with 393 urea, low-pH or low-salt buffers in the absence of Ca ++ [54,55], the m-CRP1-7 detected in this work may 394 have been produced by the in vitro manipulations (i.e., purification by affinity chromatography in the 395 absence of Ca ++ , transfection of EPC cells, etc). We may also speculate that t-CRP1-7 could preferentially 396 exist in fish until an stimulus triggers their conversion to m-CRP1-7 or viceversa since circulating hCRP is 397 pentameric (p-hCRP) [13] and converts to monomeric (m-hCRP) after interaction with exposed 398 phosphocholine heads and Ch-enriched lipid rafts of cellular membranes in damaged tissues [16,17,48,52].
399 The t-CRP1-7 may be functionally analogous to the circulating p-hCRP and the m-CRP1-7 could be 400 analogous to the converted m-hCRP. Alternatively, zebrafish m-CRP1-7 may be synthesized de novo as 401 monomers. We may also think on the possibility of the existence of heterologous CRP1-7 oligomers but any 402 of those possibilities remain speculative until specific reagents could be developed.