Enhancing antibody serodiagnosis using a ... - ACS Publications

Enhancing antibody serodiagnosis using a ... - ACS Publicationshttps://pubs.acs.org/doi/pdf/10.1021/acsinfecdis.8b00014and cost-effective assay to dia...
2 downloads 6 Views 2MB Size
Subscriber access provided by - Access paid by the | UCSB Libraries

Enhancing antibody serodiagnosis using a controlled peptide co-immobilization strategy Laura Sola, Paola Gagni, Ilda D'Annessa, Riccardo Capelli, Camilla Bertino, Alessandro Romanato, Francesco Damin, Greta Bergamaschi, Edoardo Marchisio, Angela Cuzzocrea, Mauro Bombaci, Renata Grifantini, Marcella Chiari, Giorgio Colombo, Alessandro Gori, and Marina Cretich ACS Infect. Dis., Just Accepted Manuscript • DOI: 10.1021/acsinfecdis.8b00014 • Publication Date (Web): 23 Mar 2018 Downloaded from http://pubs.acs.org on March 27, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Infectious Diseases

Enhancing antibody serodiagnosis using a controlled peptide co-immobilization strategy Laura Sola1§, Paola Gagni1§ , Ilda D’Annessa1§, Riccardo Capelli1, Camilla Bertino1, Alessandro Romanato1, Francesco Damin1, Greta Bergamaschi1, Edoardo Marchisio4, Angela Cuzzocrea4, Mauro Bombaci2, Renata Grifantini2, Marcella Chiari1, Giorgio Colombo,1, 3, Alessandro Gori1*, Marina Cretich1*

1: Consiglio Nazionale delle Ricerche, Istituto di Chimica del Riconoscimento Molecolare (ICRM), Via Mario Bianco, 9, 20131, Milano, Italy 2: Istituto Nazionale di Genetica Molecolare “Romeo ed Enrica Invernizzi” (INGM), Via Francesco Sforza. 35, 20122, Milano, Italy 3: Dipartimento di Chimica, Università di Pavia, V.le Taramelli 12, 27100, Pavia, Italy 4: Diagnostic Bioprobes s.r.l. (DiaPro), via G. Carducci 27, 20090 Sesto San Giovanni, Italy

§: Equally contributed * Corresponding authors [email protected] ; [email protected]

ACS Paragon Plus Environment

1

ACS Infectious Diseases 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 29

Antigen immunoreactivity is often determined by surface regions defined by the 3D juxtapositions of amino acids stretches that are not continuous in the linear sequence. As such, mimicking an antigen immunoreactivity by means of putative linear peptide epitopes for diagnostic purposes is not trivial. Here we present a straightforward and robust method to extend the reach of immune-diagnostic probes design by co-presenting peptides belonging to the same antigenic surface. In this case study focused on a computationally predicted Zika virus NS1 protein putative antigenic region, we reached a diagnostic confidence by the oriented and spatially controlled co-immobilization of peptide sequences found adjacent within the protein fold, that cooperatively interacted to provide enhanced immunoreactivity with respect to single linear epitopes. Through our method, we were able to differentiate Zika infected individuals from healthy controls. Remarkably, our strategy fits well with the requirements to build high-throughput screening platforms of linear and mixed peptide libraries, and could possibly facilitate the rapid identification of conformational immunoreactive regions.

Keywords: discontinuous epitopes; peptide bioprobes; peptide design; microarrays; serodiagnosis; arbovirus; clickable polymer; zika virus Protein and peptide microarray technology has led to a revolution in the field of biomarker discovery allowing the rapid and simultaneous screening of up to thousands of bioprobes in a single assay, thanks to its rapid and multiplexed format of analysis1,2. The generated complex analytical outcome can then be interpreted to unveil selective molecular interactions, and ultimately be translated into a diagnostic application. In this scenario, peptides are ideal bioprobes for microarray analysis since favourable synthetic accessibility and ease of chemical manipulation allow their application in high-throughput schemes as well as the fine tuning of surface immobilization parameters to increase sensor performances3. In the field of serodiagnostics, where antibodies elicited in response to a pathogen infection are used as disease biomarkers, the ability of short linear peptides to mimic the immunoreactivity of a whole protein antigen, thus acting as molecular bioprobes for selective antibody-capturing, is of broad interest4,5. However, this approach may be restricted to reproduce continuous and unstructured antigen regions, representing a major limitation to the peptide-based approach. Indeed, it is often observed that antigen-specific antibody response is directed against epitopes constituted by extended surface regions made of non-contiguous linear peptide

ACS Paragon Plus Environment

2

Page 3 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Infectious Diseases

sequences brought in close contact by the native 3D-arrangement of the antigen structure, namely discontinuous or conformational epitopes6. As a result, when approaching the design and discovery of new immunodiagnostic peptidic tools, a considerable challenge lies in reproducing the immunoreactivity of extended surface regions, as single linear epitopes may lack of such conformational properties. Indeed, even if some general strategies can be commonly applied, including stapling and cyclization techniques7, the fine conformational control of free peptides can prove synthetically demanding. In this sense, remarkable examples on antigen surface mimicry enabled by chemical approaches were reported by the groups of Liskamp8,9,10,11 and Timmerman12,13 by applying macrocyclization strategies of different peptides on bridging scaffolds to thermodynamically favour a limited set of peptide conformations. However, if highthroughput screening capacity (tens to thousands of probes) is desirable, these methods remain considerably synthetically-demanding and not accessible to groups lacking of advanced synthetic expertise. In turn, this difficulty could hamper the needs of a prompt early-stage screening. Rapid yet broadly accessible alternatives to enhance antibody-based diagnostics are therefore desirable, which should be targeted at the clinic diagnostic value rather than focusing on a close structural mimicry of putative discontinuous epitopes.

ACS Paragon Plus Environment

3

ACS Infectious Diseases 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 29

Scheme 1. Strategy to enhance immunoreactivity by spatially controlled co-presentation of peptidic probes on microarray surfaces through “click” reaction of yne-modified peptides on Copoly Azide.

Here, we present an alternative strategy for the immunoreactivity mimicking of surface antigenic regions that relies on the spatially controlled co-presentation of those peptide sequences forming the conformational antigenic surface within the protein fold, to functionally reproduce its immunoreactive properties through cooperative effects arising from the mutual conformational dynamics of constituent linear sequences. Our method takes advantage of a “click” 14 immobilization scheme on microarray surface allowed by the use of the recently developed coating polymer named copoly Azide15 (Scheme 1).

We tested this strategy in the frame of our current efforts towards the development of rapid and cost-effective assay to diagnose Zika virus (ZIKV) infections, presently one among the most needed and sought for in public health protection16,17. In particular, we focused our study on the ZIKV NS1 protein antigen, as literature evidences strongly support the role of NS1 conformational epitopes in the elicitation of ZIKV specific antibodies18. We indeed identified a putative antigenic region (PAR) of NS1 through MLCE, a novel computational approach to predict potential antibody binding epitopes from the analysis of the energetic and conformational dynamic properties of the cognate protein antigen19,20,21. Once correctly combined through our co-presentation strategy, combinations of peptides spanning the region predicted to be most prone to Ab-recognition were found effective in discriminating Zika positive vs healthy samples, whereas the corresponding single linear peptide sequences failed in doing so. Results and Discussion

Epitope prediction on ZIKV NS1 protein The genome of ZIKV encodes a unique polyprotein that is further cleaved in three structural (prM, E and C) and seven non structural (NS1, NS2A-B, NS3, NS4A-B and NS5) proteins. Besides the envelope protein E, which is the scaffold of the viral particle, NS1 is the only non structural protein secreted in the host extracellular matrix where it interacts with the complement system and functions as an immune-evasion factor22,23, thus representing a good target for epitope design. The functional unit of NS1 is made of a homodimer, that when

ACS Paragon Plus Environment

4

Page 5 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Infectious Diseases

secreted in the extracellular space further trimerizes to form a hexameric structure bound to the lipid membrane24. Each monomer is composed of three distinct regions, an N-terminal βroll, a globular Wing domain and a C-terminal β-ladder comprising ten strands25,26 (Figure 1). The fold of the monomer, as well as the quaternary arrangement of the dimeric and hexameric complexes, are conserved among other flaviviruses, such as Dengue (DENV) and West Nile (WNV) for which the three dimensional structures are known27,28.

Figure 1. Structure of NS1 protein. A: Dimer. B: Monomer. Each monomer is composed of three distinct regions, an N-terminal β-roll, a globular Wing domain and a C-terminal βladder comprising ten strands

Our calculations have been carried out by considering either the single monomer or the dimeric unit using a predictive method that combines the analysis of the structural, dynamic and energetic properties of the protein to detect those regions that are most prone to establish protein-antibody interactions (for details, see Methods and refs29,30). Potential epitope sequences are predicted based on the consensus results of MLCE analyses conducted using different confidence levels as well as using the domain-decomposition approach. Since the βroll and β-ladder together form the hydrophobic surface able to bind the lipid membrane, while the loop region facing the other side of the β-ladder forms the solvent exposed face of the domain, it is not surprising that the latter region was predicted to be a putative antigenic site by our calculations. Our calculations are further supported by previous hypotheses on the significance of this region, which display a differential surface electrostatic potential with respect to other flaviviruses 25. Indeed, the loop surface of the β-ladder has been previously

ACS Paragon Plus Environment

5

ACS Infectious Diseases 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 29

identified as a potential antigenic region by other groups that predicted different epitope sequences, both for antibody or T-cell responses

31, 32, 33, 34

partially overlapping with our

sequences. Nevertheless, in some cases the effective antigenic capacity of these predicted epitopes was not experimentally validated. Moreover, the region predicted by our calculations partially overlaps with the antibody binding site detected by X-ray crystallography for the WNV NS1 dimer 35 Some of the residues identified as main players in the WNV NS1-antibody interaction are not conserved in ZIKV, suggesting a key role of this region in guiding the specificity of the flaviviruses infection and reinforcing the meaningfulness of our prediction. Our predicted region entails chains whose directionality alternates the N→C orientations of the various sequences. We thus divided the region into four putative linear epitopes covering the whole surface. Their sequences and locations are reported in Figure 2.

Figure 2. Putative antigenic region (PAR) of the ZIKV NS1 protein. Predicted region is located on the solvent exposed face of the β-ladder domain. Alternating peptide stretches composing the accessible surface are highlighted in different colors. Relative NtCt

ACS Paragon Plus Environment

6

Page 7 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Infectious Diseases

orientation and sequence are represented on the right panel. The sequence of the entire βladder domain is shown on the top of the figure with the four epitope stretches highlighted.

Controlled peptide immobilization on “clickable” surface To enable probe conjugation onto microarray surface, peptides were modified with a shortchain PEG spacer (O2Oc) including a terminal alkyne function for chemoselective conjugation to the recently developed copoly Azide polymer15 (Scheme 1) via Copper catalysed azide-alkyne-cycloaddition (CuACC). In recent reports, we and others showed how this immobilization strategy is highly favorable to finely control ligand orientation and surface density36,

37

, ultimately resulting in optimal probe exposure on the chip and more

reliable analytical performances. Here, this immobilization strategy enabled a control of either the probe density and the mutual orientation of peptides. . The co-immobilization procedure is straightforward and robust, i.e. linear co-spotted peptides are mixed in a 1:1 molar ratio in the spotting buffer for CuAAC conjugation. To confirm an equal representation of each component within the peptides combination, peptide surface binding yields were determined by IRIS38, a label free interferometric technique that quantifies biomolecule mass immobilization at each spot. It is worth noting that the possibility to control the ligand(s) surface density is crucial, as we previously demonstrated that probe overcrowding can be detrimental even in the recognition of individual immune-reactive probes39 and could be possibly even more unfavorable in co-immobilized probes combinations. At 50 µM spotting concentration a peptide spacing in the 1-2 nm range was calculated from mass surface density as measured by the IRIS platform in dry conditions. Noteworthy, in the bioassay wet conditions we can speculate a role of the polymeric surface coating which, due to its hydrophilicity and low density (as previously characterized

40

,

41

,

42

), provides a highly

flexible 3D scaffold favorable to probe transient interplay.

Immune-response screening to Zika patient’s sera A preliminary characterization of the sera used in this work was performed by comparing their immunoreactivity towards the full NS1 ZIKV antigen in a native vs thermally denatured form. Both proteins were spotted on copoly(DMA-NAS-MAPS)43 coated silicon chips and probed with a panel of healthy controls and of ZIKV positive samples. In all sera, native

ACS Paragon Plus Environment

7

ACS Infectious Diseases 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 29

antigen provided a much higher immunoreactivity to IgG than the denatured form (Figure 1S in Supplementary Information), confirming a crucial role of protein 3D structure integrity for antibody binding and supporting the relevance of discontinuous epitopes in ZIKV NS1 immunoresponse. Negative samples provided negligible signals for both antigen forms. We then focused our attention to the four distinct linear peptides spanning the predicted NS1 immunoreactive region (P1, P2, P3, P4) (Scheme 2, Table S1). Screening of linear peptides P1-4 against the panel of 20 sera from ZIKV infected patients resulted in poor immunoreactivity and no significant discrimination towards healthy individuals in terms of mean fluorescence intensity (MFI) and recognition frequency (Figure 3A). Consistent with this, the combination of the same sequences in single extended linear construct spanning the whole region (E1, E2) (Scheme 2; Table S2) failed to give a productive discriminative response (Figure 3A). This finding supported our initial hypothesis that a strategy implementing the use of single linear peptides was needed to address immunoreactivity preservation and consequently enable diagnostic discrimination of patients samples. We therefore moved to test peptides combination that were designed to reproduce side-to-side proximal sequences within NS1 native fold via controlled directional co-immobilization (Scheme 2; Table S2).

Scheme 2: peptide sequences and epitope combinations. See Table S2.

ACS Paragon Plus Environment

8

Page 9 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Infectious Diseases

It has to be underlined that our method does not aim at closely reproducing a specific conformational state within the protein antigenic region but, rather, it takes advantage of the spatial proximity of flexible linear regions whose conformational dynamics permits to populate ensembles that, in combination and cooperatively, expose the functional groups necessary for antibody recognition. In this framework, we designed our peptide immobilization scheme so that relative directional orientation would match the same N→C directionality between proximal sequences found in the PAR of NS1. Statistically, a 1:1 ratio co-immobilization of peptides will determine an effective side-to-side alignment of the peptidic probes. We therefore co-immobilized different peptide combinations made of two (C1, C2, C3, C4, C5) or three (T1, T2) linear peptides P1-4. Combinations were designed to mimic the same spatial arrangement and relative orientation between the different peptide stretches found within the PAR (see Scheme 2; Table S2). Importantly, we could observe that the co-spotted combinations provided a generally higher immunoreactivity against ZIKV patients’ sera with respect to the IgG response detected for the single peptidic probes (Figure 3, upper panel), whereas healthy controls yielded a low response and smaller variability around the mean fluorescence value observed within this sample population; a similar effect is observable in terms of recognition frequency (see Figure 3, lower panel).

ACS Paragon Plus Environment

9

ACS Infectious Diseases 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 29

Figure 3. A) IgG response from sera of 20 ZIKV positive patients and 20 healthy (HD) controls probed on single peptide epitopes (P1-4) and extended sequences that could span broader interaction areas (E1, E2) and B) probed on epitope combination C1-5 and T1-2. Mean fluorescence intensity (top) of the tested panels with 95% CI and recognition frequencies (bottom) are reported. Fluorescence was detected with 25% laser power and 60% PMT gain. Statistical significance level of unpaired t test and Fisher’s exact test p values are showed (*, < 0.05; **,