Assessment of bio-based materials for enhanced signal detection of sporozoites using csELISA

Background: 1 Malaria is responsible for over 435,000 deaths annually, with most cases occurring in sub-Saharan Africa. 2 Detecting the presence of Plasmodium spp. sporozoites (spzs) in the salivary glands of Anopheles vectors 3 of the parasites using the circumsporozoite enzyme-linked immunosorbent assay (csELISA) is an important 4 malaria surveillance method. The addition of bio-based materials have shown the potential to improve the 5 adsorption and binding of target antigens and thus can improve the sensitivity and detection of analytes in 6 immunoassays. Here, we evaluate the use of two bio-based polymers, chitosan, and cellulose nanocrystals 7 (CNC), as antibody carriers and substrate coating on 96-well plates and on a paper substrate to determine 8 whether detection of Plasmodium falciparum (Pf), P. vivax VK210 (Pv210) and P. vivax VK247 (Pv247) 9 can be improved through assay modification. 10 Methods: 11 Modified csELISA assays were conducted using chitosan and CNC as either an antibody carrier or a well 12 coating on 96-well plates (ultra-low- and high-affinity) and results read using standard spectrophotometry 13 of 96-well plates and a quantitative image-based color analysis using photographs of paper plate assays. 14 Changes in frequency and dissipation, resulting from the adsorption of antibodies to model films in a quartz 15 crystal microbalance with dissipation monitoring (QCM-D), were followed to understand better the 16 interactions between the bio-based materials and assay proteins. 17 Results: 18 The csELISA performed on high-binding well-plates showed that chitosan, used either as an antibody 19 carrier or well coating, resulted in the greatest increase in detection for Pv210 and Pv247 positive 20 recombinant proteins, increasing absorbance readout values up to 6x for Pv210 and up to 5x for Pv247. On 21 paper csELISA (PcsELISA), chitosan as an antibody carrier yielded the greatest increase in detection 22 sensitivity for all three Plasmodium species when color intensity of positive recombinant proteins was 23 compared to blanks. Compared to controls, chitosan as a carrier resulted in a ~2.5-fold increase in color 24 intensity for Pf, a ~4-fold increase for Pv210, and a ~2-fold increase for Pv247. QCM-D showed a preferred 25 interaction between the assay antibodies and chitosan surfaces, most likely driven by electrostatic 26 interactions. The addition of bio-based materials, mainly chitosan, as shown by QCM-D interactions, absorbance readout 29 values, and image-based color analysis, increased the color intensity of positive samples run through 30 csELISA in all systems, allowing for clearer detection of Plasmodium spp visually, using a 31 spectrophotometer and quantitative color intensity. Furthermore, the adaptation of a PcsELISA coated with 32 chitosan using positive recombinant proteins shows potential as a cost-effective alternative assay platform 33 as it reduced reagent volumes by 80% and assay run time from seven hours to one hour.


43
Human malaria is a life-threatening disease caused by Plasmodium, a group of apicomplexan parasite 44 species that is vectored by female Anopheles mosquitoes (1). Detecting the presence of Plasmodium spp. 45 sporozoites in the salivary glands of Anopheles is an important malaria surveillance method, which helps 46 measure the intensity of malaria transmission and evaluate intervention methods to assess programmatic 47 are deposited directly onto the paper substrate rather than into the well of a 96-well plate, the surface area 68 of the reaction is reduced, thereby decreasing the volume of antibodies, samples and other reagents used 69 (9). Despite the numerous advantages that come with paper-based assays, there are limitations, notably in 70 those that use a peroxidase-based reaction as colorimetric indicator. Due to differences in capillary forces 71 among the reagents and analytes, uneven distribution over the test zone can result in heterogenous color 72 formation, making correct interpretation of the results difficult (10). 73 To overcome this lack of homogeneity and improve binding affinity between the antibodies and 74 the test zones, as well as to act as coating agent smoothing the surfaces (and thus capillary forces), bio-75 based materials can be used as additives in the process. Some of these bio-based materials have been 76 demonstrated to improve biomolecule stability and immobilization on the well walls (11). The goal of this 77 work was to assess if the addition of these materials can improve assay sensitivity and reduce detection 78 limits when compared to traditional 96-well plate based ELISAs. 79 One such bio-based material is cellulose. This polymer is formed by glucose monomer links in β 80 (1→4)-D-glucopyranose conformation, and it is the most abundant polysaccharide on earth, present in the 81 cell walls of plants and wood, found in tunicates and algae, and is also produced by some bacteria (12) 82 Fibrils are packed in two alternating forms within the cellulose macrostructure: a tight crystalline form, 83 where cellulose chains are highly ordered, and an amorphous one. Traditional processing of wood results 84 in lignocellulosic pulp, which is the raw material to produce paper (13). However, after an acid hydrolysis 85 step, the pre-treated lignocellulosic pulp can be processed to selectively remove the amorphous portions 86 and produce cellulose nanocrystals (CNC) (14). 87 Like cellulose, chitosan has also shown potential as a bio-based material for assay improvement. 88 Chitosan is a polycationic linear polysaccharide which is easily obtained from chitin, the second most 89 abundant natural polysaccharide after cellulose. Chitosan is produced by the partial deacetylation of chitin 90 under alkaline conditions, which is a structural component of the exoskeleton of arthropods and cell walls 91 of fungi and yeast (15-17). This biodegradable, nontoxic and biocompatible bio-based material has 92 demonstrated a key role in paper-based assays, producing uniform color signals at the detection spot when 93 used as a substrate (18,19). 94 Here, the chemistry of bio-based materials was integrated with the well-established csELISA to 95 determine whether this could potentially improve the detection of sporozoites of P. falciparum, P. vivax were tested as well-coating materials for the plates and as antibody carriers. When used as antibody carriers, 133 the polymers were mixed with the antibodies in solution before adding to the plates. 134

Preparation of bio-based materials for well coating 135
For well-coating solutions, a 0.1% chitosan solution and a 0.1% CNC suspension were prepared in 136 1% acetic acid and water, respectively. All suspensions and solutions were prepared with Milli-Q water 137 (Millipore Inc., 18.2 MΩ cm). Suspensions of CNC were sonicated for 5 minutes at 25% amplitude, with 138 the pulser set in five seconds intervals of three second "on" and two seconds "off," while maintaining the 139 suspension in an ice bath, prior to use. Wells of the 96-well polystyrene plates were filled with 200 µL of 140 the bio-based material, covered with a lid and incubated for 30 minutes at room temperature. Thereafter, 141 the solution, or suspension, was aspirated and the wells were rinsed as follows: i) wells coated with chitosan 0.1% were rinsed one time with 200 µL of 1% acetic acid and two times with 200 µl MilliQ-water; ii) wells 143 coated with 0.1% CNC were rinsed one time with 200 µL of PBS-Tween (0.05%) and two times with 144 200µL MilliQ-water. After that, the plates were left covered with plate lids overnight at room temperature 145 (ca. 20 °C) to air dry. 146

Preparation of bio-based materials for antibody carriers 147
A 1% chitosan solution was prepared in 1% acetic acid and then diluted to 0.1% with PBS, pH 7.4. 148 CNC suspensions were diluted in PBS, pH 7.4 to obtain a 0.1% concentration and sonicated, as previously 149 described. The Pf , Pv210 or Pv247 stock capture antibodies were rehydrated as described (20), and added 150 to 5 mL of the prepared bio-based material solution (in the place of PBS). Antibody-bio-based material 151 solutions were then vortexed and incubated at room temperature for one hour before use. The remainder of 152 the protocol was followed as described in the MR4 csELISA protocol (20). 153

Quartz Crystal Microbalance with Dissipation monitoring (QCM-D) 154
To better understand the interactions occurring between the bio-based materials and the csELISA 155

Development of paper ELISA template 166
A paper-mimic of 96-well plates (Figure 1) was created on chromatography paper (Whatman Grade 167 diameter, resembling the wells of 96-well ELISA well plates, were drawn onto the paper with pencil. With 169 a commercial liquid wax solution, wells were then outlined using a fine-tip paintbrush. To create a 170 hydrophobic well-barrier, the "plate" was heated on a hot plate at 100°C for three minutes to allow for 171 complete penetration of the wax through the paper matrix. Chitosan and CNC were then evaluated on the 172 paper 'plates' in different combinations as antibody carriers and well-coating. Recombinant proteins and 173 bio-based materials were prepared as described above. For well-coatings, 100 µL of bio-based material 174 solution or suspension was pipetted onto paper wells and allowed to dry overnight. For bio-based material 175 antibody carrier solutions, 10 µL was pipetted onto each well. 176

Paper cs-ELISA protocol adaptation 177
All incubation steps from the csELISA protocol (20) were reduced to 15 minutes. Excess reagent 178 was removed after 15 minutes by placing a Kimwipe (Kimtech Science Kimwipes Delicate Task Wipes 179 product code #34155) on the underside of the paper plate, which wicked liquid through. To prevent leaking 180 of the wells or mixing of the well contents on the underside of the paper, the paper-plate was suspended by 181 placing it on an empty reagent reservoir. All volumes were reduced to 20% of the volume outlined in the 182 csELISA protocol (20). Working sample and antibody solution volumes were reduced from 50 µL to 10 183 µL, blocking and wash buffers were reduced from 200 µL to 40 µL, and ABTS (2,2'-azino-bis(3-184 ethylbenzothiazoline-6-sulfonic acid)) was reduced from 100 µL to 20 µL. Following the addition of ABTS, 185 plates were allowed to incubate for 15 minutes before a cellphone camera was used to take a photograph of 186 the entire paper-plate, to create an image to assess the resulting color change. 187 showed absorbance values up to six times higher than those from the control wells (without chitosan as a well-coating or antibody carrier) ( Figure S1). Moreover, on high-affinity binding plates, the absorbance 219 values of Pv210 from wells where chitosan was used as a well-coating, antibody carrier or when used at 220 the same time with CNC as a carrier at recombinant protein concentration of 18.2 µg/µL were comparable 221 to those with Pv210 at 182 µg/µL in wells where bio-based materials were not used as a well-coating or 222 antibody carrier. Thus, using chitosan, as well-coating or antibody increased the sensitivity of recombinant 223 protein concentration detection by 10-fold (Figure Error! Reference source not found.2). 224 Circumsporozoite ELISA control assays (no bio-based materials as antibody carriers or well-225 coatings) performed on ultra-low binding plates for Pv210 yielded no increase in absorbance values when 226 recombinant protein was present over the dilution series, compared to blanks (blocking buffer only) ( Figure  227 2), indicating that little to no antibody binding occurred on the well surfaces. Nevertheless, the absorbance 228 values obtained when chitosan was used either as an antibody carrier or well-coating were between 1.2 and 229 1.7x higher than those observed for Pv210 on control wells of high-affinity binding plates. 230 For Pv210 assays performed with CNC as a well-coating, results more heavily depended on the 231 specific plate type, and did not yield an increased absorbance value over the recombinant protein or when 232 the chitosan was used alone either as an antibody carrier or well-coating. With CNC applied as a well-233 coating with chitosan as an antigen carrier on the high-binding plate a two-fold increase in absorbance was 234 observed when compared to controls. 235

Pv247 236
The results for Pv247 on high-affinity binding plates showed a similar trend to that observed for Pv210 237 (see Figure 2), with up to a five-fold increase in absorbance values when compared to controls when 238 chitosan was used as an antibody carrier or well-coating. Chitosan, either as a well-coating or antibody 239 carrier, was also capable of reducing the working concentration, and lowering the detection limit, of the 240 Pv247 recombinant protein up to five-fold without a significant reduction in sensitivity for assays carried 241 out on both high-binding plates (Figure 2). 242

Adsorption of antibodies on model film in QCM-D 244
QCM-D followed the adsorption of the capture antibodies onto model films of the bio-based materials to 245 better understand the affinity between them. Flow concentrations of the antibodies were 2 µg/µl for Pf, 182 246 µg/µl for Pv210, and 91 µg/µl for Pv247 in PBS and flow of 50 µL/min. The graphs of results are shown 247 in Figure 3, and a summary of the relevant values is presented in Table 1. Table 1

Bio-based materials as antibody carriers and well-coatings with the Paper csELISA 261
CYMK data was compared between seven different assays for three antibodies: chitosan well-262 coating and CNC antibody carrier combination, CNC well-coating and chitosan antibody carrier 263 combination, chitosan well-coating, CNC well-coating, chitosan antibody carrier, CNC antibody carrier, 264 and a control (no antibody nor carrier and no bio-based well-coating). 265

Pf 266
For Pf recombinant proteins tested at the working concentration (2 µg/µL), the combination of 267 chitosan as an antibody carrier and CNC as a well-coating yielded an approximately 4.5-fold increase in 268 CMYK color intensity (with blank/background values subtracted) compared to controls (no antibody carrier 269 or well-coating). Similar increases were observed from the other two CNC assays, with CNC as an antibody 270 carrier or well-coating alone. Chitosan as an antibody carrier alone provided an approximately 2-fold 271 increase in color intensity when compared to controls (no antibody-carrier or well-coating; Figure 4 and 272 S2). 273 When ten-fold dilutions were performed, CNC as a well-coating and chitosan as an antibody carrier 274 continued to perform best at lower concentrations, followed closely by CNC as either a well-coating or 275 antibody carrier (Figure 4). 276

Pv210 277
At the recombinant protein working concentration, the top three performing bio-based polymer 278 assays involved an antibody carrier (Figure 4). Two of them had chitosan as the antibody carrier, yielding 279 a 4-4.5-fold increase in color intensity when compared to controls (no antibody carrier or well-coating). 280 Chitosan as an antibody carrier alone increased color intensity approximately 4-fold, while chitosan as an 281 antibody carrier in combination with CNC as a well-coating and CNC as an antibody carrier alone provided 282 an approximate 4.5-fold increase in color intensity. 283 At the point following the working concentration (first dilution of antigens Pv210 tested), similar 284 increases for chitosan as an antibody carrier in combination with CNC as a well-coating as well as CNC as a well-coating alone were observed, with CNC as a well-coating providing the greatest increase in intensity. 286 Chitosan as an antibody carrier alone performed better than controls (no antibody carrier or well-coating) 287 but did not perform as well as the first two which involved CNC. 288

Pv247 289
For the Pv247 recombinant protein assayed at the working concentration, the same three assays that 290 performed optimally with Pv210 antibodies and recombinant proteins performed the best. ( Figure S2). 291 Chitosan as an antibody carrier alone yielded the greatest increase in color intensity, yielding an 292 approximate 2-fold increase in color intensity when compared to controls (no antibody carrier or well-293 coating; Figure 4). Chitosan as an antibody carrier in combination with CNC as well-coating and as CNC 294 as a well-coating alone performed similarly, yielding slightly less than a 2-fold increase of color intensity 295 when compared to controls (no antibody carrier or well-coating). 296 When ten-fold dilutions were performed, a similar trend was observed as for the other antigens: 297 CNC as a well-coating performed the best at lower concentrations, with chitosan as an antibody carrier and 298 the combination of chitosan as an antibody carrier with CNC as a well-coating performing just below. 299 300

301
In this study, two bio-based polymers, chitosan and CNC, were used to modify the csELISA and 302 any resulting differences in detection capabilities were assessed. The use of low quantities (<0.1% wt. of 303 the solutions used) of both bio-based materials was shown to improve detection signals. Chitosan as an 304 antibody carrier performed the best overall in improving detection on polystyrene plates regardless of high-305 or ultra-low binding affinity. The favored interaction of chitosan with the antibodies was also observed on 306 the QCM-D, where the surface it generated showed higher affinity for the antibodies. These results are 307 likely related to the favorable interactions between the positively-charged amino groups of the chitosan and 308 the negatively charged antibodies (24). Furthermore, due to its chemical structure, chitosan can generate 309 electrostatic interactions and hydrogen bonding with different polymers, which may explain the increased 310 affinity observed when used with the commercially treated plates. 311 Thus, these results support the addition of chitosan as either a well-coating or antibody carrier in 312 the csELISA to help increase the intensity of the signal when detecting Pv210 and Pv247 proteins. This 313 may be particularly useful when determining the presence of sporozoites in field collected mosquitoes, 314 where the number of sporozoites can be highly variable (25). 315 Furthermore, the csELISA for malaria parasite detection in mosquitoes was adapted to a paper 316 format using the same bio-based materials as well-coatings and antibody carriers, thus combining the paper-317 based csELISA, which reduced the volumes of reagents and samples necessary, with the improvements 318 observed using the bio-based materials. 319 Combining bio-based materials with paper-based assays provided advancements regarding the 320 detection and visual readability of adapted csELISAs using cost-effective materials. The data presented 321 here can be used to further develop and improve paper-based assays. Figure 5 shows photos taken of 322 PcsELISAs after 15 minutes of drying and after drying overnight. Despite the higher intensity values seen 323 in assays with CNC as a well-coating, further optimization of bio-based materials may be needed to obtain 324 consistently reproducible results. In assays where CNC was used as a well-coating (Fig 5. c-1), reagents 325 tended to leak through the hydrophobic barriers, false positives were common, and overall color intensity 326 was greater. This made results harder to distinguish using the naked eye. Using CNC with a different paper 327 substrate, for example, nitrocellulose or nanocellulose coated papers, may provide a viable alternative to 328 prevent reagent leaking while maintaining color intensity. Color homogeneity, which can best be seen 329 following overnight drying (Fig. 5 x-2), was improved for both chitosan and CNC when compared to the 330 control (no antibody carrier or well-coating) (Fig. 5 a-1). The chitosan antibody carrier assay (Fig. 5 b-1)  331 showed a much clearer, even distribution of color throughout the entire test well. The change in color 332 intensity as recombinant protein concentration increased was easily visible to the naked eye for assays with 333 chitosan as antibody carrier. This is an important consideration, particularly if plate readers are not 334 available. Here, the chitosan antibody carrier paper-based csELISA was shown to have the most potential to be further explored for future use and scalability. This is due to consistent quantitative colorimetric data 336 and clear visual results. Chitosan as an antibody carrier does not require the additional overnight incubation 337 step that is necessary to coat paper 'plates'. Future work evaluating the use of paper-based csELISAs will 338 reveal whether this method can be implemented for more economical and efficient detection of sporozoites 339 in field-collected mosquitoes in resource-limited settings. 340 The more favorable use of chitosan as an antibody carrier rather than a well-coating is likely related 341 to the more efficient interaction between the antibodies and chitosan when put in contact with the solution. 342 Specifically, more active points of chitosan can interact with the antibodies when mixed in solution instead 343 of those sites interacting with the paper or plate surfaces. This effect of groups exposed would also affect 344 the surface available to interact with the antibodies when being applied as the coating, as the surface area 345 of the bio-based materials will become more compact during the adsorption and drying (26). Furthermore, 346 the improved detection of the recombinant proteins when using the CNC as a well-coating and chitosan as 347 an antibody carrier can also been explained by the increase in surface area that the CNC coating induced 348 on the paper csELISA, which can be measured by the changes in rugosity and porosity of the paper (27). 349 This change in surface area could also increase the interaction between the chitosan and the cellulosic 350 surfaces -either the paper by itself, or the paper coated with CNC -in the presence of PBS, as the greater 351 salt content promotes adsorption between the two bio-based materials, chitosan and paper (cellulose)(28). observed when compared to controls (no antibody carrier or plate coating). This is potentially due to a 357 higher affinity between the chitosan and antibodies than the antibodies with the well surfaces. This 358 improved interaction was also demonstrated by QCM-D, where adsorption of the antibodies was higher on 359 the chitosan model films. Experiments with PcsELISAs showed the same trend, where the use of chitosan as either an antibody carrier or well-coating increased intensity values when compared to the controls (no 361 antibody carrier or well-coating). The use of CNC as a well-coating with PcsELISAs also improved the 362 signal further when compared with the test zones with no bio-based material, possibly by increasing the 363 available surface area for antibody-substrate interactions. Unlike the addition of chitosan which changed 364 the charge of the wells in polystyrene plates, CNC is a 3-D material which added volume and area, another 365 benefit to paper-based assays when compared to polystyrene plates. 366 The bio-based materials tested, particularly chitosan, both as antibody carriers and well-coatings, 367 have demonstrated their feasibility in improving the detection of the sporozoite proteins from P. polystyrene plates (both high-and ultra-low affinity binding), but also with paper-based csELISAs. 371 Although further development of PcsELISAs is needed, this could be a more convenient assay with 372 regard to cost, availability, scalability, time, and ease-of-use. Work ow showing conversion of cellphone photograph to numerical color data. CMYK changes in each well were averaged to one single value, which was then identi ed as the color intensity difference percentage. This color intensity value was then plotted against antigen concentration to determine the relationship between the two. ELISA results for Pf, Pv210 and Pv247 run on high-binding (MaxiSorp™) and ultra-low binding(Corning® 3474) plates. The conditions tested were chitosan as antibody carrier or well-coating (Chi 0.1% coating), chitosan as an antibody carrier (Chi 0.1% carrier) and chitosan as well-coating and cellulose nanocrystal as a carrier (Chi 0.1% coating / CNC 0.1% carrier). The samples and controls were run at four concentrations: the standard cs-ELISA recombinant protein working solution concentrations and three 10fold dilutions. The working solution concentrations were: 2 µg/µl for Pf, 182 µg/µl for Pv210 and 91 µg/µl for Pv247. The absorbance was measured at 405 nm. Samples and controls were run in triplicate and each assay was repeated three times. Bars represent SD, n=9.  Paper csELISA results for all antibody carrier and well-coating assays. Recombinant proteins were tested at concentrations as follows: Pf at 0 pg/µL, 0.2 pg/µL, 2 pg/µL, 50, pg/µL, and 100 pg/µL, Pv210 at 0 pg/ µL, 18.2 pg/µL, 182 pg/µL, 4500 pg/µL, and 9100 pg/µL, and Pv247 at 0 pg/µL, 9.1 pg/µL, 91 pg/µL, 2275 pg/µL, and 4550 pg/µL. Data were plotted and converted to log-scale as a graph of color intensity difference versus concentration of the recombinant protein. The change in color intensity for each concentration was calculated as the difference between the average CMYK value and the corresponding blank (in percentage).

Figure 5
Photographs of paper-plate assays were taken using a cellphone camera after drying for 15 minutes, and overnight. Signi cant differences can be seen in the distribution of color and overall well homogeneity between the three different assays. Both bio-based material assays showed a notable improvement over the control, with the chitosan carrier assay showing the clearest visible difference in increasing antigen concentration.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. csELISAmanuscriptsupportinginformation.docx