Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Improved performance of Eimeria-infected chickens fed corn expressing a single-domain antibody against interleukin-10


Antimicrobial resistance is a significant challenge for human and animal health, and developing effective antibiotic-free treatments is a strategy to help mitigate microbial resistance. The global poultry industry faces growing challenges from Eimeria-induced coccidiosis, a serious enteric disease of chickens that currently requires treatment using ionophore antibiotics. Eimeria stimulates interleukin-10 (IL-10) expression in the small intestine and caecum of infected chickens, suppressing their immune response and facilitating disease progression. Single-domain antibodies raised from llamas immunized with chicken IL-10 (cIL-10) were developed that bind cIL-10 in vitro, block cIL-10 receptor binding and induce interferon gamma (IFN-γ) secretion from cIL-10-repressed primary chicken splenocytes. Single-domain antibodies expressed in transgenic corn demonstrated significant accumulation in phenotypically normal plants. When fed to Eimeria-challenged chickens, the transgenic corn significantly improved body weight gain (equal to that of salinomycin-treated animals), normalized the feed conversion ratio (to the same level as uninfected control animals), lowered E. tenella lesion scores to those of salinomycin-treated control animals, and reduced oocyst counts below those of infected untreated control animals. Here, we propose that transgenic corn may have a role in reducing the use of antibiotics in poultry production and maintaining animal health and productivity, and may contribute to efforts against global antimicrobial resistance.


Global chicken production set a record of 99.6 Mt in 2019, up from 95.6 Mt in 2018, reflecting the continued growth in chicken meat as a desirable protein source among consumers1. Poultry is second only to swine in terms of total tonnage of meat produced2 and is an efficient source of animal protein for human nutrition. Compared to beef production, poultry utilizes 28 times less land, 11 times less water and 6 times less reactive nitrogen consumption, and has a fivefold lower carbon footprint3.

Coccidiosis, an enteric disease caused by Eimeria infection in the small intestine and caecal linings4, is reported to be present in over 90% of commercial flocks5. Clinical coccidiosis presents with diarrhoea and bloody stools, resulting in high morbidity and mortality, while subclinical coccidiosis decreases animal productivity and is often undetected. Lost productivity and increased animal prophylactic and treatment costs result in an estimated US$3 billion loss to the global poultry industry6,7,8. Disease control strategies include vaccination8,9, increased production hygiene10,11 and antibiotics, where ionophores are commonly employed12. Although these treatments provide a toolkit for managing coccidiosis, they each have shortcomings such as increased production costs, inconsistent protection or the emergence of antibiotic resistance. Due to the increasing risk of antibiotic resistance, tighter regulatory measures and consumer-driven demand to decrease antibiotic use in animal-based food production, new tools are needed to maintain productivity and animal health and welfare.

Coccidiosis stimulates the local secretion of IL-10 into the lumen of the gastrointestinal tract of chickens13,14. IL-10 can reduce inflammation and suppress immune responses by reducing cytokine secretion, antigen presentation, CD4+ T-cell activation, and the induction of tumour-necrosis factor-α and IFN-γ15,16. Suppression of the immune response in this manner allows the pathogen to colonize the intestinal tract, leading to higher morbidity and mortality among affected birds. Previous research has established the concept that Eimeria infection could be controlled by blocking IL-10 signalling in the gut. Feeding chickens lyophilized eggs from chickens immunized with IL-10 peptides has demonstrated improvements to gut health and animal performance in Eimeria-challenged chickens13,17. This has led to the proposal that egg-yolk-derived antibodies might be incorporated into feed as a general means for controlling coccidiosis or other diseases in poultry13. However, this form of implementation suffers from high production costs for egg-derived antibodies and poor thermal stability of the IgYs18, preventing adequate dosing in common feed preparation processes, which can exceed temperatures of 85 °C.

Here, we address those implementation issues of high production costs and thermal stability. We developed a single-domain antibody (VHH, or sdAb) that binds full-length chicken IL-10 (cIL-10), as the basis for an oral treatment strategy to treat coccidiosis. This approach enabled the generation and screening of sdAbs that bound cIL-10, and which were then evaluated to predict their efficacy in vivo. These antibodies have demonstrated thermal stability19, which is useful in feed pelleting. Selected sdAbs were expressed in transgenic corn, providing what, we believe, may be one of the lowest possible production costs for recombinant protein expression available in the world20,21. Once made, the transgenic corn was used to feed Eimeria-challenged chickens, demonstrating significant improvements in growth performance and gastrointestinal pathology relative to challenged, but untreated, control animals.


Generation of sdAb against cIL-10

Full-length cIL-10 (IBI Scientific) was used to inoculate a llama, followed by subsequent booster immunizations at seven and eleven weeks. Chicken and llama IL-10 homologues are only 48% identical (68% similar; Supplementary Fig. 1), providing adequate antigenic disparity to generate a library of sdAb candidates, which were isolated from the llama as described previously22. Approximately 200 different sdAbs were cloned and evaluated for their ability to bind to cIL-10 by enzyme-linked immunosorbent assay (ELISA) and found to have a variety of apparent half-maximum effective concentration (EC50) values (Fig. 1a).

Fig. 1: In vitro analysis of candidate anti-cIL-10 antibodies.

a, Evaluation of candidate sdAbs. Serial dilutions of four individual candidates were tested for binding to cIL-10 by direct ELISA22. Means of n = 2 independent experiments are shown with error bars corresponding to the s.e.m. b, Biacore experiments tested the relative ability of two candidate sdAbs to inhibit the cIL-10–cIL-10r interaction. Samples were tested in duplicate at each concentration. Means of n = 2 independent experiments are shown with error bars (smaller than the symbols for each data point) corresponding to the standard deviation. c, Cell-based evaluations of candidate sdAbs. Primary chicken splenocytes will secrete IFN-γ when incubated in the presence of ConA (dashed horizontal red line). IFN-γ secretion is suppressed in the presence of cIL-10 (dashed horizontal green line). Serial dilutions of individual sdAbs were evaluated for their ability to overcome cIL-10-mediated suppression of IFN-γ secretion when included at concentrations of 1, 30 or 1,000 nM. Two of the sdAbs shown here were able to restore IFN-γ secretion to a level above that of the ConA-only control; pAb ctl, positive control pAb targeting cIL-10; sdAb ctl, negative control sdAb that targets an unrelated bacterial protein. The data are presented as the averages from n = 3 biologically independent samples.

Source data

IL-10 forms a homodimer in solution23, and screening for sdAbs that bind to cIL-10 identified candidate sdAbs. However, depending on which antigenic sequence the sdAb bound, it may be possible that sdAb binding would not interrupt cIL-10 receptor (cIL-10r) binding. To further evaluate candidate antibodies, we expressed the soluble domain of cIL-10r as a fusion to the Fc region of human IgG1. A similar fusion protein that includes the soluble domain of human IL-10r and the IgG1 Fc region has previously been shown to bind human IL-10 in solution24. The recombinant cIL-10r–Fc fusion was immobilized on a Biacore T200 surface plasmon resonance (SPR) probe and used to detect cIL-10 binding in the absence and presence of sdAb candidates. Increasing concentrations of sdAbs, in the presence of 10 nM cIL-10, showed a steep decrease in cIL-10 binding to the immobilized cIL-10r (Fig. 1b). The measured half-maximum effective concentration (IC50) values (10–50 nM) were in close agreement with the apparent EC50 values determined for sdAb binding to cIL-10, which ranged from 1 nM to 100 nM, with AgThG11 having an IC50 value of 10 nM and an EC50 value of 1 nM (Fig. 1a).

These in vitro binding evaluations showed that the selected sdAbs could bind cIL-10, and that binding of cIL-10 could antagonize cIL-10r binding. To further demonstrate the biological significance of the in vitro interaction, we tested the sdAbs in a cell-based assay using primary chicken splenocytes25. Splenocytes treated with 5 μg ml−1 concanavalin A (ConA) secrete 4,000 pg ml−1 of IFN-γ in the absence of cIL-10. However, 1.5 ng ml−1 cIL-10 can suppress the secretion of IFN-γ by approximately 75%, down to 1,000 pg ml−1. Dosing of our sdAbs at concentrations of 30 nM or greater demonstrated recovery of ConA-induced IFN-γ secretion in the presence of cIL-10, demonstrating that the selected sdAbs could bind cIL-10 to circumvent its suppression of IFN-γ secretion in a dose-dependent manner (Fig. 1c). The primary amino-acid sequences of two candidate sdAbs (AgThG11 and AgThG611) are shown in Fig. 2.

Fig. 2: Comparison of two candidate sdAbs for transformation into corn.

The amino-acid sequences of two candidate sdAbs that were tested in in vitro assays (Fig. 1), showing the relative positions of the three complementarity determining regions (CDR). The dots indicate amino acids in the sequence of AgThG611 that are identical to those in AgThG11. The dashes indicate gaps in the alignment. The positions of synthetic linker sequences (lower case letters), c-myc, and 6-His tags are also indicated. The linker and tag sequences were excluded from the form of the AgThG11 protein that was expressed in maize.

The retention of binding affinity of microbially produced sdAb AgThG11 to cIL-10 following high-temperature incubation was evaluated by ELISA. As a control, a conventional anti-cIL-10 rabbit IgG polyclonal antibody (pAb) was used under identical conditions. The sdAb and pAb remained stable at 37 °C for the duration of 30 min of incubation. After 30 min treatment at 75 °C, the sdAb retained about 60% while the pAb retained only about 30% of the binding activity. Similarly, at 95 °C, the sdAb retained 65% activity after 20 s, about 50% after 3 min and approximately 15% after 30 min of incubation, while the pAb retained only about 10% of its binding activity after 20 s of incubation and no detectable activity afterwards (Fig. 3). These results are consistent with previous reports that demonstrate the thermal stability of sdAbs19.

Fig. 3: Temperature resilience of the candidate AgThG11 antibody.

Effects of temperature exposure on the binding affinity of the sdAb AgThG11 (left) and an IgG pAb (right) to cIL-10. Both antibodies were heated to 37 °C, 75 °C, 85 °C or 95 °C in a PCR thermocycler for 20 s, 3 min or 30 min. After 20 min renaturation at room temperature, antibodies were placed on ice. The ELISA (optical density at 450 nm (OD450 nm)) values of the antibody binding affinity to cIL-10 are expressed as the percentage of the binding affinity in an unheated sample. The results are presented as the means of n = 3 independent experiments with standard deviations.

Source data

Development of corn expressing anti-cIL-10

The gene encoding the sdAb candidate AgThG11 was transformed into corn using Agrobacterium transformation methods described previously26. Briefly, we cloned AgThG11 under the control of the maize 27kDagamma zein27, and the rice glutelin B-4 gene promoters28, providing two expression cassettes for the same gene to boost antibody production from the single transfer DNA (T-DNA) fragment (Fig. 4a). In both cassettes, the AgThG11 coding sequence was fused at the 5′ end with a maize gamma zein leader coding sequence29, and a KDEL endoplasmic reticulum retention coding sequence30 at the 3′ of the gene. Transgenic calli were selected using phosphomannose isomerase as a selectable marker on mannose-containing medium, subcultured into plantlets and then transferred to soil to grow into fully developed corn plants (events). Plants that expressed sdAbs were phenotypically normal (Fig. 4b). Grain was harvested from the mature plants, milled and the total protein was isolated to measure the sdAb expression level among different transgenic events. Event expression levels varied as assessed by polyacrylamide gel analysis of seed protein extracts (Supplementary Fig. 2). Grain from several T0 plants expressing AgThG11 was pooled and milled to provide corn meal for use in feeding trials with challenged animals.

Fig. 4: Transformation of corn with genes expressing the AgThG11 antibody.

a, A map of the transformation cassettes used to express AgThG11 in maize. RB, Agrobacterium T-DNA right border sequence; ZmZ27P, maize 27kDagamma zein promoter; Z27ss, signal sequence from the maize 27kDagamma zein gene; AgThG11:KDEL, coding sequence for AgThG11 with carboxy-terminal KDEL endoplasmic reticulum retention signal; NosT, transcriptional terminator from the Agrobacterium nopaline synthase gene; OsGluP, rice glutelin B-4 promoter; ZmUbi1P, maize ubiquitin 1 promoter with first intron; PMI, phosphomannose isomerase (selectable marker) coding sequence; LB, Agrobacterium T-DNA left border sequence. b, Greenhouse-grown maize plants expressing AgThG11 in seed develop normally.

Eimeria-challenged birds respond positively to sdAb AgThG11-expressing corn

Milled corn grain containing approximately 0.87 mg of the sdAb AgThG11 per gram (as determined by ELISA) was evaluated in a chicken model for coccidiosis. This trial compared unchallenged and untreated chickens (Cobb 500) with challenged and untreated animals, challenged animals treated with 500 g salinomycin (Bio-Cox) t−1 and challenged animals treated with one of three different doses of sdAb-containing milled corn grain (100 g, 400 g or 1,600 g of milled grain per 100 kg of a standard corn-soy diet31). Figure 5 shows the animal growth performance in body weight and feed conversion ratio (FCR), oocyst counts and lesion scores (LSs) from all treatments. As demonstrated in Fig. 5, significant differences (P < 0.05) were observed between the untreated, challenged and unchallenged control animals in performance, LSs and oocyst counts. Doses of 400 g or 1,600 g of the sdAb-expressing grain led to body weight gains (BWGs; 457 g and 453 g, respectively) and FCRs (defined as mass of feed intake divided by mass of weight gain); 1.81 and 1.70, respectively) that were both numerically and statistically similar to those of salinomycin-treated birds (BWG, 454 g; FCR, 1.80; Fig. 5a), and all were improved relative to the coccidia-challenged, untreated birds (BWG, 381 g; FCR, 1.95, P < 0.05), but only partially returned to the levels of the unchallenged birds (BWG, 524 g; FCR, 1.71). In terms of oocyst counts (Fig. 5b), neither salinomycin (4,414 oocysts per gram (OPG) of faeces) nor any of the sdAb treatments (2,730 to 4,447 OPG) appeared to reduce the number of E. maxima oocysts significantly (P< 0.05) below the challenged, untreated control birds with 3,844 OPG. However, the two higher doses of sdAb appeared to reduce oocyst counts from E. acervulina and E. tenella almost as well as salinomycin. The 1,600-g dose of AgThG11 resulted in 13,802 to 39,999 E. acervulina OPG (mean = 24,715 OPG), which could not be distinguished statistically (P > 0.05) from the salinomycin treatment, which resulted in 536 to 40,535 OPG (mean = 11,650 OPG), although both of these treatments showed significant improvement relative to the coccidia-challenged, untreated birds (18,693 to 368,768 OPG; mean = 77,962 OPG). With regard to E. tenella, results from the 400-g dose of AgThG11 (67 to 603 OPG; mean = 251 OPG) and the 1,600-g dose of AgThG11 (67 to 1,340 OPG; mean = 628 OPG) could not be distinguished (P > 0.05) from those of the salinomycin treatment (67 to 603 OPG; mean = 352 OPG). All three of these treatments showed significant improvement relative to the coccidia-challenged, untreated controls (335 to 2,948 OPG; mean = 1,189 OPG). LSs that were attributed to each of the three species of Eimeria tested in this study also showed that the sdAb-treated birds performed better than the coccidia-challenged, untreated birds (Fig. 5c). Coccidia-challenged, untreated birds showed a mean LS of 2.78 from E. acervulina, and 1.55 from E. maxima. Salinomycin appeared to control lesions of E. acervulina (mean LS = 1.35) and E. maxima (mean LS = 0.70) better than the sdAb doses tested here, from which the 100-g, 400-g and 1,600-g doses of AgThG11 resulted in mean E. acervulina LS of 2.08, 2.48 and 2.18, respectively, and E. maxima LS of 1.10, 1.38 and 1.13, respectively. However, all three of the sdAb doses appeared to control E. tenella lesions at least as well as salinomycin. Whereas the coccidia-challenged, untreated birds showed a mean LS from E. tenella of 1.13, the 1× , 4× and 16× doses of AgThG11 resulted in statistically significant (P < 0.05) reductions in mean LS from E. tenella to 0.65, 0.75 and 0.70, respectively. LSs from E. tenella following salinomycin treatment were intermediate (mean LS = 0.90) and could not be distinguished (P > 0.05) from results with either the coccidia-challenged, untreated birds or the AgThG11-treated birds.

Fig. 5: Inclusion of grain that expresses sdAb AgThG11 in the diet helps birds overcome the effects of a coccidiosis challenge.

a, Bird performance. Higher doses (4×, 16×) of AgThG11 grain improved both BWG (box and whisker plots) and adjusted FCRs (filled circles) as well as or better than salinomycin relative to untreated, challenged birds, as determined by single-factor analysis of variance (ANOVA; F = 5.52, P = 0.0008). The total pen weight was adjusted to a per-bird basis, where n = 8 replicates pens per treatment. Here and in b and c, differences between individual treatments were distinguished by the least significant difference all-pairwise comparisons test and are indicated by the labels A, B and C, wherein values that share no common letters are statistically different (P ≤ 0.05). The box and whisker plots show the second quartile (filled boxes), the third quartile (open boxes) and the range of the data (whiskers). The error bars in the FCR plots indicate standard deviations. b, The effect of treatments on oocyst counts in faeces. Single-factor ANOVA revealed that all treatments reduced oocyst counts for E. acervulina (F = 2.43, P = 0.05) and E. tenella (F = 6.56, P = 0.0002), while oocyst counts for E. maxima did not show substantial improvement with any treatment. Oocyst counts were determined from each pen, where n = 8 replicates pens per treatment; oocyst counts for each species were considered independently for statistical evaluation. The boxes and whiskers are defined as in a. c, The effect of treatments on lesion formation in the intestinal linings and caeca of treated birds. Average LSs are plotted (heavy horizontal bars) with error bars to indicate standard deviations. Single-factor ANOVA showed that all treatments reduced the incidence and severity of lesions from E. acervulina (F = 84.66, P = 0.0000), E. maxima (F = 24.46, P = 0.0000) and E. tenella (F = 11.05, P = 0.0000) in the intestines or caeca of challenged birds relative to the untreated, challenged controls. Each of the n = 40 birds per treatment was evaluated for lesions attributed to each species on the basis of their location within the intestinal tract or caeca. The number of birds that displayed lesions of a given severity (0 = no lesion, 4 = severe lesions) is reflected by the diameter of the circle shown above each treatment; for example, regarding A. cervulina lesions among birds given treatment 2, only 1 bird had an LS of 1 (mild), while 32 had LSs of 3 (more severe). The effect of salinomycin on E. tenella lesions and the effects of the 4× dose of AgThG11 on E. acervulina and E. maxima lesions were not statistically different from the untreated, challenged controls. LSs for each species were considered independently for statistical evaluation.

Source data


Despite the benefits of prophylactic antibiotic use for animal health and productivity32, antibiotic resistance remains a serious and growing problem in animal and human populations, which has led to calls for limiting antibiotic use in animal agriculture33,34. Unfortunately, Eimeria vaccines have not eliminated the problem of production losses and secondary infections35,36. Common ionophore antimicrobial therapeutics suffer from the development of resistance by Eimeria, while also preventing treated chickens from being marketed as ‘antibiotic-free’, which would be preferred by consumers and regulatory policies33.

Previous work has shown that blocking cIL-10 in Eimeria-infected chickens circumvents immune suppression, and permits IFN-γ production, which inhibits Eimeria infection and disease progression16,37. While the concept of oral delivery of anti-IL-10 antibodies was previously demonstrated to be efficacious13,17, the cost of antibody production38, and the thermal lability of full-length antibodies39, make implementation of this solution very challenging. Plants are one of the lowest cost platforms available for producing recombinant proteins21,38,40,41 and using corn allows the antibody to be delivered in poultry diets. Furthermore, by using sdAbs, which are known to have intrinsic thermal stability39, the final product should have the necessary robustness for implementation within industry feed production processes. By targeting IL-10, a host cytokine, rather than an Eimeria protein directly, the risk of antibiotic resistance developing is reduced for both Eimeria and other host pathogens that may exploit IL-10 secretion as part of their pathogenesis42,43. Recently, Virdi et al.44 demonstrated that soybean or Arabidopsis can be used to produce VHH–IgA Fc fusions that target F4 fimbriae to enable passive mucosal immunity by incorporating these ingredients into piglet feed, which prevented infection with enterotoxigenic Escherichia coli (ETEC). While this result is encouraging, the concern exists that a pathogen such as ETEC might develop resistance to the treatment through selection of altered surface epitopes, thereby preventing recognition by the antibodies. Expressing antibodies that interact with host-derived targets, rather than pathogen-derived targets, should circumvent potential problems associated with the development of antimicrobial resistance.

Previous research13 has shown that IL-10 levels are initially quite low in the intestinal lumen of young birds, but at around 7 d of age IL-10 levels increase, even in the absence of an Eimeria challenge, which suggests that IL-10 plays a role in the intestine that is unrelated to coccidiosis. This raises the question of whether suppressing IL-10 in otherwise healthy birds might have some negative effect on bird health. Addressing this question, Arendt et al.13 found that feeding anti-IL-10 antibodies to unchallenged birds had no statistically significant effects on body weights or FCRs in birds.

We have demonstrated that an infectious enteric disease can been treated with transgenic, corn-expressed antibodies. This is a significant step forward in treating Eimeria infections without antibiotics and this low-cost production platform integrates into current feed practices. The thermal stability of AgThG11 ensures the necessary robustness for use in pelleted feed when included directly in the mixer with other feed components, alleviating the need for sophisticated and expensive antibody collection and formulation work. Such production and delivery systems have the potential to be used by subsistence farmers in the developing economies, where mammalian cell culture or microbial fermentation to produce antibodies is prohibitively costly and complex for the purpose of animal production. This agricultural production system scales directly with food production, enabling the participation of farmers globally in producing antibodies. The outcome of these combined findings will ultimately impact food security, animal welfare and human health and nutrition.


sdAb discovery

sdAbs targeting cIL-10 were generated, cloned, expressed in E. coli and purified at ProSci as described elsewhere22. The relative affinity of candidate sdAbs for IL-10 was evaluated by direct ELISA, as described previously22.

Cloning, expression and purification of cIL-10r extracellular domain fused to human IgG Fc region

Sequences for the cIL-10r type 1 extracellular domain (cIL-10r sol; GenPept accession number for full receptor sequence CAJ15789.1) and the human IgG Fc domain (RCSB PDB accession number 1HZH) were used to create a fusion construct as described elsewhere24. DNA synthesis was carried out by GenScript. The gene was cloned into pGAPZαB (Invitrogen). This construct included a yeast α-mating factor secretion signal peptide and Kex2 protease cleavage site directly upstream from the Fc domain of the fusion protein. Pichia pastoris strain GS115 was transformed with the linearized plasmid and plated onto YPD with zeocin (Invitrogen). High-expressing clones were grown in a 3-l bioreactor (Infors AG CH-4103) for protein production. Starter cultures of 150 ml were inoculated and shaken for 16 h at 30 °C. Cultures were centrifuged and resuspended in 20 ml of 2× YPD. This was used to inoculate 1.5 l of BMGY medium (Invitrogen) supplemented with 1% casamino acids. The culture was maintained at 28 °C and a \(p_{\rm O_{2}}\) of 30% until the batch phase ended, as indicated by a sudden rise in dissolved oxygen. Feed medium (750 ml of 10% glucose, 5% peptone, 2.5% yeast extract, 1% casamino acids, 0.1% zeocin and 0.5% antifoam 204) was added over 18 h, with temperature and \(p_{\rm O_{2}}\) maintained at 20 °C and 30%, respectively. After a rise in \(p_{\rm O_{2}}\) indicated that the medium had been consumed, the culture was collected, centrifuged and filtered.

Half of the culture supernatant was adjusted to 0.5 M ammonium sulfate by the addition of 0.2 volume of 3 M ammonium sulfate, 20 mM Tris-HCl, 1 mM EDTA, pH 8.0. After 20 min of stirring on ice, the suspension was filtered and loaded onto 5 × 5-ml HiTrap Phenyl FF high-substitution columns (GE Healthcare) connected in series, using an Akta Purifier FPLC system. The column was washed with 6 column volumes (CV) of 0.5 M ammonium sulfate, 20 mM Tris-HCl, 1 mM EDTA, pH 8.0. The fusion protein was eluted with a gradient from 0.5 M to 0 M buffered ammonium sulfate over 5 CV, followed by 3 CV of buffer and finally 2 CV of water; 14-ml fractions were collected. This procedure was then repeated with the other half of the culture supernatant.

Fractions containing the fusion protein were pooled and concentrated to 2.5 ml using Amicon Ultra-15 centrifugal concentrators (3-kDa cutoff). A PD-10 desalting column (GE Healthcare) was used to exchange the buffer with 3.5 ml of 20 mM Tris-HCl, pH 7.0. This was loaded onto a 1-ml Protein G HiTrap column (GE Healthcare) using a syringe. After washing with 5 ml of buffer, the fusion protein was eluted with 7 ml of 0.1 M glycine HCl, pH 2.7 and collected in 1-ml fractions into tubes, each of which contained 200 μl of 1 M Tris-HCl, pH 9.0.

Fractions were pooled, concentrated to 1 ml using centrifugal concentrators, exchanged into 1.5 ml of 20 mM Tris-HCl, 1 mM EDTA, pH 8.0 using a G-25 minitrap desalting column (GE Healthcare), and loaded onto a 1-ml HiTrap MonoQ column (GE Healthcare). The column was washed with 10 CV of buffer, followed by elution over 30 CV from 0–30% buffer B (20 mM Tris-HCl, 1 M NaCl, 1 mM EDTA, pH 8.0), 10 CV from 30–100% buffer B, and held at 100% buffer B for 5 CV. Fractions of 2 ml in volume were collected.

Fractions were pooled and concentrated to approximately 1 ml. Aliquots of approximately 250 μl were loaded onto a 120-ml S-200 column (GE Life Sciences) equilibrated in 100 mM MES, 300 mM NaCl, pH 6.3. Fractions of 1 ml in volume were pooled, concentrated to approximately 400 µl and filtered through a sterile 0.2-µm syringe filter. One volume of sterile 80% (v/v) glycerol was added, and the solution was stored at −20 °C. The concentration was determined using an extinction coefficient of 1.62 ml mg−1 (82,910 M-1 monomer). Approximately 0.3 mg of protein was obtained.

SPR analysis of cIL-10r binding and inhibition by anti-cIL-10 VHH domains

SPR data were obtained at the University of Massachusetts Biophysical Characterization Facility. A CM5 chip (GE Healthcare) was activated for amine coupling using the manufacturer’s instructions on a Biacore T200 instrument. cIL-10r sol–hFc at a concentration of 0.35 mg ml−1 was diluted tenfold in 10 mM sodium acetate, 0.5 mM EDTA, pH 4.0, and 7,440 resonance units (RU) was bound to the chip. All SPR experiments were conducted at 37 °C.

Determination of K d for cIL-10 binding to cIL-10r sol

cIL-10 (Kingfisher Biotech) was dissolved in HBS-EP buffer (0.01 M HEPES pH 7.4, 0.15 M NaCl, 3 mM EDTA, 0.005% v/v Surfactant P20) at several concentrations from 0 to 300 nM. cIL-10 was injected onto the chip at 20 µl min−1 for 120 s, followed by a 300-s dissociation phase. The chip was regenerated between runs with a 10-s pulse of 10 mM sodium acetate, 0.5 mM EDTA, pH 4.0 at 20 µl min−1, followed by a 2-min re-equilibration with HBS-EP. All concentrations were assayed in duplicate. The RUmax values were plotted against [cIL-10] and fitted to a rectangular hyperbolic equation to obtain the Kd (42 + 12 nM). This is in good agreement with the literature value of 30 nM for in vitro binding of a recombinant human IL-10 to the soluble portion of IL-10r45.

Determination of IC50 values for anti-cIL-10 VHH domains

VHH domains at 0 to 1,000 nM were mixed with cIL-10 at 10 nM and injected onto the chip as described above. RUmax values were calculated and plotted against VHH concentration to obtain IC50 values. Each concentration was analysed in duplicate. VHH domains in the absence of cIL-10 showed no detectable binding to the chip at concentrations as high as 1,000 nM.

Cell-based assay

Assays with primary chicken splenocytes were carried out at Marin Biologic Laboratories, using methods described previously15. Splenocytes were treated with ConA (5 µg ml−1) and cIL-10 (1.5 ng ml−1) before adding the sdAbs.

Construction of vectors for maize transformation

The AgThG11 sdAb DNA sequence was codon-optimized for protein expression in maize and fused at its 5′ end with nucleotide sequences encoding the signal peptide of the 27 kDa maize gamma-zein and at its 3′ end with the KDEL-encoding sequence for recombinant protein retention in the endoplasmic reticulum. The final AgThG11 DNA fragment was synthesized by GenScript and cloned between either the maize 27 kDa gamma-zein or the rice glutelin B-4 gene promoters and Nos terminator sequences for developing two plant expression cassettes, which were subsequently cloned into the T-DNA part of the pAG4988 vector. The pAG4988 vector is a derivative of the pSB11 vector and is capable of recombining with the pSB1 vector in triparental mating in Agrobacterium tumefaciens strain LBA440446,47. Immature maize embryos from Hi-II A×B were transformed using the resulting Agrobacterium strain as described previously24. The manA gene encoding the enzyme phosphomannose isomerase and residing on the same T-DNA was used as a marker for selecting primary transformants in tissue culture26. The presence of the AgThG11 gene in transgenic maize events was confirmed by PCR. Two separate forward primers, OsGluB4P_F3 (5′-TCAACCAGCCCAAGTTTCCA-3′) and ZmZ27P_F2 (5′-GCACTTCTCCATCACCACCA-3′), were used in combination with the common reverse primer sdAb-Rminus6 (5′-TCCTGCAGCTGAACCTGGGAG-3′) to amplify 125-base-pair (bp) or 166-bp DNA fragments respectively from each of the two integrated AgThG11 expression cassettes. The endogenous maize gene actin 2 served as a PCR control for amplification of a 197-bp genomic DNA fragment using the primers Act2-L4 (5′-AGCTGTTGGCCATGGAGCATTGTGCA-3′) and Act2-R4 (5′-GCGACAGCCGACACCCTTGCGT-3′). Plants were grown in a greenhouse using a 14-h light and 10-h dark schedule with the temperature set at 82 °F (28 °C) during the light period, and 70 °F (21 °C) during the dark period. Plants were self-pollinated to generate the harvested grain used in the trial.

Western blotting

Endosperm samples were chipped with a blade from individual kernels and genomic DNA was extracted from endosperm to verify the presence of the transgene by PCR. Transgene-positive kernels were pulverized to a fine powder in a FastPrep-24 Grinder (MP Biomedicals) and 20–30 mg seed powder was processed for total protein isolation using 27 mM sodium carbonate/3 mM sodium bicarbonate buffer, pH 10.8. The extracted total seed protein (20 µg per lane) was resolved on a 12% polyacrylamide gel and transferred onto a PVDF membrane (Bio-Rad) according to the manufacturer’s instructions. The membrane-bound AgThG11 was detected with HRP-conjugated rabbit monoclonal anti-camelid VHH antibody (GenScript) at 1:10,000 dilution.

Thermal stability assay

The effects of temperature exposure on the binding affinity of microbially produced AgThG11 sdAb or a conventional IgG polyclonal anti-cIL-10 antibody were evaluated by ELISA. Both antibodies were used at equal molar amounts in 1× PBS + 0.01% BSA and all heat treatments were performed in PCR thermocyclers. After heat treatment, antibodies were allowed to renature for 20 min at room temperature before being placed on ice for use in ELISA. Corning Costar 3361 high-binding 96-well plates were coated overnight at 4 °C with 5 µg ml−1 NeutrAvidin (ThermoFisher) that was dissolved in 50 mM carbonate/bicarbonate buffer pH 9.6. The plates were washed four times in 1× PBS + 0.05% Tween-20 and blocked for 2 h with 1× PBS + 2% dried milk. The recombinant cIL-10 protein (IBI Scientific) was biotinylated using a Lightning-Link Rapid Biotinylation Kit (Expedeon) according to the manufacturer’s instructions and biotinylated protein at a concentration of 400–600 ng ml−1 was applied to the NeutrAvidin-coated ELISA plates. After 1 h incubation, the plates were washed as above and either the sdAb AgThG11 or a conventional non-conjugated anti-cIL-10 rabbit IgG pAb (Kingfisher Biotech) was added to the plates. The bound cIL-10 antibodies were probed with HRP-conjugated rabbit monoclonal anti-camelid VHH (GenScript) antibody or sheep anti-rabbit IgG polyclonal (ThermoFisher) antibody, respectively. The HRP signal detection was performed with TMB substrate and optical density values at 450 nm were recorded on a Biotek Synergy HTX plate reader after terminating reactions with 2 M H2SO4.

Feeding trial with coccidiosis challenge

Feeding trials were carried out at Southern Poultry Feed and Research (Athens GA). Methods for animal care, Eimeria challenge, performance measurements, oocyst counting and lesion scoring were as described previously48. Each treatment was conducted on 8 randomly selected cages, each containing 8 randomly placed chickens for a total of 64 animals per treatment. Specific treatment feed was provided to the designated animals ad libitum starting at day 0 and continuing to the conclusion of the study (day 20). Challenged animals received a mixture of E. acervulina (100,000 oocysts per animal), E. maxima (50,000 oocysts per animal) and E. tenella (75,000 oocysts per animal) by oral gavage in 1 ml of distilled water on day 14. Unchallenged animals received 1 ml of distilled water on day 14 with no inoculum. Starting at 120-h post-exposure, until 144-h post-exposure, samples of the faeces passed during this period were collected for oocyst counts. Birds and feed were weighed on days 0, 14 and 20. Lesion scoring was conducted on five birds from each pen on day 20. Mean values for pen or cage weight gain, feed consumption, feed conversion (adjusted for mortality: feed consumed/final live weight + mortality weight), coccidial LSs and mortality were calculated. Statistical evaluation of the data was performed at Southern Poultry Feed and Research using a STATISTIX for Windows program (Analytical Software). The procedures used were general linear procedures using ANOVA with a comparison of means using Tukey’s least significant difference (t-test) at a significance level of P ≤ 0.05.

Reporting Summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.

Data availability

Source data for Figs. 1, 3 and 5 are provided with the paper. Data from these studies are available through Figshare ( at


  1. 1.

    Livestock and Poultry: World Markets and Trade (USDA, 2019);

  2. 2.

    Ritchie, H. & Roser, M. Meat and Seafood Production & Consumption (Our World In Data, accessed 2017);

  3. 3.

    Eshel, G., Shepon, A., Makov, T. & Milo, R. Land, irrigation water, greenhouse gas, and reactive nitrogen burdens of meat, eggs, and dairy production in the United States. Proc. Natl Acad. Sci. USA 111, 11996–12001 (2014).

    ADS  CAS  PubMed  Google Scholar 

  4. 4.

    Laurent, F., Mancassola, R., Lacroix, S., Menezes, R. & Naciri, M. Analysis of chicken mucosal immune response to Eimeria tenella and Eimeria maxima infection by quantitative reverse transcription-PCR. Infect. Immun. 69, 2527–2534 (2001).

    CAS  PubMed  PubMed Central  Google Scholar 

  5. 5.

    Jones, P. J., Niemi, J., Christensen, J.-P., Tranter, R. B. & Bennett, R. M. A review of the financial impact of production diseases in poultry production systems. Anim. Prod. Sci. 59, 1585–1597 (2018).

    Google Scholar 

  6. 6.

    Dalloul, R. A. & Lillehoj, H. S. Poultry coccidiosis: recent advancements in control measures and vaccine development. Expert Rev. Vaccines 5, 143–163 (2006).

    CAS  PubMed  Google Scholar 

  7. 7.

    Noack, S., Chapman, H. D. & Selzer, P. M. Anticoccidial drugs of the livestock industry. Parasitol. Res. 118, 2009–2026 (2019).

    PubMed  PubMed Central  Google Scholar 

  8. 8.

    Tensa, L. R. & Jordan, B. J. Comparison of the application parameters of coccidia vaccines by gel and spray. Poult. Sci. 98, 634–641 (2019).

    PubMed  Google Scholar 

  9. 9.

    Price, K. R., Hafeez, M. A., Bulfon, J. & Barta, J. R. Live Eimeria vaccination success in the face of artificial non-uniform vaccine administration in conventionally reared pullets. Avian Pathol. 45, 82–93 (2016).

    CAS  PubMed  Google Scholar 

  10. 10.

    Blake, D. P. & Tomley, F. M. Securing poultry production from the ever-present Eimeria challenge. Trends Parasitol. 30, 12–19 (2014).

    PubMed  Google Scholar 

  11. 11.

    Hermans, P. G., Fradkin, D., Muchnik, I. B. & Morgan, K. L. Prevalence of wet litter and the associated risk factors in broiler flocks in the United Kingdom. Vet. Rec. 158, 615–622 (2006).

    CAS  PubMed  Google Scholar 

  12. 12.

    Fatoba, A. J. & Adeleke, M. A. Diagnosis and control of chicken coccidiosis: a recent update. J. Parasit. Dis. 42, 483–493 (2018).

    PubMed  PubMed Central  Google Scholar 

  13. 13.

    Arendt, M. K., Sand, J. M., Marcone, T. M. & Cook, M. E. Interleukin-10 neutralizing antibody for detection of intestinal luminal levels and as a dietary additive in Eimeria challenged broiler chicks. Poult. Sci. 95, 430–438 (2016).

    CAS  PubMed  PubMed Central  Google Scholar 

  14. 14.

    Hong, Y. H., Lillehoj, H. S., Lillehoj, E. P. & Lee, S. H. Changes in immune-related gene expression and intestinal lymphocyte subpopulations following Eimeria maxima infection of chickens. Vet. Immunol. Immunopathol. 114, 259–272 (2006).

    CAS  PubMed  Google Scholar 

  15. 15.

    Rothwell, L. et al. Cloning and characterization of chicken IL-10 and its role in the immune response to Eimeria maxima. J Immunol. 173, 2675–2682 (2004).

    CAS  PubMed  Google Scholar 

  16. 16.

    Wu, Z. et al. Analysis of the function of IL-10 in chickens using specific neutralising antibodies and a sensitive capture ELISA. Dev. Comp. Immunol. 63, 206–212 (2016).

    CAS  PubMed  PubMed Central  Google Scholar 

  17. 17.

    Sand, J. M., Arendt, M. K., Repasy, A., Deniz, G. & Cook, M. E. Oral antibody to interleukin-10 reduces growth rate depression due to Eimeria spp. infection in broiler chickens. Poult. Sci. 95, 439–446 (2016).

    CAS  PubMed  PubMed Central  Google Scholar 

  18. 18.

    He, Z., Tong, C., Sheng, L., Ma, M. & Cai, Z. Monitoring glycation-induced structural and biofunctional changes in chicken immunoglobulin Y by different monosaccharides. Poult. Sci. 95, 2715–2723 (2016).

    CAS  PubMed  Google Scholar 

  19. 19.

    Mizukami, M. et al. Highly efficient production of VHH antibody fragments in Brevibacillus choshinensis expression system. Protein Expr. Purif. 105, 23–32 (2015).

    CAS  PubMed  Google Scholar 

  20. 20.

    Egelkrout, E. et al. Enhanced expression levels of cellulase enzymes using multiple transcription units. BioEnergy Res. 6, 699–710 (2012).

    Google Scholar 

  21. 21.

    Tschofen, M., Knopp, D., Hood, E. & Stöger, E. Plant molecular farming: much more than medicines. Annu. Rev. Anal. Chem. 9, 271–294 (2016).

    Google Scholar 

  22. 22.

    Pardon, E. et al. A general protocol for the generation of Nanobodies for structural biology. Nat. Protoc. 9, 674–693 (2014).

    CAS  PubMed  PubMed Central  Google Scholar 

  23. 23.

    Zdanov, A. et al. Crystal structure of interleukin-10 reveals the functional dimer with an unexpected topological similarity to interferon γ. Structure 3, 591–601 (1995).

    CAS  PubMed  Google Scholar 

  24. 24.

    Terai, M. et al. Human interleukin 10 receptor 1/IgG1-Fc fusion proteins: Immunoadhesins for human IL-10 with therapeutic potential. Cancer Immunol. Immunother. 58, 1307–1317 (2009).

    CAS  PubMed  Google Scholar 

  25. 25.

    Shouval, D. S. et al. Interleukin 10 receptor signaling: master regulator of intestinal mucosal homeostasis in mice and humans. Adv. Immunol. 122, 177–210 (2014).

  26. 26.

    Negrotto, D., Jolley, M., Beer, S., Wenck, A. R. & Hansen, G. The use of phosphomannose isomerase as a selectable marker to recover transgenic maize plants (Zea mays L.) via Agrobacterium transformation. Plant Cell Rep. 19, 798–803 (2000).

    CAS  PubMed  Google Scholar 

  27. 27.

    Reina, M., Guillén, P., Ponte, I., Boronat, A. & Palau, J. DNA sequence of the gene encoding the Zc1 protein from Zea mays W64 A. Nucleic Acids Res. 18, 6425 (1990).

    CAS  PubMed  PubMed Central  Google Scholar 

  28. 28.

    Qu, L. Q. & Takaiwa, F. Evaluation of tissue specificity and expression strength of rice seed component gene promoters in transgenic rice. Plant Biotechnol. J. 2, 113–125 (2004).

    CAS  Google Scholar 

  29. 29.

    Prat, S., Cortadas, J., Puigdomrnech, P. & Palau, J. Nucleic acid (cDNA) and amino acid sequences of the maize endosperm protein glutelin-2. Nucleic Acids Res. 13, 1493–1504 (1985).

    CAS  PubMed  PubMed Central  Google Scholar 

  30. 30.

    Semenza, J. C., Hardwick, K. G., Dean, N. & Pelham, H. R. B. ERD2, a yeast gene required for the receptor-mediated retrieval of luminal ER proteins from the secretory pathway. Cell 61, 1349–1357 (1990).

    CAS  PubMed  Google Scholar 

  31. 31.

    Broiler Performance and Nutrition Supplement (Cobb-Vantress, 2015).

  32. 32.

    Angelakis, E. Weight gain by gut microbiota manipulation in productive animals. Microb. Pathog. 106, 162–170 (2017).

    PubMed  Google Scholar 

  33. 33.

    Ben Lagha, A., Haas, B., Gottschalk, M. & Grenier, D. Antimicrobial potential of bacteriocins in poultry and swine production. Vet. Res. 48, 1–12 (2017).

    Google Scholar 

  34. 34.

    Roth, N. et al. The application of antibiotics in broiler production and the resulting antibiotic resistance in Escherichia coli: a global overview. Poult. Sci. 98, 1791–1804 (2019).

    PubMed  Google Scholar 

  35. 35.

    Kadykalo, S. et al. The value of anticoccidials for sustainable global poultry production. Int. J. Antimicrob. Agents 51, 304–310 (2018).

    CAS  PubMed  Google Scholar 

  36. 36.

    Mathis, G. et al. Effect of lasalocid or salinomycin administration on performance and immunity following coccidia vaccination of commercial broilers 1. J. Appl. Poult. Res. 23, 577–585 (2014).

    CAS  Google Scholar 

  37. 37.

    Lillehoj, H. S. et al. A recombinant Eimeria protein inducing interferon-γ production: comparison of different gene expression systems and immunization strategies for vaccination against coccidiosis. Avian Dis. 44, 379–389 (2017).

    Google Scholar 

  38. 38.

    Yusibov, V., Kushnir, N. & Streatfield, S. J. Antibody production in plants and green algae. Annu. Rev. Plant Biol. 67, 669–701 (2016).

    CAS  PubMed  Google Scholar 

  39. 39.

    Wang, S. et al. Increasing stability of antibody via antibody engineering: stability engineering on an anti-hVEGF. Proteins 82, 2620–2630 (2014).

    CAS  PubMed  Google Scholar 

  40. 40.

    Kusnadi, A. R., Nikolov, Z. L. & Howard, J. A. Production of recombinant proteins in transgenic plants: practical considerations. Biotechnol. Bioeng. 56, 473–484 (1997).

    CAS  PubMed  Google Scholar 

  41. 41.

    Smith, M. D. & Glick, B. R. The production of antibodies in plants: an idea whose time has come? Biotechnol. Adv. 18, 85–89 (2000).

    CAS  PubMed  Google Scholar 

  42. 42.

    Lee, Y., Kim, W. H., Lee, S. J. & Lillehoj, H. S. Detection of chicken interleukin-10 production in intestinal epithelial cells and necrotic enteritis induced by Clostridium perfringens using capture ELISA. Vet. Immunol. Immunopathol. 204, 52–58 (2018).

    CAS  PubMed  Google Scholar 

  43. 43.

    Zhang, R. et al. Mutual interactions of the apicomplexan parasites Toxoplasma gondii and Eimeria tenella with cultured poultry macrophages. Parasites Vectors 11, 1–12 (2018).

    Google Scholar 

  44. 44.

    Virdi, V. et al. Yeast-secreted, dried and food-admixed monomeric IgA prevents gastrointestinal infection in a piglet model. Nat. Biotechnol. 37, 527–530 (2019).

    CAS  PubMed  PubMed Central  Google Scholar 

  45. 45.

    Josephson, K. et al. Design and analysis of an engineered human interleukin-10 monomer. J. Biol. Chem. 275, 13552–13557 (2000).

    CAS  PubMed  Google Scholar 

  46. 46.

    Komari, T., Hiei, Y., Saito, Y., Murai, N. & Kumashiro, T. Vectors carrying two separate T-DNAs for co-transformation of higher plants mediated by Agrobacterium tumefaciens and segregation of transformants free from selection markers. Plant J. 10, 165–174 (1996).

    CAS  PubMed  Google Scholar 

  47. 47.

    Ishida, Y., Saito, H., Ohta, S., Hiei, Y. & Komari, T. High efficiency transformation of maize (Zea mays L.) mediated by Agrobacterium tumefaciens. Nature 14, 745–750 (1996).

    CAS  Google Scholar 

  48. 48.

    Abdelrahman, W. et al. Comparative evaluation of probiotic and salinomycin effects on performance and coccidiosis control in broiler chickens. Poult. Sci. 93, 3002–3008 (2014).

    MathSciNet  CAS  PubMed  Google Scholar 

Download references


We acknowledge the contributions of the following scientists: A. Lee (ProSci); K. Wilhelmsen (Marin Biologic); L. Bartlett (University of Massachusetts Biophysical Characterization Facility); B. Lumpkins and G. Mathis (Southern Poultry Feed and Research).

Author information




P.A.L. led the antibody discovery programme, led the cell-based assay development and performance, managed analytical development, helped transplant, water and harvest plants, milled transgenic grain for use in the feeding studies, and reviewed and edited the manuscript. M.P. characterized the antibodies in vitro and in the cell-based assay, which required microbial antibody and IL-10r soluble-domain production, performed Biacore experiments, performed western blots and reviewed the manuscript. O.B. conducted all gene design and synthesis, constructed transformation vectors, helped genotype transgenic plants, developed the ELISA assay, performed ELISAs, measured the thermal stability of the antibodies, and reviewed and edited the manuscript. B.S. conducted western blot analysis, helped develop the ELISA assay, performed ELISAs and reviewed the manuscript. V.S. conducted plant transformation and plant tissue culture, generated all transgenic events, managed and conducted plant production, and reviewed the manuscript. J.B. led the animal challenge studies and reviewed the manuscript. X.L. helped transplant, water and harvest plants, conducted assays, and reviewed and edited the manuscript. R.M.R. developed the overall research plan and strategy, managed and coordinated studies between groups, prioritized antibodies for development, helped transplant, water and harvest plants, drafted the initial paper, and reviewed and approved the final manuscript.

Corresponding author

Correspondence to R. Michael Raab.

Ethics declarations

Competing interests

The authors declare the following competing interests: employment at Agrivida, Inc.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Supplementary Figs. 1 and 2.

Reporting Summary

Source data

Source Data Fig. 1

Numerical data for the graphs.

Source Data Fig. 3

Numerical data for the graphs.

Source Data Fig. 5

Numerical data for the graphs.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Lessard, P.A., Parker, M., Bougri, O. et al. Improved performance of Eimeria-infected chickens fed corn expressing a single-domain antibody against interleukin-10. Nat Food 1, 119–126 (2020).

Download citation

Further reading


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing