The effects of graded levels of quillaja saponins on the productivity of commercial laying hens Authors: K.W. Bafundo, M. Alberto-Tempra, L. Ferreira, B. Dehaeck, W. Schelstraete Published: 2026 ABSTRACT A study was conducted to determine the effects of feeding graded levels of quillaja saponins (Clarity Q®, Huvepharma, Antwerp, Belgium) on the performance and productivity of commercial layers. Three thousand H&N layer pullets were given one of 4 levels (0, 125, 250 and 500 mg/kg) of quillaja saponins in feed and their growth performance was monitored throughout a 5-to-15-week rearing period in a completely randomized design. At 16 weeks, pullets were moved to a laying facility where they continued to consume the same treatment feeds through 60 weeks of age. At 21 weeks, collection of egg production data began, with hen day egg production (HDEP), egg weight (EW), egg mass (EM), feed intake and feed efficiency (g feed consumed/g of egg) used as primary response variables. Data were analyzed for 2 production periods: 21 to 40 and 41 to 60 weeks of age. Additionally, eggshell thickness was measured biweekly from weeks 39 to 59 in 40 randomly selected eggs per treatment. During the rearing period, quillaja saponins significantly increased weight gain per week and improved feed conversion values (p < 0.05). Analysis of hen-day egg production (HDEP) at 40 weeks of age demonstrated that quillaja addition at 500 mg/kg significantly improved HDEP compared to the 0 and 125 mg/kg levels. EW, EM and feed efficiency were significantly improved by both 250 and 500 mg/kg compared to the lower quillaja levels, but significant differences between 250 and 500 mg/kg were not observed. By 60 weeks of age, HDEP was maximized by quillaja addition at 250 mg/kg, and EW, EM and feed conversion plateaued such that significant differences between 250 and 500 mg/kg were not observed. Eggshell thickness responded similarly, with significant differences observed at 250 and 500 mg/kg compared to the other feeding levels. Data from this study data indicate that egg production variables were positively affected by quillaja additions to feed at both 21 to 40 and 41 to 60 weeks of age and 250 mg/kg produced the most consistent responses in egg production criteria. Introduction Saponins are natural products of plant origin that are known to influence many biological processes and cell types (Cheeke, 2000). These effects are associated with molecular structures that contain both water soluble and fat soluble moieties, imparting detergent-like characteristics to these molecules. Previous research has shown that saponins affect serum cholesterol, reduce ammonia production, and modify the integrity of biological membranes (Francis et al., 2002; Augustin et al., 2011; Fleck et al., 2019). Their effects on membrane integrity have been shown to reduce bacterial replication (Fink and Filip, 2022) and diminish the effects of certain protozoal diseases (Augustin et al., 2011; Rocha et al., 2012). Recent research has focused on the effects of saponin supplementation in poultry production and studies now confirm that better performance, lower mortality rates, and improved immune responses are often observed when disease-challenged broilers are exposed to saponin feeding (Bafundo et al., 2021a and b; Blue et al., 2023; Saddoris-Clemons et al., 2024). Moreover, studies by Bafundo et al. (2021a) and Stefanello et al. (2022) demonstrate that the presence of intestinal disease is not a requisite for improved growth to occur. Their work shows that improved performance is closely tied to saponin-induced improvements in intestinal morphometrics and better nutrient digestibility. When these factors are coupled with other studies demonstrating that intestinal barrier function was enhanced by saponin administration (Blue et al., 2023; Saddoris-Clemons et al., 2024), it is clear that saponins exert a positive effect on the intestinal tract. At this juncture, research on the effects of saponins in poultry has largely focused on the broiler chicken, and the studies noted above illustrate benefits that are of significance in broiler production. However, few studies of this type have focused on laying fowl, with the work of Souza et al. (2023) in layer breeders, the only recent publication in egg-producing birds. Of importance in their study were improvements in feed efficiency (kg feed/kg egg) and eggshell characteristics that underscore the potential that saponins hold when administered to laying hens. To address the information deficit in laying fowl, our experiment was conducted to expand our understanding of the potential benefits of quillaja administration in commercial layers. Consequently, our objectives were to evaluate the effects of quillaja saponins on growth and egg production variables through the majority of the layer’s production life. To this end, feed containing quillaja saponins was fed continuously through 60 weeks of age. Materials and methods The trial was conducted at Poulpharm Poultry Research, Batangas, The Philippines. The site is a commercial research facility specifically designed for the evaluation of feeding and nutritional programs in pullets and laying hens. All procedures and methods used in the study were approved by the Ethics Committee of Poulpharm, bv. Birds, treatments and feeding logistics Three thousand one-day-old H&N Super Nick White pullets were acquired from a commercial hatchery, and upon receipt were housed in a floor pen facility divided into four brooding areas with each area accommodating 750 birds. The environmental and management conditions for each of these areas were identical. The birds were fed diets that contained one of four levels of quillaja saponins: 0, 125, 250 and 500 mg/kg of diet (administered as a proprietary, standardized quillaja saponin extract, Clarity Q®, Huvepharma, Antwerp, Belgium). The feeding plan employed in this trial was similar to the breeder’s recommendations for H&N Super Nick White layers. A starter feed was administered from placement through the fifth week, grower feeds from 6 through 10 weeks, developer feeds from 11 through 15 weeks, and pre-lay feeds for weeks 16 and 17. Layer feeds were introduced during the 18th week and were supplied until the conclusion of the study (60 weeks of age). All feeds were designed to meet the nutrient requirements recommended for H&N Super Nick White laying hens (Table 1). The integrity of the quillaja saponin treatment levels was maintained for the duration of the trial, and care was taken to ensure feeding levels were consistent during feed changes (grower, developer, pre-lay feeds, etc.) and during movement from the grow out to the laying house. A proprietary feed assay, specifically developed by the manufacturer for this quillaja-based product, was used to determine quillaja levels in feed. Samples were taken at feed manufacturing and immediately sent to the sponsoring company for assay. Average results for these assays were: 104.0, 256.5 and 514.0 mg/kg, respectively, for the 125, 250 and 500 mg/kg inclusion levels. Quillaja saponins were not detected in any control feeds.Table 1Composition and calculated nutrient analysis of the diets used from placement through 60 weeks-of-age.IngredientKg per ton feedPullet StarterPullet GrowerPullet DeveloperPre-LayLayerWeeks 0-4Weeks 5-11Weeks 12-15Weeks 16-17Weeks 18-60Corn576567587560630Soybean meal216200119231215Soy Protein Concentrate3530---Pollard10011521085-Copra meal3050505035Mono-Dicalcium Phosphate433614Coconut Oil16105810Limestone, fine1214173240Limestone, coarse---2045NaCl33334Choline Cl0.350.350.30.30.5Vitamin Premix10.250.250.250.250.25Mineral Premix20.750.750.750.750.75DL-Methionine1.41.60.91.21.6L-Lysine1.30.70.45-0.2L-Threonine-0.60.4--L-Tryptophan-0.3---Acidifier0.50.50.50.51Toxin Binder22222Anticoccidial0.50.5---Protease0.250.250.250.250.25Phytase0.10.10.10.10.06Mold Inhibitor0.50.50.50.50.5Antioxidant0.1250.1250.1250.1250.125TOTAL10001000100010001000Nutrient composition/kgME, kcal/kg29502850275028002800Crude protein %19.018.015.517.517.0Fat, %3.73.84.04.23.7Fiber, %3.63.84.03.83.3Ash, %5.45.56.08.911.8Ca, %1.01.00.92.03.5Av. P., %0.480.450.370.400.45Na, %0.180.170.160.160.16Digestible, %Lys0.980.820.540.700.77Met0.470.330.250.350.43Met+Cys0.980.820.540.700.67Thr0.650.580.380.490.54Trp0.190.200.130.150.16Arg1.141.200.570.731.01Val0.820.800.430.630.72Ile0.710.680.410.560.611The vitamin premix contained per kg: 55,000 IU vitamin A; 13,000 IU vitamin D; 150 g vitamin E; 10 g vitamin K; 10 g vitamin B1; 30 g vitamin B2; 200 g vitamin B6; 125 mg vitamin B12; 200 g niacin; 5 g folic acid; 60 g pantothenic acid and 750 mg biotin. 2The mineral premix contained per kg: 100 g Fe; 10 g Cu; 75 g Zn; 100 g Mn; 1 g I and 0.12 g Se. Table 1Composition and calculated nutrient analysis of the diets used from placement through 60 weeks-of-age.IngredientKg per ton feedPullet StarterPullet GrowerPullet DeveloperPre-LayLayerWeeks 0-4Weeks 5-11Weeks 12-15Weeks 16-17Weeks 18-60Corn576567587560630Soybean meal216200119231215Soy Protein Concentrate3530---Pollard10011521085-Copra meal3050505035Mono-Dicalcium Phosphate433614Coconut Oil16105810Limestone, fine1214173240Limestone, coarse---2045NaCl33334Choline Cl0.350.350.30.30.5Vitamin Premix10.250.250.250.250.25Mineral Premix20.750.750.750.750.75DL-Methionine1.41.60.91.21.6L-Lysine1.30.70.45-0.2L-Threonine-0.60.4--L-Tryptophan-0.3---Acidifier0.50.50.50.51Toxin Binder22222Anticoccidial0.50.5---Protease0.250.250.250.250.25Phytase0.10.10.10.10.06Mold Inhibitor0.50.50.50.50.5Antioxidant0.1250.1250.1250.1250.125TOTAL10001000100010001000Nutrient composition/kgME, kcal/kg29502850275028002800Crude protein %19.018.015.517.517.0Fat, %3.73.84.04.23.7Fiber, %3.63.84.03.83.3Ash, %5.45.56.08.911.8Ca, %1.01.00.92.03.5Av. P., %0.480.450.370.400.45Na, %0.180.170.160.160.16Digestible, %Lys0.980.820.540.700.77Met0.470.330.250.350.43Met+Cys0.980.820.540.700.67Thr0.650.580.380.490.54Trp0.190.200.130.150.16Arg1.141.200.570.731.01Val0.820.800.430.630.72Ile0.710.680.410.560.611The vitamin premix contained per kg: 55,000 IU vitamin A; 13,000 IU vitamin D; 150 g vitamin E; 10 g vitamin K; 10 g vitamin B1; 30 g vitamin B2; 200 g vitamin B6; 125 mg vitamin B12; 200 g niacin; 5 g folic acid; 60 g pantothenic acid and 750 mg biotin. 2The mineral premix contained per kg: 100 g Fe; 10 g Cu; 75 g Zn; 100 g Mn; 1 g I and 0.12 g Se. Composition and calculated nutrient analysis of the diets used from placement through 60 weeks-of-age. The vitamin premix contained per kg: 55,000 IU vitamin A; 13,000 IU vitamin D; 150 g vitamin E; 10 g vitamin K; 10 g vitamin B1; 30 g vitamin B2; 200 g vitamin B6; 125 mg vitamin B12; 200 g niacin; 5 g folic acid; 60 g pantothenic acid and 750 mg biotin. 2The mineral premix contained per kg: 100 g Fe; 10 g Cu; 75 g Zn; 100 g Mn; 1 g I and 0.12 g Se. To evaluate the effects of the quillaja treatments during rearing, the pullets were weighed at the start of the 5th week and randomized to one of 10 replicates of 75 birds each in a completely randomized design. This process ensured similar starting weights for each replicate. Birds were allowed ad libitum access to their respective feeds and water for the duration of the rearing period (5 to 15 weeks). Feed consumption was determined weekly and mortality was recorded daily. At 15 weeks of age, birds were weighed and feed consumption for the entire period was determined. Bird performance, feed intake, percent uniformity and total mortality were analyzed for the entire rearing period. At 16 weeks of age, birds were moved from the rearing pens to the laying the facility. Prior to transport, each pullet was weighed and tagged according to treatment. At the layer farm, the pullets were physically assessed so that unthrifty and unhealthy birds were removed from the study. Weights were balanced during allocation to cages. During the laying portion of the study, each of the four quillaja treatments was replicated 8 times. Each replicate contained 21 cages housing 4 hens per cage. In total, each replicate involved 84 hens. After moving, birds were allowed to acclimate to the environment from 16 to 19 weeks of age. Standard commercial laying rations (Table 1) containing the appropriate level of quillaja saponins were administered at 18 weeks of age and the collection of egg production data began at the start of week 21. The egg production portion of the trial was conducted until the birds reached 60 weeks of age, with egg production criteria evaluated over two 20-week production periods (21 to 40 and 41 to 60 weeks of age). During the laying period, percent hen-day egg production (HDEP, the number of eggs produced daily/total number of hens), egg weight (EW), and egg mass (EM = HDEP x mean EW) were measured/calculated weekly. Mortality was recorded daily. Feed consumption was determined daily and calculations of feed efficiency per gram of egg (feed efficiency = g feed consumed/g of egg produced) were made on a weekly basis. For statistical purposes, the statistical unit was the weekly average for a respective replicate. Measurements of eggshell thickness were performed every other week, beginning in week 39 and continuing through week 59. At each sampling time, 5 of 8 replicates per treatment were randomly sampled, such that 8 eggs from each replicate (40 eggs per treatment per sampling time) were randomly selected for eggshell measurements. Eggs were broken on a flat surface, rinsed with tap water and then dried. Eggshell thickness was determined by using a digital micrometer where 3 different points (top, equator, and bottom) of each egg were measured. Data collected represent the mean thickness of each eggshell sampled. As noted above, the weekly average of the respective replicate was considered the observational unit. Throughout the duration of the test (from placement through the laying period) all hens were administered standard commercial vaccinations and deworming treatments required in routine commercial egg production. These included vaccination for Infectious Bursal Disease, Newcastle disease, Infectious bronchitis, Fowl pox, Infectious Laryngotracheitis, Egg Drop Syndrome and Salmonella. Salinomycin was used for the prevention of coccidiosis during the brooding and rearing periods. Administration and timing of these vaccines/treatments were consistent with procedures normally used in commercial egg production in the Philippines. Birds and cages were examined routinely by a veterinarian; at no point during the trial were the signs of overt disease observed. Statistics During the 5-to-15-week rearing period, bird performance, feed intake, uniformity and mortality were analyzed using one-way ANOVA. The dependent variable was the average over the respective period (5 to 15 weeks) and the treatment was the only independent variable. Where the treatment effect was significant, a pairwise comparison between means was performed using the Tukey procedure where p < 0.05 was considered significant. During the laying period repeated measurements of egg production variables were taken from the same experimental replicates. To avoid errors associated with pseudo-replication, each variable was modeled using a linear mixed effects model, incorporating a random effect for each replicate and fixed effects for the treatments, the week of observation, and their interaction. The formula below was used:yitr=β0+βt*Weekt+βi*Treati+βit(Weekt*Treati)+αr+εitrWhere:•yitr is the outcome for the ith + treatment at the tth week and the rth replicate•β0 is the intercept•βt is the week (factor) effect•βi is the treatment (factor) effect•βit is the interaction (factor) effect between week and treatment•αr is the random effect for the rth+replicate•εitr is the residual error term •yitr is the outcome for the ith + treatment at the tth week and the rth replicate yitr is the outcome for the ith + treatment at the tth week and the rth replicate •β0 is the intercept β0 is the intercept •βt is the week (factor) effect βt is the week (factor) effect •βi is the treatment (factor) effect βi is the treatment (factor) effect •βit is the interaction (factor) effect between week and treatment βit is the interaction (factor) effect between week and treatment •αr is the random effect for the rth+replicate αr is the random effect for the rth+replicate •εitr is the residual error term εitr is the residual error term Since the different dependent outcomes are in essence a time series, it is reasonable to assume the residuals within a replicate have an autoregressive correlation. This means that observations closer in time are more likely to have higher correlations than those more distant in time. This was addressed by the inclusion of an autoregressive correlation structure in the lme function with the week as a covariate and the replicate ID as the grouping factor. In addition, the residual variance showed an increasing tendency with time and, to stabilize the age-associated variance over each week, the model was allowed to have a different residual variance in each week. The resulting fit was assessed by inspection of a qq-plot of the normalized residuals and a scatter plot of the normalized residuals versus the fitted values and time. Each model was fitted using the lme function in the R package nlme (Pinheiro and Bates, 2023). Each fixed effect (treatment, week, and their interaction) was tested using a likelihood ratio test (LRT). The reported means are the expected marginal means, calculated with the emmeans package in R(. The difference between means was calculated based on the model and adjusted using Tukey’s procedure to control the family-wise type-I error rate at α = 0.05. The results are expressed as the mean estimate and its 95% confidence interval. In addition, as treatments were coded as ordinal factors, linear, quadratic and cubic trends were also assessed using the contrast function of the emmeans package. The level of significance was set at 0.05 in all procedures. Results The outcomes of feeding graded levels of quillaja saponins during the 5-to-15-week rearing period are presented in Table 2. Weight gain per week and feed conversion ratio were significantly improved by saponin additions during this period. In addition, mortality tended to be lower in the saponin treatments but there was no consistent response pattern among them. Mortality for the 250 mg/kg level was significantly lower than the control.Table 2Treatment means1 for performance, uniformity, feed intake and mortality of layer pullets fed quillaja saponins from 5 to 15 weeks of rearing.Quillaja Saponinmg/kgBody WeightgWeight Gain2g/weekUniformity%Feed Intakeg/weekFCR2g/gMortality2%01046.8[1028.9, 1064.6]71.6a[70.06, 73.09]93.00[90.15, 95.85]342.5[339.19, 345.86]4.79b[4.69, 4.89]2.13b[1.56, 2.7]1251058.9[1041.0, 1076.8]75.3b[73.79, 76.81]93.7[90.82, 96.51]340.5[337.19, 343.87]4.53a[4.42, 4.63]1.46ab[0.89, 2.03]2501069.2[1051.3, 1087.0]75.7b[74.16, 77.18]92.7[89.82, 95.51]339.7[336.38, 343.06]4.50a[4.39, 4.6]0.80a[0.23, 1.37]5001079.1[1061.2, 1097.0]76.2b[74.7, 77.72]94.7[91.82, 97.51]344.9[341.55, 348.22]4.53a[4.43, 4.63]1.86ab[1.29, 2.43]1Data represent the average results collected from weeks 5 through 15 of rearing, representing 10 replicates of 75 birds each per treatment. Means were separated using the Tukey procedure, where significance was determined at p < 0.05. 2Treatment effects were significant (p < 0.01). Table 2Treatment means1 for performance, uniformity, feed intake and mortality of layer pullets fed quillaja saponins from 5 to 15 weeks of rearing.Quillaja Saponinmg/kgBody WeightgWeight Gain2g/weekUniformity%Feed Intakeg/weekFCR2g/gMortality2%01046.8[1028.9, 1064.6]71.6a[70.06, 73.09]93.00[90.15, 95.85]342.5[339.19, 345.86]4.79b[4.69, 4.89]2.13b[1.56, 2.7]1251058.9[1041.0, 1076.8]75.3b[73.79, 76.81]93.7[90.82, 96.51]340.5[337.19, 343.87]4.53a[4.42, 4.63]1.46ab[0.89, 2.03]2501069.2[1051.3, 1087.0]75.7b[74.16, 77.18]92.7[89.82, 95.51]339.7[336.38, 343.06]4.50a[4.39, 4.6]0.80a[0.23, 1.37]5001079.1[1061.2, 1097.0]76.2b[74.7, 77.72]94.7[91.82, 97.51]344.9[341.55, 348.22]4.53a[4.43, 4.63]1.86ab[1.29, 2.43]1Data represent the average results collected from weeks 5 through 15 of rearing, representing 10 replicates of 75 birds each per treatment. Means were separated using the Tukey procedure, where significance was determined at p < 0.05. 2Treatment effects were significant (p < 0.01). Treatment means1 for performance, uniformity, feed intake and mortality of layer pullets fed quillaja saponins from 5 to 15 weeks of rearing. Data represent the average results collected from weeks 5 through 15 of rearing, representing 10 replicates of 75 birds each per treatment. Means were separated using the Tukey procedure, where significance was determined at p < 0.05. 2Treatment effects were significant (p < 0.01). Production variables collected during the first 20 weeks of production are illustrated in Table 3. While feed consumption was unaffected by treatment, all egg production variables responded in a linear manner to increasing levels of quillaja saponins, but the highest level (500 mg/kg) showed similar responses to the 250 mg/kg level. During this period, quadratic and cubic trends were not significant. Generally, levels below 250 mg/kg were not as effective in improving the variables measured during this phase of production, and in each case, the 500 mg/kg level produced the maximal response. For most variables, however, there were no significant differences between this level and 250 mg/kg, and in each case the . The only exception to this statement was in HDEP, where the 250 mg/kg was not different from either 125 or 500 mg/kg.Table 3The effects1 of quillaja saponins on mean egg production variables recorded from 21 to 40 weeks of age.Quillaja Saponinsmg/kgHen-Day Egg Production2,3%Egg Weight2gEgg Mass2Feed Intakeg/bird/dayFeed Conversion Ratio2g feed/g egg078.3b[77.3, 79.4]56.6b[56.2, 57.1]44.6b[43.9, 45.2]108.5[108.0, 108.9]2.75a[2.69, 2.80]12578.9b[77.8, 79.9]57.4b[57.0,57.8]45.5b[44.8, 46.1]108.7[108.2, 109.1]2.80a[2.75, 2.86]25080.3ab[79.2, 81.3]58.7a[58.3, 59.1]47.4a[46.7, 48.0]108.6[108.2, 109.1]2.58b[2.53, 2.64]50081.5a[80.2, 82.3]59.5a[59.0, 59.9]48.8a[48.0, 49.3]108.5[108.0, 108.9]2.59b[2.54, 2.65]Standard error40.5040.2080.3230.2310.0261Each value represents the mean of 8 replicates where each replicate contained 84 hens. The replicate was considered the experimental unit for all statistical analyses. Treatment effects for all variables except Feed Intake were significant (p < 0.05). Means with different superscripts indicate significant differences among groups at the 0.05 level. The Tukey’s pairwise comparison was used to control the family-wise type I error rate at 0.05.2Significant linear effect, p < 0.001.3When the HDEP response at 21-40 weeks was compared to the HDEP response at 41-60 weeks of age (Table 4), a time period x treatment interaction was observed, p < 0.05. 4The standard error was estimated from the fitted model using the emmeans package in R.. Table 3The effects1 of quillaja saponins on mean egg production variables recorded from 21 to 40 weeks of age.Quillaja Saponinsmg/kgHen-Day Egg Production2,3%Egg Weight2gEgg Mass2Feed Intakeg/bird/dayFeed Conversion Ratio2g feed/g egg078.3b[77.3, 79.4]56.6b[56.2, 57.1]44.6b[43.9, 45.2]108.5[108.0, 108.9]2.75a[2.69, 2.80]12578.9b[77.8, 79.9]57.4b[57.0,57.8]45.5b[44.8, 46.1]108.7[108.2, 109.1]2.80a[2.75, 2.86]25080.3ab[79.2, 81.3]58.7a[58.3, 59.1]47.4a[46.7, 48.0]108.6[108.2, 109.1]2.58b[2.53, 2.64]50081.5a[80.2, 82.3]59.5a[59.0, 59.9]48.8a[48.0, 49.3]108.5[108.0, 108.9]2.59b[2.54, 2.65]Standard error40.5040.2080.3230.2310.0261Each value represents the mean of 8 replicates where each replicate contained 84 hens. The replicate was considered the experimental unit for all statistical analyses. Treatment effects for all variables except Feed Intake were significant (p < 0.05). Means with different superscripts indicate significant differences among groups at the 0.05 level. The Tukey’s pairwise comparison was used to control the family-wise type I error rate at 0.05.2Significant linear effect, p < 0.001.3When the HDEP response at 21-40 weeks was compared to the HDEP response at 41-60 weeks of age (Table 4), a time period x treatment interaction was observed, p < 0.05. 4The standard error was estimated from the fitted model using the emmeans package in R.. The effects1 of quillaja saponins on mean egg production variables recorded from 21 to 40 weeks of age. Each value represents the mean of 8 replicates where each replicate contained 84 hens. The replicate was considered the experimental unit for all statistical analyses. Treatment effects for all variables except Feed Intake were significant (p < 0.05). Means with different superscripts indicate significant differences among groups at the 0.05 level. The Tukey’s pairwise comparison was used to control the family-wise type I error rate at 0.05. Significant linear effect, p < 0.001. When the HDEP response at 21-40 weeks was compared to the HDEP response at 41-60 weeks of age (Table 4), a time period x treatment interaction was observed, p < 0.05. 4The standard error was estimated from the fitted model using the emmeans package in R.. During the 41-to-60-week period, feed consumption was slightly reduced by increasing levels of saponins, and 500 mg/kg significantly reduced intake compared to controls. Conversely, HDEP was highest and significantly different from controls when 250 mg/kg quillaja saponins were fed. Neither 125 nor 500 mg/kg was significantly different from the control or the 250 mg/kg level. When comparing the egg production response across both production periods, a different pattern was evident across the 4 dosage levels (Table 3, Table 4). As a result, a time period x treatment interaction occurred where egg production was maximized at 500 mg/kg during the initial laying period but achieved its highest level at 250 mg/kg during the second production period. Additionally, there was no marked difference between 250 and 500 mg/kg during this period, and the response in HDEP was stabilized.Table 4The effects1 of quillaja saponins on egg production variables recorded from 41 to 60 weeks of age.Quillaja Saponinsmg/kgHen Day Egg Production2,3%Egg Weight4gEgg Mass4Feed Intake4g/bird/dayFeed Conversion Ratio4g feed/g eggEggshell Thickness5Mm084.5b[83.5, 85.6]59.0bc[58.5, 59.4]49.8b[49.2, 50.5]107.6a[107.1, 108.1]2.17a[2.13, 2.20]0.40c[0.39, 0.41]12586.1ab[85.1, 87.1]58.8c[58.4, 59.2]50.6ab[50.0, 51.3]107.6a[107.1, 108.1]2.13a[2.10, 2.16]0.41b[0.40, 0.42]25086.9a[85.8, 87.9]59.6ab[59.3, 60.2]51.8a[51.2, 52.5]106.9ab[106.4, 107.3]2.06b[2.03, 2.10]0.43a[0.42, 0.44]50086.1ab[85.1, 87.2]60.1a[59.6, 60.5]51.7a[51.52.4]106.4b[105.9, 106.9]2.06b[2.03, 2.09]0.43a[0.42, 0.44]Standard error60.5040.2080.3230.2310.0260.0051Each value represents the mean of 8 replicates where each replicate contained 84 hens. The replicate was considered the observational unit for all analyses. Treatment effects for all variables were significant (p < 0.05). Means with different superscripts were significant (p < 0.05). Tukey’s pairwise comparison was used to control the family-wise type I error rate p < 0.05.2Linear and quadratic effects were significant, p = 0.03 and p < 0.04, respectively.3As described in Table 3, footnote 3, comparison of HDEP over both time periods revealed a time period x treatment interaction (p < 0.05).4Significant linear effect, p < 0.001.5Eggshell thickness was determined in 40 randomly selected eggs per treatment collected biweekly from week 39 to week 59. The pen was the observational unit. Linear effect was significant, p < 0.05.6The standard error was estimated from the fitted model using the emmeans package in R. Table 4The effects1 of quillaja saponins on egg production variables recorded from 41 to 60 weeks of age.Quillaja Saponinsmg/kgHen Day Egg Production2,3%Egg Weight4gEgg Mass4Feed Intake4g/bird/dayFeed Conversion Ratio4g feed/g eggEggshell Thickness5Mm084.5b[83.5, 85.6]59.0bc[58.5, 59.4]49.8b[49.2, 50.5]107.6a[107.1, 108.1]2.17a[2.13, 2.20]0.40c[0.39, 0.41]12586.1ab[85.1, 87.1]58.8c[58.4, 59.2]50.6ab[50.0, 51.3]107.6a[107.1, 108.1]2.13a[2.10, 2.16]0.41b[0.40, 0.42]25086.9a[85.8, 87.9]59.6ab[59.3, 60.2]51.8a[51.2, 52.5]106.9ab[106.4, 107.3]2.06b[2.03, 2.10]0.43a[0.42, 0.44]50086.1ab[85.1, 87.2]60.1a[59.6, 60.5]51.7a[51.52.4]106.4b[105.9, 106.9]2.06b[2.03, 2.09]0.43a[0.42, 0.44]Standard error60.5040.2080.3230.2310.0260.0051Each value represents the mean of 8 replicates where each replicate contained 84 hens. The replicate was considered the observational unit for all analyses. Treatment effects for all variables were significant (p < 0.05). Means with different superscripts were significant (p < 0.05). Tukey’s pairwise comparison was used to control the family-wise type I error rate p < 0.05.2Linear and quadratic effects were significant, p = 0.03 and p < 0.04, respectively.3As described in Table 3, footnote 3, comparison of HDEP over both time periods revealed a time period x treatment interaction (p < 0.05).4Significant linear effect, p < 0.001.5Eggshell thickness was determined in 40 randomly selected eggs per treatment collected biweekly from week 39 to week 59. The pen was the observational unit. Linear effect was significant, p < 0.05.6The standard error was estimated from the fitted model using the emmeans package in R. The effects1 of quillaja saponins on egg production variables recorded from 41 to 60 weeks of age. Each value represents the mean of 8 replicates where each replicate contained 84 hens. The replicate was considered the observational unit for all analyses. Treatment effects for all variables were significant (p < 0.05). Means with different superscripts were significant (p < 0.05). Tukey’s pairwise comparison was used to control the family-wise type I error rate p < 0.05. Linear and quadratic effects were significant, p = 0.03 and p < 0.04, respectively. As described in Table 3, footnote 3, comparison of HDEP over both time periods revealed a time period x treatment interaction (p < 0.05). Significant linear effect, p < 0.001. Eggshell thickness was determined in 40 randomly selected eggs per treatment collected biweekly from week 39 to week 59. The pen was the observational unit. Linear effect was significant, p < 0.05. The standard error was estimated from the fitted model using the emmeans package in R. EW and EM responded similarly, and each variable reached a plateau at 250 and 500 mg/kg. Eggshell thickness, measured biweekly from week 39 to week 59, was significantly improved by each saponin level when compared to controls; both 250 and 500 mg/kg differed significantly from the 125 mg/kg feeding level. Both eggshell thickness and feed conversion followed the same response trend that was evident for the other variables. Despite the slight reduction in feed intake observed at 500 mg/kg during this phase of production, the highest levels (250 and 500 mg/kg) of quillaja additions produced significant improvements in feed conversion when compared to both lower levels. Discussion While the beneficial effects of saponins in animal production have been recognized for many years (Cheeke, 2000; Francis et al., 2002), recent studies with these compounds have shown specific effects that are of significance in poultry production. Reports have shown that saponin supplementation positively affects the intestinal tract of chickens by minimizing the effects of intestinal pathogens (Bafundo et al., 2020; Blue et al., 2023), and by increasing villus height and reducing crypt depth (Bafundo et al., 2021a; Stefanello et al., 2022). In addition, quillaja saponins are known to modulate the intestinal cytokines that are involved in the inflammatory response to intestinal infections (Saddoris-Clemons et al., 2024). This effect helps maintain the integrity of the intestinal desmosomes and the functionality of the intestinal barrier. These findings are supported by the work of Blue et al. (2023), who showed that quillaja saponins increase the abundance of mRNA for the tight junction proteins Claudin 1 and Claudin 5, both of which are required for the mucosa to remain free from toxins and bacterial pathogens. As shown by Saddoris-Clemons et al. (2024), saponin supplementation alleviated the leaky-gut syndrome brought about by extended feed deprivation and decreased the potential for intestinal disease by strengthening the intestinal barrier (Blue et al., 2023). Thus, whether in healthy or disease- challenged chicks, quillaja saponins produce an intestinal environment which is conducive to the efficient uptake of nutrients. Both Bafundo et al. (2021a) and Stefanello et al. (2022) have shown that the digestibility of dry matter, fats, protein and minerals improves significantly with the dietary inclusion of quillaja saponins. Given the effects described above, quillaja saponins have been shown in numerous trials to improve performance and carcass yield of growing broilers (Bafundo et al., 2021b; Stefanello et al., 2022; Dai et al., 2023). In addition, these effects have been used commercially to offset the absence of antibiotic growth promoters (Saddoris-Clemons et al., 2024). Although trials dealing with the effects of saponins in laying fowl are rare in comparison to those conducted in broilers, Souza et al. (2023) administered a saponin combination to layer breeders for a 19-week period and evaluated egg production performance during this interval. These investigators noted that egg production (HDEP) was not statistically improved by saponin treatment, however, saponin supplementation improved feed efficiency (kg feed per kg egg) and increased the thickness, strength and quality of eggshells. While these effects are similar to those reported in the current study, data from our trial indicate a wider range of responses, with HDEP, EW, EM, and feed conversion responding positively (significant linear effects) to increasing saponin levels during the first 20 weeks of egg production. As noted above, these effects are in general agreement with previous research describing the body weight and carcass yield responses of broiler chickens to saponin feeding (Bafundo et al., 2021b; Dai et al., 2023). An important difference in the current study was exposure to saponin-containing feed during the rearing period, where quillaja additions significantly improved weekly body weight gain and feed conversions. As noted previously, quillaja saponins are known to improve intestinal health and nutrient digestibility in broilers and may have produced similar effects in the pullets in this study. These factors could have produced hens that were heavier and fitter and in turn, better prepared for the stresses imposed by the onset of lay. At this point, the long term effects of early saponin feeding cannot be confirmed but these data provide a suitable rationale for additional studies on the potential benefits of saponin exposure during the rearing period. In large part, the trends established for improved egg production variables during the initial 20 weeks of production were maintained from 41 to 60 weeks of age. The EW, EM and feed conversion responses from 41 to 60 weeks were similar to those measured during the initial egg production period. However, between 41 and 60 weeks, egg production was maximized at 250 rather than 500 mg/kg producing a significant quadratic effect. Comparison of this effect to that recorded during the initial period (linear effect) indicated a different response over time (a time period x treatment interaction). It is difficult to determine whether this response in egg production was directly related to the slight decrease in feed intake observed during the final production period because the other production variables (EW, EM, eggshell thickness) did not decrease in a similar manner. Moreover, these variables were greatest, or nearly so, at the highest levels of saponin feeding. When Souza et al. (2023) recorded similar reductions in feed intake in saponin-fed layer breeders, they also noted that egg production, albumin percent, Haught units, eggshell strength and cuticle quality were unchanged, and that feed conversion improved significantly. Thus, factors other than feed intake may have influenced the egg production response at 500 mg/kg, and further study is required to understand the consistency and magnitude of this effect. While it is well recognized that eggshell thickness is influenced by nutrition, hen age, genetics and overall hen health, improvements in shell thickness are often correlated with better eggshell quality and improved eggshell strength (Ketta and Tumova, 2018). These benefits are likely indicators of better shell integrity which reduces the risks of cracks during handling, transport and storage. Our results indicate that feeding quillaja saponins improves eggshell thickness, a finding that is consistent with the layer breeder data reported by Souza et al. (2023). Because cuticle quality has been also associated with lower levels of trans-shell Salmonella contamination (De Reu et al., 2006; Munoz et al., 2015), supplemental work should address the effects of saponin feeding on eggshell quality and whether levels of bacterial contamination are affected. As noted above, feed consumption was largely unaffected by quillaja administration, with only slight reductions in feed intake occurring at 500 mg/kg during the final egg production period. It is noteworthy that key production variables such as HDEP, EW, EM, feed conversion and eggshell thickness were positively affected by the higher levels of saponin supplementation at the same or slightly reduced levels of feed intake. These findings agree with those of Bafundo et al. (2021a, 2021b) who demonstrated saponin-related improvements in broiler body weight and carcass yield at equivalent or slightly reduced levels of feed intake. Although previous work has shown that these effects are likely related to improvements in protein, energy and mineral digestibility (Bafundo et al., 2021a; Stefanello et al., 2022) and/or the quillaja-enhanced activity of enterocyte nutrient transporters (Blue et al., 2023), studies of this nature have not been conducted in commercial layers. However, they do seem to be plausible explanations for the improvements in productivity observed throughout this trial. Future studies involving these assays are likely to help explain the responses produced and to shed additional light on the mechanisms of quillaja activity in the avian intestine. In summary, the results of this study demonstrate that feeding quillaja saponins to commercial layers produced positive effects in critical egg production variables. Significant improvements in HDEP, EW, EM, feed efficiency and eggshell thickness were observed when at least 250 mg/kg of quillaja saponins were provided to commercial hens. These effects were evident in all production variables when measured between 21 and 40 and 41 to 60 weeks of age. Ethics declaration This study was conducted in accordance with the ARRIVE (Animal Research: Reporting of In Vivo Experiments) guidelines. This study was approved by the Ethics Committee of Poulfarm b.v. Author Contribution Dr. K.W. Bafundo was involved in project administration, data curation, formal data analysis and interpretation, and composition of the manuscript. Dr. Miriam Alberto-Tempra established the methodology used in testing, directed the investigation, supervised the staff, and was involved in data curation and interpretation. The trial was conceptualized by Dr. Leandro Ferreira; he participated in trial supervision and data interpretation and trial site monitoring. Dr. Ben Dehaeck was the project administrator, providing supervision, resource acquisition and application, and direction to the scientists involved (listed as co-authors). Dr. Wim Schelstraete carried out the formal data analyses, data interpretation and statistical rationale for the analyses used. He contributed significantly to the composition of the text. Disclosures The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Kenneth W. Bafundo reports financial support was provided by Huvepharma NV. Kenneth W. Bafundo reports a relationship with Huvepharma NV that includes: consulting or advisory. None has patent pending to None. None If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.