Effect of Supplementation with Artemisia absinthium or Pumpkin Seed Powder on Ascaridia galli Worm Burdens, Productivity, Nutrient Digestibility, and Cytokine Gene Expression in Laying Hens Open Access

Ascaridia galli is a nematode commonly found in chickens in cage-free production systems (1,2,3). Adult worms are found free in the jejunum and ileum of chickens (4,5). Infections with A. galli can potentially result in reduced bird performance and nutrient digestibility and negatively affect egg production or quality due to inflammatory and other reactions of the intestine that impair its function (6,7). These parasites can also impair animal welfare by complete or partial obstruction of the intestinal lumen (8).

In many countries, benzimidazoles are the only drug family of which compounds are approved to treat nematodes in commercial flocks of layers with no withdrawal period for eggs (9,10). The lack of drug options raises concerns regarding the possibility of resistances of A. galli against benzimidazoles (11).

Over the last few years, there has been an increase in the market’s demand for cage-free egg production systems, which have shown a high prevalence of A. galli (3). Moreover, not only are consumers more concerned with the use of drugs on food animals, but organic producers do not have the option of using these drugs (12). This situation requires research evaluating alternative treatments.

Two phytogenic products commonly used by backyard owners have shown potential to be used against other nematodes species in other hosts: Artemisia absinthium and pumpkin seed. A. absinthium, commonly called wormwood, contains several compounds, such as terpenes, limonene, myrcene, α, and β thujone, which have shown anthelmintic properties (13). Artemesia absinthium was effective in the control of Haemonchus contortus in sheep (14). Pumpkin seeds have a diversity of compounds with anthelmintic properties such as cucurbitacin, flavonoids, terpenes, and saponins (15). Their potential as an alternative to control nematodes has been shown in ostriches (16) and mice (17).

The aim of this study was to test if A. absinthium or pumpkin seed can be used as treatment to alleviate potential negative effects of an A. galli infection on egg production and quality, nutrient digestibility, and cytokine gene expression.

A total of 96 1-day-old Hy-Line white pullets were raised on the floor until 12 wk of age and then moved to individual laying hen cages. At 16 wk, birds were divided into three groups with 32 birds each. One group was left untreated (untreated control) while the other two groups were fed mash diets containing either A. absinthium powder (Starwest Botanicals, Sacramento, CA) or pumpkin seed powder (nuts.com, Cranford, NJ). The Artemisia dose level was 0.02%, similar to what was used in sheep (14). The pumpkin seed dose level was 1%. The dose was based on anecdotal evidence from small flock owners.

At 25 wk, eight birds per treatment were necropsied. Twelve of the remaining 24 birds in each treatment group were challenged by oral gavage with 250 embryonated A. galli eggs in 1 ml distilled water. The other 12 birds received only 1 ml distilled water. This resulted in a 3 × 2 factorial design with three levels of diet treatment (untreated control, artemisia, and pumpkin seed) and two levels of worm infection (nonchallenged and challenged). Each treatment combination consisted of 12 chickens in individual cages.

At 34 wk, titanium dioxide (TiO₂) was added to the feed (0.50%) as an undigestible marker to evaluate nutrient digestibility. At 35 wk, all birds were euthanatized using carbon dioxide followed by cervical dislocation. Samples were collected as described below.

During the entire study, birds were fed diets formulated according to Hy-Line recommendations for Hy-Line white pullets and laying hens. Feed and water were available ad libitum. Birds and consumed feed were not weighed. Animal care and experimental procedures were performed in compliance with all federal and institutional animal use guidelines and approved by the Auburn University Institutional Animal Care and Use Committee (PRN 2019-3449).

Worm eggs from a naturally infected layer flock (18) were purified using a process similar to what was described by Collins et al. (19). Briefly, feces collected from the flock were flushed with tap water through sieves with pore sizes of 212, 90, and 38 µm. Five milliliters of retained material in the last sieve were added to 45 ml of saturated sodium chloride solution. The suspension was centrifuged (420 g, 5 min), and the supernatant was flushed with tap water through a 38 µm sieve. The retained material containing the eggs was finally suspended in tap water. To prevent fungal growth, 10 ml of egg suspension was stored with 1 ml of 0.5% formalin in 50 ml centrifuge tubes. The eggs were embryonated for 21 days at room temperature with frequent aeration, after which the embryonation rate was on average 67%. For the challenge, the concentration was adjusted to 250 embryonated eggs/dose with tap water. Absence of Heterakis gallinarum DNA in the egg suspension was confirmed by PCR using primers published by Cupo and Beckstead (20).

After infection, about 20 g of feces were collected weekly from trays placed under each cage. Feces of three birds that were in adjacent cages were pooled, resulting in four replicates per group. One gram of the homogenized pooled feces was dissolved in 29 ml of saturated sodium chloride and poured through a strainer. Both chambers of a McMaster slide were filled with the solution and eggs were counted at 100× magnification. Fecal egg counts were calculated following the McMaster method.

At the final necropsy in week 35, the jejunum was opened in a longitudinal direction. The content was first inspected for the presence of adult worms; adult worms were collected in 70% alcohol, counted, sexed, and measured.

All diets were analyzed for crude protein (CP) (22), calcium, and phosphorus (modified procedure using Wolf et al. (23) and CEM Application Notes for Acid Digestion) at a commercial laboratory (Dairy One Forage Laboratory, Ithaca, NY). Gross energy (GE) of feed was determined at Auburn University (Auburn, AL). Duplicate 0.75 g samples of feed were analyzed for GE using an isoperibol oxygen bomb calorimeter (model no. 6400, Parr Instruments, Moline, IA) standardized with benzoic acid.

An indigestible marker (0.50% TiO₂) was added to all diets for determination of apparent ileal nutrient digestibility. At the final necropsy in week 35, content was gently removed from the ilea of all birds starting 2 cm posterior from the Meckel’s diverticulum to 2 cm anterior from the ileal-cecal junction. Digesta of three birds that had been housed in adjacent cages were pooled, resulting in four replicates per group, to provide sufficient sample for nutrient analysis. All digesta samples were lyophilized in a Virtis Genesis Pilot Lyophilizer (SP Industries, Warminster, PA) and then ground in an electric coffee grinder (Capresso 560.4 Infinity, Montvale, NJ) on the finest setting. Thereafter, samples were sent to a commercial laboratory (Dairy One Forage Laboratory) to be analyzed for CP and minerals using methods as mentioned previously. Gross energy of ileal digesta (single 0.75 g samples) was determined at Auburn University. Due to sample size limitations, two replicates from the noninfected control and the artemisia control groups were pooled to ensure sufficient sample for nutrient analysis.

At the prechallenge necropsy in week 25 and the final necropsy in week 35, 1 g of the jejunal wall of 10 to 12 birds per group was collected approximately 10 cm upstream of Meckel’s diverticulum and snap frozen using dry ice. The samples were stored at −80 C. Relative gene expression was determined as described (27). In brief, RNA was extracted from 15 mg jejunal wall with the Qiagen RNeasy kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. DNA was digested on the column during RNA extraction with the RNase-Free DNase Set (Qiagen). Total RNA was transcribed into cDNA using the Lunascript kit (New England Biolabs, Ipswich, MA). The cDNA was diluted 1:3 in RNAse-free water, and partial TGF-β, IFN-γ, IL-4, IL-8, IL10, IL13, GAPDH, and HMBS genes were amplified in duplicates by qPCR using the Fast SYBR Green Master Mix (Qiagen). Cycling parameters consisted of an initial 10 min at 95 C denaturation cycle, followed by 40 cycles of denaturation for 30 sec at 95 C, annealing for 1 min at 60 C, and extension at 72 C for 30 sec. The specificity of PCR products was confirmed with a melting curve analysis. Primers and their efficiencies are listed in Supplemental Table S1. The gene expression was calculated relative to the expression of two housekeeping genes (GAPDH and HMBS) and the average of the uninfected birds as a calibrator as described by Vandesompele et al. (28) and Hellemans et al. (29).

Normality of all parameters was checked by Shapiro’s test or Q-Q plots. Worm eggs per gram feces (EPG) were analyzed by repeated measures two-way ANOVA for the effect of treatment (untreated control, artemisia, and pumpkin seed) and week using the anova_test function of the rstatix package (30).

Worm burden was compared by Kruskal-Wallis test since the data were not normally distributed. Worm sizes were analyzed by two-way ANOVA for treatment and worm sex as main effects and their interactions. Worm fragments with unknown sex were excluded for this analysis.

Egg production was analyzed separately by repeated measures two-way ANOVA for the effect of treatment and week with days as replicates for the effect of treatment before the challenge. Repeated measure three-way ANOVA was used to analyze egg production after challenge with treatment, infection status, and week as main effects after the challenge. Egg quality parameters except yolk color were analyzed by two-way ANOVA using the Anova function of the car package (31) with main effects treatment and week before the challenge and by three-way ANOVA with main effects treatment, challenge, and week after the challenge.

Yolk color was compared separately for each week with the Kruskal-Wallis test with the Dunn test as post hoc test and P values adjusted with the Benjamini-Hochberg method since the data were ordinal and not normally distributed.

Nutrient digestibility was analyzed with two-way ANOVA for treatment and infection status as main effects. Relative gene expression values were log₂ transformed and analyzed with two-way ANOVA for treatment and infection status as main effects.

Tukey’s honestly significant difference test was used as post hoc test for all ANOVAs. P ≤ 0.05 was considered statistically significant. All statistical analyses were performed in R v. 4.3.2 (32).

From the time of infection with A. galli until 4 wk postinfection (p.i.), Eimeria oocysts were identified in replicates of all groups. Shedding of worm eggs started 4 wk p.i. in the untreated control and artemisia groups with an average of 150 and 25 EPG, respectively. For the pumpkin seed–treated group, shedding of A. galli eggs started at 5 wk p.i. A peak of egg shedding in the untreated control group was observed in weeks 7 and 8 p.i. with an average of 675 EPG, while the peak egg shedding in the artemisia group was in week 8 p.i. with an average of 2100 EPG, and the peak shedding for the pumpkin seed group was observed in week 9 p.i. with an average of 2050 EPG. There was no interaction between treatment and week (P = 0.094), and the effect of the treatment was not significant (P = 0.674). However, the differences between weeks were significant (P = 0.001) with worm egg shedding in weeks 4 and 5 p.i. significantly lower than in weeks 8 and 9 p.i. (Table 1).

In the untreated control and artemisia groups, eight of the 12 birds necropsied at the end of the experiment had adult worms. In the pumpkin seed group nine out of 11 birds were positive for adult A. galli. The average number of adult worms per bird was 2.1 for the untreated control group (0–8 worms per bird, median = 1), 4.9 for the artemisia group (0–21 worms per bird, median = 1), and 6.7 for the pumpkin seed group (0–46 worms per bird, median = 2). There were no statistically significant differences in the worm burden between the treatments (P = 0.524).

Average length of female worms was 52.0 mm and of male worms 39.5 mm. Average worm lengths were 45.1, 46.8, and 45.2 mm in the untreated control, the artemisia group, and the pumpkin group, respectively. Sizes differed significantly between worm sexes (P < 0.001), but not between groups (P = 0.855). There was no interaction between treatments and the sex of the worms (P = 0.447).

Weekly egg production is presented in Table 2. Before as well as after the challenge differences between the treatments were not significant (P = 0.129, P = 0.317). The effect of the challenge was not significant (P = 0.617). Before the infection, differences between weeks were significant (P < 0.001) with each week significantly from each other, except that the difference between weeks 24 and 25 was not significant. After the challenge, differences between weeks were not significant (P = 0.981). There were no interactions of treatment and week before the challenge or treatment, challenge, and week after the challenge (P > 0.05).

Before the challenge, treatment had a significant effect on egg weight (P < 0.001) with eggs of the artemisia-treated group lighter than eggs of the other two groups, egg white weight (P < 0.001) with egg whites of birds fed pumpkin seed heavier than the egg whites of the birds treated with artemisia, and shell weight (P = 0.008) with egg shells of artemisia-treated birds lighter than those of the control birds (Table 3, Supplemental Table S2). Week had a significant effect on egg weight (P < 0.001), yolk weight (P < 0.001), and egg white weight (P = 0.002). For no parameter were interactions between treatment and week statistically significant (P > 0.05).

After the challenge, there were statistically significant interactions between treatment and challenge for factors except egg white weight. Eggs of artemisia-treated birds were significantly lighter than eggs of the other groups, but only in unchallenged groups (P = 0.011). Albumen and Haugh units of eggs of pumpkin seed–treated birds were higher than the albumen of birds given feed containing artemisia, but again only in uninfected birds (P = 0.009, P = 0.026). Only in the artemisia-treated group were yolks of infected birds heavier than of uninfected birds (P < 0.001). Shell weights of eggs from uninfected, artemisia-treated birds were lower than shell weights of artemisia-treated, infected birds. The difference between infected and uninfected birds was not significant in the untreated and pumpkin-treated birds (P = 0.004). Shells of birds fed pumpkin seed were thinner than eggshells of control birds, but only in uninfected birds (P = 0.011). There were no interactions of treatment and challenge for egg white weight, but egg whites of pumpkin seed–treated birds were heavier than egg whites of the other groups (P = 0.021; Tables 4, Supplemental Table S3).

Week had significant influence on weight (P < 0.001), yolk weight (P < 0.001), egg white weight (P = 0.033), and shell thickness (P = 0.033), but interactions of weeks with the other two main effects were not significant (P > 0.05; data not shown).

The Kruskal-Wallis test showed a significant difference in yolk color between groups in weeks 27, 28, 32, and 33 (P = 0.025, P = 0.017, P = 0.001, P < 0.001), but the Dunn test as post hoc test identified significant pairwise comparisons only in weeks 32 and 33. In week 32, the infected control birds had a significantly darker yolk color than infected birds given feed with pumpkin seed (P-adj 0.015). In week 33, the uninfected control birds had a significantly darker yolk color than birds given feed with artemisia or pumpkin, regardless of their infection status (artemisia, uninfected P-adj = 0.003; artemisia, infected P-adj < 0.001; pumpkin, uninfected P-adj = 0.005; pumpkin, infected P-adj = 0.013). Equally, the infected control birds had a significantly darker yolk color than birds given feed with artemisia or pumpkin, regardless of their infection status (artemisia, uninfected P-adj < 0.001; artemisia, infected P-adj < 0.001; pumpkin, uninfected P-adj < 0.001; pumpkin, infected P-adj = 0.001).

The nutrient digestibility analyses are shown in Table 5 and Supplemental Table S4. Challenge with A. galli significantly reduced AIDE (P < 0.001), but AIDE was not affected by the treatment. Apparent ileal digestible energy was lower in challenged birds in the artemisia group, while this difference was not observed in the other treatments (P = 0.010 for the interaction).

Similarly, challenge with A. galli also significantly reduced CP digestibility (P = 0.013), while apparent CP digestibility was not affected by the treatment. Crude protein digestibility significantly differed between infected and uninfected birds only in the artemisia group (P = 0.026 for the interaction).

Calcium digestibility was not affected by the infection, but the untreated control birds had a 30% lower calcium digestibility than birds fed diets containing either artemisia or pumpkin seed (P = 0.016). Calcium digestibility was lower in challenged birds fed the untreated control diet. There were no differences in the calcium digestibility between challenged and nonchallenged birds that were fed with either artemisia or pumpkin seed on the concentrations used (P = 0.034 for the interaction). Phosphorus digestibility was not affected by either challenge or treatment.

There were no statistically significant differences between the treatments and the untreated control for any of the cytokines evaluated in week 25 (data not shown). The relative gene expression of the cytokines in 35-wk-old birds is shown in Table 6 and Supplemental Table S5. IL-8 expression was significantly upregulated in birds fed pumpkin seed compared to control birds or birds treated with artemisia (P = 0.001). For IL-13 expression a significant interaction between treatment and challenge was observed (P = 0.013). Only in artemisia-treated birds was relative IL-13 expression significantly higher in infected birds, while for untreated birds or birds treated with pumpkin seed infection did not impact the gene expression. TGF-β expression was significantly higher in artemisia-treated and infected birds than in uninfected control birds (P < 0.05). Differences in relative expression of IL-10 and IFN-γ were not significant.

The aim of this study was to test the impact of A. absinthium and pumpkin seed powder on the potential negative effects of an A. galli infection on worm burden, egg production, and quality, nutrient digestibility, as well as changes in cytokine gene expression. To our knowledge, this is the first time that the impact of an infection with A. galli together with potential treatments with A. absinthium and pumpkin seeds on digestibility of nutrients were evaluated.

An impact on overall bird performance was not noticed, due to neither the challenge nor the different treatments. It was interesting to see a minor, not significant decrease in egg production of the challenged groups 2 wk after the challenge and again 5 wk p.i. These two times coincide with the beginning and end of the histotrophic phase of the parasite (33). This is not the first time that no changes in egg production were observed after an infection with A. galli (34,35).

However, most egg quality parameters were influenced by treatments and/or infection. Regarding the treatments, results tended to be less favorable for artemisia-treated birds, specifically resulting in lower egg weights and weights of the individual components. The effect of the infection was obscured by several interactions between treatment and infection status, indirectly indicating that the treatments influenced the outcome of the infection. Inclusion of more parameters like content of the eggs might have resulted in more significant differences between groups due to the infection (36).

The effectiveness evaluation of the treatments against A. galli infection showed that they were not able to affect the number of birds tested positive for worms. In fact, 81% of birds treated with pumpkin seeds and 66% of birds treated either with artemisia or untreated control had adult worms in the intestine. Additionally, worm egg shedding in both treatments had a numerically higher peak.

Overall, challenge significantly decreased digestibility of energy and crude protein. This effect was largest and significant in birds fed artemisia, so artemisia treatment seemed to make the effects of the challenge worse. Although A. absinthium has many beneficial properties, it also has some toxicity, which might have increased the negative effects of the challenge (37). The effect might also be related to the numerically higher worm counts in artemisia-fed birds, but there was no such effect in birds feed pumpkin seed; on the contrary, the negative effect of the challenge seemed to be lowest in these birds. The decreased energy and crude protein digestibility did not result in decreased egg production. Possible compensation mechanisms are increased feed intake or using body energy and nutrient resources (38).

Digestibility of calcium and phosphorus was also numerically decreased in challenged birds. Generally, the low number of replicates, necessitated by the little intestinal content in laying hens compared to broilers, allowed only a low statistical power. Further experiments with higher replicate numbers should revisit this topic, because a potentially decreased digestibility of the minerals might contribute to a negative effect of A. galli infections on egg production. It is noteworthy that birds fed artemisia had a higher calcium digestibility, which might help to compensate this negative effect of A. galli.

IL-13, related to T helper type 2 immune response, was the only cytokine that showed a more expressive upregulation in birds infected with A. galli. This agrees with previous results investigating the immune response against A. galli carrying larvae as well as adult worms (27,39,40,41). However, for unclear reasons, this difference was only observed in artemisia-treated birds. It might be related to anti-inflammatory properties of A. absinthium that have been previously observed in mice (42).

The other investigated cytokines were not affected by the infection. This can be explained by the fact that the adult worms are free floating in the lumen, in contrast to the larvae that are attached to the mucosa during the histotrophic stages, which will result in a more severe inflammation (27,43). In addition, the remaining worms found at necropsy 2–3 wk after peak egg shedding and after most worms were expulsed (4) were not sufficient to stimulate a strong immune response. A sampling time point closer to the infection might have resulted in more significant changes (39,44).

Besides the ambiguous effect of artemisia on IL-13 gene expression, the only observed immune modulation by a treatment was an upregulation of the gene expression of IL-8, a pro-inflammatory cytokine. If and how this is related to positive effects of pumpkin seed is unclear.

The study had several limitations due to the limited resources that could be dedicated to it. One limitation of the study was that the worm burden might have been underestimated across all groups because ileal ingesta was collected for the digestibility analysis and could not be tested as systematically as the jejunum. In addition, reinfection might have taken place during the trial, which could have been revealed by investigating the intestinal walls for larvae.

Finally, it is important to point out that in this study the simplest processing method for the phytogenic treatments was used, i.e., mixing the powder in the feed, to simulate the preferred procedure of most backyard producers. This means that in both treatments all compounds with a potential antiparasitic effect were present in lower concentrations than in extracts that were used in other research. This can be the reason why not many differences were seen in our experiment, but their aqueous and alcoholic extracts did show a potential in other hosts (14,17,45,46). Coccidiosis was detected up to 4 wk p.i., i.e., 6 wk before the end of the trial. Since this is a self-limiting disease and it was detected in all groups, we do not expect a distortion of the results.

In conclusion, the infection with A. galli did not affect egg production or most egg quality parameters during the period evaluated. However, the infection with A. galli did impair nutrient digestibility. Neither treatment was effective against A. galli.

Supplemental data associated with this article can be found at https://doi.org/10.1637/aviandiseases-D-24-00052.s1.

AID =

apparent ileal digestibility;

AIDE =

apparent ileal digestible energy;

CP =

crude protein;

EPG =

worm eggs per gram feces;

GE =

gross energy;

HU =

Haugh unit;

p.i. =

postinfection