Comparative study of eggshell antibacterial effectivity in precocial and altricial birds using Escherichia coli

In this study, we compared the antibacterial effectivity of the eggs of six precocial and four altricial bird species using Escherichia coli, based on their eggshell traits. The ultrastructure of eggshell was observed using a scanning electron microscope (SEM). According to SEM results, eggs from precocial birds (chicken, turkey, quail, duck, ostrich, and goose) had cuticle on the eggshells, while eggs from altricial birds (pigeon, budgerigar, munia, and canary) did not. The environment/selection pressure may induce the divergent evolution process in eggs of precocial and altricial birds. The E. coli experiment results showed that chicken, turkey, quail, duck, and goose eggs, with a high cuticle opacity, exhibited a much lower E. coli penetration rate. In contrast, the eggs with poor (ostrich) or without (pigeon, budgerigar, munia, and canary) cuticle exhibited a higher penetration rate. It is suggested that cuticle is a main barrier against bacterial penetration in precocial birds’ eggs. Turkey and quail eggs showed the lowest E. coli contamination rate (3.33% and 2.22%, respectively), probably because of the tightly connected nanosphere structure on their cuticle. As for altricial birds’ eggs, the eggs of budgerigar, munia, and canary with small pore diameter (0.57 to 1.22 μm) had a lower E. coli penetration rate than pigeon eggs (45.56%, 66.67%, 50%, and 97.78%, respectively, P < 0.05), indicating that pore diameter played a significant role in defending against bacterial trans-shell invasion. We found that eggshell thickness and pore area decreased with egg size. The cuticle quality had no relationship with egg size, but was closely related to the bird species. The E. coli penetration rate of altricial birds’ eggs was significantly higher than that of precocial birds’ eggs, mainly because the pores are exposed on the eggshell surface and cuticle protection is absent. This study provides detailed information on the eggshell cuticle, which gives insight into the cuticle evolution process that occurred in precocial and altricial bird species. Moreover, the results of E. coli penetration may help understanding the antibacterial behavior in birds.

The Experimental Unit of China Agricultural University (Beijing, China) supplied chicken eggs. The National Center of Performance Testing of Poultry (Beijing, China) supplied eggs for the remaining species. All birds used in this study were healthy and had not previously experienced antibacterial treatment. Housing, management, feeding, and husbandry conditions were consistent with the recommendations provided by the poultry companies. All birds were fed by the guidelines of the poultry companies that ensure the nutritional requirements of the individual species are met. Ninety eggs were collected per species except ostriches, for which 30 eggs were used. All eggs were collected within 24 h of laying and were tested within 48 h. Eggs were visually inspected through candling to select intact eggs (i.e., no cracks or pinholes). Then, those eggs were weighed on a small electronic scale (AL240, 0.01-210 g, Mettler Toledo, Shanghai, China) except for the ostrich eggs, which were weighed on a larger electronic counting scale (ACS-3, 10 g-3 kg, Hua Chao, Shanghai, China). A line was drawn along the long axis of the egg with a pencil. One side was marked as E (Escherichia coli inoculation) and the other side marked as S (stained with MST cuticle blue, MS Technologies Ltd, UK).

Assessment of cuticle deposition
The method for evaluating cuticle quality was described by Chen et al. [15]. Cuticle quality was evaluated based on differences in cuticle opacity before and after staining. A high alpha (α) value denoted high staining affinity, implying more cuticle deposition. Cuticle opacity was calculated using Eqs 1 to 7. Each egg was measured at three points: the blunt end, equator, and sharp end. Cuticle quality per egg was determined from the mean value of these points.
First, X, Y, and Z values of the eggshell S surface were obtained with a spectrophotometer (CM-2600d; Konica Minolta, Japan) using the XYZ color space systems. Subsequently, eggs were immersed in MST cuticle blue for 1 min, rinsed in clean water to remove excess stain, and placed on a plastic flat to dry for 24 h. Finally, X, Y, and Z values of stained eggshells were measured using the procedure described above.
where RGB is an additive color model in which red (R), green (G), and blue (B) light are added together in various ways to reproduce a broad array of colors. In the XYZ model, Y is luminance, Z is quasi-equal to blue stimulation, and X is a mix (a linear combination) of cone response curves. XYZ color space can be converted into RGB color space by the formulas above, and then converted into opacity. α is the cuticle opacity, subscript "a" indicates egg values post-staining, "b" indicates pre-staining values, and "d" indicates values for the cuticle blue dye.

Escherichia coli inoculation
The E. coli penetration assays were conducted following previously described methods [6]. The E. coli strain HB-101 K-12 with pGLO (Bio-Rad Laboratories Co., Ltd., Shanghai, China) was grown in lysogeny broth (containing 100 μg mL -1 ampicillin Na and 5 mM L-arabinose; Sigma-Aldrich, Shanghai, China), and shaken overnight at 37C. Cultures were inoculated at a dilution of 1:50 in fresh lysogeny broth, then grown to an OD 600 of approximately 0.4 at 37C. Thereafter, the broth was placed into an ice bath. Eggs were lightly swabbed with ethanol, then incubated in a sterile, plastic egg box for 2 h at 37C. Then, the E side of the egg was immersed in ice-chilled broth for 10 min inoculation. Subsequently, the inoculated eggs were placed on a sterilized bench, to allow the culture to dry on the eggshell surface for about 15 min. Finally, each egg was transferred individually into a new sterile plastic bag and incubated for 24 h at 37C.
All eggs were removed from the incubator and placed on a sterile clean bench at room temperature for 2 h. Egg contents were drained through a hole of approximately 1 cm 2 , prepared with a rotary tool (Dremel, S-B Power Tool Company, Chicago, IL, USA). The egg was cut into two halves along the longitudinal axis with the rotary tool. The presence of E. coli was determined by observing the luminescent spots on the inner eggshell surface under a longwave UV light (LUYOR-1144A, LUYOR Corporation, Shanghai, China). The number of E. coli in the egg was determined by the number of luminescent spots on the inner surface under UV light. Contaminated eggs were grouped into three levels: light (with 1-3 luminescent spots), moderate (with 4-10 luminescent spots), and severe (with more than 10 luminescent spots).

Measurement of eggshell thickness and pore density
The eggshells were characterized by the traits of eggshell thickness, pore diameter, and pore area. Eggshell thickness and pore density were measured at the equator of eggs. Eggshell thickness was determined using a digital display micrometer gauge (Mitutoyo, Kawasaki, Japan). Pores were counted following published methods [21,22]. Eggshell fragments of chicken, turkey, quail, duck, goose, quail, pigeon, and ostrich were boiled in 1% KOH to remove the inner eggshell membrane and outer cuticle. Immersion time (~3 min for chicken, 4 min for duck and turkey, 5 min for goose, 90 s for quail and pigeon, and 8 min for ostrich) was dependent on eggshell thickness. Eggshell membranes of budgerigar, munia, and canary were too thin to boil so they were carefully removed using tweezers. Subsequently, all eggshells were rinsed in clean water for 1 min, immersed in 75% ethanol for 1 min, and then immersed in 1% HCl for 20 s to enlarge eggshell pores. Finally, eggshell fragments were rinsed in clean water for 1 min and placed on a table to dry. Eggshell inner surfaces were painted with 0.2% methylene blue (Solarbio, Beijing, China). Once the dye was dry, pores within a 0.25 cm 2 area were counted under a microscope.

Eggshell ultrastructure
SEM (S-3400 N and SU8010, Hitachi, Japan,) was used to observe ultrastructural features of the outer surface layer, cross-section, and cuticle layer of eggshells [23]. Ten eggs per species were used to measure cuticle layer thickness. Pore diameter and area of single pores were determined through SEM from the same eggshells as were used for pore density calculation. To make the results more reasonable and more accurate, we measured the single pore area and the short axis diameter of the pore (in case the pore shape was not a circle) for eggs of ten bird species. Five eggs per species were prepared, and for each egg three pieces were taken around the equator. Eggshell pieces were mounted on an aluminum stub and gold sputter-coated using an EIKO IB-3 (EIKO Engineering CO., Ltd, Japan) for about 15 min. Thereafter, they were viewed and photographed under the SEM.

Statistical analysis
The pore area was measured using the software Image J. All statistical analyses were performed with the statistical software RStudio (version 3.4.0, 2017) and figures were plotted in Origin Pro 2018. Descriptive statistics and one-way ANOVA were used to analyze eggshell thickness, pore density, pore diameter, cuticle layer thickness, and cuticle opacity per species. The model used for the one-way ANOVA analysis was: where Y ij represents traits (eggs weight, eggshell thickness, cuticle opacity, cuticle layer thickness, pore density, pore area, and pore diameter) analyzed in this study, μ represents the common effect for each test, τ i represents the species-specific or bird types-specific (precocial vs altricial birds) main effect (factor), and � ij represents individual-specific random error. The significance level chosen for all analyses was P < 0.05.
Cuticle ultrastructure. Shell cross-sections of chicken, turkey, quail, duck, goose, and ostrich eggs could be divided into mammillary, palisade, and cuticle layers (Fig 1), whereas pigeon, budgerigar, munia, and canary eggshells lacked the cuticle layer (Fig 2). Compared with eggs of other species, quail eggs had the thickest cuticle layer. Ostrich eggshells had thin and patchy cuticle layers. Quail and turkey eggs had much more tightly connected cuticle nanospheres, in contrast with chicken and goose eggs. Ostrich eggshells also had cuticle nanospheres at larger magnification. Interestingly, duck eggs seemed to have highly atypical cuticle nanospheres compared with the other species. The pores would be an opening for bacteria, if there were no cuticle coverage on the eggshell (Fig 3). Eggshell pores. Goose and ostrich eggs exhibited irregularly shaped pores, while eggs of the remaining tested species had nearly round pores (Fig 4). To make the results more precise and comprehensible, the pore diameter of different species was measured in different ways. Goose egg pores were somewhat like a "crescent moon", therefore, goose eggs' pore diameters were measured by the short axis. Ostrich eggs showed many irregular porosities; hence, the pore diameter of ostrich eggs was determined by the diameter of average pore diameter of each small pores and the sum of small pores' areas was used as the single pore area. Duck eggs had the largest pore diameter (17.21 ± 5.48 μm). Budgerigar, munia, and canary eggs had honeycomb-like pores, and the pore diameter and pore area of those species' eggs was significantly smaller (0.57 to 1.22 μm, P < 0.05) than that of the eggs of other species. Ostrich eggs had the lowest pore density (7.17 ± 215 per cm 2 ) among all tested eggs, coupled with the largest pore area (23148.14 ± 9891.43 μm 2 ). Goose eggs had a larger pore area than the chicken, turkey, quail, duck, and pigeon eggs.
Eggshell thickness. The ostrich egg had the thickset eggshell (1955.51 ± 93.99 μm), followed by goose eggs (486.94 ± 60.30 μm) ( Table 1). The eggshell thickness of budgerigar, munia and canary eggs was much smaller (57.18 to 92.61 μm, P < 0.05). The eggshell of the Egg size. Eggs size was determined by egg weight as shown in Table 1. Ostrich eggs had the largest egg size (1357 ± 82.20 g), while budgerigar, munia, and canary egg weight ranged  from 1.23 to 2.02 g, which was significantly smaller than that of the other eggs (P < 0.05). The eggshell thickness and pore area were positively correlated with egg size (Fig 5). The pore density seemed to decrease with increasing egg size. Cuticle opacity and cuticle layer thickness were not significantly correlated with egg size.

E. coli penetration
Among all the investigated species, the egg penetration rates for pigeon and ostrich were 97.78% and 76.67%, respectively, which was much higher than that for the other species (Table 1). Further, most pigeon eggs and ostrich eggs were severely contaminated by E. coli (Fig 6). Quail and turkey eggs had a lower contamination rate (2.22% and 3.33%, respectively), and these eggs were lightly and moderately contaminated. The penetration rates for munia, budgerigar, and canary eggs was 68.33%, 46.67%, and 50%, respectively.

E. coli penetration and eggshell traits
The influence of eggshell traits on the antibacterial effectivity is shown in Fig 7. From Fig 7, it can be seen that eggs with a thick cuticle layer and high cuticle opacity tend to have a lower E.  Eggshell antibacterial effectivity of birds coli penetration ratio, proving that for birds, the cuticle is an effective barrier against bacterial penetration. As for the single pore area, pore density, and eggshell thickness, none of those exhibited a consistent trend with bacterial penetration among these ten bird species. Eggshell antibacterial effectivity of birds Precocial and altricial birds' eggs. At 65%, altricial birds' eggs showed a much higher E. coli penetration rate than precocial birds' eggs (13.96%) (P < 0.05) (Fig 8). Eggshell thickness, single pore area, and pore diameter of precocial birds' eggs were significantly larger than those of altricial birds' eggs (P < 0.05). There was no significant difference between precocial and altricial birds' eggs for pore density (P > 0.05).

Discussion
Chicken, turkey, quail, duck, and goose eggs with high cuticle opacity had much lower E. coli penetration rate than ostrich, pigeon, budgerigar, munia, and canary eggs with a poor or no cuticle layer (Table 1). It showed that eggshell cuticle can effectively defend against bacterial penetration [5,6,15,24]. Besides, from Fig 7, it can be predicted that if the cuticle opacity of a bird species' eggs is higher than 25.34 or their cuticle layer is thicker than 5.6 μm, the eggs' E. coli penetration rate would be lower than 15.56%. Quail and turkey eggs had a lower penetration rate, which may be due to the cover of tightly connected nanospheres on the eggshell [11]. It has also been reported that the cuticle of chicken had moderate heritability [6,25]. These results suggest that we can improve cuticle quality by breeding to enhance the antibacterial ability of eggshell.
Scanning electron micrographs indicated that precocial birds' eggs (chicken, duck, goose, turkey, quail, and ostrich) had a cuticle layer on their eggshell surface, whereas altricial birds' eggs (pigeon, budgerigar, munia, and canary) did not. It was reported that cuticle evolution was related to environmental pressure [6,8,26]. As we know, high ambient humidity is more conductive to microbial growth [27]. Birds nesting in habitats with higher infection risk (e.g., wetter and warmer) were more likely to evolve a cuticle layer on their eggshells than those nesting in lower-risk habitats [9,28]. Our results that goose and duck eggs, which are always laid in a swamp environment, have a thicker eggshell cuticle than chicken and ostrich eggs, which are usually laid in low ambient humidity, matched well with what has been reported previously. Eggshell antibacterial effectivity of birds Most altricial birds perch on trees, while most precocial birds nest on the ground. Moreover, altricial mothers, such as Passeriformes and Columbiformes, are generally so careful in hatching eggs that they use their feathers or the epidermal layer of their brood patches to inoculate shells with antibiotic agents, reducing the possibility of microbial infection [29,30,31,32], while precocial birds like Anseriformes always hatch in muddy places where the eggs are at high risk of being contaminated by pathogenic microorganism. According to the theory of evolution, after a long time of natural selection, the cuticle quality of altricial bird eggs seems to be getting worse and some even lack cuticle deposition on the eggshell surface. Certainly, this hypothesis needs to be verified with more eggs from different bird species along with an evolutionary genomics analysis.
Absence of a cuticle layer on the eggshell surface totally exposes the pores (Fig 3), which would increase the possibility of microbe contamination [7,18], leading to higher E. coli contamination rates in altricial bird eggs. Budgerigar, munia, canary, and pigeon eggs all lack cuticle deposition on the eggshell; however, eggs of the former three species had a far lower penetration ratio than pigeon eggs. The results can be explained by the relationship between pore diameter and average bacteria size. The average diameter of spherical bacteria is 0.5 to 2.0 μm [33,34], and rod-shaped or filamentous bacteria are 1-10 μm long, with diameters of 0.25 to 1.0 μm [35]. Pore diameters for budgerigar, munia, and canary ranged from 0.57 to 1.22 μm (Table 1). Bacteria have more difficulty in penetrating through the eggshell if the pore diameter is quite close to or even smaller than their size, implying low contamination. In other words, when the eggs' pore diameter or single pore area is quite close to or even smaller than the size of bacteria, it could be an effective protection against bacterial trans-shell penetration. A large pore diameter on eggshell would make bacteria penetrate more easily. This is why goose eggs, with a large pore area, have a higher bacterial penetration ratio than quail and turkey eggs despite the high cuticle opacity of goose eggshell.
Ostrich eggs, with large pore diameter and large pore area, have a quite high contamination rate in this study. However, in natural habitats, the ostrich's bacterial contamination risk is not so high [36]. This can be attributed to the low ambient humidity and high ambient temperature in the ostrich habitat, which is not conducive to bacterial reproduction [37,38]. Additionally, the thick ostrich eggshell (almost 2 mm) may be an effective physical barrier against trans-shell bacterial contamination [15,19]. A thick eggshell indicates a long trans-shell distance (Fig 3), which can increase the difficulty and penetration time for bacterial trans-shell invasion.
Pigeon eggs, which lack cuticle and have a large pore diameter, exhibited the largest E. coli penetration ratio among all tested eggs, and 86.67% of pigeon eggs were severely contaminated in this study. Egg antibacterial behavior is usually achieved through a very complex mechanism. Embryo protection is not only ensured in the first place by a physical barrier, the shell, but also by a complex system of chemical defenses including antibodies and a variety of antimicrobial proteins in egg albumen and the eggshell membrane. Our team's previous research found that pigeon egg albumen contained higher concentrations of lysozyme C, ovalbumin, ovotransferrin, and heptoglobin than quail, turkey, duck and goose eggs [39]. Lysozyme, ovalbumin, and ovotransferrin play an important role in antimicrobial activity [40,41,42]. Heptoglobin, a natural bacteriostat [43,44], was found only in pigeon eggs, not in any other avian eggs [39].
In this study, the eggshell's antibacterial effectivity in precocial and altricial birds was investigated. As discussed above, we can argue why altricial bird's eggs have higher E. coli contamination rates than precocial bird's eggs in this study. First, and most importantly, altricial birds' eggs lack the protection of cuticle. Furthermore, the lack of cuticle leaves the pores of altricial birds' eggs open on the eggshell surface, increasing the risk of trans-shell invasion by microbes.
In addition, their thin eggshell might be another reason for the high contamination rate in altricial birds' eggs.

Conclusion
This comparative study measured the eggshell antibacterial effectivity in six precocial and four altricial birds' eggs by focusing on morphological features (cuticle quality, pore density, pore diameter, and eggshell thickness). It was found that the cuticle can be an effective barrier against bacterial penetration. Precocial birds' eggs with the protection of cuticle had lower E. coli penetration rates than altricial birds' eggs with no cuticle layer on the eggshell. The evolution of cuticle may be related to the environment/selective pressure. Besides cuticle quality, the pore diameter plays an important role in the antimicrobial process of bird eggs as well. The results in this study provide insight into the eggshell antibacterial effectivity in precocial and altricial birds' eggs, and can be a reference for antibacterial studies in eggs of various bird species in the future.