DISCUSSION

To evaluate the CEGA model as an in vivo NAM, it is critical to assess target tissue exposure. Since the injections in CEGA are done in the air space of fertilized eggs, there is a layer of eggshell membrane which xenobiotics must cross in order to reach the target tissue, which is liver. To demonstrate target tissue exposure and sufficient metabolism after reaching target tissue multiphoton microscopy, metabolite analysis, and genomic studies were conducted. For multiphoton microscopy, acridine orange was selected as an appropriate fluorescence dye since the molecular weight of acridine orange is in the range of B(a)P. The results indicated that even though the chemical was injected into the airspace, it penetrated the eggshell membrane and reached the liver (Figure 2), resulting in sufficient liver uptake. To study metabolism in CEGA, B(a)P was used as a chemical of choice, since it is known to produce its genotoxic effect post metabolic activation. Moreover, previous studies in CEGA confirmed that B(a)P forms DNA adducts and DNA strand breaks in the livers of chicken embryo-fetuses (Williams et al., 2014). B(a)P-7,8 dihydrodiol-9,10-epoxide is the reactive metabolite known to covalently bind to DNA forming adducts and producing adverse effects (Figure 1). Additionally, metabolism of B(a)P has been well documented in animal studies, which allows for comparison between the metabolites formed in the livers of chicken embryo-fetuses and other species. The characterization of metabolites formed in CEGA following exposure to B(a)P confirmed that the majority of these six metabolites were also observed in the rat (Figure 4), Two additional metabolites were formed in the chicken livers. These were identified as M378 and M300, and formed at 7.2 and 2.1%, respectively. M378 metabolite had an additional of three oxygen molecules (3O), removal of two hydrogen groups (-2H) and addition of one sulfate group (SO3), whereas M300 metabolite had additional three oxygen (3O) on B(a)P ring. However, M378 formation is also justified and follows similar pathway as mentioned in IARC document (IARC 1983) which states that B(a)P is metabolized initially by the microsomal CYP systems to several arene oxides. Once formed, these arene oxides may rearrange spontaneously to phenols, undergo hydration to the corresponding trans-dihydrodiols in a reaction catalyzed by microsomal epoxide hydrolase, or react covalently with GSH, either spontaneously or in a reaction catalyzed by cytosolic GST (IARC 1983). Overall, the findings in the metabolism study confirmed that B(a)P metabolism in CEGA aligns with the established metabolic pathway in rodents (Decker et al. 2009). Another similarity was observed with the animal study of orally administered B(a)P in F344 rats, which found a half-life of B(a)P in rat liver to be 12 hours, suggesting that unmodified/unmetabolized B(a)P will be 100% converted to its metabolites 24 hours post-exposure (Ramesh et al., 2001). In CEGA, 100% of administered B(a)P was converted to its relevant metabolites. The toxic precursor, (B(a)P-7,8-dihydridiol), was formed at 23.1% 48-hour post treatment in embryo-fetal chicken livers. In rat liver, however, B(a)P-7,8-dihydridiol was only present at 10% 48-hour post administration with its peak liver concentration at ~30% 24-hour post treatment after single dose of 100 mg/kg bw of B(a)P (Ramesh et al., 2001). Formation of the reactive metabolite, B(a)P-7,8 dihydrodiol-9,10-epoxide, which in CEGA was formed at 20%, is likely to be responsible for formation of DNA adducts observed in the embryo-fetal chicken livers (Williams et al., 2014). The difference in the amount of metabolite formed in CEGA and rodent study mentioned above may also be due to single oral administration in Ramesh et al., 2001 study as compared to three different dose in CEGA studies. The analyses of the differential gene expression in the embryo-fetal chicken livers following exposure to B(a)P also confirmed that the compound upregulated the expression of genes responsible for its bioactivation (Tables 2 and 3, Figure 7). Specifically, out of 10 significantly deregulated genes with >log 2-fold changes, three were involved in the activity of CYP1A1 and CYP1A2 isoenzymes. Other identified genes were also involved in the activity of DNA strands break mediated gene H2AX and DNA repair mediated gene RNA polymerase II. These results support the conclusion that B(a)P at a total dose of 250 ug/ egg upregulates expressions of CYP1A genes which affects the activity of CYP1A1 (Suppl. Figure 1), leading to formation of reactive metabolite, BPDE, resulting in DNA damage, and activating DNA repair mechanisms. In addition to the upregulation of genes involved in CYP1A1 activity, an increase was also observed in the expression of genes which regulate CYP1A1 by negative feedback loop. Specifically, upregulation of AhR repressor and TCDD Inducible Poly (ADP-Ribose) Polymerase (TiPARP) genes (Table 2) can negatively regulate CYP1A1 activity. This also add to the fact that with B(a)P treatment there was significant increase in CYP1A1 activity, which is a critical enzyme for metabolizing B(a)P leading to toxic metabolite, but gene regulating CYP1A1 expression by negative feedback mechanism were also present in chicken livers. Similarities between the expressions of CYP1A1 and CYPA2 genes in the embryofetal chicken livers following dosing with B(a)P were observed with the published data in rodents and human cells. For example, expressions of CYP1A1 and CYP1A2 genes in the livers of Wistar rats that received B(a)P at a single dose of 150 mg/kg bw by oral gavage was significantly increased by 2990 and 27.7 folds, respectively(Dračínská et al., 2021) . In the study with human hepatocellular carcinoma cell line (HepG2) which was incubated with 2 µM of B(a)P, CYP1A1 showed a 93-fold and 79-fold increase in expression on microarray 12- and 24-hours post-dosing, respectively, whereas RNA-seq demonstrated a 199-fold (at 12 hours) and 214-fold (at 24 hours) increases in CYP1A1 expression (Van Delft et al.; 2012). In the human tissue organoid cultures, differentiated liver had significantly higher (24-fold) CYP1A1 levels compared to undifferentiated samples, at basal level. After exposure to 50 µM of B(a)P, induction of CYP1A1 in differentiated liver organoids was around double compared to that in undifferentiated organoids (~4500- and 2000-fold, respectively), relative to control. At 12.5 µM, a 287-fold change compared to undifferentiated control was observed only in differentiated organoids. Induction of CYP1A1 was also significant at both concentrations of B(a)P compared to differentiated control (Caipa Garcia et al., 2023). These results are consistent with findings in CEGA which demonstrated a fold change of 1024 folds for CYP1A1 activity combined (ENSGALG00000013402 and ENSGALG00000001325) and 16 folds for CYP1A2 activity (Table 2).