Association of single nucleotide polymorphism at long non-coding RNA 8138.1 with duration of fertility in egg-laying hens

Ethics statement

The protocols for all animal experiments were approved by the Scientific Ethic Committee of Huazhong Agricultural University with approval number HZAUCH-2018-005.

Egg-laying hens and duration of fertility trait

Commercial Jing-Hong laying hen populations were utilised in this study. Laying hen population 1 (P1) was obtained from the poultry farm of Huadu Yukou Poultry Industry, Co. Ltd, Beijing, China and population 2 (P2) from Jingzhou Yukou Poultry Industry, Co Ltd, Jingzhou, China. Briefly, the egg-laying hen populations were from two elite breeding lines, line 1 (white egg) and line 2 (brown egg). All hens were kept under standard conditions from 25 weeks aiming to study their duration of fertility. Hens were inseminated once with 2 ×10⁸ pooled sperms ejaculates collected from viable rooster flocks. Eggs were collected and marked daily from day 2–20 after artificial insemination (AI); all hens completed a reproductive phase in three replicates and lasted 60 days. The number of egg per hen over the period was recorded, and the fertilized eggs were examined by candling on day 10 of incubation (dead embryos were considered as fertile). The reproductive history of all hens was recorded daily based on DF-trait: EN (number of eggs), FN (the number of fertile eggs after a single AI), and DN (the number of days post-insemination until last fertile egg). At the end of the reproductive season, a total of 1,868 hens had a record of EN, FN, and DN respectively.

DNA extraction and quality assessment

For DNA experiments, genomic DNA was isolated from 0.5 ml blood obtained from the egg-laying hen populations (P1, n = 1,042 and P2, n = 826) by phenol-chloroform extraction method (Sambrook & Russell, 2006). The quality of genomic DNA samples was rated using Nanodrop 2000 Spectrophotometer (Thermo Scientific™ ND2000USCAN; Waltham, MA USA). DNA samples were diluted to a working concentration of 50 ng/µL and stored at 20 °C for further analysis.

Primer design, PCR amplification, and SNP detection

LncRNA8138.1 is a double-exon and a single-intron gene at chromosome 18: (4936949–4937389, 4937470–4938607 bp) (GenBank accession number: MK336169). Primer pairs (AACCTGAGCTTTCAACAGAC & CCCAACTGCTCCAACATTAG) were designed using Primer Premier Software version 5.0 for amplification of lncRNA8138.1. Assays were performed by PCR in DNA samples of 60 hens selected randomly to construct a DNA pool with equal DNA concentration of 50 ng/µl for each hen. PCR amplifications were performed in a 25 µl volume containing 50 ng pooled DNA, 2.5 µl of 10 × PCR buffer, 5 mM of dNTPs, 10 pmol of forward and reverse primer, 0.625U Taq DNA polymerase (Takara Biomedical Technology Co., Ltd. Beijing, China) and ddH2O. The reactions were performed under the following conditions: an initial pre-denaturing at 95 °C for 5 min; 35 cycles of 95 °C (20 s), 55 °C–60 °C (20 s) 72 °C (20 s); and a terminal incubation step at 72 °C for 5min. The PCR products were sequenced directly with Applied Biosystems 3730xl DNA Analyzer (Applied Biosystems^®, Waltham, MA USA), and SNP discovery was conducted by the Seqman program of DNASTAR 7.1.0 software (DNASTAR, Inc., Madison, WI, USA).

PCR amplification and polymorphism genotyping

The region containing the lncRNA8138.1 polymorphism was amplified in a single PCR system with the necessary reagents, and the reaction conditions were identical to those described above. Genotyping of PCR products was performed by PCR-RFLP assay in both P1 and P2 individuals using the primer pairs (AGTCACAGACCAGTAGTTTT & CCTCTAAAA

TCTTAGCAGCA). The PCR-RFLP condition for amplifying was pre-denaturing at 95 °C for 3 min, and 32 cycles of denaturation at 95 °C for 30 s, annealing at 64 °C for 30 s, extension at 72 °C for 20 s, and elongation at 72 °C for 7 min (Adeyinka et al., 2017). Next, PCR-RFLP products were digested with an enzyme (Hind III) and electrophoresed using 3% agarose gel.

Mutational analysis and structure prediction

Predicting the common secondary structure of lncRNA is an important step towards understanding the function (Blythe, Fox & Bond, 2016). Here, we experimentally predict the secondary structure of lncRNA8138.1 r.4937159A>G polymorphism in chicken, which may be used to upgrade sequence/structure alignments for other species. DNA samples of hens were selected randomly to construct a DNA pool with equal DNA concentration of 50 ng/µl for 60 individuals having AA and GG genotypes. PCR amplification reactions were performed in a 20 µl volume containing 50 ng pooled DNA, 2.5 µl of 10 × PCR buffer, 5 mM of dNTPs, 10 pmol of forward and reverse primer, 0.625U Taq DNA polymerase (Takara Biomedical Technology Co., Ltd. Beijing, China) and ddH2O. The reactions were conducted under the following conditions: an initial pre-denaturing at 95 °C for 5 min; 35 cycles of 95 °C (20 s), 55 °C–60 °C (20 s) 72°C (20 s); and a terminal incubation step at 72 °C for 5min. The PCR products were sequenced with ABI3730xl DNA analyzer (Applied Biosystems^®, Waltham, MA, USA). The resulting nucleotides from the genome of AA and GG genotypes were used to predict the common secondary structure of lncRNA8138.1 in chicken with the use of web-based structure prediction programs known as Dynalign (http://rna.urmc.rochester.edu/RNAstructureWeb/Servers/dynalign/dynalign.html) (Reuter & Mathews, 2010).

Statistical Analysis

Descriptive statistics for the DF-trait were calculated using an online based descriptive statistics calculator. The SNP genetic parameters including homozygosity (Ho), heterozygosity (He; Ho + He = 1) and polymorphism information content (PIC) were estimated according to methods developed by (Nei & Roychoudhury, 1974). PIC is an indicator of polymorphism that can be used to estimate genetic diversity. According to PIC values, the genetic variations were classified as high PIC (>0.5), medium (0.25 < PIC < 0.5) and low (PIC < 0.25) (Botstein et al., 1980). Genotypes were tested for Hardy-Weinberg equilibrium using the Chi-squared tests (R function HWE.chisq) implemented in the R environment (v 3.0.2) (Pouresmaeili et al., 2013). The statistical significance level was set at P <0.05. The relationship between lncRNA8138.1 polymorphism and DF-trait in P1 and P2 were estimated based on the general linear model procedure of SAS v. 9.2 (SAS Institute Inc., Cary, NC, USA) (Littell, Stroup & Freund, 2002). Briefly, the following linear model was used: ${\displaystyle Y_{\mathrm{ijklm}}=\mathrm{\mu }+B_{\mathrm{i}}+G_{\mathrm{j}}+A_{\mathrm{k}}+L_{\mathrm{l}}+e_{\mathrm{ijklm}}}$

where: Y_ijklm represents the dependent variable (DF-trait), µ is the overall population means, B_i is the fixed effect of ith breed (i = 1), G_j is the fixed effect of jth genotype (j = AA, AG, and GG), A_k = is the fixed effect of kth age (k = 1), L_l is the fixed effect of lth location associated with the DF-trait, e_ijklm is the random error. Differences in means were considered significant at P <0.05 and the SAS PROC t-test (SAS v. 9.2; SAS Institute Inc., Cary, NC, USA) was used to estimate the differences in the LS Means between genotypes. Additive and dominance effects were estimated using REG procedure in (SAS v. 9.2 (SAS Institute Inc., Cary, NC, USA)). Statistically significant differences in the investigated populations for additive and dominance effects were estimated as a t-test for LS Means.

Descriptive statistics of laying hens duration of fertility

During the reproductive season, DF-trait was defined by the EN (number of eggs), FN (the number of fertile eggs after a single AI) and DN (the number of days post-insemination until last fertile egg). The descriptive statistics of the recorded DF-trait in P1 (n = 1, 042) are shown in Fig. 1A. In the P1, the median EN, FN and DN were 14.67, 9.33 and 11.33 respectively and interquartile range (IQR) for the EN, FN, and DN were (13.33–15.67 eggs), (7.67–10.67 eggs) and DN, (10.67–11.67 days) respectively (Fig. 1A). Figure 1B illustrates the descriptive statistics of the recorded DF trait in the P2 (n = 826); the median EN, FN and DN were 15.33, 9.33 and 10.33, respectively. The IQR for the EN, FN and DN were (14.00–20.00 eggs), (7.33–11.17 eggs), and (8.33–12.33 days) respectively (Fig. 1B). In general, DF-trait showed high individual variability in both P1 and P2, even when we observed the same and optimal multiple insemination conditions.

LncRNA8138.1 genetic polymorphisms

In this study, approximately 1.6 kb nucleotides of lncRNA8138.1 were analysed. The analyses revealed a set of four possible SNP candidates and 7 nt indel in egg-laying hens that were detected through sequencing, including SNP r.4937159A>G; r.4937219T>C; r.4937258G>C; r.4937318C>G and g.4937319_4937325delinsTGTGTGG (Fig. 2). Several previous studies have reported that variants associated with lncRNA cannot be undervalued (Bhartiya & Scaria, 2016; Hrdlickova et al., 2014). Therefore, r.4937159A>G was detected by PCR-RFLP; the relationship with chicken DF-trait was further investigated in large egg-laying hen populations.

Genetic parameters of the detected polymorphism

Table 1 showed the genotype frequency of lncRNA8138.1 r.4937159A>G polymorphism in egg-laying hen populations. The χ² test indicated that r.4937159A>G genotype frequency was not in agreement with HWE (P <0.05) in both P1 and P2 (Table 1). In the P1, data collected for r.4937159A>G polymorphism showed that “G” allele (0.591) was more frequent than “A” allele (0.409); and “G” allele (0.530) was often seen compared to “A” allele (0.470) in P2. The genotype frequencies of the majority of hens in P1 were either heterozygous AG (0.607), while (0.106) were homozygous AA and (0.287) were homozygous GG. In the P2, the majority were also heterozygous AG (0.708), (0.176) were homozygous AA, and (0.116) were homozygous GG (Table 1). In addition, polymorphism information content values revealed the genetic diversity of lncRNA8138.1 r.4937159A>G polymorphism (PIC = 0.366 and 0.374) in P1 and P2, respectively (Table 1).

Common secondary structures of lncRNA8138.1 mutation in chicken

To determine the common secondary structures of lncRNA8138.1 in chicken, the mutational profile of lncRNA8138.1 gene was utilized for the structure prediction. This is done over individuals having AA and GG genotypes separately, and the results showed that individuals with AA genotype differed in secondary structure by 2-point mutations when compared with the individuals with GG genotype. The 2-point mutations represent the AG_deletion in the AA individuals (Fig. 3A) and AG_insertion in the GG individuals (Fig. 3B); however, a consistent pattern of similarity was noticed in both AA and GG genotypes secondary structures (Fig. 3).

Analyses of associations between lncRNA8138.1 nucleotide variation and DF trait

Associations between lncRNA8138.1 r.4937159A>G polymorphism and DF-trait were investigated in the egg-laying hen populations. For all DF-trait (EN, FN and DN), LS Means and standard error of mean (SEM) values were tabulated for the genotype effect (Table 2). In the P1, association analysis confirmed that SNP r.4937159A>G was significantly related to DF-trait (EN: P < 0.0001, FN: P < 0.0001, and DN: P ≤ 0.05; Table 2). In the P2, a significant association was observed for the EN (P < 0.0001) and FN, (P < 0.01); Table 2). Besides, additive and dominance effects were significant for all DF-trait except DN in the P1 (P < 0.05, Table 2), whereas in the P2, additive effects were significant for EN and FN traits (P < 0.01, Table 2). Additionally, comparison of lncRNA8138.1 r.4937159A>G genotypes revealed significant differences between genotypes, i.e., (GG vs AA, GG vs AG, and AA vs AG) respectively (Table 2).