Association between sperm mitochondrial ND2 gene variants and total fertilization failure

ABSTRACT

The objective of this study was to explore the association of sperm mitochondrial ND2 (MT-ND2) gene variants with total fertilization failure (TFF). A retrospective comparative study of 246 cases of fresh in vitro fertilization (IVF) cycles or half-intracytoplasmic sperm injection cycles in the Han Chinese population was performed from July 2011 to May 2017. A total of 59 cases undergoing TFF, and 187 control cases with normal fertilization (fertilization rates >50%) were included. The sperm mitochondrial genovariation was determined using nested sequencing. A total of 32 homoplasmic variants and 47 heteroplasmic variants of MT-ND2 gene were observed in this study. There were no significant differences in the frequencies of the 32 homoplasmic variants of MT-ND2 gene between the TFF and control groups. A total of 53 pair-wise comparisons were performed, and the general characteristics of the IVF failure and control subjects were adjusted in logistic models. Data suggested that there were no significant differences in the frequencies of point 4914, 5320, and 5426 heteroplasmic variants of MT-ND2 gene between the TFF and control groups. In addition, no significant difference was observed in the frequency of mtDNA haplogroup D or haplogroup G between the IVF failure group and the normal fertilization group. This study suggests that the MT-ND2 gene variants might not be associated with TFF.

Abbreviations: ATP: adenosine triphosphate; dNTP: deoxy-ribonucleoside triphosphate; FADH2: flavin adenine dinucleotide; FDR: false discovery rate; FSH: follicle-stimulating hormone; IVF: in vitro fertilization; LH: luteinizing hormone; MTATP6: mitochondrially encoded ATP synthase 6; MTCYB: mitochondrially encoded cytochrome b; mtDNA: mitochondrial DNA; MT-ND2: mitochondrial ND2; NADH: nicotinamide adenine dinucleotide; ND2: NADH dehydrogenase subunit 2; OXPHOS: oxidative phosphorylation; PCR: single nucleotide polymorphisms; SNPs: single nucleotide polymorphisms; TFF: total fertilization failure

KEYWORDS:

• Chinese
• total fertilization failure
• mitochondrial DNA
• MT-ND2
• variant

Introduction

Infertility is currently a global health problem with an average estimated incidence of 10–15% among reproductive age couples (Chan et al. 2016). In China, its prevalence is 7–10% (Chen et al. 2016). With the development of assisted reproductive technologies, in vitro fertilization (IVF) is a way to help establish pregnancy, but total fertilization failure (TFF) still occurs in 5–15% of the cases (Krog et al. 2015). Fertilization is determined 16–18 h after insemination, by the presence of two pronuclei. The cases in which none of the oocytes are found to contain two or more pronuclei are defined as TFF (Bungum et al. 2010). Recently, TFF research has focused on factors such as sperm motility and quality, oocyte quality, and mitochondria. However, the association between the concrete mechanism of fertilization failure and specific variants of sperm mitochondrial genes has rarely been reported. A study showed that the sperm deformity index, abnormal acrosome rate index, and DNA fragmentation might cause fertilization failure (Jiang et al. 2016). Another study demonstrated that fertilization failure was associated with sperm parameters and oocyte quality (Peultier et al. 2015). In addition, MT-ND4 might be associated with Chinese male infertility (Ni et al. 2017). Our previous study suggested that MT-ATP6 variants might be possible causes of IVF failure, but MT-ND3 homoplasmic variant 10397 might help decrease the risk of IVF failure (Mao et al. 2015, 2016). In this study, we discuss the relationship between mitochondrial genes and fertilization failure, laying a foundation for studying the association between fertilization failure and the sperm MT-ND gene variants.

The sperm mitochondria are located in the sperm mid-piece. A single sperm contains approximately 22–72 mitochondria (Bahr and Engler 1970; Otani et al. 1988; St John et al. 2000; Gabriel et al. 2012). Mitochondria are the energy factories of a cell, playing a key role in energy production and maintenance of sperm motility (St John et al. 2000). In 1981, S. Anderson discovered that human mtDNA contains 16,569 base pairs, encoding 37 genes: 22 tRNA, 2 rRNA, and 13 peptide genes. The 13 proteins were the catalytic subunits of the enzyme complex containing ND1-ND6, ND4L, COXl- COX3, Cytb, ATPase6, and ATPase8, which are required for oxidative phosphorylation (OXPHOS), and they are a part of the electron transport chain (St John 2014). The OXPHOS system catalyzes the transfer of electrons from nicotinamide adenine dinucleotide (NADH) or reduced flavin adenine dinucleotide (FADH2) to molecular oxygen, the final electron acceptor. Electron transport is coupled to the transfer of protons that establishes an electrochemical gradient across the inner mitochondrial membrane, and this gradient is used to generate adenosine triphosphate (ATP) (Fromm et al. 2016). The ATP necessary for sperm motility is primarily derived from OXPHOS in mitochondria; therefore, OXPHOS and the ability to generate ATP directly affects sperm motility. Mitochondrial DNA (mtDNA) variants may lead to abnormalities in mitochondrial energy metabolism, thus, reducing sperm motility. The NADH dehydrogenase subunit 2 (ND2) is an mtDNA-coding gene and a subunit of NADH dehydrogenase. NADH dehydrogenase is the main component of complex I; it is directly involved in the electron and proton transfer in the respiratory chain to produce ATP through OXPHOS. In recent years, studies have also shown that both the deletion of and presence of variants of mitochondrially encoded cytochrome b (MTCYB) and mitochondrially encoded ATP synthase 6 (MTATP6) in sperm mitochondria might affect the sperm motility in adults (Feng et al. 2008). However, the specific mechanism of fertilization failure, and whether a mitochondrial gene-specific locus variant is a fertilization failure risk factor or an avoiding factor, is not clear.

Single nucleotide polymorphisms (SNPs) often determine individual phenotypes, drug reactions, or susceptibility to a disease. Studies have suggested a relationship between mitochondrial gene variants and haplotypes in many diseases. A previous study showed that mtDNA variants (m.4435A>G) had a correlation with hypertension (Lu et al. 2011). It is also important to note that mtDNA haplogroups have been associated with various diseases, including Alzheimer’s disease (van der Walt et al. 2004), Parkinson’s disease (van der Walt et al. 2003; Huerta et al. 2005; Coskun et al. 2012), osteoarthritis (Fernández-Moreno et al. 2012), type 2 diabetes (Achilli et al. 2011), and various cancers (Czarnecka and Bartnik 2011). In this study, we try to further investigate the relationship between the human sperm mitochondrial ND2 (MT-ND2) gene variants and TFF, in an effort to aid the clinical prediction of IVF failure.

Results and discussion

Relationship of MT-ND2 gene variants with TFF

There were no termination codon and frameshift variants detected in the MT-ND2 gene analyzed in this study. Furthermore, the polymerase chain reaction (PCR) product was sequenced and analyzed using BLAST, revealing that the sequence of the PCR product was identical to that of MT-ND2 gene. This suggested that we could rule out co-amplification of nuclear pseudogenes that might interfere with this study. Our research found 32 homoplasmic variants in the MT-ND2 gene, including 7 missense variants and 25 synonymous variants. However, there were no significant differences in the frequencies of the 32 homoplasmic variants in MT-ND2 gene between the TFF group and control group (Table 1). The m.5178C>A variant was a diagnostic single-nucleotide polymorphism marker for haplogroup D. A previous report indicated that the m.5178C>A variant was associated with increased longevity (Tanaka et al. 1998). In addition, several studies have shown that the m.5178C>A variant caused resistance to the role of diabetes in patients of atherosclerosis (Matsunaga et al. 2001), hypertension (Kokaze et al. 2004), myocardial infarction (Takagi et al. 2004), and male Parkinson’s disease in the Han Chinese population (Gusdon et al. 2015). Furthermore, m.5178C>A may confer increased metabolic flexibility in adaptation to cold climates or seasonal variants (Nishimura et al. 2012). However, the 5178 homoplasmic variant showed no association with TFF risk in this study.

Table 1. MT-ND2 homoplasmic variants and TFF.

			Study group (n = 59)		Control group (n = 187)
Locus	Nucleotide change	Amino acid change	n	Frequency (%)	n	Frequency (%)	p-Value
4655	G-A	syn	1	1.69	1	0.53	ns
4736	T-C	syn	1	1.69	0	0	ns
4769	A-G	syn	59	1	187	1	ns
4820	G-A	syn	0	0	1	0.53	ns
4824	A-G	T-A	4	6.78	5	2.67	ns
4833	A-G	T-A	2	3.39	2	1.07	ns
4883	C-T	syn	5	8.47	29	15.51	ns
4958	A-G	syn	2	3.39	1	0.53	ns
4973	T-C	syn	0	0	1	0.53	ns
5033	A-G	syn	0	0	2	1.07	ns
5046	G-A	V-I	1	1.69	0	0	ns
5048	T-C	syn	0	0	1	0.53	ns
5108	T-C	syn	1	1.69	2	1.07	ns
5147	G-A	syn	2	3.39	2	1.07	ns
5153	A-G	syn	0	0	2	1.07	ns
5178	C-A	L-M	6	10.17	26	13.90	ns
5201	T-C	syn	0	0	1	0.53	ns
5237	G-A	syn	1	1.69	0	0	ns
5301	A-G	I-V	1	1.69	5	2.67	ns
5351	A-G	syn	3	5.08	4	2.14	ns
5360	C-T	syn	1	1.69	0	0	ns
5387	C-T	syn	1	1.69	0	0	ns
5390	A-G	syn	0	0	1	0.53	ns
5417	G-A	syn	2	3.39	1	0.53	ns
5426	T-C	syn	0	0	3	1.60	ns
5441	A-T	syn	0	0	2	1.07	ns
5460	G-A	A-T	1	1.69	4	2.14	ns
5465	T-C	syn	0	0	2	1.07	ns
5466	A-G	T-A	0	0	3	1.60	ns
5474	A-G	syn	1	1.69	0	0	ns
5498	A-G	syn	1	1.69	0	0	ns
5508	T-C	syn	0	0	1	0.53	ns

ns: not significant; syn: synonymous.

Nucleotide: the four canonical bases were adenine (A), cytosine (C), guanine (G), and thymine (T). Amino acid: alanine (A), threonine (T), valine (V), isoleucine (I), lysine (L), and methionine (M). The values show 7 missense variants and 25 synonymous variants. Data were analyzed using the Chi-squared test, but there were no significant mtDNA homoplasmic variants associated with TFF.

There were 47 heteroplasmic variants in the MT-ND2 gene, including 36 synonymous variants (Table 2) and 11 missense variants (Table 3). Data were analyzed using the Chi-square test, and the analysis indicated that the frequencies of point 4914, 5320, and 5426 heteroplasmic variants in the TFF group were significantly lower than those in the control group (18.64% versus 32.09%, 25.42% versus 41.71%, 18.64% versus 33.16%, respectively, p < 0.05) (Tables 2 and 3). We then applied the false discovery rate (FDR) method to account for multiple comparisons. After FDR correction, the point 4914, 5320, and 5426 heteroplasmic variants showed no significant association with TFF, as shown in Tables 2 and 3. Due to the small sample size and natural variability in human samples (and phenotypes), the ‘significant’ p-values presented in this article did not survive the FDR threshold. With a larger dataset in the future, perhaps a strong statistical argument can be made. In addition, in multivariate logistic analysis between the TFF group and the control group, after adjusting for confounding variables, we observed that these heteroplasmic variants were not independent protective factors for IVF (Data not shown). Thus, we tried to investigate the cumulative effects of SNPs in the MT-ND2 gene on IVF in the logistic regression model, also adjusted for the same confounding variables. Interestingly, we observed that there was a cumulative effect between 4914 and 5320 heteroplasmic variants with respect to protection against TFF (p < 0.05) (Supplemental Table 1). In addition, there was another significant cooperative effect among the point 4914, 5320, and 5426 heteroplasmic variants with respect to protection against TFF (p < 0.05) (Supplemental Table 1). However, there were significant differences in age, number of oocytes retrieved, and number of MII oocytes between the two groups, which might affect the final results of IVF. Thus, 53 pair-wise comparisons were also performed, and the general characteristics of the IVF failure and control subjects were adjusted in logistic models (Table 4). For each couple, the probability of TFF (i.e., propensity score) was estimated from a logistic regression model that considered multiple maternal baseline covariates (age (male and female), follicle-stimulating hormone (FSH), luteinizing hormone (LH), sperm concentration, motility grade A and B, number of retrieved oocytes, numbers of Metaphase II oocytes). We then implemented one-to-one matching without replacement on the closest propensity score of the IVF failure group and normal fertilization group (nearest neighbor matching). The process was repeated until each case in the IVF failure group was matched with one case in the normal fertilization group. Data suggested that there were no significant differences between the TFF group and control group in the frequencies of point 4914, 5320, and 5426 heteroplasmic variants of MT-ND2. Furthermore, there were no cooperative effects among the point 4914, 5320, and 5426 heteroplasmic variants with respect to protection against TFF (p > 0.05) (Table 5). The data suggested that the MT-ND2 gene variants might not be associated with TFF.

Table 2. MT-ND2 heteroplasmic synonymous variants and TFF.

			Study group (n = 59)		Control group (n = 187)
Locus	Nucleotide change(Wild type/variant)	Amino acid change	n	Heteroplasmy levels (%)	n	Heteroplasmy levels (%)	p-Value(Raw/FDR adjusted)
4646	T/C	syn	1	1.69	0	0	ns
4670	C/T	syn	1	1.69	0	0	ns
4733	T/C	syn	0	0	1	0.53	ns
4736	T/C	syn	14	23.73	59	31.55	ns
4850	C/T	syn	1	1.69	2	1.07	ns
4856	T/C	syn	17	28.81	63	33.69	ns
4883	C/T	syn	5	8.47	29	15.51	ns
4895	A/G	syn	0	0	1	0.53	ns
4914	C/T	syn	11	18.64	60	32.09	0.047/0.736
4940	C/T	syn	17	28.81	65	34.76	ns
4958	A/G	syn	21	35.59	81	43.32	ns
4973	T/C	syn	0	0	1	0.53	ns
4991	G/A	syn	16	27.12	56	29.95	ns
5021	T/C	syn	0	0	1	0.53	ns
5033	A/G	syn	0	0	2	1.07	ns
5042	A/G	syn	0	0	1	0.53	ns
5054	G/A	syn	0	0	1	0.53	ns
5093	T/C	syn	0	0	1	0.53	ns
5108	T/C	syn	3	5.08	9	4.81	ns
5147	G/A	syn	17	28.81	57	30.48	ns
5153	G/A	syn	0	0	2	1.07	ns
5198	A/G	syn	0	0	1	0.53	ns
5222	A/G	syn	0	0	2	1.07	ns
5231	G/A	syn	4	6.78	5	2.67	ns
5237	G/A	syn	0	0	2	1.07	ns
5351	A/G	syn	24	40.68	91	48.66	ns
5360	C/T	syn	0	0	2	1.07	ns
5387	C/T	syn	17	28.81	78	41.71	ns
5417	G/A	syn	0	0	6	3.21	ns
5426	T/C	syn	11	18.64	62	33.16	0.033/0.736
5456	C/T	syn	0	0	1	0.53	ns
5459	C/T	syn	0	0	1	0.53	ns
5465	T/C	syn	0	0	3	1.60	ns
5471	G/A	syn	10	16.95	49	26.20	ns
5474	A/G	syn	25	42.37	84	44.92	ns
5498	A/G	syn	20	33.90	74	39.57	ns

ns: not significant; syn: synonymous.

Nucleotide: the four canonical bases were adenine (A), cytosine (C), guanine (G), and thymine (T). The values showed 36 synonymous variants. Data were analyzed using the Chi-squared test, and the data indicated that the frequencies of point 4914 and 5426 heteroplasmic variants in the TFF group were significantly lower compared with the control group (p < 0.05). FDR p-value is the false discovery rate corrected p-value using the BH (Benjamini & Hochberg) method and the p-value less than 0.05 was determined as statistically significant.

Table 3. MT-ND2 heteroplasmic missense variants and TFF.

			Study group (n = 59)		Control group (n = 187)
Locus	Nucleotide change(Wild type/variant)	Amino acid change	n	Heteroplasmy levels (%)	n	Heteroplasmy levels (%)	p-Value(Raw/FDR adjusted)
4680	C/A	L/M	0	0	1	0.53	ns
4767	A/G	M/V	0	0	1	0.53	ns
4824	A/G	T/A	1	1.69	9	4.81	ns
4833	A/G	T/A	2	3.39	9	4.81	ns
5041	T/C	M/T	17	28.81	52	27.81	ns
5178	C/A	L/M	5	8.47	32	17.11	ns
5263	C/T	A/V	1	1.69	3	1.60	ns
5301	A/G	I/V	1	1.69	14	7.49	ns
5320	C/T	T/I	15	25.42	78	41.71	0.024/0.736
5442	T/C	F/L	2	3.39	4	2.14	ns
5460	G/A	A/T	0	0	4	2.14	ns

ns: not significant.

Nucleotide: the four canonical bases were adenine (A), cytosine (C), guanine (G), and thymine (T). Amino acid: alanine (A), threonine (T), valine (V), isoleucine (I), lysine (L), methionine (M), and phenylalanine (F). The values showed 11 missense variants. Data were analyzed using the Chi-square test, which indicated that the frequency of point 5320 heteroplasmic variant in the TFF group was significantly lower compared with the control group (p < 0.05). FDR p-value is the false discovery rate corrected p-value using the BH (Benjamini & Hochberg) method and the p-value less than 0.05 was determined as statistically significant.

Table 4. Background characteristics of matched and unmatched cases.

	Unmatched			Propensity matched
Variable	TFF group (n = 59)	Control group (n = 187)	p-Value	TFF group (n = 53)	Control group (n = 53)	p-Value
Female age (years)	29.3 ± 4.7	31.0 ± 5.3	0.0173	29.6 ± 4.8	30.2 ± 5.0	0.5649
Male age (years)	31.0 ± 4.8	32.6 ± 5.7	0.0678	31.3 ± 5.3	31.8 ± 6.2	0.6475
Female basal FSH (mIU/ml)	6.3 ± 3.2	6.1 ± 2.3	0.5626	6.1 ± 3.2	6.5 ± 2.2	0.4876
Female basal LH (mIU/ml)	5.4 ± 4.4	5.4 ± 3.4	0.9747	5.0 ± 2.7	5.2 ± 3.4	0.6782
Retrieved oocytes (n)	9.8 ± 6.2	12.6 ± 6.5	0.0048	11.4 ± 6.3	11.5 ± 5.6	0.9610
Metaphase II oocytes (n)	8.7 ± 5.5	11.5 ± 6.2	0.0026	9.3 ± 5.5	9.5 ± 5.5	0.7928
Infertility type
Female infertility (n)	42	165	0.0047	38	40	0.8722
Male infertility (n)	3	3		3	2
Both male and female infertility (n)	2	2		2	0
Unknown infertility (n)	12	17		10	11
Ovulation induction protocols
Long protocol (n)	53	170	0.4843	49	47	0.4312
Ultra-long protocol (n)	2	11		2	2
Other (n)	4	6		2	4
Sperm
Sperm concentration (×10^6/ml)	55.2 ± 25.3	57.2 ± 25.0	0.5809	56.4 ± 26.3	56.0 ± 23.9	0.9263
Motility grade A (%)	29.8 ± 12.0	30.1 ± 15.2	0.8692	30.4 ± 15.1	29.0 ± 11.6	0.6136
Motility grade B (%)	20.9 ± 9.5	26.5 ± 11.7	0.0009	22.1 ± 9.2	20.3 ± 7.6	0.2585

Values presented are mean ± SD. The general characteristics of each group are displayed. A total of 53 pair-wise comparisons were performed, and the general characteristics of the IVF failure and control subjects were adjusted.

Table 5. Results of the cumulative effects of heterozygous variants with respect to protection against TFF in the logistic regression model after propensity matching (1:1, both 53 cases in each group).

Heterozygous site	β	Standard error	Wald Х^2	p-Value	OR (95% CI)
4914	−0.4102	0.4557	0.8103	0.3680	0.66 (0.27–1.62)
5230	−0.4426	0.4232	1.0936	0.2957	0.64 (0.28–1.47)
5426	−0.5895	0.4487	1.7259	0.1889	0.55 (0.23–1.34)
4914 and 5320	−0.7027	0.4943	2.0213	0.1551	0.50 (0.19–1.30)
4914 and 5426	−0.8296	0.5437	2.3279	0.1271	0.44 (0.15–1.27)
5320 and 5426	<0.001	−0.7486	2.5089	0.4726	0.47 (0.19–1.19)
4914, 5320, and 5426	−1.1658	0.6206	3.5286	0.0603	0.31 (0.09–1.05)

Association between TFF and the haplogroups D and G

A previous study showed that mitochondrial haplotype was unlikely to be a reliable genetic marker of male factor infertility for a population of men in the UK (Mossman et al. 2012). However, another study showed that the mtDNA haplogroup R was a strong, independent predictor of sperm motility in the Han Chinese population, conferring a 2.97-fold decreased chance of asthenozoospermia compared with those without haplogroup R (Feng et al. 2013). In addition, our previous study indicated that men with haplogroup Z might inherit a higher risk of IVF failure in the Han Chinese population (Mao et al. 2016). Haplogroups D and G are subgroups of haplogroup M, with diagnostic single-nucleotide polymorphism markers m.5178C>A and m.4833A>G, respectively. The haplogroups were inferred, and association tests were performed through Chi-square test and multivariate logistic analysis in this study. There was no significant difference in the frequency of mtDNA haplogroup D or G in the IVF failure group compared with the normal fertilization group (p > 0.05). This result suggests that mtDNA haplogroups D and G might not be associated with TFF in the Han population.

In summary, this study showed that the MT-ND2 gene variants and haplogroups might not be associated with TFF in the Han population. However, this may be, in part, because of the limited sample size and insufficient study power. The results of this study help to characterize the relationships between the MT-ND2 gene variants and TFF. However, amplification bias might have occurred during nested PCR, and functional studies with larger sample sizes may be required for further assessment of these genetic markers.

Materials and methods

Subjects

From July 2011 to May 2017, we included 246 cases of fresh IVF cycle or half-intracytoplasmic sperm injection cycle for a retrospective comparative control study. The indications for performing IVF among the patients were mainly comprised female factors and unexplained infertility. There were 59 couples (mean female age, 29.27 ± 0.61 years old; mean male age, 31.07 ± 0.70 years old) with TFF in the IVF failure group and 187 couples (mean female age, 30.98 ± 0.35 years old; mean male age, 32.60 ± 0.41 years old) with normal fertilization (50% fertilization rate) in the control group. Fertilization is determined 16–18 h after insemination by the presence of two pronuclei. The cases in which none of the oocytes are found to contain two or more pronuclei are defined as TFF (Bungum et al. 2010). We obtained approval for this study from Zhengzhou University’s research ethics committee and received informed written consent from each couple in the fertilization failure and control groups for the genetic studies.

Sperm mtDNA extraction, nested PCR, and sequence analysis

The protocol for sperm preparation and total genomic DNA extraction was performed as described previously by our group (Mao et al. 2015, 2016). The primer sequences, along with their nested PCR reaction conditions and GenBank Accession serial numbers, are shown in Table 6. In the first round of amplification, a total reaction volume of 50 μL included 2 mM MgCl2, 200 nM of each deoxy-ribonucleoside triphosphate (dNTP), 200 nM of each primer, 2 U Taq DNA polymerase, and 2 μL DNA template. A second round of amplification was performed with the same reaction mixture, but a different annealing temperature than in the first round. All sample genes underwent direct sequencing using 3730 DNA Sequencing Analysis Gene Mapper® Version 4.0 (Applied Biosystems, Foster City, CA, USA). In the forward and reverse sequencing of two positive sequences, if the second wave height was greater than 10% of the same site crest, it was considered a heteroplasmic variant (Lacan et al. 2009; Kloss-Brandstätter et al. 2010; Mao et al. 2016).

Table 6. Nested PCR primers, reaction conditions, and GenBank accession numbers.

MT-ND2	Primer	Product (bp)	Annealing temperature (°C)	No. of cycles	Accession numbers
Outer	F:CGCTACGACCAACT CATACACC R:ATGGCTGAGTGAAG CATTGGAC	1739	60	30	NC-012920
Inner	F:CCCCATCCTAAAGTA AGGTCAG R:GCTTTGAAGGCTCTT GGTCTGT	1160	59	30	NC-012920

Outer: first round of amplification. Inner: second round of amplification.

Sequence comparison

The Mitomap database (www.mitomap.org) and the dbSNP database (https://www.ncbi.nlm.nih.gov/projects/SNP/) provided the human sequence of Cambridge (rCRS) to compare the sequencing results. We detected the nucleotide variant, then confirmed the peak figure again and checked whether the variant in amino-acid-encoding loci had changed.

Selection of mitochondria polymorphism haploid type classification

Mitochondria polymorphism haplogroups were identified using a human mtSNP database (http://www.mitomap.org/bin/view.pl/MITOMAP/HaplogroupMarkers). Haplogroup D was characterized by C5178A, and haplogroup G was characterized by A4833G.

Statistical analysis

Statistical analysis was performed as described previously by our group (Mao et al. 2015, 2016). Continuous data were expressed using mean ± standard deviation, and the classification of the data was expressed using numbers and percentage (%). We used the independent sample t-test to analyze the measurement data, and Chi-squared test or Fisher’s exact test with FDR correction to analyze the count data. Multivariate logistic analysis was also used to analyze association between the mitochondrial variants and TFF, adjusted for all confounding variables (age (male and female), FSH, LH, sperm concentration, motility grade A and B, number of retrieved oocytes, numbers of Metaphase II oocytes). Furthermore, 53 pair-wise comparisons were performed and the general characteristics of the IVF failure and control subjects were adjusted in logistic models. A two-sided p-value less than 0.05 was considered statistically significant.

Notes on contributors

Analyzed the data and wrote the manuscript: ZJL, MGH; Performed the experiments: ZJL, HXH; Analyzed the data: CHY, ZY, CX.

Acknowledgments

The material contained in the manuscript is original, has not been published, has not been submitted or is not being submitted elsewhere. This work was supported by the Health Department of Henan Province, China (Grant 4117), the Foundation of the He’nan Educational Committee (Grant 13A320637) and the science and technology project of Henan Province (Grant 172102310131).

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary material

Supplemental data for this article can be accessed here.

Additional information

Funding

This work was supported by the Health Department of Henan Province, China [Grant 4117];Foundation of the He’nan Educational Committee [Grant 13A320637]; the science and technology project of Henan Province [Grant 172102310131].