Abstract
Objective:
To determine if the levels of sperm very-long chain polyunsaturated fatty acids (VLC-PUFA) are correlated with sperm parameters and the outcome of live birth following conventional therapy for unexplained infertility.
Design:
Cohort analysis of the Reproductive Medicine Network's Assessment of Multiple Intrauterine Gestations from Ovarian Stimulation (AMIGOS) randomized controlled trial.
Subjects:
Male partners of 185 couples with unexplained infertility with baseline semen samples available for analysis
Exposure:
We determined the levels of VLC-PUFA in total lipid isolated from sperm membranes using liquid chromatography-mass spectrometry/mass spectrometry analyses.
Main outcome measures:
Sperm concentration, motility, morphology, total motile count, and live birth following standard treatment for unexplained infertility
Results:
Total VLC-PUFA percentage was positively correlated with sperm concentration (Spearman r=0.56, p<0.0001), total motile count (r=0.40, p<0.0001) and morphology (r=0.26, p=0.0005). Following adjustment for male BMI, age, and race, a one standard deviation increase in the percentage of total VLC -PUFA was associated with a 62% increase in the geometric mean of sperm concentration [GM Ratio: 1.62 (95% CI 1.45, 1.82)] and a 43% increase in the geometric mean of total motile count [GM Ratio: 1.43 (95% CI 1.24,1.63)]. While no evidence of association was observed for sperm motility, a positive relationship was also observed between the percentage of total VLC-PUFA and sperm morphology [adjusted incidence rate ratio (IRR) for one standard deviation increase in total VLC-PUFA: 1.18 (95% CI: 1.02, 1.36)]. Following adjustment for female age and treatment group, the probability of a live birth outcome was 72% more likely among those in the third tertile of hydroxylated VLC-PUFA percentage compared to the first tertile [RR 1.72 (95% CI 1.01, 2.94)].
Conclusions:
The positive correlation between sperm VLC-PUFA percentage and sperm parameters, and the significant association between hydroxylated VLC-PUFA percentage and the outcome of live birth, strongly suggest that this class of fatty acids is essential to normal sperm structure and function.
Keywords: Unexplained infertility, Fatty acids, Sperm lipids, very long chain polyunsaturated fatty acids, Ovarian stimulation, Live Birth
Capsule
The percentage of sperm very long chain polyunsaturated fatty acids is positively associated with sperm parameters and live birth among couples with unexplained infertility undergoing ovarian stimulation with intrauterine insemination.
Introduction
The lipid composition of the sperm plasma membrane is essential to maintaining sperm structure and function (1, 2) . The most relevant lipid species include phospholipids, neutral lipids, and glycolipids, of which phospholipids are the most abundant. Of importance in this study is the fatty acid composition of sperm sphingomyelin (SM) and ceramides (CM) that are uniquely enriched in very-long chain polyunsaturated fatty acids (VLC-PUFAs, ≥ 28 carbons) of both n-6 and n-3 series (3) .These unique fatty acids are derived from elongation and desaturation of dietary C18-C22 LC-PUFA by a series of fatty acid elongase and desaturase enzymes, particularly the Elongation of Very Long Chain Fatty Acid-4 enzyme that is expressed in few tissues, including the brain, retina, and testes (4-6).
Interestingly, although VLC-PUFAs only account for approximately 5% of the lipid composition of the sperm membrane (7), previous studies using transgenic mouse models suggest they are essential for male fertility (8, 9). Animal studies using a homozygous fatty acid desaturase-2 (FADS2) knockout mouse model found that the resulting FADS2 enzyme deficiency and subsequent inability to synthesize arachidonic acid (AA, 20:4n6), docosahexaenoic acid (DHA), and n6-docosapentaenoic acid (DPAn6, 22:5n-6) led to almost complete infertility with significantly decreased sperm count and abnormal morphology(8). Dietary supplementation with DHA and subsequent elongation to VLC-PUFA resulted in a restoration of fertility, sperm count, and morphology(8). Zadravec, et al. (9) showed that deletion of Elovl2, the enzyme that converts C-22 PUFA to C-24/26 PUFA (precursors of VLC-PUFA), also resulted in complete arrest of spermatogenesis, with seminiferous tubules displaying only spermatogonia and primary spermatocytes without further developing into germinal cells in mice. In these mice, providing DHA did not restore fertility due to lack of the ELOVL2 enzyme necessary for elongating the DHA to C24/26 precursors for ELOVL4 to make VLC-PUFA, indicating that DHA itself could not restore fertility unless it is elongated to VLC-PUFAs (9).
In humans, studies by our group and others suggest an important role for sperm VLC-PUFAs and their precursors for fertility (10-13). A 2019 study including 70 men undergoing fertility testing found that lower levels of VLC-PUFAs were correlated with decreased total sperm count and total motile sperm (14). This study also noted that VLC-PUFAs comprised 0-6.1% of sphingomyelin present, and that there was no association between VLC-PUFA percentage and DHA concentration, suggesting that a lack of precursor was not the primary cause of low levels (14) . Although these findings support the critical role of these fatty acids in male fertility, we currently lack a direct link between sperm VLC-PUFAs and infertility in humans, as well as fertility treatment outcomes.
The purpose of this study was to determine if the percentage of sperm membrane VLC-PUFAs is correlated not only with sperm parameters, but also with the outcome of live birth following conventional therapy for unexplained infertility.
Materials and Methods
This investigation included a cohort of participants from the Assessment of Multiple Intrauterine Gestations from Ovarian Stimulation (AMIGOS) clinical trial conducted by the NICHD’s Cooperative Reproductive Medicine Network (RMN)(15). The goal of the AMIGOS study (clinicaltrials.gov NCT01044862) was to determine whether the rate of multiple gestations among women with clinical pregnancies in couples with unexplained infertility treated with up to four cycles of ovarian stimulation (OS) with intrauterine inseminations (IUI) could be reduced with letrozole (n=299) compared to conventional therapy with gonadotropin (n=301) or clomiphene (n=300) (15, 16). Details on the study protocol and baseline characteristics of the study participants have been described previously (15, 17). Of the 900 couples participating in the trial, 185 males had baseline semen samples obtained at enrollment available for analysis of VLC-PUFA measures, of which 49 (26.5%) of their partners had a live birth.
Baseline semen samples from participants in the AMIGOS trial were utilized for the analysis. The semen samples were subjected to lipid extraction and sphingomyelin (SM) fatty acid profiling using tandem mass spectrometry as previously described [12]. Briefly, samples were pelleted by centrifugation to remove the seminal fluid. Total lipids were extracted from the cell pellets using the method of Folch et al. (18), subjected to mild base hydrolysis to hydrolyze ester-bound fatty acids, and extracted into chloroform under basic conditions to isolate intact sphingolipids. Isolated sphingolipids were introduced into a TSQ Ultra triple quadrupole mass spectrometer (Thermo Scientific, Waltham, Massachusetts) using a Nanomate chip–based nano-ESI source (Advion Biosciences, Ithaca, New York) operating in infusion mode and SM species were measured using precursor ion scanning of m/z 184 (19). We extracted total lipids from all samples and processed them for VLC-PUFA identification and quantification at the same time to minimize both inter and intra-specimen variations due to processing and analyses. During the processing, we also added internal standards to all the samples and the values of VLC-PUFA were quantified by normalizing to the internal standards and VLC-PUFA values presented as relative mole percent. Relative abundances of SM molecular species were calculated using the lipid mass spectrum analysis (LIMSA) software (University of Helsinki) and are represented as a relative percent of the sum based upon their response values (14). This study received approval from the University of Oklahoma Health Science Center’s Institutional Review Board (IRB).
Data Analyses
Hydroxylated and total (hydroxylated + non-hydroxylated) SM-VLC-PUFA percentages were evaluated as both continuous variables and as tertiles defined by percentiles observed in the overall study population. Bivariate comparisons of baseline characteristics, semen parameters, and VLC-PUFA measurements by outcome of live birth were estimated using Wilcoxon Rank Sum tests, Chi-Square test or Fisher’s Exact test, as appropriate. Bivariate associations of continuous VLC-PUFA measurements, duration of infertility and semen characteristics were estimated using Spearman correlation coefficients. Associations between VLC-PUFA percentages and outcomes of total motile count (millions) and sperm concentration (million/mL) were evaluated using linear regression models. To meet the model assumptions of normally distributed residuals and equal variance, log transformations were applied to the total motile count and sperm concentration outcomes and the exponentiated regression coefficients and 95% confidence intervals provided the geometric mean ratios (GMR). Log-transformation of the outcomes of sperm motility (%) and morphology (%) were not sufficient to meet the model assumptions for linear regression. After assessing dispersion parameters and model fit using Pearson chi-square p-values, negative binomial regression models for over-dispersed count data were used to estimate incidence rate ratios (IRR) and 95% confidence intervals for associations between VLC-PUFA and outcomes of sperm motility (%) and morphology (%). Risk ratios (RR) and 95% confidence intervals for associations between VLC-PUFA and the outcome of live birth were estimated using modified Poisson regression models with robust standard errors.
Models assessing relationships with semen parameters evaluated male age (<30 years, 30-35 years, ≥36 years), BMI [normal (18.5kg/m2 to <25kg/m2), overweight (25kg/m2 to <30kg/m2), obese (≥30kg/m2)], and race/ethnicity (Non-Hispanic White, Non-Hispanic Black, Hispanic, Other) as potential confounders. Models assessing the outcome of live birth evaluated the same male characteristics as potential confounders, as well as female age (<30 years, 30-35 years, ≥36), female race/ethnicity (Non-Hispanic White, Non-Hispanic Black, Hispanic, Other), treatment group (Clomiphene, Letrozole, Gonadotropin), duration of infertility (1 year or less, 2 years, 3 years, >3 years), and history of live birth (yes/no). Covariates were retained in the model if they met the criterion of changing the VLC-PUFA estimate by 10% or more upon exclusion. None of the covariates met the change-in-estimate criterion for models assessing semen parameter outcomes; thus unadjusted estimates are reported. Female age and treatment group met the criterion for models assessing the outcome of live birth and were included as covariates in the adjusted models.
Results
Baseline characteristics of couples contributing data to the present investigation are shown by live birth outcome in Table 1. No differences were observed by treatment, sperm concentration, total motile count, motility, semen volume, male or female age, female BMI, live birth outcome or history of live birth. Couples who achieved a live birth reported a shorter duration of infertility and were more likely to receive treatment with gonadotropins when compared to those that did not achieve a live birth. Notably, the distributions of male BMI, age, and baseline sperm parameters were similar between outcome groups. Additionally, median total and hydroxylated VLC-PUFA percentages were similar between those with and without live births, but when evaluated by tertiles (Table 1) those with live births were observed to have numerically lower proportions in the upper tertile of total VLC-PUFA measures (22.4% vs 37.5%, p=0.08) and higher proportions in the upper tertile of hydroxylated VLC-PUFA (44.9% vs 29.4%, p=0.14) when compared to distributions among those who did not achieve a live birth.
Table 1.
Descriptive characteristics of the sample.
| Live Birth (N = 49) |
No Live Birth (N = 136) |
p- valuea |
|
|---|---|---|---|
| Median (IQR) | Median (IQR) | ||
| Male age (years) | 33.0 (31.0, 38.0) | 34.0 (30.0, 37.0) | 0.85 |
| Female age (years) | 31.0 (30.0, 33.0) | 33.0 (29.0, 37.0) | 0.14 |
| Male BMI (kg/m2) | 28.6 (25.1, 32.5) | 28.9 (25.7, 33.6) | 0.69 |
| Female BMI (kg/m2) | 24.7 (21.3, 29.5) | 26.8 (22.6, 31.5) | 0.17 |
| Duration of infertility (months)* | 24.0 (15.0, 45.5) | 30.0 (19.0, 48.0) | 0.06 |
| Sperm concentration (millions/mL) | 47.3 (25.0, 95.0) | 45.9 (24.5, 76.7) | 0.62 |
| Sperm total motility (%) | 55.0 (50.0, 61.0) | 57.0 (50.0, 66.0) | 0.55 |
| Sperm morphology (%)* | 9.0 (5.0, 19.0) | 12.5 (5.0, 22.0) | 0.31 |
| Total Motile Count (millions)* | 73.2 (38.6, 108.9) | 52.2 (29.6, 110.6) | 0.40 |
| VLC PUFA (Total) Percentage (%) | 1.5 (1.0, 2.3) | 1.8 (0.9, 3.1) | 0.33 |
| VLC PUFA (OH only) Percentage (%) | 0.2 (0.1, 0.3) | 0.2 (0.1, 0.2) | 0.08 |
| N (%) | N (%) | p-value | |
| Male Race/Ethnicity | 0.81c | ||
| Non-Hispanic White | 37 (75.5) | 94 (69.1) | |
| Non-Hispanic Black | 3 (6.1) | 13 (9.6) | |
| Hispanic | 6 (12.2) | 22 (16.2) | |
| Other | 3 (6.1) | 7 (5.1) | |
| Female Race/Ethnicity | 0.65c | ||
| Non-Hispanic White | 35 (71.4) | 86 (63.2) | |
| Non-Hispanic Black | 3 (6.1) | 14 (10.3) | |
| Hispanic | 7 (14.3) | 27 (19.8) | |
| Other | 4 (8.2) | 9 (6.6) | |
| Male Partner Smoking | 0.79b | ||
| Yes | 8 (40.0) | 22 (36.7) | |
| No | 12 (60.0) | 38 (63.3) | |
| N missing = 105 | |||
| Male Partner Alcohol Use | 0.08b | ||
| Yes | 43 (93.5) | 102 (82.9) | |
| No | 3 (6.5) | 21 (17.1) | |
| N missing = 16 | |||
| History of live birth | 0.87b | ||
| Yes | 11 (22.5) | 29 (21.3) | |
| No | 38 (77.5) | 107 (78.7) | |
| Treatment Group | 0.0006 b | ||
| Clomiphene | 14 (28.6) | 56 (41.2) | |
| Letrozole | 9 (18.4) | 48 (35.3) | |
| Gonadotropin | 26 (53.1) | 32 (23.5) | |
| VLC PUFA (Total) Percentage Tertiles | 0.08b | ||
| Tertile 1 (< 1.310%) | 16 (32.6) | 45 (33.1) | |
| Tertile 2 (1.310% - < 2.440%) | 22 (44.9) | 40 (29.4) | |
| Tertile 3 (≥ 2.440%) | 11 (22.4) | 51 (37.5) | |
| VLC PUFA (OH Only) Percentage Tertiles | 0.14b | ||
| Tertile 1 (< 0.139%) | 13 (26.5) | 48 (35.3) | |
| Tertile 2 (0.139% - < 0.205%) | 14 (28.6) | 48 (35.3) | |
| Tertile 3 (≥ 0.205%) | 22 (44.9) | 40 (29.4) |
Wilcoxon
Chi-square
Fisher’s Exact
Duration of infertility is missing n=3 for no live birth and n=1 for live birth, measures of sperm morphology are missing n=10 for no live birth and n=3 for live birth, and total motile count measures are missing n=2 for no live birth.
Counts of missing values are reported but not included in calculations of proportions or when conducting comparisons
The median VLC-PUFA percentage was 1.693% for total VLC-PUFA (ranging from 0.066% to 6.847%) and 0.175% for hydroxylated VLC-PUFA (ranging from 0.001% to 2.196%. Total VLC-PUFA percentage was moderately positively correlated with sperm concentration and total motile count, and a weak positive correlation was observed with morphology (Table 2). Hydroxylated VLC-PUFAs were also positively correlated with sperm concentration and total motile count; however, the strength of these correlations was modest. Neither total nor hydroxylated VLC-PUFAs were correlated with sperm motility (Table 2) and the percentage of hydroxylated VLC-PUFAs was not correlated with sperm morphology. Furthermore, correlations between continuous duration of infertility (months) and total VLC-PUFA and hydroxylated VLC-PUFA percentages were weak and negative and did not achieve statistical significance (total VLC-PUFA ρ= −0.07, p=0.38; hydroxylated VLC-PUFA ρ= −0.11, p=0.16).
Table 2.
Spearmen Correlations with continuous VLC-PUFA percentage (%) and semen parameters
| Total VLC PUFA percentage (%)(n=185) |
Hydroxylated VLC PUFA percentage (%)(n=185) |
|||
|---|---|---|---|---|
| Correlation Coefficient (ρ) |
p-value | Correlation Coefficient (ρ) |
p-value | |
| Sperm Motility (%) | 0.05 | 0.5018 | 0.03 | 0.6794 |
| Sperm Concentration (millions/mL) | 0.56 | <.0001 | 0.36 | <.0001 |
| Morphology (%) | 0.26 | 0.0005 | 0.04 | 0.5633 |
| Total Motile Count (millions) | 0.40 | <.0001 | 0.29 | <.0001 |
Unadjusted estimates assessing the relationship between VLC-PUFA percentage and log-transformed sperm concentration, and total motile count are shown in Table 3. A one standard deviation increase in sperm total VLC -PUFA percentage was associated with a 61% increase in the geometric mean of sperm concentration [GM Ratio: 1.61 (95% CI 1.44, 1.80], and a 39% increase in the geometric mean of total motile count [GM Ratio: 1.39 (95% CI 1.22, 1.59)]. Associations with a one standard deviation increase in hydroxylated VLC-PUFA were also positive, but more modest for both sperm concentration and total motile count. When VLC-PUFA was examined in tertiles, geometric mean sperm concentration was three times greater in those with total VLC-PUFA values in the highest tertile compared to the lowest tertile [GM Ratio: 3.01 (95% CI 2.29, 3.96). Similarly, sperm concentrations were approximately 2.2 times greater in those with hydroxylated VLC-PUFA in the highest tertile compared to the lowest (GM Ratio: 2.18 (95% CI 1.62, 2.94). Geometric mean total motile counts were approximately two times greater in the top tertiles of both total and hydroxylated VLC-PUFA when compared to the lowest tertiles.
Table 3.
Unadjusted associations between VLC-PUFA (continuous and tertile measures) and sperm parameters
| (Log) Sperm Concentration (millions/mL) GM Ratio (95% CI) |
(Log) Total Motile Count (millions) GM Ratio (95% CI) |
Sperm Motility (%) IRR (95% CI) |
Morphology (%) IRR (95% CI) |
|
|---|---|---|---|---|
| Total VLC-PUFA (%) | 1.61 (1.44, 1.80) | 1.39 (1.22, 1.59) | 1.01 (0.97, 1.05) | 1.17 (1.02, 1.36) |
| Hydroxylated VLC-PUFA (%) | 1.17 (1.03, 1.33) | 1.15 (1.00, 1.33) | 1.00 (0.96, 1.04) | 0.93 (0.77, 1.12) |
| Total VLC-PUFA Tertiles | ||||
| Tertile 1 (< 1.310%) | 1.00 | 1.00 | 1.00 | 1.00 |
| Tertile 2 (1.310% - < 2.440%) | 2.06 (1.57, 2.71) | 2.29 (1.67, 3.15) | 1.02 (0.93, 1.13) | 1.54 (1.10, 2.14) |
| Tertile 3 (≥ 2.440%) | 3.01 (2.29, 3.96) | 2.25 (1.63, 3.10) | 1.00 (0.91, 1.10) | 1.45 (1.04, 2.01) |
| Hydroxylated VLC-PUFA Tertiles | ||||
| Tertile 1 (< 0.139%) | 1.00 | 1.00 | 1.00 | 1.00 |
| Tertile 2 (0.139% - < 0.205%) | 1.43 (1.06, 1.92) | 1.32 (0.95, 1.85) | 1.02 (0.93, 1.13) | 0.73 (0.52, 1.01) |
| Tertile 3 (≥ 0.205%) | 2.18 (1.62, 2.94) | 1.95 (1.39, 2.72) | 1.02 (0.93, 1.13) | 1.05 (0.75, 1.46) |
Linear regression model estimating geometric mean (GM) ratio for one-standard deviation change in VLC-PUFA (or tertile comparisons).
Negative binomial regression model estimating incidence rate ratio (IRR) for a one-standard deviation change in VLC-PUFA (or tertile comparisons).
No evidence of association was observed between total or hydroxylated VLC-PUFA and sperm motility, with all IRRs near 1.0 (Table 3). Positive associations, however, were observed between total VLC-PUFA and sperm morphology, where a one standard deviation increase in total VLC-PUFA increased the percent of sperm classified with normal morphology by 17% [IRR: 1.17 (95% CI 1.02, 1.36)]. When compared to the lowest tertile of total VLC-PUFA, normal sperm morphology increased by approximately 50% when total VLC-PUFA was in the middle tertile or in the highest tertile. Evidence for association between hydroxylated VLC-PUFA and sperm morphology was less clear, with a 27% decreased rate of normal morphology observed among those with hydroxylated VLC-PUFA in the middle tertile [IRR: 0.73 (95% VI 0.52, 1.01)] but no differences revealed between the upper and lower tertiles [IRR 1.05 (95% CI 0.75, 1.46)] (Table 3).
When evaluating live birth outcomes, following adjustment for female age and treatment group, an increase in total VLC-PUFA by one standard deviation was not significantly associated with the outcome of live birth (Table 4). In contrast, a one standard deviation increase in hydroxylated VLC-PUFA percentage was associated with a 15% increase in likelihood of live birth [adjusted RR: 1.15 (95% CI: 1.04, 1.26)]. When evaluated as exposure tertiles, the probability of a live birth outcome was also 72% more likely among those in the third tertile of hydroxylated VLC-PUFA percentage compared to the first tertile [RR 1.72 (95% CI: 1.01, 2.94)] (Table 4).
Table 4.
Modified Poisson regression models for associations between one standard deviation change in continuous VLC-PUFA percentage and live birth
| Unadjusted RR (95% CI) 1-SD Change |
Adjusted RR (95% CI)b 1-SD Change |
|
|---|---|---|
| Total VLC PUFA (%) | 0.83 (0.65, 1.06) | 0.77 (0.58, 1.01) |
| Hydroxylated VLC PUFA (%) | 1.14 (1.08, 1.20) | 1.15 (1.04, 1.26) |
| PUFA Tertiles | ||
| Total VLC PUFA | Unadjusted RR (95% CI) |
Adjusted RR (95% CI) |
| Tertile 2 (1.310% - < 2.440%) vs. 1 (< 1.310%) | 1.35 (0.79, 2.32) | 1.12 (0.67, 1.85) |
| Tertile 3 (≥ 2.440%) vs. 1 (< 1.310%) | 0.68 (0.34, 1.34) | 0.55 (0.29, 1.07) |
| Hydroxylated VLC PUFA | ||
| Tertile 2 (0.139% - < 0.205%) vs. 1 (< 0.139%) | 1.06 (0.54, 2.06) | 1.10 (0.59, 2.07) |
| Tertile 3 (≥ 0.205%) vs. 1 (< 0.139%) | 1.67 (0.93, 3.00) | 1.72 (1.01, 2.94) |
Modeled as the probability of Live Birth = Yes
Adjusted for female age and treatment group
Discussion
In the present investigation, we identified a positive correlation between semen parameters, including sperm concentration and total motile count, and the percentage of sperm lipid VLC-PUFAs. This association was strongest for the total VLC-PUFAs, but was also present when evaluating the hydroxylated version of VLC-PUFAs. These findings confirm our prior findings from a single center (14) in a larger and more diverse group of men participating in a multicenter clinical trial. Additionally, following adjustment for female age and treatment group, we identified a positive association between the hydroxylated version of VLC-PUFAs and the outcome of live birth in couples undergoing treatment for unexplained infertility.
The sperm membrane plays a critical role in key fertilization events. The fatty acid profile has been associated with the performance not only of sperm motility but also membrane fluidity, flexibility, fusion, fission, and morphology [15]. There are several studies which have evaluated the fatty acid composition of human semen and its correlation with sperm parameters[8, 10, 16-20], and one study which investigated the relationship between sperm lipid profiles and outcomes of IVF with ICSI. Most of these studies are limited by small numbers and lack of adequate control group. To our knowledge, no prior study has explored specifically the association of VLC-PUFAs and treatment outcomes.
Recent knockout mice models as well as analyses of human genetic disorders have unveiled several important enzymes in the biosynthesis pathways of VLC-PUFAs; however, the mechanism of how these fatty acids affect the male reproductive system is not well understood (20). The process of fertilization requires reorganization of lipids and proteins to the lateral portion of the sperm head to allow for proper binding to the oocyte’s zona pellucida(1). Other than their critical role in the membrane, some studies suggested that these fatty acids improve antioxidant activity of human seminal fluid and sperm quality (21). Additionally, the finding of a relationship between sperm membrane lipid profile and pregnancy rates in couples undergoing IVF-ICSI suggest these specialized lipids may have a role in normal fertilization beyond simply binding to the zona pellucida and penetrating the oocyte plasma membrane (22).
It has been previously recognized that VLC-PUFAs exist in hydroxylated and non-hydroxylated forms; however, the significance of each is not well understood (23). Our findings of a positive association between hydroxylated VLC-PUFA percentage in sperm and live birth suggests that the hydroxylated form may be more biologically significant. Indeed, previous studies showed enrichment of 2-hydroxylated VLC-PUFA during differentiation of sperm from round spermatids to mature spermatozoa as a novel a novel hallmark of germ cell differentiation (3). This association could be a consequence of the positive relationship between hydroxylated VLC-PUFA percentage and total motile count (TMC) and sperm concentration, which was also noted in our data. However, if this were true, it would be expected that total VLC-PUFA percentage, which correlated more strongly with sperm parameters, would be associated with live birth outcome, which was not the case. More likely, since sperm parameters are in the normal range in both groups, decreased hydroxylated VLC-PUFA percentage results in intrinsic impairments in the lipid modifications that occur during the maturation and fertilization processes. In a mouse model, it was shown that spermatid maturation involved notable amounts of VLC-PUFA containing sphingolipids primarily in the sperm head, where they potentially stabilize the high-curvature membrane (9, 24). This is highlighted in ELOVL2 ablated mice in which the heterozygous Elovl2+/− males produce sperm with abnormally round heads and infertility, thought to be secondary to disruption in the portion of the head containing the acrosome (9).
There are several strengths of our study, including the relatively large number of sperm samples available for evaluation, along with the availability of the treatment course and pregnancy outcomes for couples participating in the multicenter AMIGOS clinical trial. Additionally, the participants were well characterized at baseline. Our lab has also been able to successfully isolate and measure the relative percentage of VLC-PUFAs among sphingolipids, both hydroxylated and not, despite the overall low amount present within cells. Limitations include that only a subset of AMIGOS participants had baseline sperm samples available for VLC-PUFA analysis, which could introduce bias if factors related to selection also affected VLC-PUFA and semen quality or were a consequence of both VLC-PUFA quantity and semen quality. Little is known about factors that may affect the relative percentage of total and hydroxylated VLC-PUFAs among sphingolipids, although statistical adjustment for BMI and race/ethnicity had little impact on the point estimates. Another limitation is that only a single baseline semen sample was available per patient, which may have been several months removed from the sample provided for the IUI that resulted in pregnancy. The variability in sperm membrane lipid VLC-PUFA percentages over time has not been determined. Additionally, samples used in our study were frozen at −80 C for up to ten years prior to analysis, which could result in sample degradation. While we cannot rule out some minimal level of PUFA degradation, since the samples were properly stored and were all analyzed at the same time, any degradation effects would be similar among all the samples analyzed. In addition, the levels of VLC-PUFA species present in these samples were not very different from VLC-PUFA composition of fresh sperms that we analyzed and reported in our recent studies (14). Finally, the AMIGOS trial enrolled couples with unexplained infertility, and the strengths/magnitude of the associations observed in this population may be different than those seen in more significant male factor infertility.
In conclusion, in a large prospective multicenter trial, increased sperm total and hydroxylated VLC-PUFA percentages were associated with improved sperm parameters, and greater sperm hydroxylated VLC-PUFAs were associated with improved outcomes of standard treatment in couples with unexplained infertility. Additional investigations are needed to further define the role of these specialized lipids in normal male fertility. These findings may open new possibilities for the development of male diagnostic tools and therapeutic targets.
Attestation Statement:
Data regarding subjects in the study has been presented in part at the ASRM 2021 Scientific Congress, Baltimore, MD
Data will be made available to the editors of the journal for review or query upon request.
Support:
The AMIGOS trial was supported by grants from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) [U10HD39005, U10HD38992, U10HD27049, U10HD38998, U10HD055942, U10HD055944, U10HD055936 and U10HD055925] and by the American Recovery and Reinvestment Act. The present work was also supported by the National Institute of General Medical Sciences U54GM104938, the University of Oklahoma College of Medicine Alumni Association (COMAA) to KRH, and Research to Prevent Blindness unrestricted grant to the Dean McGee Eye Institute. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NICHD or NIH.
Footnotes
Clinical trial registration: NCT01044862
References:
- 1.Flesch FM, Gadella BM. Dynamics of the mammalian sperm plasma membrane in the process of fertilization. Biochim Biophys Acta. 2000;1469:197–235. [DOI] [PubMed] [Google Scholar]
- 2.Carro M, Luquez JM, Penalva DA, Buschiazzo J, Hozbor FA, Furland NE. PUFA-rich phospholipid classes and subclasses of ram spermatozoa are unevenly affected by cryopreservation with a soybean lecithin-based extender. Theriogenology. 2022;186:122–34. [DOI] [PubMed] [Google Scholar]
- 3.Oresti GM, Reyes JG, Luquez JM, Osses N, Furland NE, Aveldano MI. Differentiation-related changes in lipid classes with long-chain and very long-chain polyenoic fatty acids in rat spermatogenic cells. J Lipid Res. 2010;51(10):2909–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Shan S, Xu F, Hirschfeld M, Brenig B. Sperm Lipid Markers of Male Fertility in Mammals. Int J Mol Sci. 2021;22(16). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Yu M, Benham A, Logan S, Brush RS, Mandal MNA, Anderson RE, et al. ELOVL4 protein preferentially elongates 20:5n3 to very long chain PUFAs over 20:4n6 and 22:6n3. J Lipid Res. 2012;53(3):494–504. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Agbaga MP, Brush RS, Mandal Md NA, Henry K, Elliott MH, Anderson RE. Role of Stargardt-3 macular dystrophy protein (ELOVL4) in the biosynthesis of very long chain fatty acids. Proceedings of the National Academy of Sciences of the United States of America. 2008;105:12843–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Furland NE, Oresti GM, Antollini SS, Venturino A, Maldonado EN, Aveldano MI. Very long-chain polyunsaturated fatty acids are the major acyl groups of sphingomyelins and ceramides in the head of mammalian spermatozoa. J Biol Chem. 2007;282(25):18151–61. [DOI] [PubMed] [Google Scholar]
- 8.Roqueta-Rivera M, Stroud CK, Haschek WM, Akare SJ, Segre M, Brush RS, et al. Docosahexaenoic acid supplementation fully restores fertility and spermatogenesis in male delta-6 desaturase-null mice. J Lipid Res. 2010;51(2):360–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Zadravec D, Tvrdik P, Guillou H, Haslam R, Kobayashi T, Napier JA, et al. ELOVL2 controls the level of n-6 28:5 and 30:5 fatty acids in testis, a prerequisite for male fertility and sperm maturation in mice. J Lipid Res. 2011;52(2):245–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Gulaya NM, Margitich VM, Govseeva NM, Klimashevsky VM, Gorpynchenko II, Boyko MI. Phospholipid composition of human sperm and seminal plasma in relation to sperm fertility. Arch Androl. 2001;46(3):169–75. [DOI] [PubMed] [Google Scholar]
- 11.Tang LX, Yuan DJ, Wang QL, Jiang F, Guo J, Tang YG, et al. Association of decreased spermatozoa omega-3 fatty acid levels and increased oxidative DNA damage with varicocele in infertile men: a case control study. Reprod Fertil Dev. 2016;28(5):648–54. [DOI] [PubMed] [Google Scholar]
- 12.Safarinejad MR, Hosseini SY, Dadkhah F, Asgari MA. Relationship of omega-3 and omega-6 fatty acids with semen characteristics, and anti-oxidant status of seminal plasma: a comparison between fertile and infertile men. Clin Nutr. 2010;29(1):100–5. [DOI] [PubMed] [Google Scholar]
- 13.Esmaeili V, Shahverdi AH, Moghadasian MH, Alizadeh AR. Dietary fatty acids affect semen quality: a review. Andrology. 2015;3:450–61. [DOI] [PubMed] [Google Scholar]
- 14.Craig LB, Brush RS, Sullivan MT, Zavy MT, Agbaga MP, Anderson RE. Decreased very long chain polyunsaturated fatty acids in sperm correlates with sperm quantity and quality. J Assist Reprod Genet. 2019;36:1379–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Diamond MP, Legro RS, Coutifaris C, Alvero R, Robinson RD, Casson P, et al. Letrozole, Gonadotropin, or Clomiphene for Unexplained Infertility. N Engl J Med. 2015;373:1230–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Diamond MP, Mitwally M, Casper R, Ager J, Legro RS, Brzyski R, et al. Estimating rates of multiple gestation pregnancies: sample size calculation from the assessment of multiple intrauterine gestations from ovarian stimulation (AMIGOS) trial. Contemp Clin Trials. 2011;32:902–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Diamond MP, Legro RS, Coutifaris C, Alvero R, Robinson RD, Casson P, et al. Assessment of multiple intrauterine gestations from ovarian stimulation (AMIGOS) trial: baseline characteristics. Fertil Steril. 2015;103(4):962–73 e4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Folch J, Lees M, Sloane Stanley GH. A simple method for the isolation and purification of total lipides from animal tissues. J Biol Chem. 1957;226(1):497–509. [PubMed] [Google Scholar]
- 19.Busik JV, Reid GE, Lydic TA. Global analysis of retina lipids by complementary precursor ion and neutral loss mode tandem mass spectrometry. Methods Mol Biol. 2009;579:33–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Iizuka-Hishikawa Y, Hishikawa D, Sasaki J, Takubo K, Goto M, Nagata K, et al. Lysophosphatidic acid acyltransferase 3 tunes the membrane status of germ cells by incorporating docosahexaenoic acid during spermatogenesis. J Biol Chem. 2017;292(29):12065–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Safarinejad MR, Safarinejad S. The roles of omega-3 and omega-6 fatty acids in idiopathic male infertility. Asian J Androl. 2012;14(4):514–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Rivera-Egea R, Garrido N, Sota N, Meseguer M, Remohi J, Dominguez F. Sperm lipidic profiles differ significantly between ejaculates resulting in pregnancy or not following intracytoplasmic sperm injection. J Assist Reprod Genet. 2018;35(11):1973–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Poulos A Very long chain fatty acids in higher animals--a review. Lipids. 1995;30(1):1–14. [DOI] [PubMed] [Google Scholar]
- 24.Rabionet M, Van Der Spoel AC, Chuang CC, Von Tumpling-Radosta B, Litjens M, Bouwmeester D, et al. Male germ cells require polyenoic sphingolipids with complex glycosylation for completion of meiosis: A link to ceramide synthase-3. Journal of Biological Chemistry. 2008;283(19):13357–69. [DOI] [PMC free article] [PubMed] [Google Scholar]