Journal of Men's Health,2024,20(11):78-87 DOI:10.22514/jomh.2024.187
Original Research

Structural insights into cortical morphometry and erectile dysfunction: a Mendelian randomization and genetic correlation study

Chen Fu1,2, Tianhong Wang1, Qihai Wan1, Jiacheng Jiang1, Xiaofeng Ou3,*,

1Department of Anesthesiology, West China Hospital, Sichuan University, 610041 Chengdu, Sichuan, China

2West China School of Nursing, Sichuan University, 610041 Chengdu, Sichuan, China

3Department of Critical Care Medicine, West China Hospital, Sichuan University, 610041 Chengdu, Sichuan, China

*Corresponding Author(s):ouxiaofeng@wchscu.edu.cn (Xiaofeng Ou)

History Submitted: 11 May 2024 | Accepted: 07 June 2024 | Published: 30 November 2024
Copyright:  ©2024 The Author(s). Published by MRE Press.
This is an open access article under the CC BY 4.0 license (https://creativecommons.org/licenses/by/4.0/).

Collapse table of contents

Abstract

Previous studies have disclosed that abnormal brain dynamics are involved in erectile dysfunction (ED). However, the association of cortical morphometry with ED remains unclear. To investigate the association of the surface area and thickness of the total cerebral cortex and 34 regions with ED, this Mendelian Randomization (MR) study was performed. Summary statistics of the surface area and thickness of cerebral cortex and ED were retrieved from previous genome-wide association studies. Genetic correlations between cortical morphometry and ED were evaluated by high-definition likelihood and linkage disequilibrium score regression (LDSC). Causal inference was made by the inverse variance weighting (IVW) estimator, supplemented by the MR-Egger, Weighted median, Maximum likelihood, MR. robust adjusted profile score, and MR-Pleiotropy Residual Sum and Outlier methods. The effect size of the surface area was scaled according to the standard deviation (SD). For surface area, the IVW estimator showed that a one SD increase of the entorhinal cortex and superior frontal cortex corresponded to a 1.25-fold (95% confidence interval (CI) = 1.03–1.53, p = 0.026) and a 1.24-fold (95% CI = 1.02–1.50, p = 0.031) risk of ED, respectively. LDSC regression found a significant genetic correlation between the surface area of the entorhinal cortex and ED (rg = 0.093, p < 0.05). For thickness, the IVW method revealed that genetically proxied one millimeter increase of the inferior temporal cortex and paracentral cortex led to a 6.58-fold (95% CI = 1.16–37.14, p = 0.033) and a 7.99-fold (95% CI = 1.17–54.49, p = 0.034) risk of ED, respectively. The other five methods also supported these findings. No heterogeneity or pleiotropy was detected in the instrumental variables. This study supports that genetically proxied cortical morphometry is associated with an elevated risk of ED. This study provides new insights into the etiology of ED.

Keywords:Cerebral cortex;Cortical morphometry;Erectile dysfunction;Mendelian randomization;Surface area;Thickness
PDF(8.29 MB)|EndNote (RIS)|BibTeX|RefMan|RefWorks

Cite this article

Chen Fu, Tianhong Wang, Qihai Wan, Jiacheng Jiang, Xiaofeng Ou. Structural insights into cortical morphometry and erectile dysfunction: a Mendelian randomization and genetic correlation study.Journal of Men's Health,2024,20(11):78-87 DOI:10.22514/jomh.2024.187

1. Introduction

Erectile dysfunction (ED) is a prevalent male urological condition characterized by the continual or repeated inability of the penis to become erect and/or achieve satisfactory sexual intercourse [1]. The latest epidemiological systematic review on the prevalence of ED revealed that the global incidence of ED among young men is as high as 30% [2]. Risk factors such as cardiovascular diseases, smoking, obesity, obstructive sleep apnea syndrome, dyslipidemia, diabetes, metabolic syndrome, stress, anxiety, and depression all contribute to an increased incidence of ED [1, 3, 4].

Penile erection is a physiological and psychological event induced by the coordinated regulation of the central nervous system, endocrine system, and peripheral neurovascular system in response to sexual stimulation received by the individual [5]. Sexual stimuli received through sensory modalities such as touch, vision, hearing, and smell are ultimately transmitted to the cerebral cortex and further relayed to the hypothalamic nuclei via the amygdala [6]. The hypothalamus integrates various sexual stimuli and further transmits them to the lower central nervous system-spinal cord (T11–L2 and S2–4), which controls psychologically induced erection via the peripheral nerves [7, 8].

Previous studies using functional magnetic resonance imaging (fMRI) have explored the mechanisms of sexual arousal (SA) and identified several cortical or subcortical activation sites associated with penile erection. These include the prefrontal cortex (PFC), parietal cortex, lateral occipitotemporal cortex, cingulate cortex, inferior temporal cortex, insula, caudate nucleus, putamen, thalamus, amygdala and hypothalamus [9, 10]. The activations within these circuits contribute to a comprehensive neurobehavioral framework of male SA, encompassing cognitive, incentive, affective, and autonomic elements [11]. Research indicates that in the absence of sexual stimuli, certain brain areas, such as the medial and left lateral orbitofrontal cortex (OFC), angular gyri, lateral temporal cortex, and posterior cingulate cortex are involved in exerting continuous inhibitory control over the SA [9, 12]. Conversely, these areas become less active during the onset of SA to reduce such inhibition [10].

In the realm of male sexual dysfunction, several functional neuroimaging investigations have identified anomalies in the central mechanisms governing male sexual functioning [13, 14]. Specifically, individuals with psychogenic ED have exhibited abnormal increases or prolongations of activity in brain areas such as the OFC [10], the medial and basal PFC [15], the lateral hypothalamus [16], and the anterior middle cingulate cortex [17], which are recognized to regulate the inhibitory processes underlying male SA [10]. Conversely, there was reduced activity in the inferior temporal cortex, which is implicated in the cognitive aspects of male SA [18, 19], as well as in the insula, hippocampus and anterior cingulate cortex, which are implicated in modulating autonomic, cognitive, and motivational aspects related to male SA [13]. Although ED is predominantly viewed as a functional issue, its neuroanatomical underpinnings have received limited consideration [10]. Nonetheless, a structural MRI study concentrating on subcortical structures detected gray matter atrophy in the nucleus accumbens and hypothalamus among individuals with ED, which is thought to impact the motivational aspect of male sexual arousal [20]. Another investigation employing diffusion-based imaging disclosed microstructural alterations in white matter tracts across several regions in ED patients, suggesting deficits in neuroanatomical connectivity in ED patients [21]. Zhao et al. [11] investigated brain structure in 40 ED patients and 39 healthy controls, and revealed that reduced cortical thickness in regions associated with disrupted male SA, which correlated with diminished sexual functioning in males. These findings suggest that in addition to functional deficits, structural changes in the brain may play a role in ED.

Observational studies can only reflect whether there is an association between the two, making it challenging to elucidate the complex causal interplay between cortical morphology and ED in a clear and intuitive manner [22]. Mendelian randomization (MR) serves as a technique for evaluating causal relationships [23]. MR utilizes genetic variants as instrumental tools to overcome confounding factors and reverse causation issues encountered in traditional observational epidemiological research because alleles follow Mendel’s random allocation law during gamete formation, and the relationship between genes and outcomes remains unaffected by common confounding variables such as postnatal environmental factors and behavioral lifestyles [24]. Additionally, it meets the temporal requirement of causality inference, known as the “cause precedes effect”, making it a natural randomized controlled trial [25]. Therefore, we employed a two-sample MR approach using genome-wide association study (GWAS) databases to investigate potential causal relationships between cortical morphometry and ED.

2. Materials and methods

2.1 Study design and included samples

The study design and schematic diagram are illustrated in Fig. 1. The total surface area (mm2) and average thickness (mm) of the total cerebral cortex and 34 regions were used as exposures and ED was used as an outcome.

Schematic diagram of the study design and analysis strategy. ED: erectile dysfunction; GWAS: genome-wide association studies; IVW: inverse variance weighting; IV: instrumental variables; MR: Mendelian randomization; MR-PRESSO: MR Pleiotropy Residual Sum and Outlier; MAF: minor allele frequency; MR.RAPS: MR robust adjusted profile score; SNP: single nucleotide polymorphisms; LDSC: linkage disequilibrium score regression; HDL: High-Definition Likelihood.

Fig. 1.Schematic diagram of the study design and analysis strategy. ED: erectile dysfunction; GWAS: genome-wide association studies; IVW: inverse variance weighting; IV: instrumental variables; MR: Mendelian randomization; MR-PRESSO: MR Pleiotropy Residual Sum and Outlier; MAF: minor allele frequency; MR.RAPS: MR robust adjusted profile score; SNP: single nucleotide polymorphisms; LDSC: linkage disequilibrium score regression; HDL: High-Definition Likelihood.

Summary-level data regarding the surface area and thickness of the total cerebral cortex and 34 regions were retrieved from Grasby KL’s study [26]. This consortium collected data from 60 cohorts globally and increased the sample size to 51,665 individuals. Of them, 94% were of European descent. To avoid bias from the mixed population, we only used the data from individuals of European descent (33,992 participants) as exposures. The 33,992 participants included 23,909 individuals from 49 cohorts participating in the Enhancing NeuroImaging Genetics through Meta-Analysis Consortium (ENIGMA) and 10,083 individuals from the UK Biobank. The surface area and thickness of the human cerebral cortex were detected by structural brain magnetic resonance imaging. The 34 regions were defined by the Desikan-Killiany atlas, which is commonly used in clinics. For one specific region, the global measure of surface area or thickness was adjusted to identify region-specific genetic variants. Finally, 35 phenotypes for surface area and 35 phenotypes for thickness were generated. Detailed information about these data sources can be found in Table 1.

Table 1.Characteristics of included GWASs.
TraitsSample sizeAuthorsDescentPMIDnSNPF statisticsUnit
Entorhinal cortex33,992Grasby KL et al. [26]European321932963023.71mm2 (surface area)
Superior frontal cortex33,992Grasby KL et al. [26]European321932963023.98mm2 (surface area)
Inferior temporal cortex33,992Grasby KL et al. [26]European321932962721.35mm (thickness)
Paracentral cortex33,992Grasby KL et al. [26]European321932962323.86mm (thickness)
Erectile dysfunction223,805 (6175 cases and 217,630 controls)Bovijn J et al. [27]European30583798--logOR

SNP: Single nucleotide polymorphism; PMID: PubMed Identifier; OR: odds ratio.

The genetic association of ED was obtained from one previous GWAS generated by Bovijn J and his colleagues [27]. This GWAS combined individuals from three cohorts: the UK biobank, Estonian Genome Center of the University of Tartu cohort, and hospital-recruited Partners HealthCare Biobank cohort. A total of 6175 cases and 217,630 controls of European descent were genotyped. The diagnosis of ED relied on the international classification of diseases-10 codes (F52.2 or N48.4), oral medications, surgical intervention for ED, or self-reports from the respondents.

2.2 Selection of genetic instruments

As shown in Fig. 1, the IVs were first selected as a genome-wide significance threshold of p < 1 × 10−5. Given the correlations of single nucleotide polymorphisms (SNPs), we further clumped the retrieved IVs at a window size of 1 Mb. The 1000G phase III reference genome was used as the reference genome and IVs with r2 ≥ 0.001 were filtered to ensure independence [24]. Generally, SNPs with minor allele frequencies <0.01 were considered rare SNPs and were excluded from our analyses. To satisfy the basic assumption of MR, the remaining IVs were subjected to weak IVs and pleiotropic outlier tests. The strength of the IVs was evaluated by F statistics (F statistics = (β/Se)2) and weak IVs were defined as F statistics < 10, as in previous studies [24, 28]. In our study, no weak IVs were detected. In addition, pleiotropic outliers were investigated by the radial MR and MR Pleiotropy Residual Sum and Outlier (MR-PRESSO) approaches [25]. If outliers were identified, these SNPs were excluded and not used as IVs. Finally, to reduce bias from reverse causality, the Steiger-MR test was adopted to confirm that SNPs explained more variance in exposures than in the outcome [29].

2.3 Statistical analyses

To infer the causal associations between the cerebral cortex and ED, an inverse variance weighting (IVW) estimator was used as the primary method to combine the effects of IVs. This method assumes that all IVs are valid IVs and thus has greater statistical power than other estimators. In addition to the IVW method, we also employed five other methods for sensitivity analyses, namely, MR-Egger, Weighted median, Maximum likelihood, MR.RAPS and MR-PRESSO. The MR-Egger [30] and Weighted median [31] methods have the ability to yield consistent causal estimations when invalid IVs exist. Additionally, the intercept of ME-Egger regression can be used to quantify horizontal pleiotropy. Given the relatively limited sample size of the exposures, we also adopted Maximum likelihood estimator as a supplement for its high statistical power in limited sample datasets. Furthermore, to mitigate bias from pleiotropic outliers, the MR-PRESSO [32] method was used. If pleiotropic outliers were identified, this method first excluded the outliers and then used the same technique as IVW to combine the effects of the IVs. Finally, the MR.RAPS estimator was supplemented due to its robustness when weak IVs and both systematic and idiosyncratic pleiotropy exist.

Given the tiny unit for the surface area of the cortex (mm2), the effect size, namely the β value, seems to be very small. Thus, we scaled the effect size according to the standard deviation (SD). The odds ratios (ORs) for the entorhinal cortex and superior frontal cortex corresponded to a 1 SD alteration in surface area. For the thickness of the inferior temporal cortex and paracentral cortex, the effect size was not scaled. The ORs corresponded to a one-unit alteration (mm) in thickness. For exposures with significant associations with ED, we further evaluated the genetic correlation using the linkage disequilibrium score (LDSC) [33] regression and the High-Definition Likelihood (HDL) [34] method.

All the analyses were performed with R 4.0.1 software (R Foundation for Statistical Computing, Vienna, Austria). In this study, p < 0.05 was used as a significance threshold for statistical inference.

3. Results

As summarized in Fig. 2, only the entorhinal cortex and superior frontal cortex were significantly associated with an increased risk of ED across the 35 phenotypes of surface area. For the 35 phenotypes of thickness, only the inferior temporal cortex and paracentral cortex were significantly associated with an increased risk of ED. The other areas of the cortex were not statistically associated with ED in our results.

Summary of results on the impact of 70 cortical phenotypes on ED. The size of the p value is mapped to the depth of colors in the heatmap. SA: surface area; TH: thickness. IVW: inverse variance weighting; MR: Mendelian randomization; MR.RAPS: MR.robust adjusted profile score.

Fig. 2.Summary of results on the impact of 70 cortical phenotypes on ED. The size of the p value is mapped to the depth of colors in the heatmap. SA: surface area; TH: thickness. IVW: inverse variance weighting; MR: Mendelian randomization; MR.RAPS: MR.robust adjusted profile score.

3.1 Causal association between the surface area of the entorhinal cortex and ED

As shown in Table 2, there was a significant association between the surface area of the entorhinal cortex and ED. The IVW estimator showed that a 1 SD increment in the surface area of the entorhinal cortex corresponded to a 1.25-fold (95% CI = 1.03–1.53, p = 0.026) risk of ED, in line with the scatter plot in Fig. 3A. The ORs were 1.26 (95% CI = 1.04–1.52, p = 0.016) for the Maximum likelihood, 1.25 (95% CI = 1.03–1.53, p = 0.034) for the MR-PRESSO and 1.26 (95% CI = 1.04–1.53, p = 0.020) for the MR.RAPS estimators, respectively. The statistical inferences by the MR-Egger and Weighted median estimators were insignificant (p = 0.187 for MR-Egger and p = 0.177 for Weighted median). However, the effect size and direction remained consistent with those of the other four methods. No horizontal pleiotropy was detected by the MR-Egger intercept (p > 0.05) and Global test (p > 0.05). In Fig. 3B, Cochrane’s Q test identified no heterogeneity (p > 0.05). Of note, LDSC regression found a significant genetic correlation between the surface area of the entorhinal cortex and ED (rg = 0.093, p < 0.05, Fig. 3C). The correlation coefficient was 0.216 as revealed by the HDL method.

Table 2.Causal estimates between cortical morphometry and ED.
Exposuresβ (95% CI)OR (95% CI)pp for Eggerp for Global test
Entorhinal cortex
IVW0.00369 (0.00044–0.00694)1.25 (1.03–1.53)0.0260.5230.208
MR-Egger0.00671 (−0.00302–0.01643)1.50 (0.83–2.72)0.187
Weighted median0.00296 (−0.00120–0.00713)1.20 (0.92–1.56)0.177
Maximum likelihood0.00382 (0.00076–0.00688)1.26 (1.04–1.52)0.016
MR-PRESSO0.00369 (0.00044–0.00694)1.25 (1.03–1.53)0.034
MR.RAPS0.00380 (0.00059–0.00701)1.26 (1.04–1.53)0.020
Superior frontal cortex
IVW0.00054 (0.00005–0.00104)1.24 (1.02–1.50)0.0310.8330.651
MR-Egger0.00068 (−0.00068–0.00205)1.31 (0.76–2.24)0.336
Weighted median0.00038 (−0.00031–0.00106)1.16 (0.88–1.53)0.296
Maximum likelihood0.00055 (0.00005–0.00106)1.25 (1.02–1.52)0.030
MR-PRESSO0.00054 (0.00008–0.00101)1.24 (1.03–1.49)0.029
MR.RAPS0.00055 (0.00002–0.00108)1.24 (1.01–1.53)0.041
Inferior temporal cortex
IVW1.88332 (0.15199–3.61465)6.58 (1.16–37.14)0.0330.9030.717
MR-Egger2.11341 (−1.94472–6.17155)8.28 (0.14–478.93)0.317
Weighted median2.53569 (0.13113–4.94024)12.63 (1.14–139.80)0.039
Maximum likelihood1.91834 (0.14201–3.69466)6.81 (1.15–40.23)0.034
MR-PRESSO1.88332 (0.31218–3.45446)6.58 (1.37–31.64)0.027
MR.RAPS2.26860 (0.39534–4.14187)9.67 (1.48–62.92)0.018
Paracentral cortex
IVW2.07854 (0.15897–3.99810)7.99 (1.17–54.49)0.0340.6570.894
MR-Egger0.81288 (−5.01110–6.63686)2.25 (0.01–762.70)0.787
Weighted median1.51203 (−1.17314–4.19719)4.54 (0.31–66.50)0.270
Maximum likelihood2.16316 (0.20584–4.12049)8.70 (1.23–61.59)0.030
MR-PRESSO2.07854 (0.54078–3.61629)7.99 (1.72–37.20)0.015
MR.RAPS2.14370 (0.10090–4.18650)8.53 (1.11–65.79)0.040

Given the tiny unit for the surface area of the cortex (mm2), the effect size, namely the β value, seems to be very small. Thus, we scaled the effect size according to the standard deviation (SD). The ORs for the entorhinal cortex and superior frontal cortex corresponded to one SD alteration. For the thickness of the inferior temporal cortex and paracentral cortex, the effect size was not scaled. The ORs corresponded to a one-unit alteration (mm). F statistics = (β/Se)2. IVW: inverse variance weighting; MR: Mendelian randomization; MR-PRESSO: MR Pleiotropy Residual Sum and Outlier; ME.RAPS: MR.robust adjusted profile score; CI: confidence interval; OR: odds ratio.

3.2 Causal association between the surface area of the superior frontal cortex and ED

In Table 2, the IVW method found that one 1 SD increase in the surface area of the superior frontal cortex was associated with a 1.24-fold (95% CI = 1.02–1.50, p = 0.031) risk of ED, in line with the results of the Maximum likelihood (OR = 1.25, 95% CI = 1.02–1.52, p = 0.030), MR-PRESSO (OR = 1.24, 95% CI = 1.03–1.49, p = 0.029), and MR.RAPS (OR = 1.24, 95% CI = 1.01–1.53, p = 0.041). The scatter plot in Fig. 3D also revealed a positive correlation between the surface area of the superior frontal cortex and ED. Like for the entorhinal cortex, the MR-Egger and Weighted median estimators identified increased risks of ED (OR = 1.31 for MR-Egger and OR = 1.16 for Weighted median) with insignificant p values (both p > 0.05). This lack of significance may be attributed to the relatively lower statistical power relative to that of the IVW method. No horizontal pleiotropy was detected by the MR-Egger intercept (p > 0.05) and Global test (p > 0.05). In Fig. 3E, Cochrane’s Q test identified no heterogeneity (p > 0.05). LDSC did not show a significant genetic correlation between the surface area of the superior frontal cortex and ED (rg = −0.066, p > 0.05), nor did HDL (rg = 0.120, p > 0.05, Fig. 3C).

Scatter and funnel plots of the surface area of cortical morphometry with ED. (A,B): scatter and funnel plots visualizing the SNP effect on the surface area of the entorhinal cortex and erectile dysfunction, and the heterogeneity in the instrumental variables. (C) genetic correlation between cortical morphometry and ED. (D,E): scatter and funnel plots visualizing the SNP effect on the surface area of the superior frontal cortex and erectile dysfunction, and the heterogeneity in the instrumental variables. In Fig. 3C, the size of the correlation coefficient (rg) is mapped to the depth of colors in the heatmap. *: p &lt; 0.05. IVW: inverse variance weighting; MR: Mendelian randomization; MR.RAPS: MR.robust adjusted profile score; IV: instrumental variables; SNP: Single nucleotide polymorphism. LDSC: linkage disequilibrium score regression; HDL: High-Definition Likelihood.

Fig. 3.Scatter and funnel plots of the surface area of cortical morphometry with ED. (A,B): scatter and funnel plots visualizing the SNP effect on the surface area of the entorhinal cortex and erectile dysfunction, and the heterogeneity in the instrumental variables. (C) genetic correlation between cortical morphometry and ED. (D,E): scatter and funnel plots visualizing the SNP effect on the surface area of the superior frontal cortex and erectile dysfunction, and the heterogeneity in the instrumental variables. In Fig. 3C, the size of the correlation coefficient (rg) is mapped to the depth of colors in the heatmap. *: p < 0.05. IVW: inverse variance weighting; MR: Mendelian randomization; MR.RAPS: MR.robust adjusted profile score; IV: instrumental variables; SNP: Single nucleotide polymorphism. LDSC: linkage disequilibrium score regression; HDL: High-Definition Likelihood.

3.3 Causal association between the thickness of the inferior temporal cortex and ED

A greater thickness of the inferior temporal cortex is linked to a greater risk of ED (Table 2). Genetically proxied a one mm increase in the thickness of the inferior temporal cortex led to a 6.58-fold (95% CI = 1.16–37.14, p = 0.033) risk of ED, in line with the scatter plot in Fig. 4A. The result was further verified by the Weighted median (OR = 12.63, p = 0.039), Maximum likelihood (OR = 6.81, p = 0.034), MR-PRESSO (OR = 6.58, p = 0.027), and MR.RAPS (OR = 9.67, p = 0.018) estimators. The funnel plot in Fig. 4B showed a symmetric distribution of IVs, revealing no heterogeneity. There was no pleiotropy as revealed by the MR-Egger intercept (p = 0.903) and Global test (p = 0.717). In Fig. 3C, the correlation coefficients of LDSC and HDL were 0.013 and 0.376, respectively (both p > 0.05).

3.4 Causal association between the thickness of the paracentral cortex and ED

In Table 2, the IVW model found that a genetically predicted one mm increase in the thickness of the paracentral cortex was associated with a 7.99-fold risk of ED (95% CI = 1.17–54.49, p = 0.034). The scatter plot showed that with the increase of the SNP effect on the thickness of paracentral cortex, the SNP effect on ED also ascended (Fig. 4C). The ORs were 2.25 (95% CI = 0.01–762.70, p = 0.787) for the MR-Egger, 4.54 (95% CI = 0.31–66.50, p = 0.270) for the Weighted median, 8.70 (95% CI = 1.23–61.59, p = 0.030) for the Maximum likelihood, 7.99 (95% CI = 1.72–37.20, p = 0.015) for the MR-PRESSO, and 8.53 (95% CI = 1.11–65.79, p = 0.040) for the MR.RAPS, respectively. Although the results from the MR-Egger and Weighted median were not statistically significant (both p > 0.05), the directions were consistent with those of the IVW estimator, reinforcing the causal link between the paracentral cortex and ED. No pleiotropy or heterogeneity was detected (all p > 0.05, Table 2, and Fig. 4D). In Fig. 3C, neither LDSC nor HDL identified genetic correlation between the thickness of the paracentral cortex and ED (p > 0.05).

Scatter and funnel plots of the thickness of the cortical morphometry with ED. (A,B): scatter and funnel plots visualizing the SNP effect on the thickness of the inferior temporal cortex and erectile dysfunction, and the heterogeneity in the instrumental variables. (C,D): scatter and funnel plots visualizing the SNP effect on the surface area of the paracentral cortex and erectile dysfunction, and the heterogeneity in the instrumental variables. IVW: inverse variance weighting; MR: Mendelian randomization; MR.RAPS: MR.robust adjusted profile score; IV: instrumental variables; SNP: Single nucleotide polymorphism.

Fig. 4.Scatter and funnel plots of the thickness of the cortical morphometry with ED. (A,B): scatter and funnel plots visualizing the SNP effect on the thickness of the inferior temporal cortex and erectile dysfunction, and the heterogeneity in the instrumental variables. (C,D): scatter and funnel plots visualizing the SNP effect on the surface area of the paracentral cortex and erectile dysfunction, and the heterogeneity in the instrumental variables. IVW: inverse variance weighting; MR: Mendelian randomization; MR.RAPS: MR.robust adjusted profile score; IV: instrumental variables; SNP: Single nucleotide polymorphism.

4. Discussion

This MR study utilized extensive population-based genetic data concerning cerebral cortical structure alongside MR techniques to delineate directional and biological links between human cortical morphometry and ED. Our study’s substantial sample size enhanced the statistical power compared to that of prior investigations into brain structure and ED. Our findings suggest that genetically mediated increases in surface areas of the entorhinal and superior frontal cortex are associated with a heightened risk of ED, with a notable genetic correlation detected between the entorhinal cortex surface area and ED. Furthermore, genetically predicted greater thickness of the inferior temporal and paracentral cortex is implicated in an elevated risk of ED. These results carry significance, as a deeper comprehension of the underlying neuroanatomical associations is likely to enhance both the diagnosis and treatment strategies for ED.

Research exploring the link between cortical morphometry and male sexual dysfunction remains limited. Guo et al. [35] discovered cortical thickening in various brain regions in premature ejaculation patients, with stronger associations found in specific areas such as the right medial OFC, right precentral gyrus, and left superior frontal cortex. However, inconsistent with our research findings, Zhao et al. [10] investigated cortical thickness abnormalities using structural MRI in 40 patients with psychogenic ED and 39 healthy controls. Their findings revealed significantly reduced cortical thickness in multiple regions associated with male SA dynamics, including the medial prefrontal, orbitofrontal, cingulate, inferotemporal, and insular cortices, which were associated with a decline in male sexual function. The discrepancy in results may stem from variations in sample size and demographic characteristics among the study populations, consequently impacting the statistical power and applicability of the findings. This emphasizes the complexity of ED as a multifaceted pathological process involving extensive alterations in the cerebral cortex.

To the best of our knowledge, this study represents the first to uncover a genetic association between increased surface area of the entorhinal cortex and superior frontal cortex and elevated risk of ED. As individuals age, cortical volume, thickness, and surface area across various regions of the frontal lobe notably decline [36]. Specifically, the volume, thickness, and surface area of the entorhinal cortex exhibit an upward trend with advancing age, peaking around 32, 40 and 50 years, respectively, before declining thereafter [37]. The entorhinal cortex serves as a crucial link between the temporal neocortex and hippocampus and is closely tied to cognitive functions [38]. Few investigations have explored the associations between regional cortical thickness and surface area and brain function [39]. The scarcity of studies employing various imaging modalities can be ascribed in part to the intricacies involved in aligning or registering datasets containing structural (thickness) and functional (activation) data among and between participants [39]. Indeed, as far as we are aware, no studies have directly associated fMRI signals with the underlying anatomical features acquired from structural MRI. Research on the connection between the entorhinal cortex and sexual function is scarce. In a neuroimaging study conducted on women, women with no history of sexual dysfunction exhibited markedly greater activation in the bilateral entorhinal cortex than did those with hypoactive sexual desire disorder when viewing erotic video materials [40]. Consistently, the superior frontal cortex exhibited the most substantial decline associated with age in terms of volume, thickness, and surface area [41]. This region is believed to play a significant role in higher cognitive functions, with a particular emphasis on working memory [42]. Regarding the association between the superior frontal cortex and ED, previous research by Yin et al. [43] found that the incidence of psychogenic ED was linked to irregular functional connectivity (FC) between the left dorsolateral superior frontal cortex and angular gyrus, along with FC between the posterior cingulate cortex and precuneus. A recent study discovered reduced fractional amplitude of low-frequency fluctuation (fALFF) and FC values in the dorsolateral superior frontal cortex of psychogenic ED patients, suggesting that dysfunction in this brain region might lead to difficulties in attentional focus during sexual activities and a diminished capacity to focus on pleasure and sensations, contributing to lower sexual satisfaction among patients [44]. However, the underlying mechanisms behind the variability in the surface area of the superior frontal cortex observed in ED patients require further investigation.

Our study also revealed that higher thickness in the inferior temporal cortex and paracentral cortex was associated with an increased risk of ED. Stoléru et al. [9] found that the inferior temporal cortex region responds to visual sexual stimuli through deactivation, which is believed to aid in the cognitive processes of SA. Functional neuroimaging investigations focusing on normal male SA have suggested that the activation pattern of the inferior temporal cortex may be associated with distinguishing stimuli as either sexual or nonsexual, and the degree of inferior temporal cortex activation appears to correlate with the intensity of erection [16, 45]. However, contrary to our findings, Zhao et al. [10] discovered cortical thickness loss in both inferior temporal cortex regions among patients with psychogenic ED. Further research and replication studies are needed to clarify these discrepancies and establish more robust conclusions. The paracentral cortex governs both motor and sensory functions of the contralateral lower extremity, as well as cortical regulation of micturition and defecation [46]. Allen et al. [47] demonstrated that activation in a specific area of the paracentral cortex was observed following mild self-stimulation of the penile shaft. A resting-state fMRI study conducted in healthy subjects demonstrated functional connectivity between the paracentral cortex and other frontal and parietal regions, supporting motor function and spatial attention [48]. While our study suggested that altered paracentral cortex thickness is a potential indicator of ED, further investigations into specific motor dysfunctions and disturbances in attentional subdomains are warranted to determine its predictive value for ED.

This study has several limitations. Firstly, the data utilized were aggregated GWAS statistics data in public databases, lacking detailed information at the individual outcome level. This deficiency could introduce an unavoidable bias into our findings, and there is an inherent challenge in avoiding sample overlap. Due to the polygenic or omnigenic nature of intricate diseases and characteristics, widespread pleiotropy exists in the human genome, making it possible to pinpoint and take into account all conceivable confounding variables. Secondly, our study only conducted a unidirectional MR analysis. We did not explore the causality between ED and cortical morphometry. Thirdly, our study solely established a correlation between alterations in brain cortical structure and ED. Although we have offered potential interpretations for certain observed cortical structural changes associated with ED in the discussion section, further exploration is needed to uncover the underlying mechanisms. Fourthly, in this study, both the groups exposed to the factor and those experiencing the outcome were limited to individuals of European descent, with no examination of other racial or ethnic groups. This omission results in ongoing uncertainties when extending conclusions about the relationship between cortical morphometry and ED risk to non-European populations. Acknowledging this, more endeavors verifying the association between cortical morphometry and ED in other races are beneficial to provide more robust conclusions.

5. Conclusions

Based on genetic data from the cerebral cortex, this study provides evidence that genetically proxied higher surface areas of the entorhinal and superior frontal cortex are associated with an elevated risk of ED. There is a significant genetic correlation between the surface area of the entorhinal cortex and ED. In addition, genetically predicted higher thickness of the inferior temporal and paracentral cortex is associated with an increased risk of ED. This study investigates the influence of the central cerebral cortex on ED and provides a better understanding of the etiology of ED.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Author contributions

CF and XFO—Conceptualization; writing-review & editing. CF and THW—Data curation. THW—Formal analysis. CF, JCJ and QHW—writing-original draft.

Ethics approval and consent to participate

Ethical review and approval were waived for this study, all the data from Mendelian randomization is publicly accessible. Informed consent was obtained from all subjects in the original genome-wide association studies.

Acknowledgment

The authors sincerely thank Tao Lin in statistical assistance.

Funding

This research received no external funding.

Conflict of interest

The authors declare no conflict of interest.

References

MacDonald SM, Burnett AL. Physiology of erection and pathophysiology of erectile dysfunction. Urologic Clinics of North America. 2021; 48: 513–525.

[Google Scholar]

Nguyen HMT, Gabrielson AT, Hellstrom WJG. Erectile dysfunction in young men-a review of the prevalence and risk factors. Sexual Medicine Reviews. 2017; 5: 508–520.

[Google Scholar]

Kim H, Kim KT, Shin WC, Yang KI, Ha JY, Cho YW. Erectile dysfunction in patient with obstructive sleep apnea: effects of continuous positive airway pressure. The Aging Male. 2024; 27: 2317165.

[Google Scholar]

Zhang X, Zhu Z, Tang G. Global prevalence of erectile dysfunction and its associated risk factors among men with type 1 diabetes: a systematic review and meta-analysis. International Journal of Impotence Research. 2024; 36: 365–374.

[Google Scholar]

Xu J, Chen Y, Gu L, Liu X, Yang J, Li M et al. Hypothalamic-pituitary-adrenal axis activity and its relationship to the autonomic nervous system in patients with psychogenic erectile dysfunction. Frontiers in Endocrinology. 2023; 14: 1103621.

[Google Scholar]

Thomas C, Konstantinidis C. Neurogenic erectile dysfunction. Where do we stand? Medicines. 2021; 8: 3.

[Google Scholar]

Bogacki-Rychlik W, Gawęda K, Bialy M. Neurophysiology of male sexual arousal—behavioral perspective. Frontiers in Behavioral Neuroscience. 2024; 17: 1330460.

[Google Scholar]

MacDonald SM, Burnett AL. Physiology of erection and pathophysiology of erectile dysfunction. Urologic Clinics of North America. 2021; 48: 513–525.

[Google Scholar]

Stoléru S, Fonteille V, Cornélis C, Joyal C, Moulier V. Functional neuroimaging studies of sexual arousal and orgasm in healthy men and women: a review and meta-analysis. Neuroscience & Biobehavioral Reviews. 2012; 36: 1481–1509.

[Google Scholar]

Zhao L, Guan M, Zhang X, Karama S, Khundrakpam B, Wang M, et al. Structural insights into aberrant cortical morphometry and network organization in psychogenic erectile dysfunction. Human Brain Mapping. 2015; 36: 4469–4482.

[Google Scholar]

Zhao L, Guan M, Zhu X, Karama S, Khundrakpam B, Wang M, et al. Aberrant topological patterns of structural cortical networks in psychogenic erectile dysfunction. Frontiers in Human Neuroscience. 2015, 9: 675.

[Google Scholar]

Poeppl TB, Langguth B, Laird AR, Eickhoff SB. The functional neuroanatomy of male psychosexual and physiosexual arousal: a quantitative meta‐analysis. Human Brain Mapping. 2014; 35: 1404–1421.

[Google Scholar]

Cera N, Di Pierro ED, Sepede G, Gambi F, Perrucci MG, Merla A, et al. The role of left superior parietal lobe in male sexual behavior: dynamics of distinct components revealed by fMRI. The Journal of Sexual Medicine. 2012; 9: 1602–1612.

[Google Scholar]

Redouté J, Stoléru S, Pugeat M, Costes N, Lavenne F, Le Bars D, et al. Brain processing of visual sexual stimuli in treated and untreated hypogonadal patients. Psychoneuroendocrinology. 2005; 30: 461–482.

[Google Scholar]

Montorsi F. Brain activation patterns during video sexual stimulation following the administration of apomorphine: results of a placebo-controlled study. European Urology. 2003; 43: 405–411.

[Google Scholar]

Redouté J, Stoléru S, Grégoire MC, Costes N, Cinotti L, Lavenne F, et al. Brain processing of visual sexual stimuli in human males. Human Brain Mapping. 2000; 11: 162–177.

[Google Scholar]

Georgiadis JR, Farrell MJ, Boessen R, Denton DA, Gavrilescu M, Kortekaas R, et al. Dynamic subcortical blood flow during male sexual activity with ecological validity: a perfusion fMRI study. NeuroImage. 2010; 50: 208–216.

[Google Scholar]

Liu Q, Zhang P, Pan J, Li Z, Liu J, Li G, et al. Cerebral activity changes in different traditional chinese medicine patterns of psychogenic erectile dysfunction patients. Evidence-Based Complementary and Alternative Medicine. 2015; 2015: 503536.

[Google Scholar]

Hagemann JH, Berding G, Bergh S, Sleep DJ, Knapp WH, Jonas U, et al. Effects of visual sexual stimuli and apomorphine SL on cerebral activity in men with erectile dysfunction. European Urology. 2003; 43: 412–420.

[Google Scholar]

Cera N, Delli Pizzi S, Di Pierro ED, Gambi F, Tartaro A, Vicentini C, et al. Macrostructural alterations of subcortical grey matter in psychogenic erectile dysfunction. PLOS ONE. 2012; 7: e39118.

[Google Scholar]

Zhang P, Liu J, Li G, Pan J, Li Z, Liu Q, et al. White matter microstructural changes in psychogenic erectile dysfunction patients. Andrology. 2014; 2: 379–385.

[Google Scholar]

Zhong Q, Jiang L, An K, Zhang L, Li S, An Z. Depression and risk of sarcopenia: a national cohort and Mendelian randomization study. Frontiers in Psychiatry. 2023; 14: 1263553.

[Google Scholar]

Burgess S, Foley CN, Allara E, Staley JR, Howson JMM. A robust and efficient method for Mendelian randomization with hundreds of genetic variants. Nature Communications. 2020; 11: 376.

[Google Scholar]

Xiong Y, Zhang F, Zhang Y, Wang W, Ran Y, Wu C, et al. Insights into modifiable risk factors of erectile dysfunction, a wide-angled Mendelian Randomization study. Journal of Advanced Research. 2024; 58: 149–161.

[Google Scholar]

Verbanck M, Chen CY, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nature Genetics. 2018; 50: 693–698.

[Google Scholar]

Grasby KL, Jahanshad N, Painter JN, Colodro-Conde L, Bralten J, Hibar DP, et al. The genetic architecture of the human cerebral cortex. Science. 2020; 367: eaay6690.

[Google Scholar]

Bovijn J, Jackson L, Censin J, Chen CY, Laisk T, Laber S, et al. GWAS identifies risk locus for erectile dysfunction and implicates hypothalamic neurobiology and diabetes in etiology. American Journal of Human Genetics. 2019; 104: 157–163.

[Google Scholar]

Xiong Y, Zhang F-X, Zhang Y-C, Wu C-J, Qin F, Yuan J-H. Genetically predicted insomnia causally increases the risk of erectile dysfunction. Asian Journal of Andrology. 2023; 25: 421–425.

[Google Scholar]

Zhang Y, Xiong Y, Shen S, Yang J, Wang W, Wu T, et al. Causal association between tea consumption and kidney function: a mendelian randomization study. Frontiers in Nutrition. 2022; 9: 801591.

[Google Scholar]

Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. International Journal of Epidemiology. 2015; 44: 512–525.

[Google Scholar]

Bowden J, Davey Smith G, Haycock PC, Burgess S. Consistent estimation in mendelian randomization with some invalid instruments using a weighted median estimator. Genetic Epidemiology. 2016; 40: 304–314.

[Google Scholar]

Verbanck M, Chen CY, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nature Genetics. 2018; 50: 693–698.

[Google Scholar]

Bulik-Sullivan BK, Loh PR, Finucane HK, Ripke S, Yang J, Patterson N, et al. LD score regression distinguishes confounding from polygenicity in genome-wide association studies. Nature Genetics. 2015; 47: 291–295.

[Google Scholar]

Ning Z, Pawitan Y, Shen X. High-definition likelihood inference of genetic correlations across human complex traits. Nature Genetics. 2020; 52: 859–864.

[Google Scholar]

Guo F, Xi YB, Gao M, Liu L, Fei NB, Qin W, et al. Alterations in cortical thickness in nonmedicated premature ejaculation patients: a morphometric MRI study. Journal of Magnetic Resonance Imaging. 2018; 47: 656–662.

[Google Scholar]

Király A, Szabó N, Tóth E, Csete G, Faragó P, Kocsis K, et al. Male brain ages faster: the age and gender dependence of subcortical volumes. Brain Imaging and Behavior. 2016; 10: 901–910.

[Google Scholar]

Wang Y, Hao L, Zhang Y, Zuo C, Wang D. Entorhinal cortex volume, thickness, surface area and curvature trajectories over the adult lifespan. Psychiatry Research: Neuroimaging. 2019; 292: 47–53.

[Google Scholar]

Hasan KM, Mwangi B, Cao B, Keser Z, Tustison NJ, Kochunov P, et al. Entorhinal cortex thickness across the human lifespan. Journal of Neuroimaging. 2016; 26: 278–282.

[Google Scholar]

Hegarty CE, Foland-Ross LC, Narr KL, Townsend JD, Bookheimer SY, Thompson PM, et al. Anterior cingulate activation relates to local cortical thickness. NeuroReport. 2012; 23: 420–424.

[Google Scholar]

Arnow BA, Millheiser L, Garrett A, Lake Polan M, Glover GH, Hill KR, et al. Women with hypoactive sexual desire disorder compared to normal females: a functional magnetic resonance imaging study. Neuroscience. 2009; 158: 484–502.

[Google Scholar]

Lemaitre H, Goldman AL, Sambataro F, Verchinski BA, Meyer-Lindenberg A, Weinberger DR, et al. Normal age-related brain morphometric changes: nonuniformity across cortical thickness, surface area and gray matter volume? Neurobiology of Aging 2012; 33: 617.e1–9.

[Google Scholar]

du Boisgueheneuc F, Levy R, Volle E, Seassau M, Duffau H, Kinkingnehun S, et al. Functions of the left superior frontal gyrus in humans: a lesion study. Brain. 2006; 129: 3315–3328.

[Google Scholar]

Yin T, Liu Q, Ma Z, Li Z, Sun R, Ren F, et al. Associations between altered cerebral activity patterns and psychosocial disorders in patients with psychogenic erectile dysfunction: a mediation analysis of fMRI. Frontiers in Psychiatry. 2020; 11: 583619.

[Google Scholar]

Liu X, Liu S, Liu T, Tang L, Ji M, Xu Y, et al. Altered regional brain activity and functional connectivity in resting-state brain networks associated with psychological erectile dysfunction. Frontiers in Neuroscience. 2023; 17: 1074327.

[Google Scholar]

Ferretti A, Caulo M, Del Gratta C, Di Matteo R, Merla A, Montorsi F, et al. Dynamics of male sexual arousal: distinct components of brain activation revealed by fMRI. Neuroimage. 2005; 26: 1086–1096.

[Google Scholar]

Patra A, Kaur H, Chaudhary P, Asghar A, Singal A. Morphology and morphometry of human paracentral lobule: an anatomical study with its application in neurosurgery. Asian Journal of Neurosurgery. 2021; 16: 349–354.

[Google Scholar]

Allen K, Wise N, Frangos E, Komisaruk B. Male urogenital system mapped onto the sensory cortex: functional magnetic resonance imaging evidence. The Journal of Sexual Medicine. 2020; 17: 603–613.

[Google Scholar]

Salvador R, Suckling J, Coleman MR, Pickard JD, Menon D, Bullmore E. Neurophysiological architecture of functional magnetic resonance images of human brain. Cerebral Cortex. 2005; 15: 1332–1342.

[Google Scholar]