Gonadotropins Across the Lifespan: Their role in the Neurodevelopment-Neurodegeneration Continuum

Abstract

Gonadotropins, follicle-stimulating hormone (FSH) and luteinizing hormone (LH), mediate critical reproductive functions via the hypothalamus-pituitary-gonadal axis. Their levels fluctuate across the lifespan, particularly during puberty and menopause, and across the menstrual cycle. In addition to peripheral expression, gonadotropin receptors are widely expressed in the brain, notably in memory-associated regions such as the hippocampus and cortex. Alterations in FSH and LH during reproductive transitions correlate with structural and functional brain changes. Puberty disorders, including central precocious puberty (CPP) and congenital hypogonadotropic hypogonadism (CHH), show altered gray and white matter and functional connectivity in the default mode network (DMN), which supports memory and is disrupted early in Alzheimer’s disease (AD). Although preclinical evidence implicates gonadotropins in amyloid and tau pathology, studies of attention and memory have yielded inconsistent results. However, reproductive disorders such as primary ovarian insufficiency (POI) and polycystic ovary syndrome (PCOS) are associated with deficits in cognitive performance, altered DMN dynamics, and increased AD risk. Menopause, characterized by marked gonadotropin elevation, is also accompanied by alterations in brain structure, connectivity, amyloid and tau deposition, and cognition, with associations with FSH and LH that are underexplored. This review synthesizes a broad range of basic and clinical evidence across reproductive transitions and disorders, highlighting shared and distinct mechanisms by which gonadotropins influence brain development, aging, and AD risk, and suggesting directions for future research.

Introduction

Gonadotropins, follicle-stimulating hormone (FSH) and luteinizing hormone (LH), are central to reproductive function, yet their impact on the brain remains poorly understood. While there has been great interest in the relationships between sex hormones, cognition, and their neural correlates, most work has focused on sex steroids such as estradiol. This gap is meaningful given that FSH and LH fluctuate significantly across the lifespan and most notably at reproductive transitions such as puberty and menopause, which are times of great neurological shifts. Clinical populations with altered gonadotropin signaling, including central precocious puberty (CPP), congenital hypogonadotropic hypogonadism (CHH), primary ovarian insufficiency (POI), and polycystic ovary syndrome (PCOS), offer additional opportunities to understand how FSH and LH influence the brain. Although few studies have directly examined how FSH and LH relate to brain structure, function, protein pathology, or cognition, available evidence suggests they play neurotrophic roles during development and puberty and may act independently from each other, from sex steroids like estradiol, and across clinical cohorts.

This review synthesizes emerging evidence to address several central questions: how do gonadotropin-brain relationships vary across the lifespan and what role do they play in endocrinological and neurological clinical disorders? We highlight where significant gaps exist and how targeted research should clarify the role of FSH and LH in shaping cognitive outcomes and brain aging.

We used PubMed and Google Scholar to conduct a targeted survey of the literature, with terms including the names of the disorders of interest in combination with relevant endocrine and neuroscience terms, including “FSH,” “LH,” “human chorionic gonadotropin,” “gonadotropin-releasing hormone,” “hypothalamic-pituitary-gonadal axis,” “neuroimaging,” “cognition,” “neurodevelopment,” “neurodegeneration,” “amyloid,” and “tau.” We included studies published in any language and in all species and systems. We prioritized articles published since January 1, 2000 with an emphasis on the most recent literature, but included early, foundational studies when necessary due to limited recent data.

Gonadotropins in Neurodevelopment

Localization

Rodent brain mapping studies demonstrate extensive, yet distinct, expression patterns of FSH receptor (FSHR) and LH/Chorionic Gonadotropin Receptor (LHCGR) [1]. Both receptors are present in memory-associated regions such as the hippocampus and cortex, though they display distinct distributions. LHCGR density is highest in the olfactory and lateral ventricles, followed by the forebrain, cerebellum, olfactory bulb, and thalamus. FSHR is instead most localized in the third ventricle, rivaling its expression in reproductive tissues, followed by cerebellum, olfactory bulb, hippocampus, and cortex. The co-localization of FSH and its receptor in the same hippocampal neurons might regulate hippocampal function in a paracrine or autocrine manner, providing a potential avenue for its direct influence on synaptic plasticity or cognition [2]. See Table 1 for a summary of all neurodevelopmental findings.

Table 1. Gonadotropins and Neurodevelopment Findings.

Population	Finding	Reference
Localization
Mouse	Diffuse and distinct expression patterns of FSHR and LHCGR in ventricles	[1]
Rat	FSH and FSHR colocalize on hippocampal neurons	[2]
Human	LH transcription reduced in cortex and hippocampus in AD	[3]
Human	FSHR and LHCGR throughout hypothalamus, brainstem, thalamus, and cortex	[4]
Rat, Mouse	LH, hCG, and FSH cross the blood-brain barrier	[5-7]
Mouse	FSH activates C/EBPβ–AEP/δ-secretase pathway to induce AD pathology	[6,7]
Rat, Review	Gonadotropins use transporters to cross the blood-brain barrier	[8,9]
Rat	hCG increases neurite outgrowth and reduces DNA fragmentation	[11]
Human	Reduced LH/FSH associated with impaired ovarian steroidogenesis	[12]
Gestation
Vertebrates	GnRH neurons migrate from the olfactory placode to forebrain with Otx2 and Vax1	[14,15]
Rat	Prenatal stress reduces LH/FSH and accelerates testicular maturation	[16]
Mini-Puberty
Human, Rodent	Postnatal GnRH increases LH, FSH, testosterone, and estradiol	[18,20]
Mouse	FSH surge coincides with nitric oxide synthase activation	[21]
Sheep	FSH increases proliferation, self-renewal, and neuronal differentiation	[22]
Human	Gut microbiome relates to AD pathology and gonadotropins	[23-25]
Human	Early life stress associated with HPG axis dysregulation	[17,26-28]
Pre-Puberty
Rat	Neuropeptides and neurotransmitters modulate LH/FSH release	[29]
Mouse	GnRH neurons receive direct cholinergic input, increasing LH	[30]

Clinical work demonstrated comparable gonadotropin levels across cortex in those with and without Alzheimer’s disease (AD), with a trend toward increased FSHR expression in the frontal lobe in AD [1]. Further, LH transcription was reduced in the cortex and hippocampus of AD patients, whereas FSH expression did not differ [3]. Although comparable data in healthy adults remain limited, publicly-available expression data reinforces gonadotropins’ broad distribution [4]. LHCGR is most predominant in the brainstem, hypothalamus, thalamus, and frontal and occipital cortices. FSHR is most expressed in the hypothalamus, brain stem, thalamus, and temporal cortex. While FSH and LH levels are often tightly correlated, this anatomical differentiation highlights the need to investigate each independently rather than presuming equivalent brain actions [1].

Peripheral administration of LH, LH homolog human chorionic gonadotropin (hCG), and FSH crosses the blood-brain barrier and activates gonadotropin-sensitive neurons [5-7], a process that may be transporter-mediated [8,9]. This is clinically relevant, as peripherally circulating FSH binds to hippocampal and cortical FSHR leading to increased amyloid-beta (Aβ) and tau production through the C/EBPβ-AEP/δ-secretase pathway in mouse models [6]. Further, LHCGR activation initiated centrally or peripherally increased neurite-bearing cells and cellular protein, gonadotropin-releasing hormone (GnRH) feedback regulation, behavioral modulation, and alterations in the metabolism of pro-inflammatory mediators [10,11]. There is also limited evidence that central and peripheral gonadotropin pools may be regulated both interdependently and independently, as brain and serum LH had a trending inverse relationship in mice [3]. These data support the opportunity for peripherally delivered gonadotropin therapies to directly modulate neuropathology.

Peripheral hormone levels may fail to capture underlying dysfunction if downstream pathways are impaired. This “hypo-response” phenotype, diminished biological responses despite normal hormone levels, can arise from disruptions at any stage of the hypothalamic-pituitary-gonadal (HPG) axis [12]. This axis consists of the hypothalamus, secreting GnRH; the anterior pituitary, synthesizing LH and FSH; and the gonads, producing gametes and secreting sex hormones [13]. Early-life variations in signaling efficiency may result in persistent differences in neural responses to hormones, influencing susceptibility to neurological and reproductive disorders [12]. Disruptions at any stage of development, including specification, migration, axon guidance, or synaptogenesis, can lead to persistent alterations in HPG axis set-points throughout life [14].

Gestation

During fetal development, specialized GnRH-producing neurons originate in the olfactory placode and migrate to the hypothalamus, regulated by cell-autonomous transcription factors and a glial scaffold provided by olfactory ensheathing cells [14,15]. The placental environment critically shapes HPG-axis programming. In rats, prenatal stress suppressed LH and FSH, accelerated prepubertal testicular maturation, and lowered adult testosterone, indicating a trade-off between early reproductive timing and adult capacity [16]. Human studies mirror this pattern: one meta-analysis found that certain adversities, such as parental absence and sexual abuse, were linked to earlier puberty in girls [17].

Mini-Puberty

During the initial months after birth, infants undergo “mini-puberty,” a transient surge in reproductive hormones which facilitates synapse formation, neuronal maturation, and the development of cognitive circuitry [18]. This lasts approximately 3-6 months, though in females FSH may remain elevated for 2-4 years. This sex-specific time course may be attributed to hormonal regulation of hypothalamic neuronal nitric oxide synthase neurons, which govern GnRH release, though it is currently not well understood [19-21].

During mini-puberty, FSH promotes neurogenesis and differentiation, evidenced by increased neurosphere size and upregulation of neuronal markers [22]. Disruptions during this critical period, including prematurity, environmental toxins, or inflammation, can result in lasting “endocrine scars,” which may make some more susceptible to neurodegeneration [18]. Additionally developing during this period, the gut microbiome is linked to neuroinflammation and shows decreased diversity in AD that correlates with cerebrospinal fluid (CSF) pathology [23,24]. While there is a link between gonadotropin levels and microbiome composition in reproductive disorders, how these might relate to later life AD-related neuropathology remains unknown [25]. There is also a growing body of work investigating how early life adversity rewires HPG-axis regulation, though no longitudinal studies to date evaluate persistent FSH and LH dysregulation [26-28].

Pre-Puberty

Beginning in pre-puberty, the GnRH system integrates inputs from neurotransmitter systems to regulate hormonal output: glutamatergic excitatory signals, GABAergic inhibitory control, and nitric oxide termination signal [29]. Cholinergic projections originating from memory-associated brain regions directly contact GnRH neurons through distinct receptor subtypes [30].

Humans have 150 000 - 200 000 GnRH-producing neurons in the basal ganglia and basal forebrain, primary cholinergic sources, vastly outnumbering the roughly 2000 hypothalamic GnRH neurons [31]. Most extrahypothalamic GnRH neurons co-express choline acetyltransferase, many in the nucleus basalis of Meynert, suggesting a potential intersection between hormone signaling and the cholinergic hypothesis of AD, which implicates the acetylcholine (ACh) system as part of the core etiology of AD [31]. While ACh may influence GnRH neuron activity, direct modulation and disruption in AD remain theoretical. GnRH receptors in the hippocampus, cortex, and cerebellum regulate neuronal growth, synaptic connectivity, and cellular excitability. Therefore, age-related neuronal loss could simultaneously deplete both neuromodulatory systems critical for hippocampal-dependent memory and synaptic plasticity, contributing to cognitive deterioration [32].

Gonadotropins in Adolescence

Overview

At pubertal onset, elevated GnRH pulsatility from HPG stimulation increases release of LH and FSH, and the initiation of sexual development [33]. Not only is puberty a time of sexual and reproductive maturation, but it is also marked by immense social, behavioral, and psychological changes, which correlate with elevated gonadotropin receptor density in areas such as the hippocampus [34]. Cortical gray matter volume (GMV) decreases from synaptic pruning, and myelination increases in both cortical and subcortical regions [35]. Additionally, volumetric increases occur in the amygdala and hippocampus, regions critical for emotion, cognition, and memory [36,37].

When activation of the HPG-axis occurs in an untimely manner, disorders of pubertal development arise. CPP stems from early activation of the HPG axis, resulting in abnormally high levels of LH and FSH and sexual maturation before the appropriate age, 8 for girls and 9 for boys [38,39]. It is one of the more common pubertal disorders, comprising 0.2% of girls and <0.05% of boys in a Danish cohort [40].

In contrast, CHH manifests as late or absent pubertal onset. It stems from deficient GnRH and presents clinically as reduced LH and FSH, with a prevalence of around 1-10 per 100 000 live births [41,42]. CHH is a heterogeneous disorder which may feature midline defects, cranial nerve abnormalities, and neurodevelopmental anomalies. When anosmia or hyposmia is present, the disorder is classified as Kallmann syndrome [43]. Additionally, there is a genetic link between CPP and neurodevelopmental disorders including Autism Spectrum Disorder and Turner Syndrome, which also may show abnormal gonadotropin levels [44,45].

Genetics

Mutations in the genes encoding kisspeptin, a major excitatory neuropeptide in reactivating the HPG axis, and its receptors are associated with CPP. Additionally, the MKRN3 gene plays a role in pubertal inhibition, with mutations implicated in CPP and other early puberty disorders [46]. Mutations in the DLK1 gene, involved in differentiation and proliferation, have also been identified [47]. Major genetic variants account for ~50% of CHH cases [43]. Mutations in the gene encoding the GnRH receptor do not impact smell, while other common genetic variants often result in Kallmann syndrome, such as the ANOS1 gene, which is crucial for the proper migration of GnRH and olfactory neurons [43,48].

Neuroimaging

Investigations associating brain metrics with serum gonadotropins in puberty and pubertal disorders are rare. In one longitudinal twin study, FSH-level changes in girls were positively associated with GMV in the left hippocampus, right cerebellum, left prefrontal areas, and the left anterior cingulate and precuneus [49]. These regions are integral to major functional connectivity networks, including the default mode network (DMN) and salience network, which contribute to cognition and are implicated in neurodegenerative disorders. Another longitudinal twin study found higher LH levels in boys and girls were associated with larger global and region-specific white matter increases, including the cingulum, middle temporal gyrus, and corpus callosum [50]. While these studies suggest that gonadotropins are tied to gray and white matter increases in adolescence, whether they are also integral to the pruning that comes later in puberty is unknown. See Table 2 for a summary of all structural findings.

Table 2. Gonadotropins and Structural Neuroimaging Findings.

Neuroimaging Findings
		(+): positive correlation, (-): negative correlation
Population	Modality	↓↑: decreased/increased compared to healthy controls	Reference
Puberty
CPP	MRI	FSH (+) with GMV in the insula;	[38]
		↓GMV in L insula and L fusiform gyrus
CPP	MRI	FSH (+) with GMV in L hippocampus, R cerebellum,	[49]
		L prefrontal areas, L anterior cingulate, and precuneus
CPP	MRI	LH (+) with white matter volume in cingulum,	[50]
		middle temporal gyrus, and corpus callosum
Reproductive Years
Menstrual	MRI/DTI	LH (+) with ordered diffusion;	[80]
		FSH (+) with cortical thickness in DMN regions and
		restricted diffusion in temporal and occipital regions
POI	MRI/DTI	↓GMV in R/L olfactory cortex and R parahippocampus	[89]
		↑WMV in L middle frontal and R superior parietal gyri;
AD
At-Risk Women	MRI	FSH (-) with hippocampal volume, GMV in DMN hubs	[131]
		FSH (+) with Aβ load in frontal gyri
		FSH and LH (-) with GMV in frontal cortices
Healthy Men	MRI	LH (+) with plasma Aβ40	[132]
Men	PET	LH (+) with Aβ40, Aβ42,	[133]
(MCI, AD)		FSH (+) with Aβ40 and amyloid SUVR

Other studies leveraged functional MRI (fMRI) to investigate functional connectivity changes in CPP (see Table 3 for summary of all functional findings). Higher FSH was related to greater verbal comprehension performance in girls with CPP but not to local neural synchrony [51]. Girls with CPP also had lower GMV in the left insula and left fusiform, and decreased connectivity between the bilateral insula and middle frontal gyrus, as well as between the amygdala and left fusiform, brain regions associated with the salience and executive attention networks [38]. Further, FSH levels correlated with increased GMV in the insula. More recent work demonstrated increased functional connectivity between the right amygdala and bilateral insulae [52]. These salience network-predominant findings may relate to the social and emotional problems associated with CPP, and while functional networks shift to some extent to maintain normal neurodevelopment, higher-order executive and emotional regulation functions can remain vulnerable.

Table 3. Gonadotropins and fMRI Neuroimaging Findings.

Neuroimaging Findings
	(+): positive correlation, (-): negative correlation
Population	↓↑: decreased/increased compared to healthy controls	Reference
Puberty
CPP	↓FC in bilateral insula, amygdala, middle frontal, and L fusiform gyri	[38]
CPP	↑Functional Connectivity (FC): L inferior temporal, L superior temporal,	[49]
	L superior occipital, and R middle gyri
CPP	↑FC: bilateral insula and amygdala; ↓FC: R dorsolateral superior frontal,	[52]
	L middle frontal, R orbital superior frontal, and L anterior cingulate gyri
CPP	↑FC: calcarine gyrus and middle occipital gyrus	[55]
	↓FC: superior temporal gyrus (undergoing GnRH therapy)
CPP	No association between FSH or LH level and neural synchrony	[51]
CHH	↓FC: R gyrus rectus, orbitofrontal cortex, and R middle temporal gyrus	[53]
Gender	↑FC: L dorsolateral and bilateral rostrolateral prefrontal cortices	[57]
Dysphoria	(GnRH-treated MFs > FMs)
Reproductive Years
Menstrual	FSH and LH (+) with DMN-salience/attention networks anticorrelation	[77]
Menstrual	Largest global flexibility and DMN reorganization during ovulation	[78]
Menstrual	↑FC: R/L hippocampi with R/L superior parietal in late follicular phase	[81]
PCOS	↓activity: L middle frontal, L posterior cingulate, R middle occipital gyri	[95]
	LH (-) with paracingulate-inferior frontal gyrus FC
PCOS	↑FC: L inferior temporal, L inferior occipital, R superior frontal	[96]
	FSH and LH (+) with superior–middle frontal FC
Menopause
Menopause	FSH (+) with R IFG, L PFC, L temporal pole (verbal task)	[117]

Neuroimaging in hypogonadotropic hypogonadism has been sparse, primarily focused on the olfactory region, and lacking FSH or LH measurements. One study with idiopathic hypogonadotropic hypogonadism revealed decreased functional connectivity within the olfactory cortex compared to age-matched healthy controls [53]. Another study in adults with Kallman syndrome found hypermetabolism in fronto-limbic and default-mode networks, mirroring network changes seen in CPP, which correlated with anxiety and avoidance behaviors [54].

GnRH Treatments

GnRH agonists and antagonists are first-line treatments for puberty induction and suppression, and their use provides insight into the direct effects of LH and FSH on the brain, though research on the subject is exceedingly rare. In the sole study of girls with CPP who underwent GnRH therapy, there was increased interhemispheric connectivity in notable DMN regions compared to treatment-naive individuals [55]. This mirrors patterns of greater between-hemisphere connectivity in women compared to men, suggesting that this sex dimorphism could be at least in part due to women’s greater level of gonadotropins [56]. Literature on GnRH-mediated puberty suppression in gender-affirming care is similarly scarce and inconclusive, with studies reporting mixed effects on memory, executive functioning, and functional connectivity [57,58]. For example, there was increased activation in the left dorsolateral prefrontal cortex and bilateral rostrolateral prefrontal cortex in GnRH-treated male-to-female participants compared to GnRH treated female-to-male participants during an executive task, suggesting that gonadotropins play an important role in driving the emergence of sex-differentiated neural activation [57]. While GnRH therapeutics consistently demonstrate great psychological benefit in CPP, CHH, and gender-affirming care, reducing risks of depression, anxiety, suicidality, and suicidal ideation, longitudinal studies are needed to elucidate the effects of FSH and LH on neurodevelopment and to inform high-quality clinical interventions for patients undergoing puberty suppression or induction [59-64].

Gonadotropins in the Reproductive Years

Overview

The average woman has roughly 400 menstrual cycles in her lifetime, driven by fluctuations in steroid hormones and gonadotropins that prepare the body for pregnancy [65]. Cycle length averages 27 days but becomes increasingly variable as menopause approaches. Menstruation-related hormonal shifts can produce meaningful changes in mood and cognition, in some cases reducing quality of life. Yet, despite ~75 million women of reproductive age in the US alone who could stand to benefit [66], research on cognitive and neural changes across the menstrual cycle and in disorders of reproduction remains limited.

To capture the cycle’s dynamic effects on the brain, within-subject designs with 3-4 sampling timepoints are recommended: early follicular, with low and stable hormone levels; ovulation, with a peak in estradiol, FSH, and LH; and mid-luteal, characterized by elevated progesterone [67]. Although rare, dense-sampling studies (eg, daily scans and hormone measures) reveal more precise associations between premenstrual hormones and brain metrics. However, most existing work centers on estradiol or menstrual phase generally, and studies that address gonadotropins often rely on phase-based assumptions about hormone levels rather than direct serum measurements, obscuring the ability to isolate gonadotropin effects.

Cognition

The few studies that directly correlate gonadotropins with cognitive performance in cycling women show inconsistent, domain-specific results. One study reported modest positive associations between FSH and visuospatial and verbal sequential performance across the cycle, with LH showing negative associations, whereas another found no gonadotropin-cognition relationships [68,69]. In women using a GnRH agonist to suppress gonadotropins, there were similarly no cognitive changes, but again without direct hormone measurements [70].

While some individual studies in the broader menstrual-cycle literature show no cognitive differences by phase in healthy menstruation [71,72] or in menstrual-related disorders [73], others suggest follicular-phase advantages in language, abstraction, and working memory [74,75]. However, a meta-analysis reported improved verbal and spatial memory around ovulation, and heightened emotional processing, memory, and fear extinction in the luteal phase [76].

Neuroimaging

There is only one densely-sampled neuroimaging study to date directly correlating daily serum gonadotropins with functional connectivity in naturally cycling women. Here, Greenwell et al. (2023) identified two dominant network states that together accounted for more than 70% of all observed community configurations: one featuring opposed activation of the DMN and attention and salience networks and the other opposed activation of the DMN and control and dorsal attention networks [77]. The former was positively correlated with FSH and LH alone, while the second correlated with gonadotropins, estradiol, and progesterone, highlighting a broader influence by gonadotropins than sex steroid hormones. A second dense-sampling study without concurrent gonadotropin levels identified four network communities that were largely stable across the cycle, except for marked reorganization around ovulation when estradiol, FSH, and LH peak [78].

Among imaging studies collecting 2-4 sex hormone samples per cycle, the only functional connectivity study reported no relationship with gonadotropin levels [79]. Structural MRI showed that LH was associated with more ordered diffusion throughout the cycle, while FSH was linked to increased cortical thickness in DMN regions and restricted diffusion in temporal and occipital regions [80]. Additionally, connectivity between the bilateral hippocampus and bilateral superior parietal lobes was highest in the late follicular phase when FSH and LH are moderately elevated [81].

Primary Ovarian Insufficiency (POI)

POI is defined by the cessation of menses before age 40 with post-menopausal FSH levels and can occur naturally or following hysterectomy or oophorectomy [82]. It is a common cause of primary infertility, with a global prevalence of 3.5% and 11.3% in North America [83]. Its etiology may be viral, metabolic, autoimmune, genetic, and/or environmental [84].

No studies to date have investigated how FSH or LH relate to brain measures in POI. However, comparisons with age-matched premenopausal women with normal-range gonadotropins offer indirect insights. A recent meta-analysis reported that both early menopause (ages 41-50) and POI are linked to higher risk of all-cause dementia compared to menopause after age 50 [85]. A retrospective survival analysis demonstrated that POI increased risk for poorer verbal fluency, visual memory, verbal fluency, psychomotor speed, and cognitive decline, but not AD risk [86]. This contrasts with work demonstrating a link between surgical menopause and AD risk and neuropathology, which may be explained by age at surgery [87,88].

While rare, neuroimaging studies to date show no differences in cognition or global GMV compared to healthy controls [89]. The POI group had regionally-reduced GMV and increased white matter volume in memory and sensory processing-related areas [77]. Diffusion changes were also seen in the left precentral gyrus, supramarginal gyrus, and right temporal lobe. Younger women showed more pronounced brain changes, suggesting earlier onset may confer greater vulnerability.

Studies of women undergoing fertility treatments, which systematically alter FSH and LH, could provide a powerful model to clarify gonadotropin-brain relationships. However, research in these populations has focused primarily on stress and cognition in relation to estradiol, without analyzing gonadotropins directly [90,91].

Polycystic Ovary Syndrome (PCOS)

PCOS, a common endocrine disorder affecting 5-20% of reproductive-aged women in the US [92], is defined by androgen excess, ovulatory dysfunction, alterations in LH and LH/FSH ratio, and ovarian cyst morphology [93]. Women with PCOS are at increased risk for type 2 diabetes, obstetrical complications, gynecological cancers, depression, and cognitive dysfunction [92]. Thought to be due to genetic and environmental factors, PCOS involves altered HPG-axis signaling which drives higher LH and LH/FSH ratio in most women [94].

In newly diagnosed PCOS patients, imaging studies reveal decreased activity in key DMN regions [95]. Further, LH levels and LH/FSH ratio were associated with uncoupling of the DMN and ventral attention network in PCOS, which may alter cognition and neurodegeneration risk [95,96].

Studies comparing women with PCOS and healthy age-matched controls are more difficult to interpret given PCOS’s hormonal heterogeneity. Only one such study reported group differences in LH and FSH, finding elevated LH but not FSH in PCOS and showing greater temporal and parietal activation during the N-back task [97]. Studies without group-level gonadotropins have reported reduced diffusion metrics, poorer MoCA performance, deficits in attention and memory, as well as altered EEG spectral power [98-100].

Gonadotropins and the Menopausal Transition

Overview

Menopause is a physiological transition defined by the cessation of menstrual cycles and neurologically characterized by alterations in gray and white matter, functional connectivity, energy metabolism, amyloid and tau deposition, and cognition [101-105]. The rise in serum FSH is considered the most reliable hallmark of the menopausal transition, known as perimenopause, increasing approximately 15-fold 6 years before the final menstrual period and stabilizing by 2 years postmenopause [101-103]. This precedes an approximately 10-fold serum LH increase, which occurs approximately 3 years prior to menopause, prior to significant estrogen changes [104]. Serum FSH trajectories separate into three distinct groups showcasing different rates of increase and final levels, which may explain why menopausal cognitive decline is not universal and help women who are particularly vulnerable [106]. CSF FSH and LH levels are elevated in postmenopause, but only cross-sectional data are available [107]. Despite this, studies linking gonadotropins to cognition and brain correlates across menopause remain sparse and largely cross-sectional, complicating interpretation amid natural hormonal fluctuations.

Cognition

Cognitive changes across the menopause transition vary by domain. In one cohort, processing speed declined from pre- to peri- to postmenopause, while verbal intelligence and fluency increased [108]. Intermediate and delayed verbal episodic memory performance were negatively associated with reproductive age (years since last menstrual cycle), FSH, and LH levels. In pre- and perimenopausal women, FSH negatively correlated with working memory accuracy and positively with reaction time [109]. However, one longitudinal study reported no relationship between FSH or LH in these domains when adjusting for age [110]. Well-designed studies directly evaluating sex differences in the relationship between gonadotropins and cognition in cognitively-normal individuals are lacking, and existing sex-stratified analyses have yielded inconsistent results [111,112].

In postmenopausal women, the relationship between gonadotropin levels and cognition appears to vary by cognitive status. Among cognitively normal women, higher FSH and lower LH correlated with better performance on a multi-domain neurological battery [113,114]. In contrast, women with mild cognitive impairment (MCI) had higher FSH and FSH/estradiol ratio than healthy controls, with the latter emerging as the strongest MCI predictor [115].

Longitudinal studies clarify which hormonal factors most underlie these apparently divergent findings. A multilevel latent profile analysis grouped perimenopausal women by cognitive trajectories, and women with weaker verbal learning and memory across the transition exhibited less variability in FSH and estradiol, suggesting that hormonal fluctuations, rather than absolute levels, exert the most influence on cognition [116]. FSH was also higher in women with strength in verbal learning and memory, designated as “super-agers.” Without premenopausal data, it is unclear whether these groups reflect pre-existing differences in baseline cognitive performance and how earlier hormonal changes may have shaped these trajectories.

Taken together, while LH shows a consistent relationship with cognition across menopause, FSH exerts more variable effects, perhaps conferring neuroprotection only in cognitively-resilient subgroups. Longitudinal studies beginning in pre-menopause will be paramount for clarifying these relationships, especially given FSH’s importance as the most reliable marker of menopause and its distinct trajectories across the menopause transition.

Neuroimaging

Across the menopause transition, higher FSH related to poorer phonemic and semantic fluency as well as activation of the right inferior frontal, left prefrontal, and left temporal pole regions during a verbal fMRI task [117]. MEG studies of peri- and early postmenopausal women showed FSH associated with increased high-frequency oscillatory power in bilateral frontal cortices, affecting sensory gain control [118], whereas LH showed no effect. Likewise, EEG data revealed greater high-frequency activity in the frontal and central regions in early postmenopause compared to premenopause [119]. Beta-2 activity was particularly elevated in postmenopause and correlated positively with FSH, implicating FSH in cortical excitability, arousal, and attention.

Gonadotropins and Alzheimer’s Disease (AD)

Overview

AD is a rapidly growing global health crisis, with 78 million cases projected by 2030 [120]. It often presents clinically as a memory-predominant syndrome and is pathologically defined by tau tangles and Aβ plaques [121]. Women face twice the lifetime risk and steeper cognitive decline, likely due to a combination of sex- and gender-related factors, including lifetime hormonal fluctuations [122]. While relatively few studies examine a role for gonadotropins in cognition and neuropathology in AD, preclinical models provide a strong line of evidence implicating elevated gonadotropin levels in the accumulation of tau and amyloid plaques (Figure 1).

Figure 1.

Methods for FSH and LH manipulations in preclinical AD models.

Tau

A small body of preclinical work has linked FSH and LH to tau pathology. In 3xTg-AD mice and human neuroblastoma cell models, FSH promoted tau accumulation through the C/EBPβ/δ-secretase signaling pathway and resultant pro-inflammatory effects [6,7]. Inhibiting FSH with an antibody or viral vector reversed memory deficits and hippocampal tau, independent of sex or estrogen status; FSH was synergistically activated by ApoE4 but not ApoE3. Interestingly, they also found FSH-induced contextual memory deficits in females but not males, adding to the limited human literature on sex differences in the gonadotropin-cognition link.

Similar effects have been reported for LH. In APP mice, LH receptor knockout reduced phosphorylated tau levels in both sexes, particularly surrounding amyloid plaques [123]. In contrast, leuprolide acetate, a GnRH agonist that chronically suppresses FSH and LH, improved spatial memory and retention in ovariectomized 3xTg-AD mice compared with saline or estradiol treatment, but failed to reduce hyperphosphorylated tau [3]. This study illuminated additional pathways relevant to tau accumulation, including changes in Wnt pathway proteins such as GSK3β and beta-catenin. These discrepancies may be due to differences in age or the AD model, consistent with human evidence of a critical window for treatment.

Amyloid

Early work showed that chronic suppression of LH reduced Aβ42 and Aβ40 by 71% and 35%, respectively, in 3-month-old C57Bl/6J female mice and M17 neuroblastoma cells [124]. Here, LH treatment at adult and postmenopausal concentrations increased secreted Aβ, whereas only the postmenopausal levels elevated insoluble Aβ, suggesting that LH influences Aβ processing in early adulthood and aggregation later in life. Leuprolide also protected against cognitive decline and amyloid deposition in non-cycling female Tg2576-AD mice, outweighing estrogen loss [125]. However, in older ovariectomized 3xTg-AD mice, leuprolide improved cognition measures without improving amyloid pathology [3], a pattern replicated in males, where leuprolide treatment had no benefit on Aβ burden in gonadectomized mice and blocked the protective effects of androgen supplementation [126].

FSH inhibition using antibodies or viral vectors reduced Aβ40 and Aβ42 and plaque burden in ovariectomized 3xTg-AD mice, most significantly in ApoE4-expressing mice [6,7]. LH agonism via hCG blocked estradiol’s spatial memory benefits and increased global and hippocampal Aβ40, Aβ42, and oligomeric Aβ in ovariectomized rats [127]. Chronic LH elevation using slow-release pellets similarly worsened amyloid pathology in gonadectomized guinea pigs, with effects linked to altered hippocampal APP processing and increasing with treatment duration [128]. In contrast, in ovariectomized APP/PS1 female mice, hCG enhanced spatial memory performance, spine density, and immune gene downregulation without affecting Aβ [129]. These findings are difficult to reconcile due to variability in species, age, and sex, with the conflicting study being the only one to assess LH agonism in an AD mouse model.

Genetic models further underscore gonadotropin involvement. LH receptor knockout in APP-AD mice reduced cortical and hippocampal plaques, astrogliosis, p-tau, c-fos, and α7-nAChR expression, though females still showed greater plaque burden in all groups [123]. Similarly, FSH receptor knockout in 3xTg-AD females with estrogen supplementation improved spatial memory performance and reduced Aβ40 levels in ovariectomized and sham-surgery groups, and Aβ42 levels in the sham group [130].

Only one clinical study has examined the relationship between AD pathology and gonadotropins in women, showing that FSH, but not LH, was associated with reduced hippocampal volume and increased Aβ load in key frontal gyri [131]. Here, estradiol was not related to Aβ or cognition, suggesting effects are independent of classical sex steroid pathways. In cognitively-normal older men, LH positively correlated with plasma Aβ40 [132], and in a larger cohort spanning subjective memory complaints, MCI, and AD, LH positively correlated with Aβ40, Aβ42, and SUVR, while FSH correlated with Aβ40 [133]. These effects were driven by the subjective memory complaints group, with FSH alone related to SUVR in MCI, and neither gonadotropin was associated with amyloid in the AD group. Together, these findings suggest that FSH and LH may differentially influence amyloid biology, and most prominent in early disease stages.

AD Risk

Several reports show elevated FSH and LH in AD compared to age-matched controls, though one study found this pattern to be female-specific and dependent on hormone replacement status [134,135]. Genetically, the FSH receptor exon variant (AS/AS) conferred lower AD risk in women but not men [136].

Neuroimaging data in AD, too, imply sex differences in the relationship between gonadotropin levels and AD risk, with higher FSH and LH associated with faster temporal volume decline in women compared to men [137]. Additionally in women with genetic AD risk factors, FSH and LH correlated with reduced GMV in frontal cortices, and FSH additionally with DMN hubs such as the precuneus [131]. This aligns with postmortem work showing elevated FSH receptor expression in the frontal cortex of AD patients [1]. Conversely, another study reported lower cortical and hippocampal LH transcription in female AD patients, with no FSH differences [3].

Other Neurodegenerative Disorders

Gonadotropins have also been hypothesized to play a role in other neurodegenerative disorders via neuroinflammatory and age-related pathways. Limited available evidence suggests involvement of LH, more so than FSH, in Parkinson’s disease (PD), frontotemporal dementia, and Huntington’s disease. Interestingly, each of these diseases is more prevalent in men. Postmenopausal women with PD had similar FSH and lower plasma LH than healthy controls [138,139], while men showed no significant group differences [140]. There were also correlations between EEG β-band functional connectivity and FSH in PD males and LH in PD females, suggesting sex differences in the hormonal underpinnings of bradykinesia [141]. Men with Huntington’s disease exhibited similar serum FSH and lower serum LH than age-matched healthy controls [142], as well as frontotemporal dementia compared to AD in a cohort of estrogen-free women [135].

Discussion

We present evidence across species, developmental stages, and clinical contexts highlighting a significant role for FSH and LH in brain development and aging that is distinct from that of other sex hormones which have traditionally received more attention (see Figure 2). Gonadotropin receptors are broadly but distinctly distributed in the brain, where they influence neurogenesis and differentiation and interact with the cholinergic system. During puberty, rising gonadotropin levels are associated with increased gray and white matter volume and functional connectivity changes in high-order functional networks. In adulthood, gonadotropins are dynamic across the menstrual cycle and increase rapidly in menopause, with underexplored effects on cognition and functional connectivity. FSH and LH may play a critical role in AD: they induce tau and amyloid pathology in rodent models, and are elevated in clinical populations with AD. While these findings hint at compelling gonadotropin-brain relationships across the reproductive lifespan, major gaps in our understanding remain. Key challenges include a lack of studies using direct hormone measurements, analytical challenges in isolating the effects of individual hormones, and heterogeneity in preclinical models with limited clinical translation.

Figure 2.

Summary of FSH and LH actions across the lifespan. FSH and LH bind to g protein-coupled receptors diffusely across the brain (regions with detectable expression in color) to influence brain-based measures in the pubertal, reproductive, and menopausal stages, as well as in AD. Brain expression images were generated by the Human Protein Atlas [149,150].

The Importance of Direct Measurement of Hormones

While existing literature establishes a crucial role for gonadotropins in the brain, most studies lack biofluid FSH and LH levels. Typically, they rely on menstrual cycle phase or menopause status as proxies, which are insufficient to capture the unique contributions of gonadotropins to brain-based outcomes, particularly given their actions within a larger hormonal ecosystem [71,72]. Dense-sampling neuroimaging approaches, which include frequent hormone measurements coupled with neuroimaging, are uniquely suited to capture these dynamics, yet only one such study has been conducted [77]. This single-subject study, while pioneering, has limited generalizability. Longitudinal studies that capture hormonal shifts across the menopause transition have been informative when applied to cognitive trajectories, but studies incorporating neuroimaging and AD biomarkers are sorely needed [116]. While we present the breadth of existing work, readers should be cautious in drawing generalized conclusions, as FSH and LH behave differently depending on the biological context.

Addressing Statistical Challenges

FSH and LH are mostly tightly correlated but do not always act interchangeably. For example, in one study of women with AD, FSH alone was associated with Aβ load in frontal gyri and hippocampal volume [131], while another showed changes in hippocampal LH transcription but no changes in FSH [3]. This underscores the need for research that examines them independently and alongside sex steroids to understand their specific neurological contributions. However, disentangling their effects presents a substantial statistical challenge because hormone levels are often highly collinear. Relatively few studies have addressed this issue directly, often studying each hormone individually [137], using ratios [115], or rarely controlling for one while evaluating others [131]. Because sex hormones operate within a tightly coordinated system whose dynamics vary across the menstrual cycle and reproductive stages, sophisticated regression- or interaction-based modeling approaches are necessary to disentangle the unique contributions of each hormone. Interventional studies that leverage clinical contexts where hormone levels are directly manipulated, such as hormone replacement therapy, in vitro fertilization (IVF), or interventions with GnRH agents, would provide a still more powerful framework to assess individual hormone contributions. Such work will have significant implications for the timing and targeting of hormone-based therapeutic approaches.

Bridging the Gap: The Need for Translational Research

While FSH and LH are consistently linked to amyloid and tau pathology in preclinical pharmacological and genetic rodent models [7,123], clinical studies are rare and inconsistent [131]. The diversity of rodent AD models limits translatability, and none fully capitulate human AD pathology [143]. While rodent work has powerfully demonstrated that FSH induces neuropathology via pro-inflammatory pathways, no clinical studies have yet assessed this relationship [6]. This gap and others like it provide an opportunity for bench-to-bedside studies which apply parallel approaches to establish mechanistic pathways in humans. Recent advancements in rodent neuroimaging, such as the ability to combine rodent fMRI and Ca2+ imaging [144] and to map parallel rodent and human functional networks [145], enable direct rodent-to-human study designs that enhance translational interpretability.

Gonadotropins and AD and Related Dementias

Preclinical studies which strongly implicate FSH and LH in the development of both amyloid and tau pathology should strongly motivate researchers in ADRD, who have focused largely on the impact of estradiol on women’s increased AD risk, to explore a potential major role for gonadotropins in neurodegeneration [6,123,127]. Clinically, higher FSH in women with AD is associated with reductions in GMV in frontal regions, the hippocampus, and DMN regions, as well as increased frontal amyloid burden. LH, by contrast, is associated with reduced GMV in frontal regions in women and increased amyloid biomarkers in men with AD. LH also shows stronger relationships with other male-predominant neurodegenerative disorders [138-142]. Together with the weaker LH-cognition relationships compared to FSH across women’s lifespans, the pattern of sex differences in which gonadotropin is more influential may reflect differences in hormonal feedback: estradiol exerts potent negative feedback on FSH, more so than LH [146], whereas LH alone drives testosterone production [147]. Further work in both sexes should characterize distinct gonadotropin profiles across neurodegenerative disorders and explore whether these hormonal differences can inform sex-specific therapeutic approaches.

Next Steps

The gonadotropin-brain relationship is complex, dynamic, and appears to have major but understudied clinical implications for patients across the reproductive lifespan. To close the gap, populations with clinically manipulated or naturally shifting gonadotropin levels over short time periods, such as women receiving IVF treatments and those in the perimenopausal transition, offer valuable opportunities for longitudinal study. The IVF population is particularly well-suited for this research because specific hormones are tightly controlled while others are systematically manipulated. Densely sampling hormone levels, cognitive measures, and multimodal neuroimaging in these populations will clarify how dynamic hormonal changes influence the brain [77], with particularly important implications for brain aging. Additionally, recent advancements in awake rodent multimodal neuroimaging techniques will enhance the translational impact of preclinical mechanistic data [148]. Emerging evidence suggests several testable hypotheses. Over the course of aging, brain connectivity and cognition may track specifically with gonadotropin levels, perhaps more powerfully than with estradiol [77,106,116], and with treatment implications for neurodegenerative processes such as AD, which disproportionately impact women. Incorporating inflammatory cytokine levels and markers of microglial and astrocytic activation could further clarify the underlying processes by which gonadotropins impact brain circuitry [6]. Moving forward, integrating longitudinal human studies with mechanistic animal research will be critical for defining the role of gonadotropins in brain health and for informing new approaches to the early identification and treatment of neurological disease.

Glossary

FSH

follicle-stimulating hormone

LH

luteinizing hormone

CPP

central precocious puberty

CHH

congenital hypogonadotropic hypogonadism

POI

primary ovarian insufficiency

PCOS

polycystic ovary syndrome

FSHR

FSH receptor

LHCGR

LH/Chorionic Gonadotropin Receptor

hCG

human chorionic gonadotropin

GnRH

gonadotropin-releasing hormone

HPG

hypothalamic-pituitary-gonadal

Ach

acetylcholine

AD

Alzheimer’s disease

GMV

gray matter volume

DMN

default mode network

MRI

magnetic resonance imaging

fMRI

functional MRI

MCI

mild cognitive impairment

Aβ

amyloid-beta

Author Contributions

JAG (ORCID: https://www.orcid.org/0000-0002-9796-533X): Conceptualization (lead), Investigation (lead), Supervision (lead), Visualization (equal), Writing - Original Draft Preparation (lead), Writing – Review & Editing (lead); MZE: Investigation (equal), Writing - Original Draft Preparation (equal), Writing – Review & Editing (equal); JFR: Investigation (equal), Visualization (equal), Writing - Original Draft Preparation (equal), Writing – Review & Editing (equal); TAAM: Investigation (equal), Visualization (equal), Review & Editing (equal); CAF (ORCID: https://www.orcid.org/0000-0002-7051-6382): Supervision (lead), Writing – Review & Editing (equal), Funding Acquisition (lead).

Funding Statement

R01AG089467 to CF; T32NS131190 to JG.