Multiple genetic polymorphisms have been shown to increase Alzheimer’s disease (AD) risk without guaranteeing its onset (Bertram et al., 2007). Though the mechanisms of genetic risk are largely unknown, one possibility is that risk gene-dependent vulnerabilities and age-dependent vulnerabilities may interact to trigger neurodegeneration (Risacher et al., 2010; Reinvang et al., 2013). Genome-wide association studies have repeatedly shown that the ε4 allele of the apolipoprotein E (APOE) gene is the strongest known genetic risk factor for AD, increasing risk by 200–300% and decreasing age of onset, in a dose-dependent manner (Farrer et al., 1997; Bertram et al., 2007; Harold et al., 2009; Fei and Jianhua, 2013). A related risk factor, apolipoprotein J (APOJ, also known as clusterin, CLU), also modestly increases AD risk, with the APOJ-C risk allele increasing AD risk by 10–15% (Harold et al., 2009; Lambert et al., 2009). These two AD risk genes encode similar proteins that associate with lipoproteins in the brain (Koch et al., 2001; Elliott et al., 2010) to interact with a shared family of cell surface receptors (Kounnas et al., 1995; Leeb et al., 2014) that promote neurite outgrowth (Nathan et al., 1994; Kang et al., 2005), cholesterol metabolism (for review, see Holtzman et al., 2012), and clearance of the AD pathological hallmark, amyloid-β (Demattos et al., 2004).

Apolipoprotein E genotype is associated with accelerated atrophy in the brain during disease progression (Shen et al., 2010), and the gray matter density in the medial temporal lobe (MTL) of APOE-ε4 carriers is particularly sensitive to this type of atrophy (Thomann et al., 2008; Fei and Jianhua, 2013). While the effect of APOE genotype on MTL atrophy has been well established in older adults, effects on the young brain are less clear. Whole brain analyses in healthy APOE-ε4 carriers report increased (Alexander et al., 2012), decreased (Wishart et al., 2006; O’dwyer et al., 2012; Knickmeyer et al., 2013), or unchanged (Mondadori et al., 2007; Filippini et al., 2009; Dennis et al., 2010; Samuraki et al., 2012; Stein et al., 2012; Matura et al., 2014) MTL volume depending on the experimental parameters. Although less studied than APOE-ε4, APOJ was also not associated with MTL differences in a large scale genome-wide association study (GWAS) meta-analysis (Stein et al., 2012). As with APOE, it may be that more sensitive analyses are necessary to detect differences induced by APOJ risk. Thus, our aim was to assay gray matter volume in a particular area of the MTL that is susceptible to AD pathology very early in the disease, the entorhinal cortex (Braak et al., 2011). The entorhinal cortex facilitates communication between the hippocampus and neocortex for memory consolidation. We sought to test whether APOE and APOJ genotypes affected entorhinal cortex volume in a cohort of young adults. In addition, the putative functional similarities between the proteins these genes encode (Kounnas et al., 1995; Koch et al., 2001; Elliott et al., 2010; Leeb et al., 2014) prompted us to explore whether APOE and APOJ polymorphisms may have a cumulative effect, which would suggest the potential of a common pathway for these risk factors in a region critical for AD pathogenesis.

Apolipoprotein E is the strongest genetic risk factor for AD, and the entorhinal cortex is one of the earliest brain regions to develop neuropathological lesions and atrophy in AD (Hyman et al., 1984; Gomez-Isla et al., 1996; Braak et al., 2011). Because pathogenesis of AD begins as early as two decades prior to the presentation of clinical symptoms (Bateman et al., 2012), it is plausible that a series of genetically-induced differences in brain structure—such as entorhinal cortex volume—could accumulate with aging to increase vulnerability to AD later in life. Here, we assayed entorhinal cortex volume in a cohort of young adults to directly investigate whether APOE and APOJ genetic AD risk alters the structure of this region in early adulthood. Apolipoprotein E-ε4 carriers exhibited greater entorhinal cortex volume in the right hemisphere (Figure 1), and further analysis revealed that the APOJ-C genotype was associated with higher bilateral entorhinal cortex volume in APOE-ε4 non-carriers (Figure 2). The highest risk group (APOE-ε4, APOJ-C) exhibited greater entorhinal cortex volume compared to the lowest risk group (non-APOE-ε4, non-APOJ-C) (Figure 3). When additive odds ratios of APOE and APOJ served as regressors for volumetric measurements, higher disease risk was associated with greater right entorhinal cortex volume.

Volumetric studies report conflicting results when examining MTL volume in APOE risk carriers depending on the experimental parameters. For instance, one study of infants 1–3 months old with family histories of schizophrenia found APOE-associated differences in temporal lobe volume, including the hippocampus and entorhinal cortex (Knickmeyer et al., 2013). However, this effect was not seen in a follow up study with 6–25 month old infants with no family history of schizophrenia (Dean et al., 2014). Similarly, while some evidence has suggested that healthy APOE-ε4 positive adults exhibit decreased bilateral hippocampal volume (O’dwyer et al., 2012) and gray matter density in the right medial temporal region (Wishart et al., 2006), another study in young adults reported that APOE-ε4 carriers showed relative increases in volume in the vicinity of the hippocampus (Alexander et al., 2012). Further contributing to the controversy, other studies have suggested no differences in MTL volume by APOE genotype (Mondadori et al., 2007; Filippini et al., 2009; Dennis et al., 2010; Samuraki et al., 2012; Stein et al., 2012; Matura et al., 2014). Less research has focused on the lesser AD risk factor, APOJ, and its effect on MTL volume. The limited extant research has suggested no difference exists in the MTL by APOJ genotype (Bralten et al., 2011; Stein et al., 2012). A meta-analysis of GWAS including over 7000 cognitively normal subjects showed that APOJ genotype was not significantly associated with hippocampal volume (Stein et al., 2012). However, this study did not focus on young adults or target the entorhinal cortex specifically. While the APOJ-C risk allele has no significant effect on gray matter volume by whole brain analyses in young adults, a related risk gene, CR1, was associated with smaller local gray matter volume in the entorhinal cortex in a prior study (Bralten et al., 2011).

The inconsistency of extant evidence could be due to differences in age, gender, exclusion criteria, statistical methods, family history, and how each region of interest was defined across studies. Another explanation may be that the entorhinal cortex is a relatively small region within the MTL, and few studies explicitly target the entorhinal cortex. Thus, it is possible that changes in entorhinal cortex volume are not large enough to alter volume of the MTL as a whole. By specifically targeting the entorhinal cortex in our study, we were able to detect differences in this relatively small MTL structure that may have otherwise been below the threshold for detection.

Our results indicate that the effects of APOE and APOJ may be stronger in the right hemisphere. Medial temporal lobe asymmetry has been linked with AD and disease progression. For instance, APOE-ε4 carriers have thinner left entorhinal cortex compared to the right, while those without APOE-ε4 show this asymmetry in the presence of AD pathology (Donix et al., 2013). Individuals with subjective memory impairment had reduced volumes of the hippocampus and entorhinal cortex bilaterally (Striepens et al., 2010). However, APOE-ε4 carriers without this impairment performed better on episodic memory tests and had larger right hippocampal volumes, as opposed to those with impairment performing worse with smaller right hippocampal volumes (Striepens et al., 2011). Interestingly, differences in symmetry were also reported in delusional AD patients with greater right temporal horn sizes than left, while non-delusional patients did not show this asymmetry (Geroldi et al., 2000). Although asymmetry effects have not yet reported for APOJ, it is possible that APOE may work in conjunction with APOJ to induce asymmetry.

With aging, reductions in MTL volume in the hippocampus and entorhinal cortex are correlated with cognitive decline in the elderly (Bergfield et al., 2010; Raz et al., 2010; Varon et al., 2011). However, reductions in gray matter volume during the course of brain development in adolescence and early adulthood, correlate with increased synaptic pruning (Gogtay et al., 2004). Thus, increased entorhinal cortex volume during this developmental window could indicate a deficit in neural efficiency. Our study involved individuals of an average age of 21.8 years old, all of whom are at these later stages of brain development. Prior evidence has linked the APOJ-C risk allele to abnormal brain development through decreased white matter integrity (Braskie et al., 2011) and subsequent altered coupling between the hippocampus and prefrontal cortex during memory processing, mirroring disrupted connectivity in patients (Erk et al., 2011). Other evidence has also linked APOE-ε4 to abnormal brain structural and functional development in young risk allele carriers, suggesting that APOE and APOJ risk may impede the pruning process in the entorhinal cortex.

Alternatively, these findings could contribute to evidence suggesting that the APOE and APOJ risk genes confer an evolutionary advantage early in life, but confer a disadvantage later in life by increasing AD risk. This antagonistic pleiotropy hypothesis of APOE and APOJ risk has been proposed in other contexts (e.g., see Tuminello and Han, 2011; Stevens et al., 2014). College-aged young adults carrying the APOE-ε4 allele have an advantage in verbal fluency, decision making, and memory (Zetterberg et al., 2009; Marchant et al., 2010; Jochemsen et al., 2012; Rusted et al., 2013; Green et al., 2014), but these advantages disappear by middle age and give way to impairment in old age. These beneficial effects early in life could be due in part to increased volume in the entorhinal cortex, allowing for more effective processing of memory-associated information.

Finally, larger entorhinal cortex volume in APOE and APOJ risk carriers could indicate more neuroinflammation in the entorhinal cortex of young adults. It has been postulated that APOE-ε4 may increase neuroinflammation during aging, thus increasing susceptibility to dementia later in life (e.g., for review, see Guo et al., 2004; Kim et al., 2009). Several studies have also linked APOJ-C risk to decreased CLU levels (Schurmann et al., 2011), which are thought to serve an anti-inflammatory role (Savkovic et al., 2007). Glia can contribute to half of brain volume changes (Snell, 2009), and studies have shown that APOE-ε4 is associated with increased neuroinflammation (Guo et al., 2004; Chen et al., 2005; Maezawa et al., 2006; Zhu et al., 2012). Because the entorhinal cortex is one of the earliest affected brain areas in AD, it is possible that the increase in entorhinal cortex volume could be due to a deficit in APOE and APOJ carriers to control neuroinflammation in the young adult brain.

Overall, our findings in a young adult cohort indicate an additive effect of two of the strongest AD risk factors on a brain locus at the epicenter of AD pathogenesis. Entorhinal cortex volume increased as a function of increasing APOE and APOJ genetic AD risk. These results contribute to a growing literature characterizing genetic markers of AD risk in the young brain, long before the first overt signals of Alzheimer’s appear. Thus, these findings inform our understanding of the role of these genetic risk factors in the normal brain, and contribute to the development of biomarker targets for future preventive AD therapies.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.