The current study yielded two main findings. Despite exhibiting no significant differences in memory performance or grey matter volume, young healthy adult APOE
e4 carriers exhibited significantly greater ESA in bilateral hippocampus/PHG compared to non-carriers. Additionally, while exhibiting greater overall ESA in these MTL regions, e4 carriers exhibited globally reduced connectivity to other task-associated cortical regions relative to non-carriers (see ). Notably, exceptions to the reduced functional connectivity in the e4 carriers were in regions known to evince some of the earliest functional activation changes in e4 adults [i.e., PCC; (Chen, Reiman, Alexander, Caselli, Gerkin, Bandy et al., 2007
; Reiman, Chen, Alexander, Caselli, Bandy, Osborne et al., 2004
; Reiman, Chen, Alexander, Caselli, Bandy, Osborne et al., 2005
)]. These results argue against a positive (or anagonistic) pleiotropic role of the APOE
e4 allele in successful memory encoding in young adulthood. It is still possible that e4 carriers may derive an early beneficial effect, but results suggest that any window for benefit may be much earlier in the developmental process.
Focusing on our first finding, young adult APOE
e4 carriers, compared to non-carriers, exhibited greater ESA activation within the MTL (see ), specifically within the hippocampus and PHG. However, both groups exhibited equitable memory performance and were found to have comparable volumes of grey matter within the MTL, indicating that these functional differences cannot be explained by either differences in memory performance or MTL atrophy. Results are consistent with previous studies in older adults which also found increased MTL activity for APOE
e4 carriers compared to noncarriers (Bondi, Houston, Eyler and Brown, 2005
; Dickerson, Salat, Bates, Atiya, Killiany, Greve et al., 2004
; Han, Houston, Jak, Eyler, Nagel, Fleisher et al., 2007
; Wishart, Saykin, Rabin, Santulli, Flashman, Guerin et al., 2006
). These studies concluded that increased activations (in the absence of behavioral and volumetric differences) reflect a compensatory mechanism in APOE
e4 carriers. Specifically, that e4 carriers required additional activation, perhaps reflecting additional cognitive effort in memory-related tasks in order to maintain similar performance to noncarriers.
The increased MTL activation in e4 carriers observed in the current study may reflect an innate difference in functionality of this region compared to non-carriers, which, in turn, may be indicative of premorbid functional weakness or generally non-contributory hyperactive processing in regions known to be sensitive to late-onset AD pathology (Morishima-Kawashima, Oshima, Ogata, Yamaguchi, Yoshimura, Sugihara et al., 2000
; Selkoe, 2002
). The prospect of APOE
-driven innate differences support work from Shaw et al. (2007)
who found a linear decrease in anterior MTL grey matter volume in healthy adolescents as a function of APOE allele carrier status (i.e., APOE e2 > e3 > e4 carriers). Differences between our results and Shaw's in regards to MTL volume may be based on differences in methodology (e.g., separation of genotyped groups).
In addition to these group differences in ESA, our second finding focused on observed group differences in functional connectivity with the MTL. Despite exhibiting greater ESA in bilateral hippocampus, the APOE
e4 carriers exhibited reduced functional connectivity with these regions compared to non-carriers. Taken together results suggest that APOE
e4 non-carriers may be better able to integrate processing mediated by the MTL with that of other brain regions, regions also involved in ESA. The reduced functional connectivity exhibited by the APOE
e4 carriers may be indicative of the fact that activation in this region is modulated relatively independently of other brain regions. APOE
e4 carriers did exhibit increased connectivity with the ventral PCC, a region exhibiting reduced metabolism and altered activations in those genetically at risk for late-onset AD (Reiman, Chen, Alexander, Caselli, Bandy, Osborne et al., 2004
). In accordance with the abovementioned hyperactivation of the MTL by APOE
e4 carriers, results may reflect a change in the underlying physiology of this region that is present at an early age. Whether these changes in MTL activation and functional connectivity reflect early changes associated with the disease state or are a causal factor in developing said impairments is a research question that needs to be answered.
Again, whether these functional changes represent a compensatory mechanism of the MTL network, or a more fundamental difference in functional brain organization associated with functional variants of the APOE
genotype has yet to be determined. Bondi and colleagues (Han and Bondi, 2008
) have argued that the APOE
gene may have pleiotropic (beneficial) effects in young adults, which serve to enhance cognitive performance. Likewise, Mondadori and colleagues (Mondadori, de Quervain, Buchmann, Mustovic, Wollmer, Schmidt et al., 2007
) posit that their young adult cohort exhibited beneficially increased neural activity. The current results do not appear to support the notion of pleiotropic effects in memory function by young adulthood, as increased MTL activity and connectivity to the PCC in our e4 carriers did not result in enhanced, but equitable memory performance compared to non-carriers. Furthermore, results support recent finding from Filippini and colleagues (2009)
who concluded that the presence of the APOE e4 allele was associated in greater MTL activity during encoding in healthy young adults (25-30 year olds; 18 carriers vs. 18 non-carriers) in the absence of behavioral or structural differences between the two groups.
Rather, the current findings fit better with either an inefficiency or a compensatory account of cognitive function. That is, based on the lack of behavioral differences between genotyped groups, increased MTL activity in e4 carriers may be interpreted as either a) inefficient processing associated with dysfunction in the MTL or b) compensatory processing necessary to achieve similar cognitive output to the seen in non-carriers who are able to rely on more efficient MTL functioning. As noted, these two ideas are not incompatible. The fact that, in e4 carriers, these task-induced increases in MTL activation are, in turn, associated with reductions in functional connectivity with this region suggest that this enhanced processing isn't reflected in other ESA regions. Rather, the observed changes are limited to the MTL and the PCC, another region affected early in AD pathology (Lustig, Snyder, Bhakta, O'Brien, McAvoy, Raichle et al., 2003
; Petrella, Prince, Wang, Hellegers and Doraiswamy, 2007
; Prince, Woo, Doraiswamy and Petrella, 2008
), indicating that they may reflect regional-specific neural alterations associated with the e4 allele. While the basis for these differences require future research, the results suggest that investigation should not be limited to individuals in older cohorts alone.
Furthermore, it should also be noted that our compensatory theory (defined above as increased processing necessary to achieve similar cognitive function) differs from the that of Han and Bondi (Han and Bondi, 2008
) in that their account is, in actuality, one of enhancement, rather than compensation in young e4 carriers. That is, Han and Bondi propose that “Young e4 participants perform better on memory and other neurocognitive tasks” (p. 252). They further suggest that increased neural recruitment in older e4 carriers would help compensate for cognitive declines. Their theory does not outline whether enhanced (not comparable) performance in young carriers is accompanied by increased or comparable neural recruitment compared to non-carriers. The current results observed increased MTL recruitment and equal
performance in e4 carriers compared to non-carriers, suggesting (as describe above) either the increased activation is inefficient in some way or that young e4 carriers require increased neural processing to accomplish similar performance to non-carriers. Further research investigating other cognitive processing associated with MTL function (e.g., retrieval, novelty, source memory) will be necessary to fully elucidate the validity of the compensation theory. In doing so it will be crucial to consider not only group differences in activation levels and performance, but also whether cognitive processes may differ between groups. Additionally, it should be noted that the current study employed a sample size of 24 participants. Future studies should additionally focus on larger samples in order to validate current findings.
Taken together our findings suggest that the APOE e4 allele influences brain activity earlier in the lifespan than previously reported. Moreover, the presence of the APOE e4 allele appears to alter not only functional activations within the MTL, but also the functional connectivity of the MTL, during the memory encoding process. Brain regions implicated in this investigation are known to incur structural and functional changes in MCI and AD and have been observed to be altered in healthy middle aged adults and seniors with the e4 allele. While these commonalities exist with our project participant sample, it is unclear whether the brain activation differences observed in the young adult APOE e4 carriers reflects an adaptive response to an underlying weakness in the episodic memory network or a more fundamental, APOE-driven pleiotropism in functional brain organization. Results from our functional connectivity analyses suggest the former, as the APOE e4 carriers appear to be less able to integrate processing mediated by the MTL with other ESA regions.