apolipoprotein E(apoE) polymorphism critical to many disease

[tipirneni]

Three common protein isoforms of apolipoprotein E (apoE), encoded by the ε2, ε3, and ε4 alleles of the APOE gene, differ in their association with cardiovascular and Alzheimer's disease risk and COVID-19. 


fig. Locations of polymorphic variants in and around the APOE gene. The genomic location of the 23 DNA variants, identified by sequencing 5.5 kb in 96 individuals, is shown, below the exon-intron structure of the APOE gene. An asterisk (*) marks the new variant identified in the Mayan sample from Campeche at position 3701, and x’s show the population distribution of the observed polymorphisms (J = Jackson, C = Campeche, N = North Karelia, and R = Rochester). Variants that result in amino acid substitutions are boxed.


The global ubiquity of the three common apoE isoforms and, in particular, the persistently high relative frequency of the ε3 allele in all human populations, has puzzled investigators and prompted much speculation regarding the evolutionary forces responsible for the observed polymorphism

Therefore, although we did not find statistically significant evidence of the effects of natural selection on APOE variation, it is not unreasonable to infer that selection has acted. The value of examining this possibility closely is heightened by the variety of phenotypes with which the protein isoforms have been associated.



Precisely when the new variant at site 3937, and its associated haplotypes, began to expand in frequency relative to the ancestral allele is uncertain and is unlikely to be resolved using analytical methods that assume selective neutrality (such as those employed here). The presence of both ε3 sublineages in each of the four populations investigated, in the absence of evidence for recurrent mutation at the sites involved, is consistent with site 3937 having arisen prior to the major population expansions that accompanied the spread of anatomically modern humans <100,000 years ago. Alternatively, strong selective pressure may have facilitated the global distribution of the variant much more recently. Indirect evidence in support of the latter hypothesis is the fact that the basal haplotype of the ε3 clade (haplotype 6) is absent in the European samples from Rochester and North Karelia (hence, if present, likely to be at low frequency in those populations), suggesting that the rise in frequency of the ε3s in Europe may have occurred after the differentiation of the clade into two primary lineages. The low overall level of population heterogeneity (FST = 0.06) is also consistent with a recent rapid expansion of ε3. In either case, the lower relative frequency and patchy geographical distribution of the ε2 haplotypes, as well as their clear derivation from the ε3 clade, suggest that the variant defining these latter alleles (at site 4075) arose subsequently.



A relevant question is whether known phenotypic effects associated with the observed variation contribute to current differences in reproductive fitness. There is strong evidence that the inheritance of an ε4 allele places carriers at a higher risk of succumbing to CAD or AD, at least in European and Asian populations (Davignon et al. 1988; Roses 1996). The deleteriousness of ε4, relative to that of ε3, is consistent with the history of long-term genetic change at the APOE locus, but CAD and AD are both diseases of late adulthood or old age, and increased susceptibility to these conditions would not be expected to result in important differences in reproductive success (for an opposing argument, see Finch and Sapolsky [1999]). The widely expressed protein does play many different roles in the body, including facilitation of lipid absorption, neural growth and regeneration, and immune function (Mahley and Huang 1999), one or more of which could have a direct effect on fertility or could contribute to differential survival and reproductive success. One study has suggested that men carrying at least one ε3 allele have, on average, more children than men with other APOE genotypes (Gerdes et al. 1996a). Alternatively, APOE variation may reflect an adaptation to changing diets (Hanlon and Rubinsztein 1995), such as those which accompanied the transition from subsistence to agricultural economies (Corbo and Scacchi 1999), may play a key role in neurological response to head injury (Friedman et al. 1999), or may mediate susceptibility to lipophylic pathogens (Martin 1999).



Population-level differences in the distribution of APOE variation are relevant to a comprehensive prediction of risk and understanding of disease etiology. Despite the low overall level of polymorphism at the APOE gene, considerable heterogeneity characterizes each of the three common alleles at the sequence level, heterogeneity which helps explain previously perplexing association results. For example, in a 5-year prospective epidemiological study of AD incidence among different ethnic groups, Tang et al. (1998) found that, compared to ε3/ε3 homozygotes, the relative risk (RR) of AD associated with one or more copies of the ε4 allele was significantly increased among whites (RR 2.5) but not among blacks (RR 1.0) or Hispanics (RR 1.1). These results confirmed previous reports suggesting that the association of ε4 with AD is weaker or nonexistent among blacks living in New York City (Tang et al. 1996) and Indiana (Sahota et al. 1997), as well as among black Nigerians (Osuntokun et al. 1995) and East Africans (Sayi et al. 1997). Similar variation is found regarding lipids (Xu et al. 1999). These observations take on new significance in the light of our finding that geographic and/or ethnic differences exist in the distribution of haplotypes within the ε4 class (fig. 3). If, as our results suggest, different types of ε4 alleles are found at different relative frequencies in different geographic regions, this heterogeneity (which may be related to nonneutral forces acting on the locus) can be—indeed, must be—accounted for.



Similarly, our data provide an invaluable evolutionary context in which to interpret more circumscribed analyses. There has, for instance, recently been much interest in characterizing APOE promoter-region polymorphism and examining the association of such variants with AD and CAD risk. These analyses have met with only limited success. Either the observed variants have turned out not to explain any greater proportion of the observed variance in phenotype than explained by the common allelic variants or positive associations in one population have failed to be replicated in subsequent studies. The most problematic discrepancies in this regard have centered on the role of the −491 variant (Artiga et al. 1998b). Some workers have suggested that variation at this site is a strong determinant of AD risk (Artiga et al 1998a; Bullido et al 1998), whereas other researchers have either questioned the importance of the polymorphism relative to other regulatory region variants (Lambert et al. 1998b; Town et al. 1998), or failed to replicate the association altogether (Roks et al. 1998; Song et al. 1998). Our analysis suggests a possible explanation: −491 (site 560 here) appears to be particularly susceptible to recurrent mutation and/or gene conversion, placing it in association with different allelic backgrounds, with different functional effects, in different populations. In this context, it is unsurprising that results have conflicted. On the other hand, the prominent placement of site 832 (−219) in our inferred haplotype network, as a site defining major subtypes of both ε3 and ε4 haplotypes in multiple populations, appears consistent with the association of this site with differences in both AD risk (Lambert et al. 1998a, 1998b) and myocardial infarction (Lambert et al. 2000). The relationship of other variants with unique positions in the APOE gene tree (particularly sites 1163 and 2440 among ε3s and site 1998 among ε4s) clearly merit detailed investigation. When large samples are typed at these variable sites, and relevant phenotypes are scored, it will be possible to test directly the independence of effects of the variable sites on lipid phenotypes.
from www.ncbi.nlm.nih.gov/pmc/articles/PMC1287893/

[tipirneni]

The human ApoE protein is composed of 299 amino acids, and it is abundantly expressed both in the CNS and in the periphery [20, 21]; ApoE functions as a primary regulator of cholesterol transport and lipid metabolism. Human ApoE has three common isoforms that differ by a single amino acid at residues 112 or 158; these isoforms are ApoE2 (Cys112 and Cys158), ApoE3 (Cys112 and Arg158), and ApoE4 (Arg112 and Arg158), and their differences obviously alter protein function [22]. In addition to SARS-CoV-2, ApoE gene polymorphisms are also associated with the cellular attachment and organismal responses to several other viruses, such as hepatitis C virus (HCV), HBV, and HIV-1 [23–25]. Notably, the role of ApoE in regulating viral infections seems to occur partly due to its interaction with heparan sulfate proteoglycans (HSPGs), functioning either as a Trojan horse or competing with viral particles for binding to HSPGs [26], suggesting that ApoE has the function of binding to the key receptor of virus and affecting virus infection. Shi and colleagues observed that human-induced pluripotent stem cell (hiPSC)-derived ApoE4-expressing neurons and astrocytes are more susceptible to SARS-CoV-2 infection, and ApoE4 astrocytes exhibit a more severe response [27]. Whether ApoE affects viral infection through interaction with viruses or receptors is currently unclear.

Here, a meta-analysis was carried out and confirmed that ApoE4 is significantly associated with the incidence and severity of COVID-19. Moreover, we reported the potential interactions of different ApoE isoforms with ACE2 and the spike protein. Further, we analyzed the regulatory effects of different ApoE isoforms on the expression of ACE2, and the balance of the RAS pathway, providing plausible evidence that ApoE4 contributes to severe COVID-19.

A total of 2325 COVID-19 cases and 644063 controls from six case–control studies concentrating on the association between ApoE gene polymorphism and the incidence of COVID-19 were included. The pooled results showed that ApoE gene polymorphism (ε4 carrier genotypes VS non-ε4 carrier genotypes) is associated with a high risk of COVID-19 (P = 0.0003, OR = 1.44, 95% CI 1.18–1.76) (Fig. 2A). For the disease severity, a total of 573 COVID-19 cases and 324752 controls from five studies were analyzed. The results showed that ApoE gene polymorphism (ε4 carries genotypes VS non-ε4 carries genotypes) is associated with increased risk of disease progression (P < 0.00001, OR = 1.85, 95% CI 1.50–2.28) (Fig. 2B). Those integrate data conclude that the ε4 allele of ApoE gene is not only associated with risk but also the severity of COVID-19.



The fluorescence results showed that ACE2 was primarily expressed on the cell membrane, and the colocalization between ApoE and ACE2 were observed in the lungs, kidneys, cortices and hearts of ApoE3-TR mice (Fig. 3A). Then the colocalization of ApoE and ACE2 were further observed in the A549, HEK-293 T, SH-SY5Y, and HUVECs cell lines which corresponding to various tissues (Fig. 3B). The ApoE and ACE2 proteins were widely expressed in A549, HEK-293 T, SH-SY5Y, and HUVECs cell lines, and ACE2 was expressed at the highest levels in HEK-293 T cells (Additional file 1: Fig. S1).


fig. ApoE colocalizes with ACE2 in vitro and in vivo. A Representative coexpression of ApoE (red) and ACE2 (green) in lung, kidney, brain cortex and heart sections of ApoE3-TR mice as shown by immunofluorescence staining. The nucleus (blue) was stained with DAPI. Scale bars = 20 μm. B Representative coexpression of ApoE (red) and ACE2 (green) in A549, HEK-293, SH-SY5Y and HUVECs as shown by immunofluorescence staining. The nucleus (blue) was stained with DAPI. Scale bars = 20 μm

ApoE4 suppresses ACE2 expression in vitro
SARS-CoV-2 not only use ACE2 allowing virus entry but also downregulates ACE2 expression on cells, which may cause the imbalance between the RAS and ACE2/Mas axis and subsequently contribute to severe disease condition [33, 34]. Consistently, studies reported a negative association between ACE2 amount and COVID-19 severity and fatality at both population and molecular levels [35]. Thus, after analysing the interaction between ApoE and ACE2, we further evaluated the effect of ApoE gene polymorphisms on ACE2 expression at the cellular level, and the results showed that overexpression of ApoE4 led to a significant downregulation of ACE2 protein expression in SH-SY5Y, HEK-293 T, A549 and HUVEC cells (Fig. 6A–D). Consistent results were obtained by immunofluorescence in these cells (Fig. 6E). Considering the critical role of ACE2 in the RAS system, reduced expression of ACE2 may lead to dysregulation of RAS signalling, thereby contributing to disease progression.



ApoE4 suppresses ACE2 expression in vivo
To validated this negative expression relationship between the ApoE4 and ACE2 in vivo, here, we selected the ApoE-TR mouse model, in which expression of the human ApoE gene is regulated by the mouse ApoE promoter in mice on the C57BL/6 J background (Fig. 7A). The mouse ApoE gene was replaced by the human ApoE2, ApoE3 or ApoE4 gene so that only the human ApoE gene is expressed. Thus, ApoE-TR mice are a good model for investigating the biological function of the human ApoE protein. First, we compared the protein expression of ACE2 in different tissues and found that ACE2 was abundantly expressed in bowel and kidney tissues (Fig. 7B). Furthermore, the inhibition of ACE2 by ApoE4 was analysed in a variety of tissues, such as the cortex, hippocampus, liver, bowel, kidney, and heart (Fig. 7C–H). However, inhibitory effects were not observed in the lung tissues of ApoE-TR mice (Fig. 7I). In addition, the regulatory relationship was further verified by immunofluorescence staining (Additional file 1: Fig. S3).


fig. ApoE4 downregulates ACE2 protein expression in vivo. A ApoE-TR mice were selected to evaluate the regulatory effect of ApoE4 on ACE2 protein expression. B ACE2 protein levels in the heart, liver, lung, kidney, brain, and bowel (equivalent amounts of total proteins) of ApoE3-TR mice were measured by western blotting, n = 4 mice per group. C-I ACE2 protein levels in the heart, liver, cortex, hippocampus, bowel, kidney, and lung of ApoE2-TR, ApoE3-TR and ApoE4-TR mice were measured by western blotting. The results were normalized to the expression of a-tubulin. n = 6 mice per group. The data are expressed as the mean ± SD. Statistical differences were evaluated by one-way ANOVA. *P < 0.05, **P < 0.01, ***P < 0.001


ApoE4 decreases the conversion of Ang II to Ang 1–7
Considering that ACE2 catalyzes the conversion of Ang II to Ang 1–7 to counter-regulate the harmful effects of the RAS system, we further measured the Ang II and Ang 1–7 protein levels by ELISA. ApoE4-TR mice showed increased Ang II and decreased Ang 1–7 in the cortex, kidney, and bowel compared to other ApoE subtypes (Fig. 8A–C). In the liver, heart, and lung of the ApoE4-TR mice, the level of Ang II tends to be higher, and Ang 1–7 is slightly lower but with no statistical significance (Fig. –F). ApoE4-TR mice also showed reduced expression of Mas receptor (MasR), an essential receptor of Ang 1–7, in these tissues (Fig. 8G–J). In addition, we confirmed the regulation of ApoE4 on the RAS system in vitro. Compared with the ApoE2 and ApoE3 overexpression groups, overexpression of ApoE4 promoted the expression of Ang II and inhibited the expression of Ang 1–7 in HEK-293 T cells (Additional file 1: Fig. S4A and B). These results suggest that the ApoE polymorphism is associated with the effect of ApoE4-induced disorders in RAS signalling.





from www.ncbi.nlm.nih.gov/pmc/articles/PMC9910247/