Apolipoprotein E4 allele,antibodies against periodontal microorganisms,and cognition in older adults

Abstract

Background:

The presence of the apolipoprotein E4 (APOE4) allele and periodontal disease are independently correlated with higher levels of amyloid-β and inflammation in the brain, worse cognition, and Alzheimer's disease.

Objective:

To assess whether the presence of the APOE4 allele modifies the relationship between IgG antibodies against periodontal microorganisms and cognitive function in older adults participating in the NHANES III study.

Methods:

This cross-sectional analysis was conducted among participants of the third National Health and Nutrition Examination Survey (NHANES III) (1988 to 1994), aged 60 years and older, with measurements of IgG antibodies against 19 periodontal microorganisms and APOE4 alleles (N = 1644).

Results:

Approximately 77.5% of participants carried no APOE4 allele, 20.0% had one allele, and 2.6% were homozygous. Mean cognitive scores were 16.1, 16.0, and 15.3 for non-carriers, heterozygous, and homozygous APOE4 allele carriers, respectively (p = 0.01). Antibody groups were not correlated with APOE4 carrier status. The Orange-Blue cluster (antibodies against E. nodatum, A. naeslundii) was correlated with cognitive score (Spearman r = 0.08, p < 0.001), but not other antibody groups. In the multivariable-adjusted models, APOE4 alleles did not modify the associations between antibody clusters and cognitive score (p-interaction > 0.05).

Conclusions:

Mean cognitive scores were lower among APOE4 carriers. The APOE4 allele did not modify associations between groups of IgG antibodies against periodontal microorganisms and cognition among older aged adults without cognitive impairment. These findings need to be verified in larger prospective studies.

Keywords

Alzheimer's disease cognition IgG antibodies oral health oral microorganisms periodontal disease

Introduction

The apolipoprotein E (APOE) gene makes a protein that regulates cholesterol clearance, which is impaired among carriers of the APOE4 allele.¹ Higher Alzheimer's disease (AD) risk among APOE4 allele carriers is hypothesized to be due to impaired clearance of amyloid-β,² which accumulates in the brains of individuals with AD.³ However, the biological mechanisms are poorly understood, and APOE4 carriers have been shown to have worse cognition than non-carriers even among individuals without AD.^4,5 P. gingivalis, a microorganism associated with periodontal disease, is found in the brains of individuals with AD and is associated with proliferation of amyloid-β and inflammatory cytokines.⁶ APOE4 allele carriers are also more likely to have more periodontal disease which is hypothesized to be due to the pro-inflammatory effects of APOE4.⁷ As periodontal disease can also lead to higher systemic inflammation,⁸ additive effects of the APOE4 allele and periodontal disease on AD are plausible. Indeed, chronic inflammation has been shown to strengthen the association between APOE4 and AD risk.⁹ It is reported that empirically derived clusters of IgG antibodies against 19 microorganisms related to periodontal disease were associated with cognitive scores in participants of NHANES III who were 60 years and older.¹⁰ Empirically derived clusters of antibodies against different periodontal microbes reflect how these group together in vivo and the cluster scores represent an overall measure of all the microbes making up that cluster.¹⁰ IgG antibodies against periodontal microbes are markers of periodontal infection.¹¹ Individuals with the APOE4 allele who have higher levels of CRP develop AD sooner than those with lower CRP levels.⁹ As individuals with periodontal disease have higher CRP levels⁸ and have a higher risk of AD,¹² a synergistic interaction may be present between the APOE4 allele and antibodies against periodontal microorganisms in relation to cognition in this population.

We, therefore, evaluated effect modification by the APOE4 allele of the association between empirically derived groups of 19 IgG antibodies against periodontal microorganisms and cognition in older participants of the NHANES III study.

Methods

Data source

The data for these analyses came from the Third National Health and Nutrition Examination Survey (NHANES III) conducted by the National Center for Health Statistics (NCHS) of the Centers for Disease Control and Prevention (CDC) details of which can be found elsewhere.^13–15 Participants of NHANES III were a representative sample of the non-institutionalized civilian U.S. population who consented to participate and were surveyed between 1988 and 1994. We linked publicly available NHANES III data from questionnaires, laboratory findings, and medical and dental examinations, with restricted data on single nucleotide polymorphisms (SNPs) to classify APOE4 variants. All analyses were conducted at the Research Data Center (RDC) in Atlanta, GA in compliance with the related policies.

The Institutional Review Board of the University of South Carolina classified these analyses as Non-Human Subjects Research. The manuscript is compliant with the STROBE statement. We used SAS 9.4 (SAS Institute, Cary, NC) for data management and statistical analyses. The threshold for statistical significance was 0.05.

Population description

The study population consisted of NHANES III participants with valid data on IgG antibodies against 19 periodontal microorganisms (evaluated among participants 40 + years, N = 8153), cognitive assessment (conducted among participants 60 + years), and SNPs of the APOE gene rs429358 and rs7412 (total N = 3195). We used information from individuals with complete data on antibodies, cognition, and APOE4 genes (N = 1644).

Exposure measures

Cluster formation and naming the clusters

IgG antibody titers against 19 periodontal pathogens were grouped into four mutually exclusive groups using cluster analysis.¹⁶ The antibodies in each of the groups were: Orange-Red (P. melaninogenica, P. intermedia, P. nigrescens, P. gingivalis), Red-Green (T. forsythia, T. denticola, A. actinomycetemcomitans, E. corrodens, S. noxia, V. parvula, C. rectus), Yellow-Orange (S. intermedius, S. oralis, S. mutans, F. nucleatum, P. micros, C. ochracea), and Orange-Blue (E. nodatum, A. naeslundii). We named the clusters by adapting Socransky and Haffajee's description of periodontal microorganism groups, in which Red and Orange complexes indicated periodontal disease, while Yellow and Purple complexes indicated healthy periodontium, and the Blue complex indicated both healthy and disease state, and Green complex weakly indicated periodontal disease.^17,18

Calculating cluster scores

We calculated cluster scores by summing z-scores of standardized log transformed IgG titers for each antibody in that cluster. For example, to calculate the Orange-Red cluster score z-scores of log transformed antibody titers against P. melaninogenica, P. intermedia, P. nigrescens, and P. gingivalis were summed up.¹⁶

Outcome assessment

The cognitive score ranged from 0 to 17 with higher scores indicating better cognitive function. This was estimated from responses to the Mini-Mental State Examination (MMSE) which was administered in the home interview and at the mobile examination center.^19–21 The MMSE consisted of 17 questions (six orientation, six recall, and five attention-related, also called the Short Mental Status Examination).¹⁹ One point was assigned for each correct response and 0 for an incorrect response. The score was the sum of the responses. The first part of the shortened MMSE was assessment of orientation. In this part the interviewer asked the participant about the day of the week, the date, and the participant's complete address including street, city or town, state, and ZIP code (adult questionnaire). The second part consisted of recall evaluation. Here the interviewer told the participant the names of three items (apple, table, and penny) and asked him or her to repeat them after a short while. This exercise was repeated to give a maximum of 6 points for recall. Finally, to evaluate attention the interviewer asked the participant to serially subtract 3 from 20 and repeat this up to five times. For example, they were asked to subtract $3 from $20, $3 from $17, and so on, and assigned one point for each correct answer. We used the total MMSE score as a continuous variable in our analyses.

Apolipoprotein E (APOE) alleles

The APOE gene is located on chromosome 19 (19q13) and is responsible for synthesizing the apolipoprotein E. SNPs rs429358 and rs7412, residing within the APOE gene, affect the ApoE protein structure and are used to classify APOE alleles. APOE4 status was determined by the presence of cytosine on rs429358 and rs7412.²² Details are provided in Supplemental Table 1. We classified the APOE4 genotype in the following way: 2 APOE4 alleles (carrier), 1 APOE4 allele (carrier), and 0 APOE4 alleles (non-carriers).

Covariates

Covariates to control for potential confounding factors included age (40 to 64 years, and ≥65 years), sex (male and female), race (non-white and white), educational level (<12 years, ≥ 12 years completed education), poverty income ratio (PIR) was divided into three groups (low (≤1.3), middle (1.3 < PIR ≤ 3.5), and high (>3.5)) smoking status (current, former, never smoker), drinking status (drinker and non-drinker), body mass index (BMI) normal (≤24.9 kg/m2), overweight (25 to ≤29.9 kg/m²) and obese (≥30 kg/m²), periodontal disease, annual dentist visits, and history of diabetes, hypertension, and cardiovascular disease. Oral health measures (history of periodontal disease and annual dentist visits) were obtained from the Oral Examination component of NHANES III.²³ We defined severe periodontitis as the presence of 2 or more interproximal sites with ≥ 6 mm attachment loss (AL) (on different teeth) and 1 or more interproximal site(s) with ≥ 5 mm pocket depth (PD); moderate periodontitis as 2 or more interproximal sites with ≥ 4 mm clinical AL (on different teeth) or 2 or more interproximal sites with PD ≥ 5 mm, also on different teeth; and mild periodontitis as ≥ 2 interproximal sites with ≥ 3 mm AL and ≥2 interproximal sites with ≥ 4 mm PD (on different teeth) or 1 site with ≥ 5 mm.²⁴ In these analyses we compared individuals with severe or moderate periodontitis with those with mild or no periodontitis.

Statistical analysis

As the outcome measure (cognitive score) was highly skewed we opted to use quantile regression for the primary analyses rather than transforming the outcome and using least squares regression. The advantages of this approach were: first, the results remained on the natural scale making them more interpretable; second, quantile regression does not require assumptions about the distribution of residuals and is not sensitive to outliers²⁵; and finally, it was possible to evaluate associations between antibodies and cognition at different parts of the outcome distribution.²⁶ For the primary analyses, we used quantile regression to evaluate the relation between antibody clusters and cognition scores from the 10^th and 25^th quantiles. Confidence intervals were obtained by resampling. For the sake of completeness, we repeated the analyses using least squares regression after log transforming the cognitive score. In all analyses, we controlled for age, sex, race, educational level, PIR, smoking status, drinking status, BMI, diabetes, hypertension, other comorbidities, periodontal disease, and annual dentist visits. We calculated Spearman correlation coefficients between cognitive score, APOE4 alleles, antibody cluster scores, and mean antibody cluster scores across APOE4 carrier states.

Results

Non-carriers of the APOE4 allele compared with carriers were more likely to be older, white, and visit a dentist in the last year, and less likely to have a history of diabetes, hypertension, and cardiovascular disease (Table 1). Mean (95% confidence intervals) of cognitive scores were 16.14 (16.09, 16.19), 16.00 (15.88, 16.05), and 15.31 (14.67, 19.94) non-carriers, heterozygous, and homozygous APOE4 allele carriers respectively (p-value = 0.01). The differences in antibody cluster scores across APOE4 allele carriers and non-carriers did not reach statistical significance (Table 2), but the Orange-Blue cluster was positively correlated with the cognitive score (Table 3).

Table 1.

Characteristics of APOE4 allele carriers versus non-carriers of adults ≥60 years with information on antibodies against periodontal microorganisms and measures of cognition, NHANES III.

		APOE4 allele
	N	Non-Carrier (%)	One allele (%)	Two alleles (%)
Age *
<65 years	479	29.3	38.9	40.6
65 + years	1165	70.7	61.1	59.4
Sex
Female	894	58.6	55.4	51.4
Male	750	41.4	44.6	48.6
Race *
Non-White	315	8.5	13.4	20.3
White	1329	91.5	86.6	79.7
Educational level
≥12 years	789	34.5	38.6	36
<12 years	847	65.5	61.4	64
Income poverty level
Low	461	15.4	17.5	23.5
Middle	849	54	53.5	31.5
High	334	30.6	29	45
Smoking status
Non-smoker	806	47.5	46	38.2
Ever smoker	607	41.1	39.2	48.1
Current smoker	231	11.4	14.8	13.7
Alcohol use
Drinker	1107	60.7	58.6	50.5
Non-drinker	536	39.3	41.4	49.5
Physical activity
Vigorous	424	18.7	22	19.1
Moderate	644	39.3	44.3	45.8
Sedentary	576	42	33.7	35.1
Body mass index (BMI)
Obese	442	25.7	28.3	12.8
Overweight	676	40.4	39.1	50.1
Normal weight	523	33.9	32.6	37.1
History of diabetes*
Present	254	10.7	13.3	6.5
Absent	1389	89.3	86.7	93.5
History of hypertension*
Present	769	44.3	49	33.2
Absent	871	55.7	51	66.8
History of Cardiovascular disease (CVD)*
Present	481	31	30	15.5
Absent	1163	69	70	84.5
Annual dental visits*
Yes	551	46.3	38.1	26.1
No	1073	53.7	61.9	73.9
Periodontal disease
Yes	786	42	42.6	34.9
No	858	58	57.4	65.1

*p < 0.05

Table 2.

Antibody cluster scores among APOE4 allele carriers versus non-carriers of adults ≥60 years, NHANES III.

		APOE4 allele
	N	Non-Carrier, mean (se)	One allele, mean (se)	Two alleles, mean (se)
Antibody cluster scores
Red-Green	1644	1.46 (0.11)	1.15 (0.24)	1.00 (0.7)
Orange-Red	1644	0.04 (0.08)	−0.08 (0.16)	−0.12 (0.64)
Yellow-Orange	1644	0.74 (0.11)	0.51 (0.21)	0.67 (0.54)
Orange-Blue	1644	0.06 (0.04)	0.06 (0.072)	0.05 (0.18)

Table 3.

Correlation coefficients for cognitive score, APOE4 alleles, and antibody clusters.

	Total Cognition score	Apoe4b3 Alleles	Orange_Blue	Orange_Red	Red_Green	Yellow_Orange
Total Cognition score	1.000	−0.0290.24	0.075<0.00	−0.041 0.10	−0.013 0.61	0.0130.60
Apoe4b3 Alleles	−0.0290.24	1.000	−0.0220.38	−0.0220.38	−0.037 0.14	−0.0400.11
Orange_Blue	0.075<0.00	−0.0220.37	1.000	0.261 < 0.00	0.390 < 0.00	0.367<0.00
Orange_Red	−0.0410.10	−0.0220.36	0.261 < 0.00	1.000	0.690 < 0.00	0.731<0.00
Red_Green	−0.0130.60	−0.0370.14	0.390 < 0.00	0.690 < 0.00	1.000	0.814<0.00
Yellow_Orange	0.0130.60	−0.0400.11	0.367 < 0.00	0.731 < 0.00	0.814 < 0.00	1.000

Spearman Correlation Coefficients, N = 1644 (r, p-value)

There was no evidence of effect modification by the APOE4 allele of the association between antibody clusters and cognitive score (Table 4). The coefficients for interaction (β₄ and β₅ in Table 4) had small values and wide confidence intervals in all the models except for the model evaluating the Orange-Blue cluster. The coefficients for interaction in the Orange-Blue models had larger negative values but did not reach statistical significance (Table 4). No association was detected for the association between the antibody clusters and cognitive scores in any of the models (β₁ in Table 4). The coefficients for the Red-Green and Orange-Red clusters had negative values, while those for the Orange-Blue cluster had positive values but the confidence intervals were wide and included the null value. Compared with individuals having 2 APOE4 alleles, those with 0 or 1 allele had higher cognitive scores in the quantile regression models (Table 4). The coefficients evaluating the associations between those with 0 and 1 APOE4 alleles respectively compared with those with 2 APOE4 alleles, (β₂ and β₃ in Table 4), had positive values and reached statistical significance in most models. The strongest association between APOE4 status and cognition was seen in the model evaluating the Orange-Blue cluster. Individuals not carrying any APOE4 allele and those carrying one allele had 1.41 and 1.36 points higher cognitive scores compared with those individuals carrying 2 APOE4 alleles in the 25^th quantile of the Orange-Blue antibody score after controlling for potential confounders (Table 4). No significant associations were observed in the generalized linear models evaluating these relationships (Table 5).

Table 4.

Effect modification by APOE4 allele of association between antibody clusters and cognition in quantile regression models among adults ≥60 years, NHANES III.^a

	β₁ (95% CI) Antibody cluster	β₂ (95% CI) No APOE4 allele	β₃ (95% CI) One APOE4 allele	β₄ (95% CI) No APOE4 allele* Antibody cluster	β₅ (95% CI) One APOE4 allele* Antibody cluster
Red-Green
Quantile 0.10	−0.06 (−0.51, 0.38)	1.18 (−0.36, 2.72)	0.96 (−0.72, 2.65)	0.06 (−0.38, 0.50)	0.04 (−0.45, 0.52)
Quantile 0.25	−0.09 (−0.41, 0.24)	1.25 (0.27, 2.23)	1.26 (0.23, 2.28)	0.11 (−0.21, 0.43)	0.12 (−0.21, 0.44)
Orange-Red
Quantile 0.10	−0.15 (−0.64, 0.35)	1.36 (0.06, 2.67)	1.04 (−0.42, 2.50)	0.07 (−0.42, 0.56)	0.00 (−0.58, 0.58)
Quantile 0.25	−0.11 (−0.32, 0.10)	1.45 (0.56, 2.34)	1.40 (0.46, 2.34)	0.09 (−0.12, 0.30)	−0.06 (−0.16, 0.29)
Yellow-Orange
Quantile 0.10	0.00 (−0.49, 0.50)	1.20 (−0.60, 3.00)	0.95 (−0.96, 2.85)	0.08 (−0.40, 0.56)	0.05 (−0.46, 0.55)
Quantile 0.25	−0.10 (−0.40, 0.20)	1.12 (0.12, 2.11)	1.11 (0.06, 2.16)	0.11 (−0.18, 0.41)	0.12 (−0.18, 0.41)
Orange-Blue
Quantile 0.10	0.53 (−0.45, 1.51)	1.05 (−.22, 2.31)	0.94 (−0.44, 2.33)	−0.59 (−1.57, 0.39)	−0.42 (−1.57, 0.73)
Quantile 0.25	0.52 (−0.17, 1.21)	1.41 (0.57, 2.25)	1.36 (0.49, 2.22)	−0.60 (−1.30, 0.10)	−0.39 (−1.16, 0.37)

Models evaluating cluster score adjusted for age, sex, race, education, income, smoking, alcohol intake, physical activity, BMI, diabetes, hypertension, other comorbidities, visit to dentist, periodontal disease.

Reference group: Two APOE4 alleles.

β₄ and β₅ coefficient evaluates effect modification.

Table 5.

Effect modification by APOE4 allele of association between antibody clusters and cognition in generalized linear models among adults ≥60 years, NHANES III.^a

	β₁ (95% CI) Antibody cluster	β₂ (95% CI) No APOE4 allele	β₃ (95% CI) One APOE4 allele	β₄ (95% CI) No APOE4 allele* Antibody cluster	β₅ (95% CI) One APOE4 allele* Antibody cluster
Red-Green
	−0.01 (−0.04, 0.03)	0.04 (−0.11, 0.19)	0.00 (−0.15, 0.16)	0.00 (−0.03, 0.03)	0.00 (−0.03, 0.03)
Orange-Red
	0.01 (−0.03, 0.04)	0.03 (−0.11, 0.18)	0.00 (−0.15, 0.15)	0.00 (−0.04, 0.03)	0.00 (−0.04, 0.04)
Yellow-Orange
	0.00 (−0.04, 0.05)	0.04 (−0.11, 0.18)	0.01 (−0.15, 0.16)	0.00 (−0.04, 0.04)	−0.01 (−0.05, 0.04)
Orange-Blue
	0.02 (−0.11, 0.14)	0.04 (−0.11, 0.18)	0.01 (−0.14, 0.16)	−0.02 (−0.15, 0.11)	−0.03 (−0.16, 0.1)

Models evaluating log of cluster score adjusted for age, sex, race, education, income, smoking, alcohol intake, physical activity, BMI, diabetes, hypertension, other comorbidities, visit to dentist, periodontal disease.

β₄ and β₅ measures effect modification.

Discussion

APOE4 allele carriers had lower cognitive scores than non-carriers in multivariable models among middle and older aged individuals who did not have dementia. IgG antibody clusters against periodontal microorganisms were not associated with cognitive scores in this sample. There was no evidence that APOE4 allele carrier state modified these associations.

We previously reported that higher levels of Orange-Red and Yellow-Orange clusters of IgG antibodies against periodontal microorganisms were associated with lower cognitive scores in this population.¹⁰ In the current analysis evaluating cognitive score as the outcome, we observed non-significant but qualitatively similar associations between the cognitive score and the Orange-Red and Yellow-Orange clusters. Coefficients for the Orange-Red cluster were negative, and the coefficient associating the Yellow-Orange cluster and cognition was null at the 10^th quantile and had a negative value at the 25^th quantile, but none of the results were statistically significant. The study population for our previous report was 5160 whereas the size of the study population in this report was 1644 because genetic data was available for this subset, which is the likely reason for not detecting clear associations in this report. Our current finding that carriers of the APOE4 allele had worse cognition than non-carriers among individuals not diagnosed with dementia is consistent with previous reports.^4,5 We did not find any study evaluating effect modification by the APOE4 allele of the relationship between oral health and cognition. However, the modifying effect of the APOE4 allele of associations between various exposures and cognition is mixed and incompletely understood. The APOE4 allele added to the risk of cognitive decline in relation to air pollution,²⁷ but not with smoking.²⁸ Getting adequate sleep, on the other hand, protects against cognitive loss among APOE4 allele carriers.²⁹

Our study had some weaknesses. The short MMSE version used to assess cognitive status may not have been sensitive enough to detect small changes. The small sample size increases the chance of type-2 error (failing to find an association when it exists). This is a limitation but given that few studies on this topic have been conducted, the results could be useful by serving as a comparison for future studies on the topic or when combined in future meta-analyses. Confirmation of prior findings that APOE4 allele carriers have worse cognition, and the qualitatively similar associations seen between higher Orange-Red and Yellow-Orange antibody scores with worse cognition in this and a prior report, may give the current findings face validity. Moreover, the associations between periodontal antibodies and cognition, even if present, are likely small in magnitude. The cross-sectional study design precludes assessment of temporality and may be prone to survival bias, and unmeasured confounding. The association between APOE4 allele carrier and cognition could be underestimated because individuals with poorer cognition could be underrepresented in this cross-sectional sample. APOE4 allele carriers are at higher risk of AD, cardiovascular disease, and several other conditions increasing their risk of death^22,30,31 which would make them less likely to be participants of a cross-sectional survey (survival bias). The effects of APOE4 vary by ancestry. For example, APOE4 carriers with European and East Asian ancestries have higher AD risk than those with Nigerian Yoruban and Hispanic ancestries.^32,33 The non-white group in our analyses was heterogenous but more detailed analyses were restricted by the CDC disclosure rules to protect anonymity of participants. This remains a limitation.

The study also had some strengths. It is the first study, to our knowledge, to evaluate the effect modification by the APOE4 gene on the association between periodontal antibodies and cognition. The use of quantile regression models reduced the chances of poor model fit. Future research could use newer biomarkers of cognitive impairment as outcomes such as one predicting the probability of amyloid plaque presence by combining information from the ratio of two amyloid-β peptides, APOE proteotype, and age,³⁴ or evaluate the effect of treating periodontal disease in APOE4 carriers.

APOE4 allele carriers had worse cognition than non-carriers among middle and older aged adults without diagnosed cognitive impairment. There was no evidence that the APOE4 allele modified associations between groups of IgG antibodies against periodontal microorganisms and cognition. The findings need to be verified in larger prospective studies.

Supplemental Material

sj-docx-1-alr-10.1177_25424823251370770 - Supplemental material for Apolipoprotein E4 allele, antibodies against periodontal microorganisms, and cognition in older adults

Supplemental material, sj-docx-1-alr-10.1177_25424823251370770 for Apolipoprotein E4 allele, antibodies against periodontal microorganisms, and cognition in older adults by Anwar T. Merchant, Longgang Zhao, Eric Mishio Bawa, Amirhossein Fakhre Yaseri, Matthew Lohman, Jiajia Zhang, Alyssa Clay-Gilmour, Roger D Newman-Norlund and Julius Fridriksson in Journal of Alzheimer's Disease Reports

Footnotes

Acknowledgements

The authors have no acknowledgments to report.

ORCID iD

Anwar T. Merchant

Ethical considerations

The data was completely deidentified and was thus classified as Not Human Subjects Research.

Consent to participate

Participants had provided consent for the original survey. No new data were collected.

Author contribution(s)

Anwar Merchant: Conceptualization; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Project administration; Resources; Supervision; Writing – original draft.

Longgang Zhao: Formal analysis; Validation; Writing – review & editing.

Eric Mishio Bawa: Formal analysis; Project administration; Supervision; Validation; Writing – review & editing.

Amirhossein Fakhre Yaseri: Formal analysis; Project administration; Visualization; Writing – review & editing.

Matthew Lohman: Conceptualization; Funding acquisition; Investigation; Writing – review & editing.

Jiajia Zhang: Conceptualization; Funding acquisition; Methodology; Writing – review & editing.

Alyssa Clay-Gilmore: Conceptualization; Methodology; Writing – review & editing.

Roger Newman-Norland: Methodology; Writing – review & editing.

Julius Fridriksson: Investigation; Writing – review & editing.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: These analyses were supported by Grant number: 1R21AG070449-01 from the National Institute on Aging and the Duke endowment Grant number: 6984-SP.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

Genetic data is not publicly available because of confidentiality concerns. Data requests need to be made to the CDC.

Supplemental material

Supplemental material for this article is available online.

References

Mahley

. Apolipoprotein E: from cardiovascular disease to neurodegenerative disorders. J Mol Med (Berl) 2016; 94: 739–746.

Castellano

Kim

Stewart

, et al. Human apoE isoforms differentially regulate brain amyloid-beta peptide clearance. Sci Transl Med 2011; 3: 89–57.

Ballard

Gauthier

Corbett

, et al. Alzheimer's disease. Lancet 2011; 377: 1019–1031.

Verhaaren

Vernooij

Koudstaal

, et al. Alzheimer's disease genes and cognition in the nondemented general population. Biol Psychiatry 2013; 73: 429–434.

Michaelson

. APOE Epsilon4: the most prevalent yet understudied risk factor for Alzheimer's disease. Alzheimers Dement 2014; 10: 861–868.

Costa

MJF

de Araújo

IDT

da Rocha Alves

, et al. Relationship of Porphyromonas gingivalis and Alzheimer's disease: a systematic review of pre-clinical studies. Clin Oral Invest 2021; 25: 797–806.

Pereira

Nascimento

JCR

Rego

JMC

, et al. Apolipoprotein E, periodontal disease and the risk for atherosclerosis: a review. Arch Oral Biol 2019; 98: 204–212.

Paraskevas

Huizinga

Loos

. A systematic review and meta-analyses on C-reactive protein in relation to periodontitis. J Clin Periodontol 2008; 35: 277–290.

Tao

Ang

TFA

DeCarli

, et al. Association of chronic low-grade inflammation with risk of Alzheimer disease in ApoE4 carriers. JAMA Netw Open 2018; 1: e183597.

10.

Merchant

Vidanapathirana

, et al. Antibodies against periodontal microorganisms and cognition in older adults. JDR Clin Trans Res 2023; 8: 148–157.

11.

Papapanou

Neiderud

Disick

, et al. Longitudinal stability of serum immunoglobulin G responses to periodontal bacteria. J Clin Periodontol 2004; 31: 985–990.

12.

Borsa

Dubois

Sacco

, et al. Analysis the link between periodontal diseases and Alzheimer's disease: a systematic review. Int J Environ Res Public Health 2021; 18: 9312.

13.

Ezzati

Massey

Waksberg

, et al. Sample design: third national health and nutrition examination survey. Vital Health Stat 2 1992; (113): 1–35.

14.

National Center for Health Statistics . Plan and operation of the third National Health and Nutrition Examination Survey, 1988-94. Vital Health Stat 1994; 32: 1–407.

15.

US Department of Health, Human Services . National Health and Nutrition Examination Survey III: Oral Examination Component. Rockville, MD: Westat Inc, 1992.

16.

Merchant

Shrestha

Chaisson

, et al. Association between serum antibodies to oral microorganisms and hyperglycemia in adults. J Dent Res 2014; 93: 752–759.

17.

Socransky

Haffajee

. Periodontal microbial ecology. Periodontol 2000 2005; 38: 135–187.

18.

Socransky

Haffajee

. Dental biofilms: difficult therapeutic targets. Periodontol 2000 2002; 28: 12–55.

19.

Obisesan

Gillum

. Cognitive function, social integration and mortality in a U.S. national cohort study of older adults. BMC Geriatr 2009; 9: 33.

20.

Pfeiffer

. A short portable mental status questionnaire for the assessment of organic brain deficit in elderly patients. J Am Geriatr Soc 1975; 23: 433–441.

21.

Perkins

Hendrie

Callahan

, et al. Association of antioxidants with memory in a multiethnic elderly sample using the Third National Health and Nutrition Examination Survey. Am J Epidemiol 1999; 150: 37–44.

22.

Lumsden

Mulugeta

Zhou

, et al. Apolipoprotein E (APOE) genotype-associated disease risks: a phenome-wide, registry-based, case-control study utilising the UK Biobank. EBioMedicine 2020; 59: 102954.

23.

Centers for Disease Control and Prevention . NHANES III: Oral Examination Component, https://wwwn.cdc.gov/nchs/data/nhanes3/manuals/dental.pdf (1992, accessed February 2018).

24.

Eke

Dye

Wei

, et al. Prevalence of periodontitis in adults in the United States: 2009 and 2010. J Dent Res 2012; 91: 914–920.

25.

Bottai

Cai

McKeown

. Logistic quantile regression for bounded outcomes. Stat Med 2010; 29: 309–317.

26.

Bottai

Frongillo

Sui

, et al. Use of quantile regression to investigate the longitudinal association between physical activity and body mass index. Obesity (Silver Spring) 2014; 22: E149–E156.

27.

Franz

Gustavson

Elman

, et al. Associations between ambient air pollution and cognitive abilities from midlife to early old age: modification by APOE genotype. J Alzheimers Dis 2023; 93: 193–209.

28.

Wingbermühle

Wen

Wolters

, et al. Smoking, APOE genotype, and cognitive decline: the Rotterdam Study. J Alzheimers Dis 2017; 57: 1191–1195.

29.

Lim

Kowgier

, et al. Modification of the relationship of the apolipoprotein E ε4 allele to the risk of Alzheimer disease and neurofibrillary tangle density by sleep. JAMA Neurol 2013; 70: 1544–1551.

30.

Bennet

Di Angelantonio

, et al. Association of apolipoprotein E genotypes with lipid levels and coronary risk. JAMA 2007; 298: 1300–1311.

31.

Semaev

Shakhtshneider

Shcherbakova

, et al. Associations of APOE gene variants rs429358 and rs7412 with parameters of the blood lipid profile and the risk of myocardial infarction and death in a white population of Western Siberia. Curr Issues Mol Biol 2022; 44: 1713–1724.

32.

Belloy

Andrews

Le Guen

, et al. APOE Genotype and Alzheimer disease risk across age, sex, and population ancestry. JAMA Neurol 2023; 80: 1284–1294.

33.

Farrer

Cupples

Haines

, et al. Effects of age, sex, and ethnicity on the association between apolipoprotein E genotype and Alzheimer disease. A meta-analysis. APOE and Alzheimer Disease Meta Analysis Consortium. JAMA 1997; 278: 1349–1356.

34.

Monane

Johnson

Snider

, et al. A blood biomarker test for brain amyloid impacts the clinical evaluation of cognitive impairment. Ann Clin Transl Neurol 2023; 10: 1738–1748.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.01 MB

0.00 MB