Skip to main content
. Author manuscript; available in PMC: 2021 May 7.
Published in final edited form as: Alzheimers Dement. 2020 Aug 22;16(10):1372–1383. doi: 10.1002/alz.12104

TABLE 3.

Associations of tight junction proteins with Alzheimer’s disease (AD)-related molecules (apoE, Aβ40, Aβ42, and total tau)

Association with CLDN5 TX/CD31 TX Association with OCLN TX/CD31 TX
Unadjusted analysis Adjusting for age at death, sex, CAA score, number of APOE ε4 alleles, Braak stage, and Thai phasellnadjusted analysis Unadjusted analysis Adjusting for age at death, sex, CAA score, number of APOE ε4 alleles, Braak stage, and Thai phase
Variable Regression coefficient (95% CI) P-value Regression coefficient (95% CI) P-value Regression coefficient (95% CI) P-value Regression coefficient (95% CI) P-value
Aβ40 TBS −0.03 (−0.13, 0.06) .51 −0.03 (−0.14, 0.07) .52 0.00 (−0.02, 0.03) .67 0.01 (−0.01, 0.03) .34
Aβ40 TX −0.04 (−0.14, 0.05) .36 −0.06 (−0.16, 0.05) .32 −0.02 (−0.04, 0.00) .11 −0.02 (−0.04, 0.01) .20
Aβ40 FA 0.01 (−0.09, 0.10) .84 0.02 (−0.09, 0.14) .68 0.02 (0.00, 0.04) .086 0.04 (0.01, 0.07) .002
Aβ42 TBS −0.23 (−0.32, −0.13) <.001 −0.22 (−0.32, −0.13) <.001 0.03 (0.01, 0.05) .004 0.03 (0.01, 0.05) .0033
Aβ42 TX −0.12 (−0.21, −0.02) .017 −0.10 (−0.20, −0.01) .031 0.04 (0.02, 0.06) <.001 0.05 (0.03, 0.07) <.001
Aβ42 FA −0.04 (−0.13, 0.06) .41 −0.04 (−0.14, 0.05) .39 0.06 (0.04, 0.08) <.001 0.07 (0.05, 0.09) <.001
Aβ40/42 TBS 0.06 (−0.04, 0.15) .25 0.07 (−0.04, 0.17) .20 −0.01 (−0.03, 0.01) .51 0.00 (−0.02, 0.02) .82
Aβ40/42 TX 0.02 (−0.07, 0.12) .64 0.02 (−0.08, 0.13) .67 −0.04 (−0.06, −0.02) <.001 −0.05 (−0.07, −0.03) <.001
Aβ40/42 FA 0.03 (−0.07, 0.12) .58 0.05 (−0.06, 0.17) .39 −0.01 (−0.03, 0.01) .41 0.00 (−0.03, 0.02) .79
Total tau TBS −0.38 (−0.47, −0.30) <.001 −0.43 (−0.52, −0.34) <.001 −0.04 (−0.06, −0.02) <.001 −0.05 (−0.07, −0.03) <.001
Total tau TX −0.29 (−0.38, −0.20) <.001 −0.32 (−0.41, −0.22) <.001 0.02 (0.00, 0.04) .031 0.02 (0.00, 0.04) .13
Total tau FA −0.13 (−0.22, −0.03) .008 −0.13 (−0.22, −0.03) .009 0.03 (0.01, 0.05) .005 0.03 (0.01, 0.05) .007
p-tau TBS −0.07 (−0.17, 0.02) .11 −0.08 (−0.17, 0.02) .12 0.02 (−0.00, 0.04) .077 0.02 (0.00, 0.04) .053
p-tau TX 0.14 (0.04, 0.23) .004 0.17 (0.07, 0.27) .001 0.06 (0.04, 0.08) <.001 0.08 (0.06, 0.10) <.001
p-tau FA −0.17 (−0.27, −0.08) <.001 −0.20 (−0.30, −0.09) <.001 0.01 (−0.01, 0.03) .51 0.02 (0.00, 0.05) .048
apoE TBS −0.32 (−0.41, −0.23) <.001 −0.38 (−0.47, −0.28) <.001 −0.06 (−0.08, −0.04) <.001 −0.07 (−0.09, −0.05) <.001
apoE TX −0.18 (−0.27, −0.09) <.001 −0.21 (−0.30, −0.11) <.001 −0.06 (−0.08, −0.04) <.001 −0.07 (−0.09, −0.05) <.001
apoE FA −0.21 (−0.30, −0.11) <.001 −0.27 (−0.37, −0.16) <.001 −0.01 (−0.03, 0.01) .18 −0.01 (−0.03, 0.02) .61

Regression coefficients, 95% CIs, and P-values result from linear regression models, where CLDN5 TX/CD31 TX was considered on the original scale and OCLN TX/CD31 TX was considered on the square root scale. Regression coefficients are interpreted as the change in mean outcome level (CLDN5 TX/CD31 TX or OCLN TX/CD31 TX, on the aforementioned scales) corresponding to each 1-standard-deviation increase in the given variable (on the untransformed, square root transformed, or natural logarithm transformed scale). P-values <.0028 were considered statistically significant after applying a Bonferroni correction for multiple testing separately for each tight junction protein measure; statistically significant associations are shown in bold.

Abbreviation: APOE, apolipoprotein E; CAA, cerebral amyloid angiopathy; CI, confidence interval; CLDN5, claudin-5; FA, formic acid; OCLN, occludin; TBS, Tris-buffered saline; TX, 1% Triton X-100