Angiotensin-converting enzyme levels and activity in Alzheimer's disease: differences in brain and CSF ACE and association with ACE1 genotypes

Abstract

Angiotensin-converting enzyme (ACE) has been implicated in Alzheimer's disease (AD): ACE1 variations influence plasma ACE and risk of AD, and ACE is increased in AD brain. We measured frontal ACE level and activity in 89 AD and 51 control brains, and post-mortem CSF from 101 cases and 19 controls. Neuron-specific enolase (NSE) level and Braak stage were used to indicate neuronal preservation and disease progression. We genotyped the common ACE insertion/deletion polymorphism, rs4343, rs1800764 and rs4921. ACE activity was elevated in AD and correlated with Braak stage. Crude ACE levels were unchanged but adjustment for NSE suggested increased neuronal ACE production with Braak stage. Exposing SH-SY-5Y neurons to oligomeric Abeta1-42 increased ACE level and activity, suggesting Abeta may upregulate ACE in AD. In CSF, ACE level but not activity was reduced in AD. ACE1 genotype did not predict ACE level or activity in brain or CSF. ACE activity and neuronal production increase in AD brain, possibly in response to Abeta. Peripheral measurements do not reflect ACE activity in the brain.

Keywords: ACE1; Alzheimer's disease; Angiotensin-converting enzyme; Braak stage; cerebrospinal fluid; enzyme activity; neuron-specific enolase.

Figure 1

Representative western blot of neuron specific enolase in brain tissue homogenate prepared from the frontal cortex. Recombinant human NSE (5μg) was loaded in lane 1 and homogenates (30μg total protein) from an AD and control brain were loaded in lanes 2 and 3 respectively.

Figure 2

Bar charts showing the effect of tissue storage and lysis buffer detergents on ACE enzyme activity in frontal cortex. (A) Storage of brain tissue for up to 72h at 4°C or room temperature did not affect ACE activity. (B) Bar chart showing ACE activity in brain tissue homogenates prepared in lysis buffers containing different detergents (1% SDS, 0.5% NP-40 or 0.5% triton-X-100). Error bars indicate 1 standard error of the mean (SEM). The three detergents yielded similar measurements but measured ACE activity was lower in the absence of detergent.

Figure 3

Bar charts showing ACE level and enzyme activity in frontal cortex in relation to diagnosis and Braak stage, both before (unadjusted) and after (adjusted) adjusting for neuronal damage as measured by the reduction in NSE level below the mean control value. Error bars indicate 1 SEM. (A-B) Unadjusted ACE level did not vary with AD or Braak stage. (C-D) Adjusted ACE level increased in AD and was positively associated with Braak stage (p=0.004, Spearman rs=0.266). (E-F) ACE activity increased significantly in AD (p=0.0001) and was positively associated with Braak stage (p=0.03, rs=0.201). (G-H) Adjusted ACE activity increased even more dramatically in AD (p=0000002) and with increasing Braak stage (p=0.005, rs = 0.260). Post hoc comparison between subgroups revealed significantly increased ACE activity in Braak stage V-VI compare to 0-II (ANOVA, p=0.005).

Figure 4

Bar charts showing ACE level and activity in post-mortem CSF samples in relation to AD and disease progression. Error bars indicate 1 SEM. (A-B) ACE level was significantly lower in AD than control CSF (p=0.030) and correlated inversely with Braak stage (p=0.010, rs =-0.252). ACE levels were significantly lower in Braak stages V-VI than 0-II (p=0.04). (C-D) ACE activity was greater in AD than control CSF but the difference did not reach significance (p=0.060).

Figure 5

In vitro induction of neuronal ACE by oligomeric Aβ1-42. Human SH-SY5Y cells, differentiated in retinoic acid, were incubated with either monomeric (M) or oligomeric/fibrillar (OF) Aβ1-42. Bar charts show (A-B) ACE levels and (C-D) ACE activity in SH-SY5Y cells after incubation for 4 and 24h respectively. ANOVA revealed significant differences in ACE protein (p=0.0004) and activity (0.033) levels at 24h. Post hoc comparisons revealed that ACE protein level was significantly greater in cells incubated with oligomeric Aβ1-42 than in control cultures (p=0.00087) or in cells incubated with monomeric Aβ1-42 (p=0.0011). Error bars indicate 1 SEM.

Figure 6

Bar charts show genetic associations between ACE1 indel polymorphism (rs1799752) and ACE enzyme activity, (A) before and (B) after adjustment for neuson specific enolase, in a case-control cohort.

Figure 7

Bar charts show genetic associations between ACE1 SNPs (indel rs1799752, rs1800764, rs4291 and rs4343) and ACE protein level (A-D) and activity (E-H) in human post-mortem CSF. Error bars indicate 1 SEM.