NGF mRNA is not decreased in frontal cortex from Alzheimer\'s Disease patients

MOLECULAR BRAIN RESEARCH ELSEVIER

Molecular Brain Research 25 (1994) 242-250

ii

1

Research Report

NGF m R N A is not decreased in frontal cortex from Alzheimer's Disease patients N. Jett6, M.S. Cole J, M. Fahnestock * Department of Biomedical Sciences, McMaster University, 1200 Main Street West, Hamilton, Ont., L8N 3Z5 Canada

Accepted 5 April 1994

Abstract

Alzheimer's disease (AD) is characterized by neuronal dysfunction and degeneration in certain brain regions such as cortex, hippocampus and basal forebrain. Specific neurochemical defects such as decreases in cholinergic enzymes and in the amounts of mRNA in AD brain have also been reported. Nerve growth factor (NGF), a protein necessary for the development, regulation and survival of basal forebrain cholinergic neurons (BFCN), is synthesized in target areas of BFCN (cortex, hippocampus) and is supplied to BFCN by retrograde transport. Thus, NGF is under investigation both as a potential therapeutic agent and for its possible involvement in the pathogenesis of AD. In this study, postmortem brain tissues from both control and AD cases were investigated for amounts of poly (A) ÷ mRNA and NGF mRNA in the frontal cortex, a region rich in cholinergic afferents. Yields of poly(A) ÷ mRNA were similar from normal and AD tissues. Human NGF mRNA comigrated with murine NGF mRNA on Northern blots. Additionally, dot blot quantitation demonstrated that NGF mRNA levels do not differ in the inferior frontal gyrus of normal and AD patients. Thus, we conclude that levels of mRNA in general, and of NGF mRNA in particular, are unchanged in the frontal cortex of individuals affected by AD. Key words: Nerve growth factor; Inferior frontal gyrus; Phosphorimage analysis; Densitometry; Northern blotting; Dot blotting;

Postmortem interval; mRNA yields

1. Introduction

Alzheimer's disease (AD) is the most common type of adult-onset dementia [42]. The cause of this neurological disease is unknown, but highly characteristic pathological changes are observed at autopsy in the brain of A D patients. Some of these changes include neuronal dysfunction and degeneration of cortical, hippocampal [3,28] and basal forebrain cholinergic neurons (BFCN) [38], more specifically the cholinergic nucleus basalis of Meynert and septal nuclei [52,53,54], and the noradrenergic locus coeruleus (see [8] for review). Neurofibrillary tangles (NFT) and amyloid plaques are also characteristic of the A D brain [56,57].

* Corresponding author. Fax: (1) (905) 522-8804. Current address: Protein Design Labs Inc., 2375 Garcia Avenue, Mountain View, CA 94043, USA. 0169-328X/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0169-328X(94)00073-N

Likewise, there are biochemical and neurochemical changes associated with AD. Both overall [4,16,50] and selective [15,31] decreases in the amount of poly(A)m R N A s have been reported in AD. Furthermore, neurochemical changes in A D include a reduction in both quantity and activity of specific neurotransmitters. The most significant of these is a decrease in the neocortical activity of acetylcholine (Ach), acetylcholinesterase (AchE), and choline acetyltransferase (CHAT) [39,49, 57]; this reduction correlates well with the fact that the cholinergic cells of the basal forebrain are among the cells that degenerate in A D [53,54]. Nerve growth factor (NGF). a target-derived polypeptide, promotes the survival, regulation and growth of central cholinergic neurons of the basal forebrain. N G F elevates C h A T activity in the cortex [12], septum [18,29], and striatum [32] of neonatal rats. Many reports demonstrate that N G F induces C h A T activity in the hippocampus of adult rats with lesions and in

N. Jett~ et al. ~Molecular Brain Research 25 (1994) 242-250

neurons grafted into denervated hippocampus [17,51]. Additionally, the local administration of exogenous NGF prevents the lesion-induced degeneration of cholinergic neurons [17,24,58]. In other studies, the administration of NGF improved age-related or lesioninduced behavioural impairment [11,27,37,58]. In view of the evidence, it has been suggested [2,20,40] that the exogenous administration of NGF to AD patients might prevent the degeneration of the BFCN, which have been shown to continue to express NGF receptors in AD [13,19,33]. However, the use of NGF in the treatment of a neurological disorder such as AD requires a thorough understanding of its distribution and role in the human brain, as well as its possible alteration in AD. NGF protein has been detected in both rodent and human brain by ELISA [1,6,9,25,34] and bioassay [9]. In the adult rat CNS, NGF has been found in particularly high amounts in the hippocampus, olfactory bulb and cortex, and also in the septum, diagonal band of Broca, and nucleus basalis of Meynert [25]. Additionally, NGF injected into rat brain is taken up selectively by specific nerve terminals in the neocortex and is transported in a retrograde fashion to the nucleus basalis [44,46]. NGF mRNA can also be detected in both rodent and human tissues [7,14,21,25,47,55]. There are large differences in NGF mRNA levels in different brain regions, but NGF mRNA levels are particularly high in the hippocampus and the cerebral cortex in rodents [25,48,55]. It has been suggested that a deficit in NGF may be responsible for the neuronal degeneration observed in AD. The extremely low levels of NGF protein in human brain [6] have made this a difficult hypothesis to address. Using a two-site immunoassay, the levels of NGF protein were reported in two studies to be normal in AD cortex and hippocampus [1,34] and to be elevated in AD cortex in another study [9]. Two investigations to date have focused on the detection of NGF mRNA in human brain [14,41], with no significant differences reported between control and AD. The study by Phillips et al. [41], using in situ hybridization, demonstrated that the NGF hybridization signals were much weaker than those for BDNF, although they appeared to be similar in AD versus control tissues. However, the low signal to noise ratios for NGF mRNA in the pyramidal layer of the hippocampus did not allow quantitative analysis to be performed in this region. The investigation by Goedert et al. [14], using Northern analysis and dot blotting, examined pooled parietal and temporal cortex and found no differences in NGF mRNA levels between normal and AD individuals. The results reported in the latter paper prompted us to examine the levels of NGF mRNA in the frontal cortex, a region rich in cholinergic afferents. In our study, the use of a more narrowly defined cortical area

243

and more quantitative methods of analysis allow a rigorous examination of the relative levels of NGF mRNA in both control and AD human brain. We report no changes from controls in the levels of NGF mRNA in the inferior frontal gyrus of AD patients.

2. Materials and methods 2.1. Chemicals Guanidine thiocyanate was from Fluka (Ronkonkoma, NY). Ultra-pure urea was from Sehwarz/Mann (Cleveland, OH). Oligo(dT) cellulose Type 3 was from Collaborative Research (Bedford, MA). Nitrocellulose BA-85 membranes were obtained from Sehleieher and Sehuell (Keene, NH), whereas nylon membranes were Zeta Probe membranes from Bio-Rad (Mississauga, Ontario). [32p]dCTP was obtained from Amersham (Oakville, Ontario). Klenow polymerase, T4 polynucleotide kinase, oligo(dT)12_ts and M13 primer were obtained from New England Biolabs (Beverly, MA). DNAse I was from Promega Biotech (Madison, WI). NGF eDNA cloned in M13 was obtained from Drs. D. Clegg and L.F. Reiehardt (University of California, San Francisco). XAR-5 film was from Kodak (Rochester, NY). Phosphor screen was from Molecular Dynamics (Mountain View, CA). All other chemicals and enzymes were obtained from Sigma (St. Louis, MO). 2. 2. Tissue preparation Tissue was obtained from the Rochester Alzheimer's Disease Program (RADP) and the Rochester Alzheimer's Disease Center (RADC) at the University of Rochester (Rochester, NY). All patients were characterized neurologically, psychiatrically and histopathologieally. Diagnostic criteria for defining the AD patients were consistent with those specified by the NINCDS-ADRDA Workshop [30]. Frontal cortex tissue from 9 AD and 11 control patients without neurologic or psychiatric illness was dissected at autopsy, flash frozen in liquid nitrogen and stored at -80"C. Mean ages were 73.4 and 79.2 years old for the normal and the AD individuals respectively. Postmortem delays ranged from 1-19 h

Table 1 Age and postmortem interval data for samples used in NGF mRNA quantitation Age at time of death (years)

Postmortem interval (h)

NGF m R N A / poly (A) + mRNA × 10- 2

Normal donor 1 2 3 4 5 Mean + S.E.M.

74 83 80 73 63 74.6+3.4

9.3 14.5 15 19 6 12.8-l-2.3

15.37 + 1.54 9.95 + 0.96 10.60 + 0 12.98 + 1.06 17.34 + 4.38 13.3+1.4

AD donor 1 2 3 4 5 Mean + S.E.M.

78 75 83 74 88 79.6+2.6

10 12 12 12 7 10.6:1:1.0

12.10+ 1.86 8.36 + 1.27 18.4 + 2.40 12.0 + 0 5.96 + 0.74 11.4+2.1

V. .lett~! et al. /' Moh'cular Brain Research 25 (1994) 242-250

244

(means of 9.1 + 1.73 and 8.1 +1.15 h fi)r normal and AI) samples, respectively). Tissues used for N G F m R N A quantitation were from the inferior frontal gyrus (Brodmann areas 44 and 45). Age at the time of death and postmortem interval for each of these samples are shown in Table 1. Adult Swiss-Webster male mouse (Charles River, St. Constant, Que.) submandibular glands were similarly frozen and stored at - 8 0 ° C fk)llowing dissection.

2..5 Probe preparation A 3?p-labeled single-stranded c D N A probe wa~ prepared using Klenow polymerase and mouse N G F cloned in MI3 as a template. The single-stranded e D N A probe was separated from the unlabelled template by electrophoresis through low-melting temperature agaros~ as described previously [10].

:. ~. Hvbridization and autoradiographt 2.3. RNA extraction and purification Two different methods of m R N A purification were used, both previously described [10]. In the first, frozen tissue was homogenized without thawing in guanidine thiocyanate and 2-mercaptoethanol, followed by lithium chloride precipitation to purify total RNA. Oligo(dT) cellulose chromatography was then used to separate poly(A) + m R N A from poly (A) fractions. In the second method, frozen tissue was homogenized without thawing and incubated in proteinase K and sodium dodecyl sulfate (SDS), and poly(A) ÷ m R N A was purified directly from the homogenate by batch absorption to oligo(dT) cellulose. A poly(A)- R N A fraction was obtained from the homogenate, following batch absorption, by phenol-chloroform extraction, ethanol precipitation and D N a s e I treatment. All m R N A yields were measured by absorbance at 260 nm, and the purity estimated by the absorbance ratio at 260/280 nm, using a Beckman DU-64 spectrophotometer.

2.4. Denaturing agarose gel electrophoresis Up to 20 /.tg poly(A) ÷ m R N A per lane was electrophoresed on 0.8% a g a r o s e / 2 . 2 M formaldehyde gels as previously described [10]. R N A ladder (Gibco BRL, Burlington, Ont.) and poly(A)- lanes (containing 28S and 18S r R N A ) were stained with 2 . 5 / z g / m l ethidium bromide in distilled water for 30 min, rinsed with distilled water, destained for 2 h and then photographed under U V illumination. Poly (A) ÷ m R N A lanes were transferred to Zeta Probe nylon m e m b r a n e s or BA-85 nitrocellulose m e m b r a n e s as described previously [10]. Nitrocellulose m e m b r a n e s were vacuum baked for 2 h at 80°C whereas Zeta-probe m e m b r a n e s were cross-linked in a UV G e n e Linker C h a m b e r (BioRad, Mississauga, Ontario) at a setting of 150 mJoule.

The m e m b r a n e s were prehybridized for 24 h at 50°C in 5(E;~ formamide, 0.1% SDS, 5 x D e n h a r d t s (0.1% each bovine serum albumin, polyvinylpyrrolidone and Ficoll), 5 × SSPE (1 × SSPE: 0.18 M NaC1, 10 m M NaPO 4 pH 7.7 and l m M E D T A ) and 2 0 0 / x g / m l denatured salmon sperm DNA. The prehybridization mixture was then replaced with the same solution but containing 1 × Denhardts and a probe with specific activity > 108 c p m / / z g . The m e m b r a n e s were hybridized for 48 h at 50°C and washed twice in 2 × SSPE/0.1 ~ SDS at room temperature, twice in 0.1 × S S P E / 0 . 1 % SDS at room temperature and twice in 0.1 × S S P E / 0 . 1 % SDS at 50°C. The blots were then exposed to Kodak X A R film for 5 - 1 0 days.

2. 7. Dot blots Serial dilutions of poly(A) ~ m R N A in 80 p,l 2 0 x S S P E and 21:)/zl formaldehyde were denatured at 65°C and applied to Zeta-probe m e m b r a n e s using a blotting manifold (Bio-Rad). Mouse submandibular gland was used in each dot blot as a known standard enabling comparison between experiments. Crosslinking of m R N A to the m e m b r a n e s was at 30 m Joules. Prehybridization, hybridization and washing was as described for the Northern blots. After hybridization to the N G F probe and quantitation, the blots were stripped by immersion in boiling 0.1% SDS and either incubated for 2 h at 65°C or allowed to cool to room temperature. After ensuring that the blots were completely stripped, they were prehybridized for 4 h at room temperature in 6 x S S P E , 0.1% SDS. 1 0 x D e n h a r d t ' s reagent, 50 / z g / m l yeast t R N A and 50 / , g / m l denatured salmon sperm DNA. Hybridization to [32p]oligo(dT)12_l~, end-labeled by T4 polynucleotide kinase, was in 6 x S S P E / I % SDS at room temperature for 24-48 h. The blots were washed in 6 × S S P E / 0 . 1 % SDS four times at room temperature and then once in 6 x S S P E alone for 15 min each.

2.8. Analysis

8

< Z n.E

6

+

<

4-

_~

3-

n=11

n=9

Normal

Alzhoimer

5

0

2oo~

The dot blots were exposed to Kodak X A R film and analyzed by laser densitometry (Zeinah, CA). Some of the blots were also exposed to phosphor screens and analyzed with a phosphorimage analyzer (Molecular Dynamics, Mountain View, CA), Significance o f the differences between control and A D samples was assessed by a two sample t-test using the statistical software M I N I T A B (Minitab Inc., State College PA). For each sample, the a m o u n t of N G F m R N A was normalized to its poly(A) + m R N A content. This normalized value eliminated potential artifacts due to differences in the a m o u n t of poly(A) + m R N A loaded, and eliminateddifferences due to potential variance in overall poly(A) + m R N A content and poly(A) contamination between samples.

1

>o

Fig. 1. Yields of poly(A) + m R N A from normal and A D frontal cortex. Poly (A) + m R N A was purified either by guanidine thiocyanate or Proteinase K-SDS m e t h o d s [10]. Yields of poly(A) ÷ m R N A were determined by spectrophotometry and are expressed as /zg per g wet weight tissue. Bars represent the m e a n + S.E.M. (standard error of the mean).

3. Results 3. l. Y i e l d s o f p o l y ( A ) + m t E V A cortex tissue

AD

from

normal

and AD

Poly(A) + mRNA was isolated from both normal and frontal cortex using guanidine thiocyanate or Pro-

N. Jett6 et al. / Molecular Brain Research 25 (1994) 242-250

teinase K-SDS methods. Yields did not differ significantly between the two methods. As shown in Fig. 1, the yields of poly(A) + mRNA obtained from both normal and AD cases did not differ at the P < 0.05 level. The means obtained were 6.8 + 0.5 /~g/g and 5.9 + 1.2 /~g/g of wet weight tissue for the control (n = 11) and the AD (n = 9) samples, respectively (Fig. 1). The purity of the samples was high: 260/280 absorbance ratios ranged from 1.8-2.0.

3.2. Effect of postmortem interval and age on poly(A) + mRNA yields Regression analysis was performed to determine whether postmortem interval or age might affect the yields of poly(A) + mRNA from either normal or AD tissues. No significant correlation was obtained between the age of the individuals at the time of death and poly(A) + mRNA yields, for both the normal (r --0.34) and the AD (r = - 0 . 3 8 ) samples (data not shown). However, although no correlation was obtained between the postmortem interval and the poly(A) + mRNA yields for the normal individuals (r =

c~

a!

10.0

8.7

Total RNA

245

,'k~ x"x'q' ~.,~o ~q~

28S

rRNA 18S

Fig. 3. Native agarose gel electrophoresis of RNA purified from human frontal cortex. Total RNA was isolated using the guanidine thiocyanate method. 5/zg of the total RNA samples shown were run on a 0.8% agarose gel. N-270 is a control sample; A-18 is an Alzheimer's sample and A-85 is a degraded sample shown for comparison.

° ~

~>"

<

7.4

0.18, P = 0.583), a significant negative correlation was obtained for the AD samples (r = -0.81, P = 0.008; Fig. 2).

Z

E

6.1

+

~

4.8

O Q.

3.5

3.3. Visual analysis of purified mRNA on native agarose gels Postmortem Interval (hours)

bl

"~

"O

15 12

.m

>"

< z

rr E

9

Native agarose gels were run to check the integrity of the postmortem RNA samples prior to blotting. Electrophoresed RNA fractions were stained with ethidium bromide and photographed (Fig. 3). Both the 18S (2 kb) and the 28S (5 kb) ribosomal bands appeared undegraded in both normal and AD samples. Fig. 3 also includes a degraded sample (A-85) for comparison purposes which was not used in either yield or dot blot analysis.

s 3.4. Northern blot analysis of NGF mRNA in normal and A D frontal cortex

÷

<

a

>, O Q.

0 o

i

i

i

J

3

6

9

12

Postmortem Interval (hours) Fig. 2. Regression analysis for the correlation of postmortem interval with the yields of poly(A) + m R N A from human frontal cortex,

expressed in /zg per g wet weight tissue. Each data point represents a different sample. (a) normal (b) AD.

Denaturing agarose gels were run as an additional method of establishing the quality of the RNA and to determine optimal conditions for the detection of NGF mRNA. The assay for NGF mRNA was based on hybridization of Northern blots to a a2p-labeled single-stranded mouse NGF cDNA probe. Northern analysis demonstrated a 1.35 kb NGF mRNA band

24(~

,~. Jetty; ~'t u/. /r Mok'cular Brain Rest'arch 25 (1994) 242-250

#

•

X

•

X

5kb

2kb 1.35 --~

ira this analysis, as o p p o s e d to the 20 ( l I normal, 9 A D ) shown in Fig. 1. Initial s a m p l e s were used to work up the m e t h o d s (including t h e poly(A) + m R N A purification m e t h o d ) and, in some cases, insufficient starting m a t e r i a l was available for f u r t h e r analysis, or the specific B r o d m a n n a r e a was not identifiable. D o t blots w e r e q u a n t i t a t e d using e i t h e r a u t o r a d i o g raphy followed by laser d e n s i t o m e t r y or p h o s p h o r i m age analysis. S t a n d a r d curves of m o u s e s u b m a n d i b u l a r gland m R N A were a n a l y z e d by both m e t h o d s . Both d e n s i t o m e t r y a n d p h o s p h o r i m a g e analysis y i e l d e d s t a n d a r d curves which were l i n e a r over two to t h r e e o r d e r s of m a g n i t u d e . S t a n d a r d curves fox" N G F m R N A gave a c o r r e l a t i o n coefficient of - 0 . 8 7 w h e n a n a l y z e d by d e n s i t o m e t r y a n d a c o r r e l a t i o n coefficient of - 0 . 9 5 w h e n a n a l y z e d by p h o s p h o r i m a g e analysis. Fig. 6 shows s t a n d a r d curves o b t a i n e d for b o t h m o u s e N G F a n d p o l y ( A ) + m R N A using p h o s p h o r i m a g e analysis. D o t blots o f h u m a n s a m p l e s w e r e a n a l y z e d by b o t h m e t h ods, a n d no significant d i f f e r e n c e s were o b t a i n e d w h e t h e r s a m p l e s w e r e a n a l y z e d by d e n s i t o m e t r y or p h o s p h o r i m a g e analysis.

Fig. 4. Northern blot of RNA samples from human frontal cortex. 11 ng of mouse submandibular gland total RNA and 15 /~g of human frontal cortex poly(A) + mRNA samples were separated on ~t 0.8% agarose gel in 2.2 M formaldehyde; samples were transferred, hybridized and washed as described in Materials and methods. N-270 is a control sample; A-351 is an AD sample. Poly(A) samples of each were loaded in alternate lanes as negative controls.

"

poly A + RNA 1000 5O0 25O

from human frontal cortex (both control and AD) which c o m i g r a t e d with m o u s e N G F m R N A (Fig. 4). Specificity of h y b r i d i z a t i o n is d e m o n s t r a t e d by t h e lack o f h y b r i d i z a t i o n to poly ( A ) - s a m p l e s in a l t e r n a t e l a n e s o f t h e blot. A c a l c u l a t i o n b a s e d on t h e dilution of m o u s e s u b m a n d i b u l a r g l a n d m R N A t h a t gives a hyb r i d i z a t i o n signal of a s t r e n g t h similar to the h u m a n signal i n d i c a t e s t h a t in h u m a n brain, N G F m R N A is p r e s e n t at a p p r o x i m a t e l y 1 in 10 s m o l e c u l e s of p o l y ( A ) ÷ m R N A . This is in c o n t r a s t to t h e m o u s e s u b m a n d i b u l a r g l a n d in which N G F m R N A r e p r e s e n t s a p p r o x i m a t e l y 0.1% o f t h e p o l y ( A ) + m R N A [45]. 3.5. D o t blot analysis a n d quantitation o f N G F m R N A in n o r m a l a n d A D f r o n t a l cortex

D o t b l o t t i n g was u s e d to c o m p a r e the relative cont e n t o f N G F m R N A in c o n t r o l a n d A D s a m p l e s (Fig. 5). N G F m R N A was n o r m a l i z e d to p o l y ( A ) + m R N A for e a c h sample. Y e a s t t R N A or p o l y ( A ) - m R N A w e r e u s e d as n e g a t i v e c o n t r o l s for b a c k g r o u n d ( n o n specific) h y b r i d i z a t i o n . 10 d i f f e r e n t i n f e r i o r f r o n t a l gyrus s a m p l e s (5 n o r m a l , 5 A D ; see T a b l e 1) w e r e u s e d

125 63

"~

.,,o --

o,

~

,~.

k.v

n.

k, v

4" z-. .,.-" ...