Báo cáo khoa học: Alcohol linked to enhanced angiogenesis in rat model of choroidal neovascularization pot

However, a 4.0-fold increase in FAEES activity was observed in the choroid of alcohol-fed rats compared with regular controls Fig.. However, no dif-ference in ethyl ester production was

of choroidal neovascularization

Puran S Bora*, Sankaranarayanan Kaliappan*, Qin Xu*, Shalesh Kumar*, Yali Wang*,

Henry J Kaplan and Nalini S Bora*

Department of Ophthalmology and Visual Science, Kentucky Lions Eye Center, University of Louisville, KY, USA

Age-related macular degeneration (AMD) is the leading

cause of blindness in the Western world in people aged

over 55 Exudative (i.e wet-type) AMD is characterized

by the development of new vessels within the choroid of

the eye (i.e choroidal neovascularization; CNV) [1]

This neovasculature penetrates Bruch’s membrane and

grows beneath the retinal pigment epithelium and

neu-rosensory retina [2–7] Many different growth factors

have been implicated in the development of choroidal

angiogenesis [8–18]; including vascular endothelial growth factor (VEGF) [9,19,20] More recently, an important role for various cell populations and cellular proteins – macrophages, hematopoietic stem cells and complement proteins [21–23] has been identified Several studies have shown that the nonoxidative ethanol metabolites, fatty acid ethyl esters (FAEE), are synthesized by the esterification of free fatty acids The reaction is catalyzed by FAEE synthase (FAEES)

Keywords

adiponectin; angiogenesis; choroids; macular

degeneration; neovascularization

Correspondence

P S Bora, Department of Ophthalmology,

Jones Eye Institute, University of Arkansas

for Medical Sciences, 4301 West Markham,

# 523 Little Rock, AR 72205, USA

Fax: +1 501 686 7037

Tel: +1 501 686 5150

E-mail: pbora@uams.edu

*Present address

Jones Eye Institute, University of Arkansas

for Medical Sciences, Little Rock, AR, USA

(Received 17 January 2006, accepted 2

February 2006)

doi:10.1111/j.1742-4658.2006.05163.x

One of the pathologic complications of exudative (i.e wet-type) age-related macular degeneration (AMD) is choroidal neovascularization (CNV) The aim of this study was to investigate whether chronic and heavy alcohol consumption influenced the development of CNV in a rat model The oxi-dative metabolism of alcohol is minimal or absent in the eye, so that eth-anol is metabolized via a nonoxidative pathway to form fatty acid ethyl esters (FAEE) Fatty acid ethyl ester synthase (FAEES) was purified from the choroid of Brown Norway (BN) rats The purified protein was 60 kDa

in size and the antibody raised against this protein showed a single band

on western blot BN rats on a regular diet were fed alcohol for 10 weeks Control rats were fed water with a regular diet and pair-fed control rats were fed regular diet, water and glucose We found that FAEES activity was increased 4.0-fold in the choroid of alcohol-treated rats compared with controls The amount of ethyl esters produced in the choroid of 10 week alcohol-fed rats was 7.4-fold more than rats fed alcohol for 1 week The increased accumulation of ethyl esters was associated with a 3.0-fold increased expression of cyclin E and cyclin E⁄ CDK2; however, the level of the cyclin kinase inhibitor, p27Kip, did not change The increased accumu-lation of ethyl esters was also associated with 3.0-fold decreased expression

of APN in the choroid We also found that the size of CNV increased by 28% in alcohol-fed rats Thus, our study showed that chronic, heavy alco-hol intake was associated with both an increased accumulation of ethyl esters in the choroid and an exacerbation of the CNV induced by laser treatment These results may provide insight into the link between heavy alcohol consumption and exudative AMD

Abbreviations

AMD, age-related macular degeneration; APN, adiponectin; BN rat, brown Norway rat; CDK2, cyclin E-dependent kinase 2; CNV, choroidal neovascularization; FAEE, fatty acid ethyl ester; FAEES, fatty acid ethyl ester synthase; PEDF, pigment epithelium-derived factor; RPE, retinal pigment epithelium; VEGF, vascular endothelial growth factor.

[24–32] In addition to heart, brain, liver and pancreas,

nonoxidative alcohol metabolism also exists in the eye

[24,25] The nonoxidative ethanol metabolism pathway

and its metabolites such as FAEE may play an

import-ant role in the development and⁄ or enhancement of

CNV Cell proliferation is governed by regulatory

cell-cycle proteins, such as cyclin E and cyclin E⁄ CDK2,

Li et al [33], showed that hepatic stellate cells treated

with FAEE (linoelic acid ethyl ester) have increased

cy-clin E expression and cycy-clin E⁄ CDK2 activity,

suggest-ing that FAEE may have promitogenic effects [34]

Adiponectin (APN) is an antiangiogenesis protein and

circulating concentrations of APN decreased

signifi-cantly following chronic consumption of alcohol [35,36]

The balance between angiogenic factors and inhibitors

is important for the angiogenesis process and has not

previously been studied in the choroid of alcoholic rats

Although there have been numerous studies using

laser-induced CNV as a model of choroidal

angiogene-sis [2–7,21–23], none has studied the effect of chronic

and heavy alcohol consumption on the neovasculature

We investigated the effect of chronic and heavy

alco-hol feeding on alcoalco-hol metabolism in the choroid and

on the development of CNV in a rat model We used

ethyl alcohol (100% proof mixed with water) to feed

the rats Because it is easy to feed alcohol to rats and

they are a good source of choroidal tissue from which

to purify FAEES enzyme, we used a rat model to

induce CNV in this study

Results

Purification of FAEES

FAEES was purified to homogeneity from rat choroid

using a previously described method [29] The enzyme

was purified 9000-fold to homogeneity on a

hydroxylap-atite column with 30% yield; SDS⁄ PAGE showed a

single band of molecular mass 60 kDa (Fig 1A) The

N-terminal sequence of the purified protein matched

100% with rat adipose tissue FAEES protein (data not

shown) [29] Purified FAEES was transferred to a

nitro-cellulose membrane and the purified protein reacted

with an antibody raised in the rabbit against FAEES; a

single 60 kDa band was observed on the western blot

(Fig 1B) These results suggested that FAEES, a

60 kDa protein, is present in the choroid of the rat eye

FAEES activity

The cornea, iris and ciliary body, lens, retina and

chor-oid were collected separately from the eyes of control

rats and rats fed alcohol for 10 weeks to assay synthase

activity No significant increase in FAEES activity was observed in the cornea, iris and ciliary body, lens and retina of either group However, a 4.0-fold increase in FAEES activity was observed in the choroid of alcohol-fed rats compared with regular controls (Fig 2A) Cho-roidal FAEES activity in regular and pair-fed controls was the same (data not shown) The kinetic constants of the enzyme purified from the choroid had similar Km values compared with the heart enzyme for both oleic acid and ethanol; however, the Vmax values were four-fold lower for both substrates compared with the heart enzyme (Table 1) The calculated Kmand Vmaxvalues of the choroidal enzyme for oleic acid were 0.24 mm and

1020 nmolÆmg)1Æh)1, respectively, and the calculated Km and Vmax for ethanol were 0.41 m and 930 nmolÆ

mg)1Æh)1, respectively (Table 1)

Immunohistochemistry studies Paraffin sections of the posterior segment of eyes were immunostained for FAEES using anti-FAEES serum (raised in rabbits) The eyes of rats fed alcohol for

10 weeks showed increased staining for FAEES com-pared with control rats (Fig 2B)

FAEE production The induction of FAEES activity produced 7.2- and 7.4-fold more ethyl esters in the choroid of rats fed

60 kDa

60 kDa

94 kDa –

67 kDa –

43 kDa –

29 kDa –

20 kDa –

14 kDa –

Fig 1 (A) SDS ⁄ PAGE analysis of purified FAEES from rat choroid Purification was performed as described in the Experimental proce-dures Lane 1, molecular mass markers; lane 2, purified FAEES,

60 kDa (B) Western blot analysis of purified FAEES Antibody (raised in rabbit) against FAEES was used to react with pure FAEES protein Lane 1 shows no band (control lane) Rat albumin was used as control protein Lane 2 shows a clear band at 60 kDa.

alcohol for 9 and 10 weeks, respectively, compared

with rats fed alcohol for 1 week (Table 2) Ethyl esters

production increased as the number of weeks of

alcohol feeding increased (Table 2) However, no

control alcohol fed

cornea

A

B

10

8

6

4

2

0

iris&cb lens retina Choroid

b

a

Fig 2 (A) fatty acid ethyl ester synthase (FAEES) activity was assayed using the method described previously [28–31] Cornea, iris and cili-ary body, lens, retina and choroid were separated from the eyes of control rats (no alcohol) and rats fed alcohol for 10 weeks to assay syn-thase activity No significant increase in FAEES activity was seen in the cornea, iris and ciliary body, lens and retina However, a 4.0-fold increase in FAEES activity was observed in the choroid of alcohol-fed rats compared with controls (B) Immunohistological analysis of FAEES

in the posterior region of the rat eye Immunohistology was performed using FAEES antibody and immunoperoxidase staining kit (A) Control rats showed mild staining (brown color in the choroid) (B) Rats fed alcohol for 10 weeks showed several fold increase in staining (brown staining is marked by black arrows in the choroid) compared with controls Cell layers are marked as shown here: SC, sclera; CH, choroids; RPE, retinal pigment epithelium; OLM, outer limiting membrane; ONL, outer nuclear layer; OSL, outer synaptic layer; INL, inner nuclear layer; ISL, inner synaptic layer.

Table 1 Kinetic constants of FAEES, purified from rat choroid.

Oleic acid and ethanol were used as substrates for FAEES activity.

Vmax(nmolÆmg protein)1Æh)1) FAEES Oleic acid 0.24 ± 0.07 m M 1020 ± 5.24

(Choroid) Ethanol 0.41 ± 0.05 M 930 ± 4.35

FAEES Oleic acid 0.20 ± 0.02 m M 3950 ± 9.85

(Heart)a Ethanol 0.30 ± 0.04 M 3700 ± 8.22

a

See Bora et al [29].

Table 2 Ethyl esters produced in the choroid of alcohol-fed rats A 7.2- and 7.4-fold increase in the production of ethyl esters was observed in 9- and 10-week alcohol-fed rats, respectively, com-pared with 1-week alcohol-fed rats in the choroid However, no dif-ference in ethyl ester production was observed in 4, 8, 9 and

10 weeks of alcohol feeding compared with 1-week alcohol-fed (control) rats in the cornea, iris and ciliary body, lens and retina Weeks of

alcohol feeding

Ethyl ester formation (nmolÆg)1wet weight)

significant changes were seen in ethyl esters production

in the cornea, iris and ciliary body, lens and retina of

9- and 10-week alcohol-fed rats (Table 2)

Size of the CNV complex

Ten days after laser treatment (on day 70) alcohol-fed,

regular and pair-fed control rats were killed to evaluate

the effect of alcohol on the size of CNV There was a

28% increase (P < 0.005) in CNV complex size in

alco-hol-fed rats compared with regular and pair-fed controls

(Fig 3) However, there was no difference in CNV size

between regular and pair-fed controls (Fig 3) The

val-ues in each group were averaged from 80 laser spots

Confocal microcopy images of flat mounts (RPE–

choroid–sclera) are shown in Fig 4Aa–d There was a

significant increase in the size of the CNV complex in

alcohol-fed rats (Fig 4Ab) compared with controls

(Fig 4Aa) Interestingly, we did not observe any

signi-ficant difference in CNV complex size in rats fed

alco-hol for 8 weeks (Fig 4Ac,d) The CNV size in pair-fed

controls (Fig 4Ba) was the same as in regular controls

(Fig 4Bb) However, alcohol-fed rats (Fig 4Bd) had a

significantly higher CNV size compared with pair-fed

rats (Fig 4B) These results indicated that chronic

and heavy consumption of alcohol for 10 weeks

increased the size of the CNV in the laser-induced rat

model

Cyclin E and cyclin E⁄ CDK2

In order to explore the mechanism of the

alcohol-asso-ciated increase in CNV size we measured the levels of

cyclin E and cyclin E⁄ CDK2 by western blot in alco-hol-fed rats We found a 3.0-fold increase in cyclin E and cyclin E⁄ CDK2 levels in rats fed alcohol for 10 weeks compared with controls (nonalcohol and 1-week alcohol-fed rats) (Fig 5A) Laser photocoagu-lation alone had no effect on the expression of these proteins (Fig 5A) Interestingly, we saw no change in cyclin E and cyclin E⁄ CDK2 expression levels in rats fed alcohol for 8 weeks (Fig 5B) These rats also showed no change in CNV size after laser induction Expression of the cyclin kinase inhibitor, p27Kip, was not changed in any of these groups (Fig 5A,B)

APN study Immunohistochemistry showed that APN was expressed in the choroid of the rat eye No other tissue was stained in the rat eye (Fig 6A) RT-PCR data showed that expression of APN mRNA in the rat choroid was significantly decreased in the choroids of rats fed alcohol for 10 weeks compared with controls (Fig 6B, upper) Similarly, western blot showed a sig-nificantly decreased level of protein amount in the choroid of alcoholic rats compared with regular con-trols (Fig 6B, middle) In an in vitro study when rat choroidal endothelial cells were treated with 50 lm ethyl esters, there was significantly decreased expres-sion of APN mRNA compared with control cells (not treated with ethyl esters) (Fig 6B, lower)

Discussion AMD is the most common cause of permanent visual impairment among the elderly, and several million new cases are diagnosed each year around the world [37,38] It consists of two major forms, nonexudative (dry-type) and exudative (wet-type) AMD Approxi-mately 10–12% of patients have the wet form, which is characterized by the presence of CNV and is respon-sible for 90% of severe central vision loss from AMD [1,37–39] There is no treatment that will reliably recover lost central vision, although there are currently several clinical trials in phase II⁄ III testing using anti-angiogenic pharmaceuticals Although there is no per-fect animal model to study CNV that resembles humans, the laser-induced model may be the best available animal model to study CNV CNV can be induced by laser in mouse, rat, pig, rabbit and mon-key In this study we used rats to study CNV because rats are easy to feed alcohol and it is economical to get enough choroidal tissue to purify FAEES enzyme There are currently 12 million alcoholics in the USA and over 80 million worldwide [26] The epidemiologic

18

16

14

12

10

8

4

6

2

0

Control

3 µm)

Pair-fed Alcohol-fed Fig 3 Comparison of CNV complex size (i.e area) of regular control,

pair-fed control and 10-week alcohol-fed rats CNV complex size was

measured using IMAGE PRO - PLUS software in micron unit There was a

28% increase (P < 0.005) in the CNV complex size in alcohol-fed rats

compared with both the controls rats (regular control and pair-fed

control) The values in each group were averaged from 80 laser

spots The incidence of CNV in each group was 92%.

B

A

b

a

b

Fig 4 (A, (a–d)) Confocal microcopy

imag-es of flat mounts (RPE–choroid–sclera) from

rats perfused with fluorescein–dextran

(green color) and stained with elastin (red

color) Elastin stains Bruchs membrane

(red), the CNV complex is green There was

a significant increase in the size of the CNV

complex in alcohol-fed (10 weeks) rats

(A, b; 2500·), compared with control rats

(A, a; 2500·) However, no significant

increase in the CNV complex size was

observed in rats fed alcohol for 8 weeks

(A, d; 2500·) compared with control rats

(A, c; 2500·) (B, (a–d)) Confocal microcopy

images of flat mounts (RPE–choroid–sclera)

from rats perfused with fluorescein–dextran

(green) and stained with elastin (red) Elastin

stains Bruchs membrane (red), the CNV

complex is green There was a significant

increase in size of the CNV complex in

alco-hol-fed (10 weeks) rats (b, d; 2500·),

com-pared with pair-fed controls (B, c; 2500·).

However, there was no difference in the

CNV complex size of pair-fed controls

(B, a; 2500·) compared with regular controls

(B, bx; 2500 ·).

association between alcohol consumption and AMD is

not clear The Beaver Dam eye studies and studies by

Azen et al [40–43] suggested that a heavy alcohol

intake is associated with an increased risk of exudative

AMD [43]

Alcohol is metabolized in the eye mainly via a non-oxidative pathway using the enzyme FAEES [24,25] Although ADH4 is present in eye tissue, it does not metabolize alcohol because of the high Km value for alcohol Retinol oxidation is the main function of the enzyme in the eye [44–47] Purified rat choroidal FAEES is 60 kDa in size The Km values for two sub-strates, oleic acid and ethanol, were similar to the value for the enzyme purified from the human heart; however, the Vmax values were fourfold lower than that of human heart FAEES This may be due to a species difference [29,44,45]

In this study, we showed that feeding a rat alcohol (8 gÆkg)1) for 10 weeks increased FAEES activity in the choroid by 4.0-fold No change in FAEES activity was noted in the cornea, iris and ciliary body, lens or retina Immunohistological staining showed a qualitat-ive increase in FAEES in the choroid of alcohol-trea-ted rats compared with controls The increased FAEES activity resulted in increased production and accumulation of ethyl esters in the choroid of alcohol-fed rats, with the size of the increase related to the number of weeks of alcohol feeding

The increased concentration of ethyl esters within the choroid corresponded to a statistically significant increase in the size of CNV in alcohol-fed rats (10 weeks) compared with regular and pair-fed control rats There was no difference in the CNV size of regu-lar controls compared with pair-fed controls The data suggest that either one or both of the controls may be used for the study We believe that this is the first report to show an effect of alcohol consumption on experimental choroidal angiogenesis We also observed increased expression of cell-cycle proteins, cyclin E and cyclin E⁄ CDK2, in the choroid after 10 weeks of alco-hol feeding These proteins are known to play an important role in the proliferation of cells, including endothelial cells [33,34,48,49] The change in expres-sion of these cell-cycle proteins may be related to the amount of ethyl ester in the choroid Previous studies have shown that ethyl esters have promitogenic activity and can increase the expression of cyclin E and cyclin E⁄ CDK2 [33] As shown in Table 2, the accu-mulation of ethyl esters increased with the number of weeks of alcohol feeding The amount (17.10 nmolÆg)1 wet weight) of ethyl ester accumulated after 10 weeks

of alcohol feeding may be enough (just above the threshold level) to increase the expression of cyclin E and cyclin E⁄ CDK2 by 3.0-fold, which may have con-tributed to the 28% increase in the size of CNV Because laser treatment in rats on a normal diet did not change cyclin E and cyclin E⁄ CDK2 expression in the choroid, we believe that chronic and heavy alcohol

Cyclin E

Cyclin E/CDK2

p27 kip

Cyclin E

Cyclin E/CDK2

p27 kip

A

B

Fig 5 (A, B) Western blots of cyclin E, cyclin E ⁄ CDK2 and p27Kip.

Rat cyclin E and p27Kip antibodies were used to blot these

pro-teins Immunoprecipitation of cyclin E ⁄ CDK2 was performed using

rat cyclin E antibody Choroid was removed from the following

groups of rats, homogenized in solublizing buffer and the

superna-tant were used to run the gels (A) Expression of cyclin E,

cyclin E ⁄ CDK2 was increased threefold in rats fed alcohol for

9 weeks compared with rats fed alcohol for 1 week and rats not

fed alcohol (lane 1, no alcohol control; lane 2, 1 week of alcohol;

lane 3, 9 weeks of alcohol; lane 4, 9 weeks of alcohol and laser

photocoagulated) However, no change was observed in the

expression of p27Kip in these rats (B) No change in the expression

of cyclin E, cyclin E ⁄ CDK2 and p27Kip was observed in the rats fed

alcohol for 8 weeks compared with those fed alcohol for 1 week

(lane 1, 1 week of alcohol; lane 2, 8 weeks of alcohol) The amount

of protein used to run the gel was (A) cyclin E, lane 1–4, 6.0 lg;

cyclin E ⁄ CDK2, lane 1–4, 6.0 lg; p27Kip, lane 1–4, 10.0 lg (B)

Cyclin E, lane 1–2, 6.5 lg; cyclin E ⁄ CDK2, lane 1–2, 10.0 lg; p27

Kip, lane 1–2, 8.5 lg.

intake was important in the increased expression of

these proteins Our hypothesis that a certain level of

ethyl esters was needed to increase the size of CNV

complex was further supported by our observation

that cyclin E and cyclin E⁄ CDK2 expression did not

change in the choroid of alcohol-fed rats after 8 weeks

and we found no significant difference in the size of

the CNV complex after 8 weeks of alcohol

consump-tion

Cyclin E and cyclin E⁄ CDKs can be regulated by

growth-promoting signals on cyclin synthesis and on

the assembly of cyclin⁄ CDK complexes The

complexes can be further regulated by cyclin kinase

inhibitors [33,48,49] Although we observed a

3.0-fold increase in the expression of cyclin E and

cyclin E⁄ CDK2 in the choroid of alcohol-fed rats at

10 weeks, we did not see a change in the expression of

p27Kip, a cyclin kinase inhibitor Thus, the increased

expression of cyclin E and cyclin E⁄ CDK2 appeared

to be the result of increased production and not

decreased cyclin kinase inhibition

Studies have shown that actively growing

healthy or pathological tissues express high levels of

angiogenic factors It is also known that high levels

of angiogenic factors may not be sufficient to induce

angiogenesis, i.e VEGF is expressed at high levels in

several quiescent adult tissues that lack active angiogen-esis Thus, downregulation of antiangiogenic factors may help to enhance angiogenesis In a mouse tumor model, Brakenhielm et al [35] showed that APN was a direct inhibitor of angiogenesis Xu et al [36] showed that circulating APN was decreased in alcoholics We have also shown that choroidal APN expression and protein levels were decreased in 10-week alcoholic rats compared with controls Thus, downregulation of APN may be another factor that helped to enhance the cho-roidal angiogenesis in 10-week alcoholic rats We did not see any affect of alcohol feeding in the expression of pigment epithelium-derived factor (PEDF, an angiogen-esis inhibitor) in the laser spots of alcohol-fed rats com-pared with controls (data not shown) There is no evidence in the literature to show the affect of alco-hol in the expression of other endogenous angiogenesis inhibitors

Fig 6 (A) Immunohistological analysis of choroidal APN

Immuno-histology was performed using adiponectin antibody (primary

anti-body) in paraffin-embedded rat eyes with CY3-labeled goat

(anti-mouse IgG) serum (secondary antibody), mounting with

Aqua-Mount (Lerner Laboratories, PA) Sections were examined under

fluorescence microscope (Zeiss) Only choroids showed staining

(red in the choroid) No other tissue was stained in the eye APN is

expressed specifically in the choroid (B) (Upper) Adiponectin (APN)

mRNA expression in alcohol-fed (lane 3) and control (lane 4) BN rat

choroids Lanes 1 and 2 are GAPDH lanes The figure shows

ethi-dium bromide-stained bands for PCR product after UV exposure.

Equal amounts of total RNA (0.2 lg) were used to detect the

mRNA levels of GAPDH and APN APN expression in 10-weeks

alcoholic rat choroid was significantly decreased compared with

control choroid (Middle) Western blot of rat choroid APN antibody

was used to blot choroidal proteins Choroids were removed from

the alcoholic and control rats, homogenized and the supernatant

was used to run the gel to blot with APN antibody APN protein

was significantly decreased in 10-week alcoholic rats (lane 1)

com-pared with controls (lane 2) The amount of protein loaded in each

lane was 8 lg (Lower) APN mRNA expression in rat choroidal

endothelial cell Lane 3, 50 l M ethyl ester-treated cells; lane 4,

con-trol cells, not treated with ethyl esters Lanes 1 and 2 are GAPDH

lanes The figure shows ethidium bromide-stained bands for PCR

product after UV exposure Equal amounts of the total RNA

(0.2 lg) were used to detect the mRNA levels of GAPDH and APN.

APN expression in 50 l M ethyl ester-treated cells was significantly

decreased compared with control cells.

Rat choroid RT-PCR

A

B

Rat choroid western blot

Rat choroidal endothelial cell RT-PCR

In conclusion, in an experimental rat model of

laser-induced CNV, we have shown that heavy and

pro-longed alcohol consumption was associated with

increased FAEES expression and increased ethyl ester

concentration in the choroid Furthermore, the

increased cell-cycle regulatory proteins, cyclin E and

cyclin E⁄ CDK2 and decreased APN mRNA and

pro-tein in the choroid correlated with an increased CNV

size after 10 weeks of alcohol feeding We have

experi-mental evidence that heavy and chronic alcohol

feed-ing in rats enhances choroidal angiogenesis with

implications for the prognosis of AMD, and possibly,

tumors

Experimental procedures

Animals

Male Brown Norway (BN) rats (4–6 weeks old, 250–300 g)

were purchased from Harlan (Indianapolis, IN) The

ani-mals were maintained in accord with guidelines established

by the Committee on Animals at the University of

Louis-ville Medical School and University of Arkansas for

Med-ical Sciences

Purification of FAEES

Male BN rats (4–6 weeks old) were killed, the eyes were

phosphate buffer solution, minced and homogenized These

centrifuged and the resulting supernatant was collected

Protein was estimated using the Bradford method with

gamma-globin as standard

The enzyme, FAEES, was purified using the method

des-cribed previously The molecular mass and purity of the

were run at 150 V (stacking) and 200 V (separating), after

which they were fixed and stained with Coomassie Brilliant

Blue Molecular mass was calculated using polypeptide

standards of known molecular mass [29,30]

Western blot analysis

poly(vinylidene difluoride) membrane using Trans-Blot

sem-idry electrophoretic transfer cell (Bio-Rad, Richmond, CA)

After electroblotting the gels were stained with Coomassie

Brilliant Blue to ensure equal loading and equal transfer

Blots were stained at room temperature with appropriate

Control blots were treated with the same dilution of normal

goat or rabbit serum After washing and incubating with

(1 : 1000 dilution), blots were developed using the enhan-ced chemiluminescence western blotting detection system

‘ECL + Plus’ (Amersham Pharmacia Biotech, Arlington Heights, IL) Immunoprecipitation and western blots for

to the methods described by Li et al [33] or as described above Antibodies for cyclin E and p27Kip were purchased

Antibodies for APN was purchased from BioVision, Inc (Mountain View, CA)

Alcohol feeding

Animals were divided in three groups The control group was fed regular diet and water for 10 weeks, the pair-fed group was fed regular diet, water and glucose for 10 weeks

for 10 weeks Alcohol was mixed with water and the bottles were changed everyday One rat was housed per cage The amount of alcohol consumed was measured everyday [50,51] The alcohol and water mixture contained 20 mL of

water Rats were drinking 31–35 mL of the mixture every-day, assuming 10% of alcohol was evaporated everyday from the mixture, the amount of alcohol received by each

feeding the alcohol group rats with only water for 2 h every-day before next round of alcohol and water mixture was started The comparable amount of alcohol in humans will

be about six glasses of whiskey or rum We included a pair-fed group as a second control group Each rat from alcohol

coming from alcohol and the rest from the regular diet

feed-ing regular diet (12 kcal) and glucose (56 kcal) The

water bottle, 14 g of glucose was dissolved in 100 mL of water and one bottle was supplied to each cage After the glucose water was consumed by the rats the bottle was replaced by a regular water bottle

Rates of FAEE synthesis

The animals in both groups were sacrificed after 1, 4, 8, 9, and 10 weeks of alcohol feeding and the eyes were enucleated

to remove cornea, iris and ciliary body, lens, retina and chor-oid to perform enzyme assay Rates of FAEE synthesis was determined by incubating samples containing enzyme with

ethanol in 60 mm sodium phosphate buffer, pH 7.2, in a total

the incubation interval, the reaction was terminated by the

addition of 2 mL of cold acetone containing a known

ole-ate Volumes were reduced by evaporation under stream of

chroma-tographed on silica OF plates, developed with petroleum

visu-alization of lipids with iodine vapor, fatty acid ethyl ester

spots were scrapped, and the lipid was eluted with acetone

subtrac-tion of blanks, results were expressed as nmoles of fatty acid

ethyl ester formed per milliliter per hour [29–31]

Measurement of ethyl ester production

Accumulated ethyl esters were measured using a method

des-cribed previously [26] Briefly, homogenates of cornea, iris

and ciliary body, lens, retina and choroid were incubated

end of the incubation interval, the reaction was terminated

by the addition of 2 mL of cold acetone containing a known

ole-ate Volumes were reduced by evaporation under stream of

chroma-tographed on silica OF plates, developed with petroleum

visu-alization of lipids with iodine vapor, fatty acid ethyl ester

spots were scrapped, and the lipid was eluted with acetone

Immunohistochemistry studies

Paraffin embedded sections of the eyes from 10-week

alco-hol-fed and control rats were used for immunostaining of

FAEES Briefly, slides were deparaffinized twice for 5 min in

xylene, twice for 3 min in absolute ethanol, 3 min in 95%

meth-anol solution for 20 min to inactivate endogenous peroxidase

incubated for 20 min with serum from the same species as

the origin of secondary antibody for blocking nonspecific

binding After rinsing, they were incubated with rabbit

anti-rat FAEES (1 : 200 dilution) for 60 min at room

tempera-ture The slides were stained using anti-(rabbit IgG)

immunoperoxidase staining kit (Vector, Burlingame, CA)

according to the manufacturer’s instructions The sections

were treated with DAB for 10 min, counterstained with

Mayer’s hematoxylin for 10 min, washed thoroughly in cold

tap water, and covered with coverslip with a mounting media

for viewing by light microscopy For APN

immunohisto-chemistry, APN antibody (1: 200 dilutions) was used as

pri-mary antibody and CY3-labeled anti-(goat IgG) serum was

used as secondary antibody RT-PCR of APN for mRNA expression was performed according to the methods used elsewhere [23] The following primers (GAPDH and APN) were used for RT-PCR

GATTTGGC-3¢, reverse 5¢-CATGTAGGCCATGAGGTC CACCAC-3¢); APN (forward 5¢-ATGGGCTATGGGTA GTTGCAGTCA-3¢, reverse 5¢-TAGCTTCATGCTTTGG GTCCTCCA-3¢

Laser photocoagulation

10) animals were laser photocoagulated on day 60 by

tropicamide Krypton red laser photocoagulation (50 lm spot size, 0.05 s duration, 350 mW) was used to generate four laser spots in each eye surrounding the optic nerve The animals were killed on day 70 to evaluate the presence

of CNV, the eyes were removed and choroid–scleral flat mounts were stained for elastin using a monoclonal anti-body specific for elastin (Sigma, St Louis, MO) followed

by a Cy-3-labeled secondary antibody (Sigma) The size of the CNV was determined by confocal microcopy [23,32]

Measuring neovascularization

average molecular mass, Sigma) The eyes were removed and fixed for 1 h in 10% phosphate-buffered formalin The cornea and lens were removed and the neurosensory retina was carefully dissected from the eyecup Five radial cuts were made from the edge of the eyecup to the equator; the sclera–choroid–RPE complex was flat-mounted, with the sclera facing down, on a glass slide in Aqua mount Flat mounts were stained with a monoclonal antibody against elastin (Sigma) and a CY3-conjugated secondary antibody (Sigma) and examined using a confocal microscope (Zeiss LSM510, Carl Zeiss AG, Jena, Germany) The CNV stained green, whereas the elastin in the Bruch’s membrane stained red The incidence of the CNV complex was determined by confocal microscopy The size of the neovascular complex was measured by using image pro-plus software [23,32]

Primary culture of rat choroidal endothelial cells

Rat (BN, 2 weeks old) eyes were removed, cleaned from connective tissue and blood vessels and were dipped in

cor-nea, lens, RPE and sclera were removed Choroid was stored in HBSS buffer with out Ca and Mg Enzyme

diges-tion was performed using collagenase⁄ dispase, 1 mgÆmL)1

and the cells resuspended in HBSS (with out Ca and Mg)

After washing five times with same buffer the cells were

separated by 40 lm strainer and cultured in fibronectin

coated plates After washing with HBSS, Dulbecco’s

modi-fied Eagle’s medium was added to the plates The culture

and 95% air) Cultures were checked for the purity of the

endothelial cells by positive labeling with CD31 [52] The

medium was changed 24 h after plating and every 48 h

thereafter until reached to confluence Methods described

by Penfold et al [50] and Li et al [33] were followed for

further growth of the cells Linoleic acid ethyl ester (Sigma)

was dissolved in dimethylsulfoxide and 50 lm was added to

the cells Control cells were treated with dimethylsulfoxide

only

Statistical analysis

Differences between groups were evaluated by Student’s

t-test

Acknowledgements

We thank Drs Douglas Borchman, Sean Kuntz and

Purushottam Jha for critical review of the manuscript

and Guirong Liu and Dr Jose M.C Cruz for helping

in APN staining This study was supported by

Com-monwealth of Kentucky Research Challenge Trust

Fund; Research to Prevent Blindness, Inc., New York

and grants from NEI and NEI core grant,

1R24EY015636-01

References

1 Ferris FL III, Fine SL & Hyman LA (1984) Age related

macular degeneration and blindness due to neovascular

maculopathy Arch Ophthalmol 102, 1640–1642

2 Grossniklaus HE, Hutchinson AK, Capone A Jr,

Woolfson J & Lambert HM (1994) Clinicopathologic

features of surgically-excised choroidal neovascular

membranes Ophthalmology 101, 1099–1111

3 Grossniklaus HE & Gass JDM (1998) Clinicopathologic

correlation of surgically-excised type 1 and type 2

chor-oidal neovascular membranes Am J Ophthalmol 126,

59–69

4 Hans E, Grossniklaus HE & Green WR (2004)

Choroi-dal neovascularization Am J Ophthalmol 137, 496–503

5 Frank G, Holz D, Pauleikhoff RK & Bird AC (2004)

Pathogenesis of lesions in late age-related macular

dis-ease Am J Ophthalmol 137, 504–510

6 Kliffen M, Sharma HS, Mooy CM, Kerkvliet S & Jong

PT (1997) Increased expression of angiogenic growth

factors in age-related maculopathy Br J Ophthalmol 81, 154–162

7 Frank RN (1997) Growth factors in age-related macular degeneration: pathogenic and therapeutic implications Ophthalmic Res 29, 341–353

8 Abraham JA, Mergia A, Whang JL, Tumolo A, Fried-man J, Hjerrild KA, Gospodarowicz D & Fiddes JC (1986) Nucleotide sequence of a bovine clone encoding the angiogenic protein, basic fibroblast growth factor Science 1, 545–548

9 Amin R, Puklin JE & Frank RN (1994) Growth factor localization in choroidal neovascular membranes of age-related macular degeneration Invest Ophthalmol Vis Sci

35, 3178–3188

10 Antoinette C, Lambooij KHM, van Wely DJ, Linden-bergh-Kortleve RWAM, Kuijpers MK & Cornelia MM (2003) Insulin-like growth factor-I and its receptor in neovascular age-related macular degeneration Invest Ophthalmol Vis Sci 44, 2192–2198

11 Smith LE, Kopchick JJ, Chen W, Knapp J, Kinose F, Daley D, Foley E, Smith RG & Schaeffer JM (1997) Essential role of growth hormone in ischemia-induced retinal neovascularization Science 276, 1706–1709

12 Jin M., Chen Y, He S, Ryan SJ & Hinton DR (2004) Hepatocyte growth factor and its role in the pathogen-esis of retinal detachment Invest Ophthalmol Vis Sci 45, 323–329

13 Laterra J, Nam M, Rosen E, Rao JS, Lamszus K,

hepatocyte growth factor gene transfer enhances glioma growth and angiogenesis in vivo Lab Invest 76, 565–577

14 Leibovich SJ, Polverini PJ, Shepard HM, Wiseman

DM, Shively V & Nuseir N (1987) Macrophage-induced angiogenesis is mediated by tumour necrosis factor-alpha Nature 329, 630–632

15 Oh H, Takagi H, Takagi C, Suzuma K, Otani A, Ishida

K, Matsumura M, Ogura Y & Honda Y (1999) The potential angiogenic role of macrophages in the forma-tion of choroidal neovascular membranes Invest Ophthalmol Vis Sci 40, 1891–1898

16 Tsutsumi C, Sonoda K-H, Egashira K, Qiao H, His-atomi T, Nakao S, Ishibashi M, Charo IF, Sakamoto

T, Murata T et al (2003) The critical role of ocular-infiltrating macrophages in the development of choroi-dal neovascularization J Leukoc Biol 74, 25–32

17 Killingsworth MC, Sarks JP & Sarks SH (1990) Macro-phages related to Bruch’s membrane in age-related ma-cular degeneration Eye 4, 613–621

18 Dastgheib K & Green WR (1994) Granulomatous reac-tion to Bruch’s membrane in age-related macular degen-eration Arch Ophthalmol 112, 813–818

19 Shikun He, Man LJ, Worpel V & Hinton DR (2003) A role for connective tissue growth factor in the pathogen-esis of choroidal neovascularization Arch Ophthalmol

121, 1283–1288